NASA SP-2001-4227

UPLINK-
D LINK

Douglas J. Mudgway

The NASA History Series

National Aeronautics and Space Administration

Office of External Relations
Washington, DC
200 1
 ISBN 0-1 6-066599-X

For sale by the Superintendent of Documents, U.S. Government Printing Office

Internet: bookstore.gpo.gov Phone: toll free (866) 5 12- 1800; DC area (202) 5 12-1 800
Fax: (202) 5 12-2250 Mail: Stop SSOP, Washington, DC 20402-0001
ISBN 0-16-066599-X
Library of Congress Cataloging-in-Publication Data

Mudgway, Douglas J. 1923-

Uplink-Downlink: A History of the NASA Deep Space Network , 1957-1997 1 Dowlas J. Mudgway.
p. cm. - (NASA SP ; 2001-4227) (The NASA history series)
Includes bibli~~raphical
references and index.
1. Deep Space Network-History. I. Tirle. 11. Series. 111. Series: The NASA history series
 This book is dedicated to the memory of

Ae Remetti
Nichs%as
19141998

Throughout his entire career at JPL (1959-1996),

Nicholas A. Renzetti brought the strength and
dynamism of his own character to the Deep Space
Network. Some of it passed to all of those he touched
and to every task he undertook. Due in no small
part to his continued advocacy there exists a con-
tinuous record of the technical achievements of the
Network from its inception to 1997, in the form
of the "TDA Progress Reports" series. He gave rise,
also, to many other publications that reflected the
breadth of activity in which the Network engaged.

For many years, Renzetti envisioned a project that

would distill the essence of materials collected into
a single book-length narrative covering the full story
of the Deep Space Network-scientific, technical,
operational, personal, and political. My pleasure at
having undertaken the task is tempered only by my
regret that he was unable to see it completed.

Doagh J Mudgwdy
Sonoma, Cal$ornia
 Dedication
Foreword
Acknowledgments
Preface
Introduction
The Solar System
Size and Composition
Terrestrial Planets
Jovian Planets
Inner and Outer Planets
Asteroids
Comets
Earth and Its Reference Systems
Earth Motions
xxxv I Time Conventions
xxxv I InterplanetaryTrajectories
m i i 1 Deep Space Communications
m i i Bands and Frequencies
xxxviii Doppler Effect
miii Cassegrain Focus Antennas
xxxk Uplink and Downlink
xl Signal Power
xl Coherence
di Decibels
xliii Modulation and Demodulation
xliv Spacecraft Radio System
xliv An Essential Part of the Answer

Goldstone, California
The Pioneer Lunar Probes
The Echo Balloon Experiment
The Venus Radar Experiment
Research and Development Antenna
Washington, DC
Woomera, Australia
 UPLINK-DOWNLINK: OF THE DEEPSPACE
A HISTORY NETWORK

Johannesburg, South Africa

The Mariner Era Mission Set

The Ranger Lunar Missions
Mariner Planetary Missions
Mariner 1762
Mariner 1764
Mariner 1967
Mariner 1967
Mariner 1971
Mariner 1973
Pioneer Interplanetary Missions
Pioneer Jupiter Missions
Surveyor Lunar Lander Missions
Lunar Orbiter Missions
Apollo Missions
The Deep Space Instrumentation Facility (DSIP)
The Facility (DSIF) Becomes a Network (DSN)
L-band to S-band
Improvements for the Mariner Mars Missions
The Need for a Second Network
Larger Antennas are Needed
Incremental Improvements in the Network
Flight Project Requirements Become Formalized
The DSN Becomes a Multimission Network
Openings and Closings
Looking Back
Endnotes

The Viking Era Mission Set

The Network
Network Operations
The Helios Mission
The Viking Mission
 The Voyager Mission
Pioneers 6-1 I
Pioneer Venus
Network Engineering and Implementation
The DSN Mark I11 System Design
Implementation for Viking
DSN Mark 111-75 Model
The 64-meter Stations
The 26-meter Stations
High-Power Transmitter
Mars Radar
The Introduction of X-band
The 64-meter Antenna Problems
DSN Mark I11 Data System Project
Mark I11 Data System (MDS) Implementation
DSN Mark 111-77 Model
DSN Mark 111-77 Telemetry System
DSN Mark 111-77 Command System
DSN Mark 111-77 Tracking System
DSN Mark 111-77 Ground Communications
DSN Mark 111-77 Arraying Capability
DSN Mark 111-77 Decoding Capability
The 26-meter Antenna S-X Conversion Project
Oveniew
Endnotes

The Voyager Era

Deep Space Missions
Earth-Orbiting Missions
Network Operations
Prime Missions
Voyager at Jupiter
Voyager at Saturn
Venus Balloon/Pathfinder
Voyager at Uranus
Uranus Encounter
Errtended Missions
 Helios
Viking at Mars
Pioneers 6 9
Pioneers 10, 11
Pioneer 12 (Venus Orbiter)
Highly-elliptical Orbiters
ICE
AMPTE
Reimbursable Missions
Network Engineering and hnplementation
New 34-meter High-Efficiency Antennas
The 64-meter to 70-meter Antenna Extension
Rehabilitation of DSS 14 Pedestal
DSS 14 Azimuth Radial Bearing
DSS 43 Pedestal
Antenna Performance Upgrade
Microwave Design Considerations
Kicker Braces
Subreflector
Surface Panels
Madrid Site Implementation
Canberra Site Implementation
Goldstone Site Implementation
Networks Consolidation
Mark TVA Implementation
Signal Processing Centers
Mark IVA Network Operations Control Center
Mark IVA Ground Communications Facility
DSN Mark IVA System Upgrades
Mark IVA Telemetry
Mark IVA Command
Mark IVA Tracking
1 An Abrupt Transition
Endnotes

The Galileo Era

A Defining Moment
 Mission Set
Overview
Network Operations
Deep Space Prime Missions
The Comet Halley Missions
Giotto
Phobos
Background for MagelIan
Magellan Mission to Venus
Voyager at Neptune
Background for Galileo
Galileo Mission to Jupiter
Background for Ulysses
Ulysses Solar Mission
Background for Mars Observer
Mars Observer Mission
Deep Space Extended Missions
The pioneer Missions
International Cometary Explorer (ICE)
Voyager Interstellar Mission
Earth Orbiter Missions
Generaf
Emergency Mission Support
Network Engineering and Implementation
The 70-meter Antennas
Elevation Bearing Failure
Gearbox Rehabilitation
Subreflector Drive Problems
Interagency Arraying
Parkes-Canberra Telemetry Array
VLA-Goldstone Telemetry Array
The X-band Uplink
Block V Receiver
The DSN Galileo Telemetry (DGT) Subsystem
Beam Waveguide Antennas
Signal Processing Center Upgade Task
Background
Signal Processing Centers Upgrade
SPC Telemetry
 SPC Command
SPC Tracking
Network Operations Control Center
Ground Communications Facility
A Successful Conclusion
Endnotes

The Cassini Era

Winds of Change
The Cassini Era Mission Set
Deep Space Missions
General
Galileo
Ulysses
Voyagers I and 2
Pioneer 10
Near-Earth Asteroid Rendezvous
Mars Global Surveyor
Mars Pathfinder
Cassini
Earth-Orbiting Missions
General
Space VLBI Observatory Program
The Network
Complexes and Antennas
Ka-band Downlink
Orbiting VLBI Subnetwork
Emergency Control Center
Other Aspects
Endnotes

CHAPTERTHE ADVANCEOF TECHNOLOGY

IN THE DEEP
SPACENETWORK

458 The DSN Technology Program

459 The Great Antennas of the DSN
467 Forward CommandlData Link (Uplink)
468 Return TelemetrylData Link (Downlink)
 Low-Noise Amplifiers
Phase-Lock Tracking
Synchronization and Detection
A Digital Receiver
Encoding and Decoding
Data Compression
Arraying of Antennas
Radio Metric Techniques
Doppler and Range Data
Timing Standards
Earth Rotation and Propagation Media
Radio Science
VLBI and Radio Astronomy
The Global Positioning System
Goldstone Solar System Radar
Telecornrnunications Performance of the Network
Cost-Reduction Initiatives
Ka-band Development
Other Technologies
Optical Communication Development
1 DSN Science

The TDA Science Office

Radio Science
Celestial Mechanics
Solar Corona and Solar Wind
Radio Propagation and Occultation
Relativistic Time Delay
Gravitational Waves
Radio Astronomy
Internationd Cooperation
Tidbinbilla Interferometer
Orbiting VLBI
Host Country Programs
Cross-Support Agreements

1 Guest Observer Program

Antenna Utilization for Radio Asrronomy
 The DSN Also Benefits
Significant Advances
Search for Extraterrestrial Intelligence (SETI)
Crustal Dynamics
Planetary Radar
Asteroids
Radar Astrometry
Renaissance
Significant Events
Endnotes

In the Beginning (1958 to 1963)

Background
The Rechtin Years
The Formative Years (1964 to 1994)
A DSN Manager for Flight Projects
The Space Flight Operations Facility
The BayleyILyman Years
The DurnasIHaynes Years
The JPL Strategic Plan: 1994
Birth of the TMOD
Reinventing the Future (1994 into the New mennium)
Reengineering TMOD
Data Capture Process
Activity Plan Process
Reinventing the Future
"Bridging the Space Frontier"
T M O D Primary Challenge: 1997
Endnotes

List of Figures
Appendix
About the Author
Index
Other Books in the NMA History Series
The Sun sitting on his throne commands all things
To tend downward toward himself, and does not
allow the chariots of the heavenly bodies to move
Through the immense void in a straight path, but
hastens them a l l along
In unmoving circles around himself as center.

From "Ode on This Splendid

Ornament of Our Time and
Our Nation, the Mathematice
Physical Treatise by the Eminent
Isaac Newton"
by Edmond Ei'alley, Astronomer Royai
Lonhn, 1684
FOREWORD

From the very beginning of its association with NASA in 1958, the Jet Propulsion
Laboratory UPL) received its fair share of public recognition for its successes and f d -
ures in pursuing the exploration of deep space. It started with the Explorers, the first
American satellites to orbit Earth. Later there came the Rangers, the first spacecraft to
reach the surface of the Moon; the Mariner spacecraft, first to visit Venus and Mars;
and the Voyagers that pushed the boundaries of deep space communication further out
to Jupiter and Saturn, and eventually to Uranus and Neptune. There were other space-
craft that put landers, probes, or orbiters into planetary orbits or atmospheres, or onto
planetary surfaces. There were probes whose mission was to explore the composition and
dynamics of the interplanetary medium, and probes to observe the physics of the Sun.
There were the huge missions, such as Viking to Mars, Galileo to Jupiter, and Cassini
to Saturn, and there were small missions like Pathfinder to Mars and the New Millennium
missions to asteroids and comets. There was also science that did not require a space-
crafi for its experiments such as radio astronomy, radar astronomy, and the search for
extraterrestrial intelligence.

The public accolades that were engendered by the bountill science returned from all
of these NASA projects were shared by NASA and the scientists whose exquisite instru-
ments and innovative interpretation of the data produced the new knowledge reflected
in their results. However, what the press conferences, news releases, and media coverage
did not reveal was the incredibly complex infrastructure that made each of these mar-
velous deep space missions possible. This infrastructure, which had been built over the
years at JPL, included the Deep Space Network (DSN), an essential, integral part of
every mission. There was, in effect, a relationship between the planetary missions, the
spacecraft that carried them out, and the Deep Space Network that enabled such mis-
sions to be planned in the first place.

Without the remarkable improvement in performance of the DSN, scientific missions

to the distant planets would have been impossible. In 1964, when Mariner N flew past
Mars and took a few photographs, the limitation of the communication link meant that
it took eight hours to return to Earth a single photograph &om the Red Planet. By
1989, when Voyager observed Neptune, the DSN capability had increased so much that
almost real-time video could be received from the much more distant planet, Neptune.

It is timely that, some 40 years after its inception, the Deep Space Network should be
recognized for its remarkable litany of progress in radio communications over vast dis-
tances, thereby allowing planetary scientists to collect data from sites throughout the
solar system. This book succeeds in bringing the history of the DSN forward for the
attention of curious, generdly informed, or technical specialist readers.

NASA is to be commended for commissioning this book as part of its History Series,
for the NAWJPL Deep Space Network is the world's largest and most advanced facili-
ty for tracking, navigating, and acquiring data fiom interplanetary spacecrdi. Worldwide
in scope, and international in concept, the DSN has supported not only NASA space
missions, but also those of the space agencies of Japan, Germany, Russia, France, and
Canada. Also, in concert with a NASA policy of international cooperation, the resources
of the Network have been made available to support qualified enterprises fiom all nations.

In planning this book, New Zealand-born author Douglas Mudgway was faced with the
formidable problem of making an extremely complex technology comprehensible to the
curious or generally informed reader, while at the same time presenting for the special-
ist reader a historically accurate account of the advance of the technology that made
possible more sophisticated planetary missions. The task was further complicated by the
fact that the DSN was in a constant state of change, changes that were in consonance
with the requirements of the space missions it was supporting.

For the former type of reader, Mudgway describes what the DSN actually did, within
the framework of several overarching eras, each corresponding to a period of time dur-
ing which a major NASA deep space mission dominated the public scene. The Mariner,
Viking, Voyager, and Galileo Eras are examples. He provides an inside view of what it
took to design, build, and operate those tenuous radio communication links between
the controllers, engineers, and scientists at computer terminals at JPL and billion dol-
lar spacecraft about to land on Mars, orbit around Jupiter, or fly through the rings of
Saturn.

For the latter reader, the author provides an excellent review of the growth of the spe-
cialized technology that underlaid the remarkable expansion of Network capability that
enabled the design of increasingly ambitious planetary missions. A comprehensive appen-
dix provides help for the technical researcher.

The unique capabilities of the great antennas of the DSN, together with a generous NASA
policy of making them available for non-NASA scientific research, attracted radio and
radar astronomers from the United States and many other countries to the extent that
the requests for observing time far exceeded the time that could be made available for
these ground-based scientific purposes. Chapter 8, "The Deep Space Network as a Scientific
Instrument," brings this important aspect of the work of the DSN to the attention of
the reader, and provides a basic review of the published work that resulted.

Uplink-Downlink transforms the technical records of a major NASA facility, unique in

the world, into a viable historical narrative covering 40 years of its critical involvement
in the United States space program. The Deep Space Network emerges f ~ o mthis study
not only as a complex, human-machine system of worldwide dimensions, but also, more
convincingly, as a focus for the aspirations of the NASA scientists for ever-bigger science,
and of the JPL engineers for ever-greater innovation and enterprise in navigating to dis-
tant targets and communicating at ever-greater data-rates, in spite of the fluctuations in
available NASA funding for both, driven in some measure by the conflicting priorities
of the piloted versus unpiloted programs within NASA itseK

Throughout the narrative we observe the interaction of these powerful currents, not from
the lo+ heights of a dispassionate historical observer, but from the eye-level of a dedi-
cated participant, for the author's long career at JPL was played out at the vortex of these
often conflicting currents of self-interest. As a consequence, the engineers, technicians,
scientists, and managers at NASA, JPL, and its partner institutions in Spain and Australia,
whose names appear from time to time in &is book, are presented as real-life people con-
tributing their various talents to the milieu in which they found themselves. It was the
totality of these individual efforts, driven by a common inquiring interest in space and
focused toward the common goal of its exploration, that produced the remarkable enti-
ty known throughout the world as the NASA Deep Space Network.

I am confident that this work will be a valuable addition to the documented history of
the United States space program.
ACKNOWLEDGMENTS

To have been part of the history recorded here is sufficient reason in itself to acknowl-
edge my indebtedness to many colleagues in the DSN who supported and assisted me
throughout my long and rewarding career at the Jet Propulsion Laboratory UPL). That
said, there remains the need to recognize the important contributions rhat were made
to the writing of this history by many of those engineers and scientists, and by other
persons less directly associated with my 1ge in the Deep Space Network.

I must begin with Nicholas A. Renzetti, for it was he who brought me from Australia
to the United States in 1962 to begin a career at the Deep Space Instrumentation Facility
(DSIF), and it was he who, in 1996, after I had retired from JPL, stimulated my per-
sonal interest in producing a history of the Deep Space Network. When the project
began to falter for lack of hnding support, MacGregor Reid provided much needed
encouragement, and Paul Wesunoreland responded with the limited resources available
to him to keep it going.

Later, when those resources expired, the project came to the attention of the NASA
Chief Historian, Roger Launius. His encouragement, backed with adequate resources,
moved the project rapidly f o m d to completion. Without his enthusiastic support it is
unlikely this book would have been published. Along the way, I was ably assisted by
Louise Alstork and other members of the NASA History Ofice staff.

As the work got under way, my access to historical documents, files, and photographs
was eased immeasurably by the generous help of members of the JPL Archives and
Records section, notably John Bluth, Elizabeth Moorthy, and Robin Morris.

At all times, Shirley Wolff of the Telecommunications and Missions Operations (TMO)
Outreach Office was my lifeline to the daily pulse of the Network. I came to depend
on her patience and energy for transmitting documents and other technical material pro-
vided, at my request, by various engineers and scientists associated with the Network.
She, too, helped in bringing this project to life.

Last and longest, but by no means least, were the contributions in the form of inter-
views, discussions, technical briefings and materials, narrative reviews, and encouragement
on various DSN-related topics that were provided by: Catherine Thornton on geodesy;
Michael Klein on the search for extraterrestrial intelligence; Martin Slade on radar astron-
omy; James Hodder on network operations; Thomas Kuiper, Marvin Wick, and Pamela
Wolken on radio astronomy; George Textor on Voyager; Leslie Deutsch on DSN teleme-
try for Galileo; Joseph Wackley for DSN systems; Joseph Statman on the Big Viterbi
Decoder; Robert Wallace on 34-m antennas; Dan Batbker on microwaves; Robert Clauss
on masers; Charles Stelzried on system noise temperature, Venus Balloon, and Giotto;
Bob Preston and John Ovnick on orbiting VLBI; Fred McLaughlin on the 70-m anten-
nas; Dale Wells on 70-m antenna maintenance; James Layiand on coding and arraying;
Patrick Beyer on Galileo; Dennis Enari on Ulysses and Mars Pathfinder; Marvin Traxler
on Mars Observer; Allen Berman on Magellan; Robert White on 34-m antennas; Thomas
Wynne for photographs; Ronald Gillette on Cassini; John McKinney on Mars Missions;
Ed Massey on Voyager and Ulysses; Nick Fanelli and Joe Goodwin on the Earth orbiters
and reimbursables; Bob Ryan and David Lozier (Ames Research Center) on the Pioneers;
Michael Stewart on Magellan; Don Mischel, Tom Reid, Richard Mallis, and Robertson
Stevens for early background material; George Schultz for an early draft; and finally,
Olivia Tyler, Bobby Buckmaster, and Lynda McKinley for miscellaneous but neverthe-
less indispensable help.

The onerous task of reviewing the draft version of the book was undertaken by Larry
Dumas, Michael Hooks, Roger Launius, MacGregor Reid, Gael Squibb, and Jose Urech.
Their insighdid comments and suggestions greatly enhanced the accuracy, consistency,
and quality of the narrative.

The families of Nicholas Renzetti and Merrick kindly provided background

material on the personal lives of these two important figures in the history of the Deep
Space Network. Their contributions are gratemy acknowledged.

Dough J Mudpay
Sonomu, Calfornia
PREFACE

Although the subtitle for this book, '? I of the Deep Space Network" has an air
History
of finality about it, the suggestion of a task ended or a work completed, that does not
represent the true state of affairs. The Deep Space Network (DSN) is really a work in
progress. It had a beginning, of course, and a life that, at the time rhis book was writ-
ten, had spanned four decades. That is what this book is about. The task of recording
"what happened next" will f d to future historians.

Eloquent histories of NASA's planetary explorers have been written by others, and it was
not my intent to revisit those magnificent enterprises. To them belonged the high drama
associated with spacecraft encounters and landings on distant planets, spectacular launch-
es, startling new science, and stunning color images from distant corners of the solar system.

As a key element in all of those dramatic spacecraft events, the DSN shared their excite-
ment, but saw a quiet, unreported drama of its own. There were occasions when the
success or failure of a multimillion dollar spacecraft, the reputation of NASA, or the
recovery of critical science data from a far planet, lay in the hands of the DSN and,
not infrequently, those of a crew member at a distant tracking station. Such events appear
throughout the narrative. More often though, the determined, sometimes heroic, efforts
of engineers and technicians in laboratories at JPL, and in control rooms and antennas
at remote tracking stations, provided drama enough for those of us who were aware of
it. Struggling to meet the seemingly insatiable demands of the planetary space missions
for more, bigger, or better capability, and all of it sooner, those individuals bore the
brunt of the burden of change that characterized progress in the DSN from its begin-
ning. These events, too, are identified in the narrative.

In these pages, the informed reader will discover a simple description of what the DSN
is about, and how it worle-an aspect of NASA's spectacular planetary program that
seldom found its way into the popular media coverage of those major events. Future
historical researchers will find a complete record of the origin and birth of the DSN,
its subsequent development and expansion over the ensuing four decades, and a descrip-
tion of the way in which the DSN was used to the purpose for which it was
created. At the same time, the specialist reader is provided with an abundant source of
technical references that address every aspect of the advanced telecommunications tech-
nology on which the success of the DSN depended. And finally, archivists, educators,
outreach managers, and article writers will have ready recourse to the inner workings
of the DSN and how they related to the more publicly visible events of the planetary
space program.
Despite the way it might have appeared to many of us within it, the DSN was not cre-
ated as "an end unto itseK7' it was created by NASA to serve the needs of the NASA
planetary spaceflight program. The spaceflight projects like Mariner, Surveyor, Pioneer,
Viking, Ulysses, Galileo etc., were its "customers." It seemed appropriate, therefore, to
record the history of the DSN against a timeline based on the great planetary explor-
ers. Indeed it was their requirements for tracking and data acquisition services, to support
more ambitious flight missions, that drove the DSN to the heights of excellence and
achievement described herein.

At some point in the hture it will be timely to produce another history of the DSN.
The present work will serve as a clear point of departure for what follows, for by
that time the rapidly diminishing "corporate" memory reflected herein will have sure-
ly been lost. There is a final point to be made. At a conservative estimate, NASA
invested $4-5 billion over 40 years in developing and operating the Deep Space
Network as a major NASA facility. An investment of public money of that magni-
tude deserves a public record of how the money was spent and an account of the
return on the investment. Hopefully, this history will suffice for that purpose also.
Those were my objectives in writing this book.

Before proceeding further, a few words about internal communications practices within
the DSN organization at the Jet Propulsion Laboratory (JPL) during the period covered
in this history is in order. Perhaps in keeping with the very nature of JPG academic
associations, communications within the various departments, offices, laboratories, and
individuals that constituted the DSN at JPL were always very informal. Most interac-
tions between engineers and managers and their people were conducted "eyeball to eyeball"
in offices, corridors, bathrooms, or cafeterias, as the need arose. When two or more peo-
ple were together, a vacant ofice was used as an ad hoc conference room. Larger groups
met in a designated conference room, where the overhead projector and transparencies
were the communications medium, and a handout, consisting of hard copies of the
transparencies, on which the attendee made notes, was the only record of the material
presented and the decisions reached. Interoffice memos were generally used to address
issues that went beyond the writer's immediate organizational boundary. Anyone could
write a memo on any subject, and did, when a meeting would not suffice. There never
was a centralized, or even unified, filing system within the DSN organization. Everybody
was responsible for keeping their own files in any way they chose. However, through-
out the laboratory, there was always a paper reduction or "records roundup program in
effect. People were continually encouraged to discard inactive files in the interests of sav-
ing space. Furthermore, when people retired, they took files with them or discarded
them. So, for one reason or another, all DSN interoffice memos and personnel files,
with few exceptions, have disappeared.

Fortunately, however, this loose approach did not apply to documents that were tight-
ly controlled by the formal DSN Documentation System. It was generally considered
that anything worth preserving should be published in a formal document, and per-
sonnel were strongly encouraged to do so. As a result, a most complete record of DSN
activity in research, engineering, and deep space tracking "operations" is to be found in
the series of documents called "Progress Reports." Such reports are referenced extensively
in the chapters that follow, and are readily available from the JPL Archives Section.

The regrettable absence of interoffice memos, meeting minutes, verbatim records of dis-
cussions, letters, personnel records, etc., relating to DSN activity through the years accounts
for the paucity of anecdotal material carried in the narrative.

The NASA Deep Space Network is a crucial element in the mission to explore our solar
system. Its main purpose is to provide a high-quality communications link between an
Earth-based mission control center and the many spacecrat? of the United States and
several foreign countries that are, or were, engaged in that mission. Over the forty years
since the inception of the Deep Space Network, its people have been engaged in a con-
tinuous, pervasive drive to refine communications links. The remarkable record of
achievement described herein is the result of their effort.

In choosing Uplink-Downlink as the primary title of this book, I endeavor to convey to

the reader the key characteristic of the Deep Space Network and the essential focus of
the peopIe who make it work.

This theme, Uplink-Downlink, runs through the book and is used to show how the
DSN has developed from a single-mission network of limited capability to a giant mul-
tiple-mission network. Together with a series of innovative improvements in spacecraft
communications systems, the Network provides a deep space communications capabdi-
ty which now exceeds its modest beginnings by 12 orders of magnitude.

While the DSN is a prime example of exquisite telecommunications technology and

engineering, it is a fully operational, global system of people and machines supporting
NASA and other programs of planetary exploration. The way in which the DSN has
supported these programs through the years, and the part it has played in aU of the
momentous and historical scientific discoveries resulting fi-om this endeavor, forms the
timeline against which the history of the DSN is described.
The spectacular successes, and occasional failures, of NASA's ventures into space are well
covered by the media in various forms. We are quite accustomed to live television cov-
erage of new spacecraft launches; scientists presenting their latest "amazing facts or
pictures; images from remote spacecraft flying by, onorbit about or on the surface of
distant planets; and NASA animation movies of complex inflight or orbital maneuvers.
The complex infrastructure of tracking stations, intercontinental communication links,
data processing services, and highly skilled engineers and technicians that make it all
possible are seldom mentioned, much less shown, on screen.

That infrastructure, so essential to the NASA program of Planetary Exploration, is known

as the Deep Space Network. The fascinating history of its evolution, from a simple,
gound-based antenna and receiver recording the signal from a telemetry transmitter
onboard the first simple spacecraft attempting to reach the Moon, to the present world-
wide network of giant antennas keeping track of spacecraft at the edge of the solar
system, is the subject of this book.

Like the NASA space program itsetf, the history of the Deep Space Network is one of
continual change, of performance enhancement, of striving for excellence and, of going
where none has gone before.

To provide the general reader with a basis for understanding the significance of these
changes in the context of the Deep Space Network, we begin with a simple discussion
of the dynamics of spaceflight and why a tracking station is needed to maintain con-
tact with an artificid object in space. The function of a deep space station in keeping
track of one or more spacecrafi is addressed, and is followed by an explanation of how
a tracking station does what it does. Then, since the course of history in the Deep Space
Network was determined at various times by decisions based on radio frequency tech-
nology, it seemed that a short discourse on the principles involved would be in order.

The early phase of DSN history, from its inception in 1958 to the mid-1970s, was asso-
ciated with NASKs first attempts to fly spacecraft to the Moon, Venus, Mars, and Jupiter.
I have chosen to call this the Mariner Era because most of the major DSN develop-
ment was driven by the requirements of the Mariner missions. By 1974, the Mariner
Era was drawing to a close and the Viking Era of the bigger, more complex, more cost-
ly missions to revisit the planets, was beginning. The Viking Era opened in 1975 with
the Viking mission to Mars.

Dependent on the NASA planetary exploration program for its hnding, the DSN con-
tinued to evolve in size, capability and cost at a rare commensurate with the level of
NASA interest in unmanned exploration of the solar system. For the next twenty-five
years, the Voyager, GEaileo, and Cassini programs were the principal drivers for expan-
sion and development in the Deep Space Network. Each program effectively determined
a unique era of influence in the overall story of the DSN.

From the outset, the esoteric technologies of deep space telecommunications and infor-
mation processing were dominant features of research and development programs in the
DSN. Key characteristics of this work are described in chapter 7, "The Advance of
Technology in the Deep Space Network."

Much of the technology in the Deep Space Network shares a lot in common with that
of the giant radio telescopes of the world. It should not be surprising that the anten-
nas of the Deep Space Network were used, both individually and in collaboration with
various radio astronomy observatories, for radio astronomy investigations as an adjunct
to their principal task of spacecraft tracking. These investigations, and others related to
scientific experiments that involved the DSN but did not involve a spacecrafi, are described
in chapter 8, "The Deep Space Network as a Scientific Instrument."

To organizations born of the space age, and particularly those involved directly with the
NASA space program, growth and change were almost synonymous. Through the years
covered here, organizational changes at NASA Headquarters, at Caltech's Jet Propulsion
Laboratory (parent body of the DSN), and within its own management structure, were
a continual and essential part of its evolution. The changes that took place and their
impact on the DSN ability to cope with conflicting challenges for new planetary mis-
sions, while continuing support for ongoing missions within a fluctuating level of NASA
funding support, is the substance of chapter 9, "The Deep Space Network as an
Organization in Change."

I have elected to address the general reader in describing the history of the DSN, with
the expectation that a specialist reader will take advantage of the references I have given
to follow-up in specific areas of interest using the rich repository of related technical
material contained in the series of Traclung and Data Acquisition (TDA) Progress Reports
now available on the World Wide Web at http://tdag;Dpl.na~.gov/progresss_report.
INTRODUCTION

Surely there can be few ~ e o p l ein this

day and age who are not familiar with
the television images of a NASA space-
craft blasting off fiom Cape Kennedy in
a splendid gout of smoke, fire, and steam
to a rendezvous with an asteroid, comet,
or planet in some remote corner of the
solar system. From the c o d o r t of our
living rooms we watch the launch rock-
et rise majestically off the pad, pass
through the local cloud cover at the
launch site and soon become an ever-
diminishing point of light in the center
of the television screen. Sometimes we
observe the booster rockers burn out and
drop away before the main rocket stage
engines accelerate the vehicle and its del-
icate planetary payload out of sight. The
Titan-Centaur launch of T/oygeer I,
shown in Figure Intro-1, was a typical
example. What happened next? you may
ask and, Where did that whole thing go?

Months or even years after a launch, an Figure tntro-I. Voyager 1 outward bound for
item on the evening television news cov- Jupiter, September 1977. Voyager 1 was launched
ering a NASA press release about an by a Titan-Centaur vehicle from Cape Kennedy on
impending rendezvous of a NASA 5 September 1979. After encountering Jupiter in
spacecraft with Mars, Jupiter, Saturn, or February 1979, Voyager 1 used a gravity-assist
an asteroid, followed by a video clip or maneuver to reach Saturn in November 1980. It then
the latest set of pictures from the plan- continued on a trajectory which eventually took it
etary object, might have caught the beyond the boundary of the solar system. Twenty
attention of viewers. For certain, the pic- years later it had become the remotest of all space-
tures showed areas and details never craft and was still being tracked by the Deep Space
before seen and wid be said to have risen Network in 1998.
new questions about origins and devel-
opment of the solar system, and possibilities for water, maybe even life, on the planet
or one of its satellites. The Wyager 1 view of Jupiter as the spacecraft approached the
Figure Intro-2. Voyager 1 views Jupiter close up, February 1979. One of the most spectacu-
lar planetary photographs ever taken, this close-up picture of Jupiter was taken by Voyager 1 on
its approach to Jupiter in February 1979. Passing in front of the planet are two of the Galilean
satellites, lo and Europa. Both satellites are larger than Earth's Moon, and the Great Red Spot
seen on Jupiter's surface is larger than Earth itself.

planet, shown in Figure Intro-2, is a prime example of science data return with high
value for public interest.

There will be some with enquiring minds who may connect new scientific evidence with
a spacecraft launch, recalled from the distant past (see Figure Intro-l), and ask the ques-
tion, How did "ity' get from there to what I am seeing now? (See Figure Intro-2.)

Well, the answer depends on what is meant by "it," but an essential part of the answer
is the subject of this book, namely, the Deep Space Network. To properly understand
that part of the answer, some explanation of the name of the problem and the termi-
nology used to describe it will be necessary. The discussion that follows is directed to
that end, to readers who have a general interest in the subject but whose field of pro-
fessional interest lies elsewhere. For the expert reader, numerous technical references are
provided throughout the book for followup on topics of specific interest.

Before b e g ' i g a discussion of radio communication with spacecraft traveling in deep

space, we shall briefly review the environment of space and the motions of planetary space-
craft within that environment. This review and the discussion of space communications
that follows it will exp1ain various technical terms commonly used in describing the envi-
ronment of space, the motions of Earth and spacecraft, and space communications. It is
hoped that the general reader, who may not be familiar with many of these terms, will
find this helpful in understanding the terminology used in the main chapters of the book.

NOTE: The following disksion makes extenlive use of materid2 contained in "The Barics
of Space FIkht," u ledrning document produced by the Jet Propuhion Labomtoy for use in
its pcejight operations training propm.

The solar system has been studied for religious or scientific reasons from the very ear-
liest times. For most of that time, studies of the solar system have had to rely on indirect
measurements of the various objects in the solar system such as the visible light emit-
ted by, or reflected from the objects, or later, by the radio waves emitted by the Sun.
However, with the emergence of space flight, instruments can be sent to many objects
in the solar system to make direct measurements of their physical properties and dynarn-
ics, at close range. Data collected from such measurements have resulted in an
unprecedented increase in our knowledge of the solar system.

Size and Composirion

The solar system consists of an "average" star we call the Sun and the planets Mercury,
Venus, Earth, Mars, Jupiter, Saturn, Uranus, Neptune, and Pluto. It also includes numer-
ous satellites of the planets, comets, asteroids, meteoroids and of course, the interplanetary
medium. The planets, most of the satellites of the planets, and the asteroids, revolve
around the Sun in the same direction, in nearly circular orbits. The Sun and the pian-
ets rotate on their individual axes. Except for Pluto, all the planets orbit the Sun in or
near the same plane, called the plane of the ecliptic.

The most common unit of measurement for distances within the solar system is the
astronomical unit (AU). One AU equals the mean distance of the Sun from Earth, about
150,000,000 km. Prompted by the need for a more accurate figure for spacecraft nav-
igation purposes, the DSN refined the value of the AU in the 1960s using radar echoes
from Venus. Distances within the solar system, the distance between a DSN antenna
on Earth and a planetary spacecraft for instance, are often indicated in terms of the dis-
tance light travels in a unit of time at the speed of 300,000 krn per second.

For example:
(,
...:-~..~.-.~ii..i..,"
." - ....
i....-i.i..-.i...i.., i<<A ,iiiiiii .,.,
i..i.i.i.,.i "iii.,iiiii i
iii
iii
ii.i.i.i,".,..i< i.i.i.iii.iii.ii ,.*.".;., i.i.i,i,,.i,
i i'i.i.
>
.,./ ...,.,...,. ...,.>/ ..,..-.,. f
r,J
..:.. .:_.<_.<, r,. ..:.2 .i.2
r2 1a-.:. ~.%
/-T.2:-~.T~.-T.

Light Time Light Travels Approx. Distance

1 second 300,000 km 0.75 Earth-Moon
1 minute 18,000,000 km 0.125 AU
8.3 minutes 150,000,000 km 1 AU, Earth-Sun
1 hour 1,000,000,000 km 1.5~ Sun-Jupiter

Although the Sun is characterized as a typical star, it dominates the gravitational field
of the solar system; it contains 99.85 percent of the mass of the solar system. All the
planets combined contain only 0.135 percent of the total mass with the satellites of the
planets, asteroids, meteoroids and interplanetary medium making up the balance of 0.0 15
percent. Even though the planets account for only a small portion of the total mass of
the solar system, they retain the greater part of the angular momentum of the solar sys-
tem. This storehouse of energy can be utilized by interplanetary spacecraft to make
so-called "gravity assist" changes in their interplanetary trajectories.

In terms of volume, nearly all of the solar system appears to be an empty void. Far from
being just nothingness, however, this void comprises the interplanetary medium and
includes various forms of energy and at least two material components: interplanetary
dust and interplanetary gas. The dust consists of microscopic particles which can be
measured by special instruments carried by interplanetary spacecraft. Interplanetary gas
is a tenuous flow of gas and charged particles called plasma, which when streaming from
the Sun is called solar wind. The solar wind can also be measured by interplanetary
spacecraft. The point at which the solar wind meets the interstellar medium, that is the
stellar wind streaming from other stars, is called the heliopause. This boundary, which
marks the edge of the Suns influence, is theorized to lie perhaps 100 AU from the Sun.
At the time of this writing, the Voyager and Pioneer spacecraft were making their way
toward this remote limit of the solar system. The magnetic field of the Sun extends out-
ward into interplanetary space and it, too, can be measured by instruments carried on
planetary spacecraft.
 JUPITER

LIGHT MINUTES I ASTRONOMICALUNITS

Figure Intro-3. Mean distances of the terrestrial planets from the Sun. The orbits in this fig-
ure are drawn approximately to scale. The orbit of Jupiter is shown for scale reference only.

The Terrestrial Planets

The four terrestrial planets of the solar system-Mercury, Venus, Earth, and Mars-
have a firm rocky surface similar to that of Earth. The orbits and mean disrances of the
terrestrial planets from the Sun are shown in Figure Intro-3.

None of the terrestrial planets has rings. Earth has a layer of rapidly moving charged pard-
cles known as the Van Allen Belt, &st detected by the JPL Explorer 1 space probe in 1958.

The Jovian Planets

The four Jovian planets-Jupiter, Saturn, Uranus, and Neptune-are all much larger
than the terrestrial planets and have gaseous natures like that of Jupiter. They all have
satellites and rings, although the size and number of each varies considerably between
them. The orbits and mean distances of the Jovian planets from the Sun are shown in
Figure Intro-4.
 LIGHT HOURS I ASTRONOMICAL UNlW

Figure Intro-4. Mean distances of the Jovian planets from the Sun. The orbits in this figure
are drawn approximately to scale. Pluto, at a mean distance of 39.5 AU from the Sun, is not dis-
cussed in this book and is omitted from the figure to accommodate the scale of the other planets.

Inner and Outer Planets

Frequently, the planets whose orbits lie inside Earth's orbit-namely, Venus and
Mercury-are referred to as the inner planets. Mars, Jupiter, Saturn, Uranus, Neprune,
and Pluto are generally known as the outer planets.

Asteroids

Asteroids are rocky objects orbiting the Sun at a distance of about 2.7 AU, between the
orbits of Mars and Jupiter, and moving in the same direction as the planets. They vary
from the size of pebbles to objects measured in hundreds of kilometers. Some have orbits
which cross the orbit of Earth from time to time. The DSN Solar System Radar at
Goldstone is used to investigate the properties of asteroids with Earth crossing orbits.

Comets

Comets are believed to be composed of rocky material and water ice. Their highly ellip-
tical orbits bring them very close to the Sun and swing them deeply into space often
beyond the orbit of Pluto. Comet structures are diverse and very dynamic, but they all
develop a cloud of a s e material called a coma, that usually grows in size and bright-
ness as the comet approaches the Sun. A small bright nucleus is usually visible in the
middle of the coma. The coma and nucleus together constitute the head of the comet.
As a comet approaches the sun it develops an enormous tail of luminous dust materi-
al that extends for millions of kilometers from the head away from the sun. It dso
develops a tail of charged particles and an envelope of hydrogen. Several comets have
been investigated by planetary spacecraft flying close to or through the coma.

Without a system of coordinates to consistently identify the positions of observers, plan-

ets, and interplanetary spacecrafi, exploration of the solar system would not be possible.
Because space is observed from an Earth platform, a system of Earth coordinates is
required to establish the position of the observer. The locations of DSN antennas on
Earth's surface are specified in terms of latitude and longitude.

To establish a coordinate system for the sky, the concept of a celestial sphere whose center
is at the center of the Earth, is used. The celestial sphere has an imaginary radius larger
than the distance to the f h e s t observable object in the sky, that is to say, it extends far
beyond the limits of the solar system. The extended axis of Earth intersects the north and
south poles of the celestial sphere. The direction of a spacecrafi, planet, or star, or any other
celestial object, can be specified in two dimensions on the inside of this sphere using a sys-
tem of coordinates analogous to Earth's latitude and longitude system. Although the reference
origins are different, the andogous terms latitude and longitude on the celestial sphere are
declination (DEC, latitude) and right ascension (RA, longitude). When used in connection
with a specific location such as a DSN antenna, at a particular time of day, the term RA
is replaced by a different term cailed hour angle or HA,

A somewhat simpler system for describing the position of a distant spacecxafi relative to
a particular antenna and time of day uses the local horizon and true north as its refer-
ences. Its measurements are azimuth (AZ), measured in degrees clockwise around the horiin
firom true north and elevation (EL), measured in degrees above the local horizontal datum.
Optical telescopes, radio telescopes, and the DSN antennas are designed with mountings
that make best advantage of either the HA-Dec or Az-El coordinate systems.

In a HA-Dec system the HA axis is parallel to Earth's axis of rotation. Thus an anten-
na built on a HA-Uec mount has the advantage that motion is required mostly in only
one axis, HA, to track an object like a spacecraft, as Earth rotates. The disadvantage is
 f 3
HA-Dec MOUNT Az-El MOUNT
TYPICAL OF MORE RECENT
61, AND DSS-42 DESIGNS

HOUR-ANGLE
BEARING
BEARlNG
HOUR-ANGLE
COUNTERWEIGHT DECLINATION
BEARINGS AZIMUTH (ELEVATION
COUNTERWEIGHT
(DECLINATION STRUCTURE OMIilED
HOUR-ANGLE AXIS COUNTERWEIGHT FOR C w t r r v )
SUPPORTEO SO AS STRUCTURE OMRED
TO ALIGN WITH FOR CLARITY)
EARTH'S AXlS

L .
J
'
Figure Introd. HA-Dec and Az-El mounted antennas of the DSN.

that it requires an asymmetrical design, which is unsuited to the support of very heavy
structures. By contrast, Az-El mounted antennas are basically symmetrical structures which
can support heavy weights, but to track celestial objects they require driving in two axes,
AZ and EL simultaneously. However, the advent of high speed compurers has obviated
the problems formerly associated with converting coordinates from one system to the
other. The essential differences between the HA-Dec and &-El types of mount for DSN
antennas are illustrated in Figure Intro-5.

Earth Motions

Earth rotates on an axis inclined at 23.5 degrees to the plane of its orbit around the Sun.
Its period of rotation is 24.0 hours mean solar time. This motion is of prime importance
to the conf~gurauonof the Deep Space Network, Due to the rotation of Earth, a space-
craft moving away from Earth along its trajectory in space appears to an observer on
Earth to set toward the west. To an observer further west, the spacecrafi will appear to
rise in the east and travel across the sky until, 8 to 10 hours later, it sets. The process
repeats until 24 hours later the spacecrafi appears, rising in the East for the first observ-
er. It follows that three observers, each located 120 degrees of longitude apart on rhe
surface of Earth, would have the spacecraft continuously in view as the view period passed
from one to the other. This is the basic idea underlying the location of the three Deep
Space Communication Complexes of the Network at Goldstone, California, near Madrid,
Spain, and near Canberra, Australia. From these locations, approximately 120 degrees of
longitude apart, their great antennas, some HA-Dec, others Az-El, are able to maintain
continuous radio view of planetary spacecrafi as Earth performs its daily rotations.
The ability of the DSN antennas to view the planetary spacecrafi is not changed by the
annual rotation of Earth around the Sun. It does, however, introduce a cyclic change to
the distance between Earth and spacecraft depending on the orientation of the space-
craft trajectory relative to Earth's orbit.

Time Conventions

In addition to local time and Greenwich Mean Time, there are several other systems for
measurement of time that are used for tracking planetary spacecraft in the Deep Space
Network. Three of the most important are Universal Time Coordinated (UTC), Earth
Received Time (ERT), and Round-Trip Light Time (RTLT).

UTC is a worldwide scientific standard of timekeeping based upon caref-irlly maintained

atomic clocks that are accurate to a few microseconds via the addition or subtraction of
leap seconds as necessary at two opportunities every year. Its reference point lies on the
Earth prime meridian at Greenwich, England. All spacecrafi operations in the DSN are
conducted on the basis of UTC.

The time (in UTC) at which a DSN tracking station observes an event associated with
a planetary spacecrafi is called the Earth Received Time (ERT). The time (in UTC) that
a tracking station observed the loss of signal due to a spacecrafi occultation by a plan-
et would be an example of such an event.

The elapsed time that it takes a radio signal (traveling at the speed of light) to travel
from a DSN station to a spacecraft or planetary body, and after retransmission by the
spacecrafi or reflection from the body, return to the tracking station, is known as the
Round-Trip Light Time (RTLT). It is used by the DSN to measure spacecraft range and
other navigational and scientific parameters. The RTLT from Earth to the Moon is about
3 seconds; to the Sun, about 17 minutes. In 1993, the RTLT for the Voyager 1 space-
cr& was 14 hours, 13 minutes. By the end of 1998, this had increased to 20 hours,
18 minutes, making Vyager 1 the most distant spacecraft in the solar system.

A spacecraft sitting atop a launch vehicle at Cape Cmaveral can, in one sense, be con-
sidered as already in orbit around the Sun by virtue of the motion of the Earth orbit
around the Sun. To send such a spacecraft to an outer planet, Mars for example, the
spacecraft's existing orbit must be adjusted to cause it to intercept the orbit of Mars at
a single point. The portion of the new orbit between Earth and Mars is called the inter-
planetary trajectory.

To achieve such a trajectory, the spacecrd is lifted off the pad by the launch vehicle, rises
above Eh&"s atmosphere, orients itself to the right attitude and is accelerated in the direc-
rion of F!arth's orbit around the Sun to the extent that it becomes fiee of Earth's gravitational
effect. The magnitude and direction of the accelerating force is carefully calculated to
achieve a new orbit which will have an aphelion (firhest distance) equal to rhat of the
orbit of Mars. After injection into its new orbit, the spacecraft simply "coasts" the rest of
the way to its destination. Of course, to get to Mars itself, rather than just to Mars' orbit,
the spacecraft must be inserted into its interplanetary trajectory at precisely the right time
to reach the orbit of Mars at the same time that the planet itself reaches the point where
the spacecraft will intercept the orbit of Mars. This calls for very precise timing indeed,
and results in constrained opportunities for launch called "launch windows." A spacecraft:
interplanetary trajectory can be adjusted or trimmed, to a limited degree, by means of
midcourse maneuvers which use the onboard thrusters to change the spacecraft speed and
direction, or by means of a "gravity assist," which uses the gravitational force of a nearby
planet for the same purpose. The motions of Earth that play an important part in the
interplanetary orbit insertion process are shown in Figure Inrro-6.

f \

LAUNCH USING EARTH'S ROTATlONALAND ORBITAL VELOCITIES

ROTATIONAL MOTION
J

Figure Intro-6. Insertion into interplanetary orbit. For interplanetary launches, the spacecraft takes
advantage of Earth's orbital motion to supplement the limited energy available from the launch vehi-
cle itself. In the diagram, the launch vehicle is in addition to using Earth's rotational speed, accelerating
generally in the direction of Earth's orbital motion, which has an average velocity of approximately
f 00,000 km per hour along its orbital path. The launch vehicle is used to increase or decrease
this velocity in the tangential direction to insert the spacecraft into the desired transfer orbit.
The vital DSN operation of "initial acquisition'' is performed during, or as close as pos-
sible to, the interplanetary trajectory injection maneuver right after launch. After that,
spacecraft and mission controllers are entirely dependent upon the DSN for communi-
cations with the spacecrafi throughout the life of the mission.

The foregoing paragraphs have provided a greatly simplified answer to the question of
what happens to a spacecraft after it leaves the launch pad. They have also described
the environment in which it will move for the rest of its life. An answer to the next
question, How do spacecraft and mission controllers communicate with their spacecraft
during its long passage through the environment of deep space? requires some under-
standing of the narure of deep space communications.

The Deep Space Nerwork makes use of electromagnetic radiation for communicating
with interplanetary spacecrafi. All planetary spacecrafi are equipped with radio trans-
mitters and receivers for sending signals to and receiving signals from the Earth-based
antennas of the DSN. In addition to signals from spacecraft, DSN antennas and receivers
are capable of detecting signals fiom natural emitters of electromagnetic radiation, such
as the stars, the Sun, molecular clouds, and giant gas planets such as Jupiter. The "sig-
nals" hom these sources appear as random noise to the sensitive receivers of the DSN,
and their study and interpretation is the field of radio astronomy. The DSN supports
many research projects in radio astronomy.

Bands and Frequencies

Electromagnetic radiation from the natural emitters combines with radiation from arti-
ficial sources to create a background level of electromagnetic noise fiom which the spacecraft
signals must be detected. The ratio of the signal level to the noise level is known as the
signal-to-noise ratio (SNR). SNR is one of most common terms used in the DSN to
describe the quality of a communication link.

Electromagnetic radiation with frequencies between about 10 kHz and 100 GHz are
referred to as radio frequency (RF)radiation. For convenience in managing its use, RF
radiation is divided into groups called bands, such as S-band and X-band. The bands
are further divided into small ranges of frequencies, or channels, some of which are allo-
cated by international agreement, for the use of deep space telecommunications. The
following table shows the approximate range of frequency and wavelength for each band.
Radio Comunication Bands vs. Wavelength and Frequency
v > ~ ~ ~ ~ s ~ ~ ~ ~ ~ w w r ~ ~ / ~ # a r m r r r f ~ m ~ ~ r / a x r r r ~ ~ r r m > m ~ m r r ~ f l r r ~ z ~ r ~ ~ ~ ~ ~ ~ ~ z r r ~ ~ r ~ r ~ ~ r r ~ ~ ~ r

Band Name Wavelength (cm) Frequency (GHz)

L 30-1 5 1-2
S 15-7.5 2-4
C 7.5-3.75 4 8
X 3.75-2.4 8-12

Early spacecraft used L-band for deep space communications. Within a few years, S-
band replaced L-band, and more recently X-band came into general use for deep space
communications. Experiments to demonstrate the advantages of telecommunications sys-
tems using K-band were in progress at the time of this writing (1997). Generally speaking,
the higher frequency bands offer greater advantages for space communications, although
these tend to be offset by other factors, such as increasing losses at the highest bands.

Doppler Effect

The Doppler effect is routinely observed as changes in the frequency of spacecrafi radio
signals received by a DSN tracking station. This is caused by the relative motion between
a spacecraft and the tracking station due to the spacecraft trajectory, its orbit around a
planet, Earth's orbit around the Sun, or the daily rotation of Earth about its axis. When
the distance decreases, the frequency decreases proportionally to the rate of change of
distance, and vice versa. If two widely-separated tracking stations observe a single space-
craft, they will have slightly different Doppler signatures. This information, described as
the Doppler type of radio metric data, is generated at the DSN tracking stations and
used by spacecraft navigators to describe the path of the spacecraft through deep space,
very precisely and in three dimensions.

Cassegrain Focus Antennas

Whether the mount is of the HA-Dec or Az-El type, all DSN antennas, irrespective of
their size, use the Cassegrain focus system to concentrate the electromagnetic energy inci-
dent upon their s h c e s . The incident energy may originate &om a distant spacecraft, a
celestial body, or the station's own uplink transrnirter. It works equally well for either uplink
or downlink. A diagram of a cassegrain focus antenna is shown in Figure Intro-7.

In a Cassegrain antenna, a secondary reflector is added to the structure to fold the elec-
tromagnetic beam back to a prime focus near the primary reflector in the manner shown
r \
CASSEGRAIN FOCUS ANTENNA

Incoming
Electromagnetic

Reflecting Surface

Primary Reflecting Surface

(Parabolic Section)
L 1

Figure Intro-7. Cassegrain focus antenna.

in the figure. Incoming electromagnetic waves are focused by the prime reflector on to the
secondary reflector, or hyperboloid, which refocuses them into the receiver feed horn locat-
ed at the prime focus. Usually, several feed horns are mounted on a single cone structure,
and by rotating the slightly ofbet secondary reflector, the main beam can be directed to
any receiver horn as required. For transmitting, the system works in the reverse way.

This design accommodates very large-diameter antennas and allows bulky, heavy receiv-
ing and transmitting equipment to be located near the center of gravity of the composite
structure.

Uplink and Downlink

The radio signal transmitted from a DSN antenna to a distant spacecraft is known as
an uplink. The radio signal transmitted by the spacecraft to the DSN is known as a
downlink. Uplinks or downlinks may consist of a pure RF tone called a carrier, or a
carrier which has been modulated to carry information in each direction, including com-
mands to the spacecrafi or telemetry data horn the spacecrdi. A spacecraft communications
link that invoIves o d y a downlink is called a one-way link. When an uplink is being
received by a spacecraft at the same rime that a downlink is being received, the com-
munications mode is said to be two-way. These two distinct modes of operation play a
significant part in the operation of deep space communications.

Signal Power

Typically, a I o d broadcasting station with a transmitting power of 50,000 watts will deliv-
er an acceptable radio signal to a portable receiver at a distance of 100 km. How then
can a spacecraft transmitter, limited to a power of 20 watts, deliver an acceptable signal
to a DSN receiver across hundreds of millions of kilometers of interplanetary space? The
first step involves concentrating all of the available energy into an extremely narrow beam
pointed in one direction, rather than spreading it in a l l directions, as in a broadcast sta-
tion. At the spacecraft, this is done with a small parabolic dish antenna, typically one to
five meters in diameter. Even so, when these concentrated signals reach Earth they have
vanishingly small power, perhaps as small as I x watts. The rest of the solution is
provided by the receiving power of the large apermre antennas of the DSN and the extraor-
dinary sensitivity of its cryogenically-cooled low-noise receivers. Aided by special coding
and decoding schemes to discriminate against radio noise, the DSN can extract the sci-
ence and engineering data f?om these unimaginably weak signals and deliver it in red time
to the intended users. As this history will show, the ability of the DSN to do this has
improved by many orders of magnitude over the forty or so years since its inception.

Coherence

In addition to its use as a conveyor of modulated telemetry data, the downlink carrier is also
used by the DSN for tracking the spacecraft and for carrying out some types of radio sci-
ence experiments. Each of these applications requires the detection of minute changes (fractions
of 1 Hz) in many GHz of carrier fi-equency over many hours. This can only be accom-
plished when the frequency of the downlink carrier itself is extremely stable and is known
with very great precision. Since the spacecraft itself could not carry the massive equipment
needed to do this, it makes use of the uplink, which does have the requisite stability, as a
fiequency reference for the downlink In effect, the spacecraft simply retransmits the DSN
uplink after modulating part of it with the desired telemetry data When this is done, the
downlink is said to be two-way and coherent, that is, in-phase with the uplink

At each Deep Space Complex, a hydrogen maser-based fiequency standard provides the
reference for generating an extremely stable uplink fi-equencyfor transmitting to the space-
craft. The resulting spacecraft downlink, based on and coherent with the uplink, has practically
the same high fi-equency stability as the original reference frequency standard. Comparison
of one with the other at the phase level of individual cycles produces the desired tracking
or scientific data. The enrire process ultimately depends for its success on the "phase coher-
ence" of the uplink and downlink carriers, and the accuracy with which corrections can
be made for the many sources of error in the end-to-end system.

Under some operarional conditions the spacecraft may not have an uplink. To cover such
cases, all spacecraft carry a small auxiliary oscillator, which serves as a reference for gen-
erating a non-coherent downlink in the absence of a DSN-generated uplink. It is not
highly stable and its fiequency is affected by temperature variations on the spacecraft.
Nevertheless, there have been numerous instances in the long history of the DSN when
a spacecraft's local oscillator played a major role in saving an otherwise doomed space-
craft. Some spacecraft also carry a thermally stabilized ultra-stable oscillator (USO) which
is used for precision radio science experiments involving planetary occultations. Because
of stringent frequency requirements for spacecraft tracking navigation and radio science,
the DSN remains at the forekont of the technology for frequency and timing standards.

Decibels

Perhaps the most commonly used technical term in the DSN is the ubiquitous decibel.
Since the term decibel (or dB) appears frequently in the following history, a simplified,
non-rigorous explanation is in order. The decibel is a unit of measure used to describe
the ratio between two power levels. For example, the power delivered by a 100-watt
audio amplifier is said to be 10 dB higher than the power delivered by a small 10-watt
amplifier. The ratio of the powers is 10 to 1. Inversely, where the ratio is 1 to 10 (1110)
the equivalent term would be -10 dB.

Because of its mathematical origin, the decibel scale is not linear, it is logarithmic; the
10 dB corresponds to a ratio of 10, 20 dB corresponds to a ratio of 100, 30 dB to a
ratio of 1000 and so on. For ratios less than 10, 3 dB corresponds to a 2 to 1 power
ratio, 6 dB to a ratio of 4, etc. When the ratios are inverted they are represented by a
negative sign before the corresponding dB value, so that a ratio of 1 to 2 or 112 is rep-
resented by -3 dB.

Much of the technical progress in the DSN is reflected in terms of dB values for high
power transmitters, gain of large antennas, threshold of receiving systems, gain of maser
amplifiers, signal-to-noise ratios, bit-error rates for decoding systems, etc.

The remarkable improvement in downlink performance that took place during the Mariner,
Viking, and Voyager Eras, the first three great eras of DSN growth, is shown graphi-
cally in Figure Intro-8.

xli
 I I
LEGEND:
FREQUENCY IN GHz (8.4)
ANTENNA DIAMETER
IN M f f ERS (64)
- * SYSTEM NOISE
TEMPERATURE
IN KELVINS (30)

-
AT SATURN FOR
44.8 KBIS DATA RATE

60 1970 1980
CALENDAR YEAR

Figure Intro-8. Improvement in DSN downlink performance during the first twenty years. In
this graphic, published by C. T Steizried et al. in 1982, downlink performance is stated in terns
of the parameter M which relates the two key elements of downlink performance: the gain (size)
of the receiving antenna G, and the noise temperature (sensitivity) of the receiving system, Top. It
is expressed in dB as described above.

In the context of this figure, improved downlink performance implies an increased

capability for a DSN tracking station to return more data at the same range, the
same data at a greater range, or some other enhanced combination of data and range
from a planetary spacecraft. The effect of changes in operating frequency, from L-
band to S-band to X-band, is evident in the stepwise improvements, as is the
contribution of the larger diameter antennas both individually and in array with
other antennas. The parameter called system noise temperature is a measure of the
sensitivity of the maser receiving systems that are a key element in the overall per-
formance of the downlink.

Afier the early 1980s, improvements in terms of dB were less dramatic, not because of
any less incentive for hrther improvement, but because by that time the technology of
deep space communications was approaching the practical limits of physical realization
for that era. More esoteric approaches, such as advanced data coding methods, were
called upon, and although significant improvements continued to be made, they were
much smaller in terms of dB. This issue, and how the DSN dealt with it, is discussed
in chapter 7, "The Advance of Technology in the Deep Space Network."

Modulation and Demodulation

Scientific and engineering data is generated by a planetary spacecraft in digital form and
modulated to an S-band or X-band carrier signal for transmission to the antennas of
the DSN. Modulation is a two step process in which the bits of raw data are first applied
to a subcarrier of much lower frequency. The data-modulated subcarrier is then applied
to the RF carrier for radiation, via the spacecrafi's transmitter and antenna, to Earth.
At the DSN receiving station, the process is reversed. Sensitive receivers detect the car-
rier signal, extract the subcarrier and pass it to a subcarrier demodulator. The subcarrier
demodulator extracts the data and conditions it for recording and forwarding to JPL.
Most spacecraft apply a complex code to the raw data to protect it against noise-induced
errors during transmission. Special equipment at the tracking stations performs the extra
step of decoding the data before it is processed for recording and delivery. This end-to-
end function of a deep space communication system; and the data associated with it, is
called telemetry.

There is an analogous function that uses the S-band or X-band uplink for transmission
of command data to a spacecraft. It is called command. In this case, the subcarrier mod-
ulated with a bit-stream of command data. At the spacecraft, the carrier is detected, the
subcarrier is demodulated and the command data bits are eventually passed to the com-
mand system for immediate execution or for storage for later action.

The distance from Earth, or range, of a distant spacecraft can be measured by modu-
lating the uplink with a special code or sequence of digital characters, which, when
turned around by the spacecrafi and remodulated on the downlink, allow the time delay
between transmission and reception to be determined with great precision. The

xliii
Earth/spacecraft range is then calculated from knowledge of the time delay and the prop-
agation velocity of the modulated radio signal.

Spacecraft Radio System

Intentionally, the spacecraft itself was not among the topics discussed up to this point.
While a planetary spacecraft is a miracle of modern technology, the history of planetary
spacecraft development is beyond the scope of this book. It is important to note, how-
ever, that the spacecrafi radio system is an integral part of a deep space communications
system although not an integral part of the DSN organization during the period of time
covered here. The spacecraft radio system was ~rovidedby the flight project organiza-
tion as part of the spacecrafi. Nevertheless, responsibility for RF compatibility between
the spacecraft and the DSN was always a DSN responsibility. In the ensuing chapters,
where a spacecraft radio system is discussed, the general explanations given above will
suffice for an understanding of the matter.

In the preceding discussion, I elected ro use generalities in describing some of the basic
concepts of deep space communications. The reader will encounter all of these terms
and topics many times as the history of the Deep Space Network unfolds. These con-
cepts also allow us to answer h e questions prompted by the launch pictures, and by
the remarkable new images shown at the press conference. An essential part of the answer
is provided by NASA's Deep Space Network. Worldwide in concept, continuous in oper-
ation, the DSN is the link between the two images. The global scale of the Network is
illustrated by the map shown in Figure Intro-9.

In terms of the foregoing discussion, the Deep Space Network may be characterized as
a scientific instrument of worldwide proportions that uses a single up- and downlink
radio signal, in combination with a coherent spacecraft transponder, to perform three
basic functions. Interplanetary spacecraft depend on the first two for their communica-
tions with Earth. The third and no less significant function brings the DSN into the
science community in a more direct way. Each of these functions is associated with one
of the unique types of data carried on the up- and downlinks.

The first, and perhaps the most important, function is that of generating radio metric
data. Consisting of Doppler data, ranging data, and intederometric comparisons between
two tracking stations, radio metric data are used by spacecraft navigators to determine
the precise location of a spacecraft along its trajectory at all times. Radio metric data
Figure Intro-9. An essential part of the answer; global map of the Deep Space Network, 1992. The three Deep
Space Communication Complexes are located at intervals of 120 degrees of longitude around Earth, and are identified
by the local name of the area and the generic name of the Complex. The antennas at each Complex are shown as they
were in 1992; more have been added since then. For the sake of clarity, the NASA communications system linking each
Complex to the Network Operations Control Center have been omitted.
are aIso key to the task of pointing the DSN antennas in the right direction to estab-
lish, and continuously maintain, a radio link between spacecraft and Earth.

The second function, telecommunications, makes use of appropriate modulation impressed

on the uplink (by a tracking station), and on the downlink (by a spacecraft), to con-
nect a spacecraft with its Earth-based controlfers. Command instructions are sent to the
spacecrafi on the uplink, while engineering and science data are telemetered back to
Earth, on the downlink.

A third function relates to the use of the DSN as a scientific instrument for research
in radio and radar astronomy. Scientists were quick to recognize the potential of the
state-of-the-art capabilities of the DSN as a powerfd new resource for advancing their
experiments. Encouraged by a generous NASA policy for making time observation
available, the large antennas that could be pointed with great precision, sensitive receivers,
and extremely stable timing systems of the DSN soon attracted the attention of many
of the world's leading researchers. Using spacecraft radio signals, Extra-Gdactic Radio
Sources or echoes from powerful radar transmitters at Goldstone, scientists continue
to use the DSN as a scientific instrument to widen our knowledge of distant regions
of the universe.

What follows is an end-to-end scenario that brings all the pieces together to describe
the deep space communications process for a typical planetary mission.

Soon after the launch vehicle was lost to view on rhe television screen, the first stage
booster burned out and fell away, leaving the injection vehicle, with the spacecraft attached,
to coast for a short time until it reached the exact position and time for the trajectory
injection maneuver to begin. At about that same time, the injection vehicle had moved
far enough around Earth to be in radio view for the DSN tracking stations in Australia.
The Canberra stations tuned their receivers to the spacecraft's auxiliary oscillator fre-
quency, making allowance for the one-way Doppler effect, and locked-on their receivers
to automatically follow the spacecraft. The telemetry data on the one-way downlink car-
ried engineering data that confirmed that the spacecrafi survived the stresses of the launch
events. At the appointed time, the injection vehicle engine was fired to accelerate the
spacecraft fiee of Earth gravity and on to its interplanetary trajectory. The small initial
acquisition antenna at Canberra had been following the spacecrafi and now steered the
more powerful, but less agde, 34-meter antenna to point at the spacecraft where it too,
locked up its receivers.
Telemetry data, demodulated and decoded at the tracking station, continued to flow
to mission controllers at JPL, but they needed to send commands to the spacecrafi
immediately to orient it to the Sun and to reconfigure it for its long cruise to the
planet. Besides, spacecraft navigators were anxious to have the better, more-stable
Doppler data that was associated with the two-way, coherent mode. The transmitter
at the tracking station was turned on, and making allowance for the uplink Doppler
offset, its ultra-stable S-band (or X-band) uplink was tuned across the frequency of
the spacecraft receiver. On detecting the uplink, the spacecrafi automatically switched
over to the coherent mode of operation and began retransmitting the uplink back to
the tracking station. The DSN receivers were then in the two-way, coherent mode,
the two-way Doppler was reaching the navigators, and commands could be trans-
mitted at any time as required.

As the spacecr& continued to move away from Earth on its new trajectory, Earth rota-
tion brought the other DSN Complexes into view, and the stations there passed the
spacecraft in turn, from one to the other. This involved a complicated operational pro-
cedure to avoid losing telemetry data, but the net result was that communication with
the spacecrafi continued, uninterrupted by Earth's rotation, for the approximately two
remaining years of the mission.

During the long cruise period, the Navigation Team used two-way Doppler and rang-
ing data generated at the tracking stations to verrfy, and when necessary adjust, the
trajectory of the spacecraft to ensure that it reached the planet at the correct position
and time for a successful encounter. Because the distance between the spacecraft and
Earth had been steadily increasing, the power level of the carrier signal received by the
34-meter antennas gradually dropped to the point where the signal-to-noise ratio was
too low for the DSN receivers to operate properly. At that point, the tracking stations
changed over to the big 70-m antennas for tracking this spacecrafi. The 70-m antennas
dowed them to receive telemetry data at a higher data rate, a capability that would be
needed at the greater range of the forthcoming planetary encounter.

Encounter day finally arrived. All of the complex sequences of operation that the space-
craft would carry out during encounter had been transmitted to the spacecrafi from the
DSN and were stored in its computer memories. For some time, the spacecraft cameras
and other scientific instruments had been trained on the approaching planet; engineer-
ing and the telemetry data rate had been increased to the maxirnum possible. All the
terrestrial communications circuits linking the three Deep Space CompIexes to JPL were
on high alert. The 3eep Space Network was then doing what it does best: providing a
real-time, two-way, coherent, digital data communications link between NASA's Jet

xlvii
Propulsion Laboratory in Pasadena, California, and a planetary spacecraft encountering
one of the Jovian planets.

Responding to its stored commands, the spacecraft executed the encounter sequence of
events according to a pre-determined plan that included all of the onboard science instru-
ments. As the spacecraft entered occultation (was obscured by the planet), radio science
experiments were carried out using the spacecraft USO. The immense amount of data
collected in the short time of its a c d encounter was stored by the spacecraft on its
tape recorder for later playback via the downlink. Only the most important data could
be sent back in real time, but this was sufficient to give anxious mission scientists their
first glimpse of the planet, up close. Even after demodulating and decoding the down-
link data stream that contained the images, the tracking stations could not view them.
They had to be separated from the other science data and further processed by com-
puters at JPL before they could be analyzed by the imaging team and presented for
public viewing.

It was those pictures that the public first saw on television at the early NASA press con-
ferences firom JPL. Later, as the remainder of the encounter data were retrieved by the
DSN, scientists conducted much more detailed analyses of their data and their findings
were ultimately presented to the world at NASA press conferences in Washington. The
DSN stations continued to track the spacecraft as it pursued its orbit around the Sun,
returning science data at an ever decreasing rate and conducting radio science experi-
ments as the opportunity arose. So long as the spacecraft continued to Eunction properly,
scientists continued to press for more data from deep space, and for the support of the
Deep Space Network to obtain ir.

The foregoing end-to-end scenario established the DSN as an essential part of the
answer to the questions invoked by a search for a less obvious connection between the
launch of a spacecraft and science data presented at a NASA press conference. Perhaps
that answer begets krther the question of how the DSN came into being and how it
developed. With that question in mind, what a timely open to the history of the Deep
Space Network.

xlviii
GENESIS:
1957-1961 UPLINK-DOWNLINK: OF THE DEEP
A HISTORY NETWORK
SPACE

GOLDSTONE, CALIFORNIA

A fine, wide, blacktop road runs past the salt-encrusted bed of dry lake toward the Mars
site at Goldstone, deep in California's Mojave Desert. Opposite the lake bed off to the
right, a disused minor road leads through a narrow defr in the desiccated, brown and
brick red-colored landscape to another dry lake bed, surrounded by low, barren hills and
also encrusted with dried salt. There is nothing else here except for a small brick build-
ing and an unusual, spidery, three-legged structure supporting a large, perforated metal
dish. The open side of the dish faces the sky, and high above its center, a smaller dish
is suspended at the apex of four tall struts protruding from the surface of the large dish.
Although the immensity of the totally barren landscape dwarfs the whole structure, it
is, in fact, well over thirty meters to its highest point. A chain link fence encloses the
area, and a bronze tablet on the nearby stone marker informs the occasional visitor that
this is the site of the NASA Pioneer Deep Space Station. The inscription on the plaque,
overlaid on the photograph of the Pioneer site shown in Figure 1-1, identifies the sta-
tion's association with the NASA space program.

Figure 1-1. View of the Pioneer tracking station site, Goldstone, California, 1978. The road
entrance to the site is on the far side of the picture, while the antenna, seemingly diminished by
the immensity of surrounding landscape, stands in the center near the edge of the lake bed.
Figure 1-2. Prominent personalities of the Deep Space Network unveiling the commemorative
plaque at the Goldstone Pioneer Site, 28 October 1978. Eberhardt Rechtin, principal founder of the
Deep Space Network, is on the right. William H. Bayley, the original General Manager of the Network,
is on the left. Richard K. Mallis, DSIF Operations Manager, is standing next to Bayley. Charles Koscielski,
Director of the Goldstone Complex at the time of the dedication ceremony, is next to Rechtin.

Several of the people who played a prominent part in the establishment of the original
Network attended the dedication ceremony. They appear in the unveiling picture shown
in Figure 1-2. Their role in the chronicle of the Deep Space Network (DSN) will become
apparent in later chapters.

This antenna, built in 1958, known through most of its working life as Deep Space
Station 11 (DSS 1I), was the first antenna in what evenrually became the NASA Deep
Space Network. As the inscription of the plaque implies, it supported alI of the NASA
deep space missions during the twenty years of its operational life.

At deactivation in 1981, Station Director Lou Butcher sent a message to his colleagues
throughout the Network which neatly summarized the Station's long association with
the NASA space pragram: "The closing of the Pioneer (DSS 11) station marks the end
of 23 years of spaceflight tracking. DSS 11 was built when the Jet Propulsion Laboratory
UPL) was under contract to the Army and was the first 26-meter antenna to be built
in the DSN. DSS 11 was the leader for all DSN srations to follow in tracking the first
JPL space mission (Pioneer 3) in December 1958. It was the station that tracked Suwqor
I onto the Lunar surface, and the station that tracked the lunar module (Apollo 11)
onto the Lunar surface. It has played a major role in most of the NASA Lunar and
Planetary spaceflight achievements." The United States Department of the Interior des-
ignated the Pioneer site as a Na~onalHistoric Landmark on 27 December 1985. In a
very literal sense, rhis is where the history of the Deep Space Network began.

The International Geophysical Year (IGY)l, 1957, a cooperative international enterprise

to advance the state of scientific knowledge about Earth and its environment, commenced
in July and was running its course. In October, the Russians launched the world's &st
Earth-orbiting satellite, called Sputnik, as part of their contribution to the IGY activities.
About a month later, Soviet scientists launched a larger version of Sputnik containing a
live dog, named Laika, into a high elliptical orbit around the Earth. Both in and out of
government, scientific, and engineering circles, the effect of the Sputnik launch was one
of complete surprise, followed by damaged pride and embarrassment. From a military
viewpoint, the implications were obvious: the Soviets demonstrated an operational, inter-
continental ballistic missile (ICBM) capability ahead of the United States. W e to the
scientific world at large, Soviet science was clearly leading the "race to space," the hopes
of U.S. scientists were pinned on Project Vanguard, a U.S. Navy project to place a grape-
fruit-sized scientific package in Earth orbit. In full view of a national television audience,
the first Vanguard launch at Cape Canaverai in December 1951 was a complete failure,
enhancing even more the prestige of the Soviet accomplishment with Sputnik.

The U.S. government was engaged in a desperate "catch-up" game with its Soviet coun-
terpart in the newly emerging field of space research.

The events that followed the December Vanguard debacle are the st& of which the his-
tory of JPL is made, but to properly appreciate what happened next, one needs to make
a short backtrack to 1955.

As a consequence of a l+y experience with the U.S. Army as a conmctor on various guid-
ed missile development programs at the White Sands Missile Range,JPL had already developed
a great deal of arpertise in the techniques and technology of guidance and imcking systems for
large rockers and ICBMs. While worlung for the b y as long ago as 1955, JPL, in compe-
tition with the Navy and the Air Force, submitted a proposal for an M-orbiting satellite b r
the U.S. contribution to the IGY JPLls satellite was to be called Orbiter. The Department of
Defme (DOD), however, selected the Navy's proposal called V@ and JPIA Orbiter pro-
pod was shelved. Perhaps it was serendipity, but the U.S. Army encouraged JPL to continue
with development of Orbiter, at a low level, b r the next couple of years. Hq$-speed upper
stages for the booster rockers were developed and guidance and telemetry systems were designed
and tested. Of paticuIar spficance to the t h e n b t h m e of a deep s p a . rraclang net-
wok was the traclang system designed by JPL to track the lugh altitude test rockets. It was
called Microlock, and w o r k in conjunction with a minimum w+t radio transmitter car-
ried by the @t test unit, d d provide telemetry and positional dam to a range of several
thousand kilometers. A phase-lock tracking receiver, a key fkacure developed at JPL by Eberhardt
Rechtin and Richard J&e some years earlier, was operationat at each of the Microlock stations.

Toward the end of 1957, perhaps sensing that the Vanguard program was in trouble,
Secretary of Defense Neil H. McElroy authorized the Army to reactivate the Orbiter
program and proceed with all deliberate speed to an earliest possible launch. The Army
provided a launch vehicle based on an existing Redstone rocket design and JPL pro-
vided a suitable satellite carq4ng a small scientific payload, including a radiation-measuring
instrument designed by Dr. James A. Van Allen of the University of Iowa. The satellite
was renamed Explorer and the Army-JPL team was committed to launch in 90 days.

To track the new satellite and to receive its downlink telemetry signals, JPL expanded
its existing Microlock ground-based tracking facilities to include stations at Cape Canaveral;
Singapore; Nigeria; and San Diego, California. Primitive communications services to the
overseas sites sometimes relied on native "runners" to transport messages and tapes to
the nearest telegraph ofice. Alrhough the U.S. stations had interferometfic tracking anten-
nas, a single, helical antenna at each overseas site provided only telemetry and Doppler
data. Primitive though it was, JPL had acquired its first taste of worldwide network
deveIopment and operations.2

On 31 January 1958, less than 60 days after the Vanguard explosion on the launch pad,
and just 84 days after receiving approvd from Secretary McElroy, Explorer I lifted off
the launch pad at Cape Canaveral atop a Juno I Launch vehicle and, to the acclaim of
a national television audience, became America's first Earrh-orbiting satellite. Telemetry
data received by the Microlock ground stations from the Van Allen Geiger-counters
revealed the presence of a high altitude band of radiation encircling Earth. Eventually
named for Van Allen, this became one of the most important discoveries of the IGY.

Suddenly the "catch-upyyroles of the previous year were reversed and JPL never looked
back. Henceforth, attention was focused not on the near-Earth region but outward,
toward deep space, and its chief advocate was the ambitious and far-sighted director of
JPL, William H. Pickering.
Knight Commander of the British Empire, conferred by England's Queen Elizabeth II
in 1975, is among the many honors and citations bestowed upon William Pickering by
prestigious scientific and technical organizations throughout the world. Knighthood not
only recognized his scienfific achievement., but also symbolized his British heritage and
association with his native counuy of New Zealand. Pickering received his formal high
school education in Wellington, the capital city of New Zealand, prior to embarking on
a career in electrical engineering at Catifornia Institute of Technology (Caltech) in Pasadena,
California, in 1929. Seven years later, he had earned a Ph.D. in physics and an appoint-
ment to the Caltech faculty, where, in addition to teaching, he was engaged in cosmic
ray research with Robert A. Milii and H. Victor Neher.

During World War 11, Pickering organized and taught electronics courses for military
personnel at Cdtech, and became acquainted with the Radiation Laboratory at MIT
and its director, Lee A. Dubridge. By the time Dubridge was named to the presidency
of Caltech in 1946, Pickering was working at Caltech's Jet Propulsion Laboratory on
the design and development of telemetering systems for rocket research vehicles. Pickering
was appointed director of JPL in 1954 and immediately began moving the laboratory
toward the forefront of applied engineering research and deveIopment. Within a few
short years, the results of this move became apparent to the world-at-large in a spec-
tacular way. The Eplorer I Earth-orbiting satellite, the United States' initial response to
the Soviets' Sputnik, made its appearance in January 1958, thanks to the combined
efforts of the teams led by Pickering (satellite), von Braun (launch rocket), and Van
Allen (radiation measuring experiment).

Following this spectacular success, historian CargiII Hall described Pickerkg as, "spare,
intense, reserved, and in a quiet way, implacable . . . determined to mount a JPL pro-
gram of lunar flights. To his mind," he said, "Iunar flights might once have been a
subject fit only for science fiction, but now they were on the reachable frontier of engi-
neering science, exactly the frontier where Pickering wanted JPL to be." Prophetic words
indeed. As director of what soon became a NASA Field Center, Pickering led JPL through
the highs and lows of the first U.S. spaceflights to the Moon, and later, to Venus and
Mars. Under his leadership, NASNJPL, part of which included the Deep Space Network,
quickly established and maintained a position of pre-eminence in the emerging, new art
of science and technology for deep space planetary exploration.3

He retired from JPL in 1976 and served for two years as director of the research insri-
tute of the University of Petroleum and Minerals in Saudi Arabia. He later returned to
Pasadena to enter private consulting practice.
On 7 February 1958, responding to President Eisenhower's initiative for a United States
Space Program, the Department of Defense estabhshed an org&rion called the Advanced
Research Projects Agency (ARPA) to promote, coordinate, and manage all existing mil-
itary and civilian space activities. ARPA was to h c t i o n only as an interim organization
pending congressional establishment of a civilian space management agency.*

Toward the end of March 1958, the Secretary of Defense announced approval of a lunar
space program, to be directed by ARPA as part of the United States' participation in
the International Geophysical Year (IGY) activities. The program was named Pioneer.

The Pioneer lunar program involved three launches by the U.S. Air Force, using an
existing Thor-Able booster rocket combination, and rwo U.S. Army launches utilizing
the new Jupiter Intermediate Range Ballistic Missile (IRBM), designated as Juno 11. The
scientific objectives of the Pioneer program were to measure cosmic radiation in the
region between Earth and the Moon, and to make a more accurate determination of
the mass of the Moon. At the same time, valuable experience in the design of a lunar
probe trajectory and a tracking and telecommunications system could be obtained. Because
of its long prior association with the Army's missile deveIopment programs at White
Sands Missile Range, JPL was to be involved in the rwo Army launches, Pioneers 3 and
4. The target launch dates were set for 11 November and 14 December 1958.

Research into the tracking and communications systems required to support a space pro-
gram of lunar and planetary spacecraft had been in progress at JPL for some time.

Confronted with the problem of tracking and communicating with the two Pioneer
probes, JPL put that fund of knowledge to good account.

At that time, JPL was fortunate to have, as the head of its small missile guidance research
division, a visionary whose advocacy of the long term approach determined the course
of U.S. solar system exploration far into the future.

Td, imposing in appearance, of quiet disposition and authoritative manner, Eberhardt

Rechtin was the epitome of the top-level executive. Arriculate, with a brilliant engi-
neering mind, he seemed to be way ahead of other the speaker in a conversauon, or
the presenter at a conference. He was patient with those less acute in their thinking,
and courteous in pointing out errors of fact or judgement. A dark suit and tie com-
pleted his picture. He was also, not surprisingly, an accomplished violinist.
Eberhardt Rechtin brought a B.Sc. degree in electricd engineerixlg (1946) from the CaWornia
Institute of Technology with him when he joined JPL in 1346 as a member of the very
talented cornmunicarions and radio guidance team on the Corporal and Sergeant pro-
grams. He completed a Ph.D. in electrical engineering at Caltech in 1950 and along with
Richard J&e, went on to study the theory and design of "phase-lock" circuits. Walt
Victor's help, the basis of the "phase-lock" receivers became an integral part of the DSN.5

As the first head of the Tracking and Data Acquisition Office at JPL, it was Rechtin's
vision of a worldwide network of tracking stations, and his considerable powers of per-
suasion and tenacity in pursuing his ideas with NASA, that gave impetus and direction
to the DSN in the early formative years. It was his endorsement that gave substance to
the far-reaching proposals of Victor, Stevens, and Merrick for the new 64-meter anten-
nas that would be much larger and perform better than any existing antennas designed
for tracking distant spacecraft. He introduced the concept of the ccsumof the negative
tolerances" as the standard criterion for the margin of safety in the design of upiinks
and downlinks for planetary spacecrafi. To control the electrical intedaces (uplink and
downlink) between the DSN and the many different JPL and non-JPL designed space-
craft that he perceived would require tracking support from the DSN in the years ahead,
he mandated fked, definable specification for the DSN radio links. This would become
a permanent and vital feature of DSN management in the years ahead.

Father of the phase-lock loop and architect of the DSN were his legacies. Rechtin retired
from JPL in 1967 to head the Advanced Research Projects Agency of the U.S. Department
of Defense.

It was the perception of Eberhardt Rechtin that solar system exploration should continue
to evolve in coming years and eventually become a major part of the overall NASA space
program. He envisioned not only a lunar program which included soft landings on the
Moon's surface, but also a program rhat involved a photographic survey of Venus and Mars.

A long-range vision such as his required not one, but three antennas spaced at 120 degrees
of longitude around Earth so that they could maintain continuous contact with the space-
crafi as Earrh rorated about its axis. AIthough time and h d i n g precluded the full-scale
plan, for the immediate task of tracking the Pioneer lunar probes, one antenna sufficed.
It did not allow for continuous contact with the probe, but the time of launch and tra-
jectory could be arranged so that the actual flyby occurred when the probe was in view
of the single antenna site. Later, assuming approved appropriate funding, the additional
antennas could be added to create the full network and give the U.S. the capability to
communicate with hture spacecraft travelling to the edge of the solar system.
To lend credence to his ideas, Rechtin gathered experts in ali the fields of technology
necessary to accept a challenge of those dimensions.

Victor, Stevens, Merrick, and Bell, each an expert in his own field, had the knowledge
and drive to lead the way in developing and building the super-sensitive receivers, teleme-
try and guidance systems, hrgh power transmitters and large precision antenna structures,
all needed on a very short time scale.

Immediate decisions were needed on basic communications parameters, such as operat-

ing frequency, antenna gain, antenna diameter, beamwidth, angular pointing accuracy,
slew rate, receiver sensitivity, Doppler tracking rates, bandwidth, transmitter power, and
signal-to-noise ratios. Similar parameters were needed for the probe radio system so that
the probe and Earth communications systems continued to work together as the dis-
tance between them stretched as far as the Moon and beyond.

The team had already determined that the existing ground-based tracking systems, which
were being used by JPL for the Earrh-orbiting Eplorer flights in progress at thar time,
were not suitable for tracking lunar probes. A more efficient, high-performance, Earth-
based antenna was needed. After considerations of antenna gain, automatic tracking
performance, minimum galactic radio noise and the state-of-the-art radio communications
at that time, a frequency of 960 MHz was chosen as near-optimum for the communica-
tions link between the lunar probe and Earth. The Pioneer radio transmitter transmitted
at 960 MHz and the Earth-based tracking station received on that frequency. A large-
diameter, steerable, parabolic dish antenna capable of operating at 960 MHz was needed
to receive the transmissions fiom the Pioneer probes and be available for the support of
possible follow-on missions, perhaps even for missions beyond the Moon to the nearest
planets. The launch dates were set for yearend. Cost and construction time were vital fac-
tors in the search for a suitable antenna. W e incorporating advanced and reliable principles
of design, the antenna was to be built and operating in approximately six months.

The choice of 960 MHz as the operating frequency for the Pioneer probe brought with
it another most significant advantage. It allowed JPL to make use of an exisring design
for a large radio astronomy antenna developed some years earlier for the burgeoning sci-
ence of radio astronomy by the Naval Research Laboratory, with assistance fiom the
Carnegie Institute, Associated Universities, and the Blaw-Knox Company. As a result of
this prior work, the Blaw-Knox Company could provide a large parabolic antenna that
would meet the fmed price and construction time constraint of six months. The
antenna, 26 meters (85 feet) in diameter, was designed for radio astronomy applications,
and required substantial modifications to the antenna drive system to make it suitabIe
for precision tracking of lunar space probes.

Confident in JPfi ability to make the necessary changes and get the antenna built and
operating in time for the Pioneer 3 launch, AMA decided to purchase not one, but
three, of the 26-meter antennas. One of the antennas would be used for the imminent
Pioneer launches, while the other two were intended for ARPA longer range plans for
what it termed its World Net. It was a calculated risk on the part of ARPA, for even
though the design was complete, none of the antennas had yet been built. With the
pressure of the Pioneer launch schedule, however, there was little choice in the matter.

With the operating frequency and antenna decisions already made, attention turned to
a location for the antenna site. To avoid contaminating or obscuring the very weak radio
signals received fi-om the distant spacecraft with artificial radio interference, a location
remote from any metropolitan area was desired. At the same time it needed to be close
enough to an established community to be practical for the staff that would be required
to operate the equipment. The combined constraints of funding and schedule strongly
influenced a search for a suitable site on Government-owned property. A convenient
location that met dl of these criteria was found in a natural bowl-shaped area surrounded
by low hills near the Goldstone Dry Lake on the Fort Irwin military reservation, 72
kilometers northwest of the city of Barstow, in the Mojave Desert, California.

In March 1958, a JPL radio interference survey team certified the area as fi-ee firom radio
interference. A month later a consrruction company began work on access roads, facil-
ities, services, and buildings. While work was in progress at Goldstone, a mobile tracking
station, using a three-meter diameter antenna, was erected near Mayaguez, Puerto Rico
to obtain initial trajectory data downrange from the Florida launch site.

In early June, the steel components for the antenna started to arrive onsite ar Goldstone.
Assembly of the antenna commenced in mid-August and was completed in November
1958. By that time, all the receiving, recording, communications, servo, and antenna
drive modifications and other electronics had been installed and tested. After a short peri-
od of crew training, the sration was ready for its first operational mission, but there had
been no rime to spare. A photograph of the first Goldstone antenna, as it looked short-
ly after beginning operation in 1958, is shown in Figure 1-3.

While the antenna construction work was proceeding on an accelerated schedule at

Goldstone, the direction of the space program was changing rapidly in Washington.
President Eisenhower's Executive Order No. 10783 officially established the National
Figure 1-3. The 26-meter antenna and tracking station (DSIF 11) at Pioneer site, Goldstone,
California, 1958. Pioneer, the first Goldstone antenna as it looked shortly after beginning opera-
tion in 1958. The antenna was deactivated in 1981 and has been designated a National Historic
Landmark by the U.S. Department of the Interior.

Aeronautics and Space Administration (NASA) on 1 October 1958. Discussions regard-

ing the status of the Jet Propulsion Laboratory and its possible acquisition by NASA as
a component Field Center quickly followed.

NASA outlined its proposal for the transfer of JPL from the Army on 15 October 1958.
A few weeks later, the Department of Defense indicated acceptance of the proposal and
on 3 December 1958, Presidential Order No. 10793 transferred the functions and facil-
ities of the Jet Propulsion Laboratory of the California Instirute of Technojogy fiom the
Department of the Army to NASA

NASA launched Pioneer 3 on a trajectory toward the Moon three days later, on
6 December 1958. Due to a rocket booster problem, the probe only reached an altitude of
63,500 miles before f i g back to Earth, somewhere in c e n d Africa. The small antenna
in Puerto Rico tracked it for 14 hours (round-trip) and was able to maintain telemerry con-
tact to at least 60,000 kilometers. The much larger Goldstone antenna was able to acquire
telemetry for the entire h e that the Pioneer 3 probe was above its horizon. The telemetry
received from the herto Rico station proved to have sigmficant scientific value because it
contained the data &om two passes through the then-mysterious Van Allen radiation belts.

Although the Pioneer 3 flight was a disappointment, three months later, the next lunar
probe, Pioneer 4, became the first United States spacecraft to leave Earth's gravitational
field. Launched on 3 March 1959, on a mjectory similar to that of Pioneer 3, the probe
traveled 435,000 miles toward destination before its batteries became depleted and its trans-
missions to Earth ceased. M e r 41 hours of flight, Pioneer 4 passed 37,300 miles kom the
Moon's surface on a flight path that was a good deaf ahead of and below the planned tra-
jectory. Althoug!n one of the key experiments to observe the hidden side of the Moon was
defeated by the penurbed trajectory, data &om the onboard radiation counters identified
a third belt of radiation and sensed sigdicant changes in rhe intensity of the Van Allen
radiation belts discovered by Pioneer 3. With irs large antenna working well, the Goldstone
station recorded over 24 hours of telemetry data fi-om three successive 10-how passes over
the site, before the signal ceased during the fourth pass on 6 March 1959.

While the Pioneer 3 and 4 missions had not been entirely successful, the performance
of the new antenna exceeded all expectations. It demonstrated the eficacy of irs design
and M y vindicated the confidence that M A had shown in JPCs expertise and abili-
ty to accomplish a challenging task. It would henceforth be known as the Goldstone
Pioneer tracking station, identified as DSIF 11.

The Echo Balloon Experiment

Important and highly visible as they were, the Pioneer 3 and 4 lunar probes were not the
only projects that engaged the attention of Rechtin and his engineering team at JPL in
the spring of 1959. In January, JPL had agreed to cooperate with NASA and Bell Telephone
Laboratories (BTL) of Holmdel, New Jersey in an experiment to test the feasibility of
long-range communications between two distant points on the surface of the Earth by
means of a reflected signal from the surface of a large orbiting balloon, Based on separate
/ \

ECHO 1 BALLOON

JET PROPULSION LABORATORY BELL TELEPHONE LABORATORIES

GOLDSTONE, CALIFORNIA HOLMDEL' NEW JERSEY
GODDARD SPACE FUGHT CENTER
COMPLmR
WASHINGTON, D.C. POINTING
DATA & r DATA

t 1

Figure 1-4. Essential features of the Project Echo Experiment, 1960.

and independent proposals from scientists at BTL, who were interested in studying the
use of "passive repeaters" for radio communications, and scientists at the Langley
Aeronautical Laboratory in Harnpton, Viiginia, who were interested in air density meas-
urements in Earth's upper atmosphere, chis experiment, named Project Echo, was endorsed
several years earlier by the United States IGY Committee. Scientists had agreed on a design
for an experiment to satisG both scientific objectives. A radio signd would be transmit-
ted from the east coast of the United States, reflected off the meralized sdace of a large
(100-foot diameter) Earth-orbiting balloon, and received on the west coast. A similar link,
using a much higher frequency, would be set up in the west-to-east direcrion. The plan
called for setting up separate links at 960 MHz (east to west) and 2,370 MHz (west to
east). The essential features of the experiment are shown in Figure 1-4.

In the f d of 1758, while considering how the Echo balloon experiment might be imple-
mented, it became apparent to engineers at JPL that the new antenna, then under
construction at Goldstone, could serve as the west coast receiving terminal for the 960-
MHz, east-to-west leg of the experiment. For its part, BTL would provide a 960-MHz
transmitting station and a 2,390-MHz receiving station at its east coast facilities in New
Jersey. However, one major problem remained unsolved. There was no transmitter for
Figure 1-5. Az-El transmitting antenna for the Echo Balloon Experiment, Goldstone, 1960.
The antenna was constructed between July and December 1959, and the complete station became
operational in July 1960.

the 2,390-MHz for the west-to-east leg of rhe experiment. The new Goldstone anten-
na was designed as a receiving station only. It had no transmit capabiliry.

To solve that problem, a second 85-foot antenna was built at Goldstone in 1959 using
funds provided by the NASA Communiations Satellite program. To avoid mutual radio
interference between the two antennas, a sire some miles away but similar to the Pioneer
site was selected. A low range of hills separated the two sites and provided the neces-
sary level of radio isolation between them. JPL named the site "Echo" for its obvious
association with the Echo experiment.

An early photograph of the transmitting antenna at the Goldstone Echo site is shown
in Figure 1-5.

Unlike rhe Pioneer antenna, which used a polar or hour angle-declination (HA-Dec)
drive system, the Echo antenna used a high-speed, azimuth-elevation (Az-El) drive which
was more suitable for tracking Earth satellites, such as Echo, than the more astronomy-
oriented polar drive used for deep space probes. The Echo station also included a 10-
kilowatt transmitter operating at 2,390 MHz, which at the time represented a very
advanced, state-of-the-art development. In addition to supporting the Echo balloon exper-
iment, the advanced features of the M o station allowed JPL engineers to evaluate the
relative merits of large antenna drive systems and the performance of high-power trans-
mitters and low-noise receivers, operating at S-band, for h m e applications in deep space
tracking stations.

Less than a month after the 111 station was completed, on 12 August 1960, Echo I was
launched. Within two hours of launch, a recording of President Eisenhower's voice was
transmitted from the Echo site at Goldstone to BTL in New Jersey. Echo had scored a
"first" in long distance passive reflector communications. Ironically, this distinction was
almost immediately eclipsed by the informal Goldstone to Woomera "moon-bounce"
experiment described later in this chapter. Experiments in long distance communiarions
and upper atmospheric air density soon followed.

At the conclusion of the Echo experiment, ownership of the Echo antenna was trans-
ferred to the new NASA Office of Tracking and Data Acquisition and became an official
part of the Deep Space Instrumentation Facility. Its ultimate destiny, however, lay in a
different direction from that of Pioneer, the other DSIF antenna at Goldstone. The first
step along this path began with the Venus Radar Experiment in March 1961.

The possibility of detecting an Earth-based radio signal reflected from the surface of
Venus had intrigued radio astronomers through most of the 1950s. In fact, investiga-
tors at Lincoln Laboratory, Massachusetts Institute of Technology, and at Jodrell Bank
Experimental Station, University of Manchester, had attempted to detect echoes from
Venus in 1958 and 1959 without success.

A successful experiment of this kind could offer great potential for a plethora of new
scientific knowledge about Venus and its physical properties, and the possibility, for
example, to determine the reflective properties of the surface for electromagnetic radia-
tion, the orientation of the planet's axis of rotation, and its rate of rotation. Also, much
could be learned about the characteristics of the interplanerary medium along the trans-
mission path between Earth and Venus. Most importantly though, it could offer the
possibility to independently determine a value for the astronomical unit (AU) with much
greater accuracy than had been previously possible.
The astronomical unit (AU, used by astronomers as a measure of the mean distance of
Earth from the Sun, is a vita parameter in the calculation of the ephemerides, or the
paths of planets around the solar system. It was therefore of immense significance to
]Plus long-range plans for sending spacecraft on missions to close encounters with dis-
tant planets. JPL engineers viewed the Venus radar experiment as a technological challenge
for the type of equipment required for future missions to the planets and for personnel
to operate it. Without the very highest level of performance in both areas, the hope for
success would be minimal, and would establish a standard against which the perform-
ance of future operational DSIF stations could be judged.

During the experiment, a 13-kilowatt, continuous wave, S-band signal was transmitted from
the new Echo antenna toward the planet, while the reflected s g r d was received by the
Pioneer antenna fitted with a Maser receiver, specially designed for reception of very weak
signals. The first undisputable radar returns from the planet Venus were obtained with the
Goldstone radar on 10 March 1961. Using advanced, s p e m analysis techniques, the radar
team, led by Richard Goldstein, measured the delay to, and changes in, the h d a m e n d
character of the returned signal due to its round-trip to Venus and back to Earth. These data
were recorded for subsequent scientific analysis. During the t h e that Venus remained with-
in 50 and 75 million miles &om Earth, over 200 hours of good radar data were obtained.

M e r analysis, this impressive bank of scientific data yielded the much sought after "improved
value for the AU" in addition to satisfying the several other scientific objectives of the
experiment. The new value for the AU was determined to be 149,598,500 500 km,a *
most significant improvement in the accuracy of previous values.6 Subsequently, JPL esti-
mated that Mariner 2, the first mission to Venus the following year, would have missed
the planet by too great a distance to have been of any scientific value, had the old value
of the AU been used in the trajectory computations.

The Venus radar experiment thus represented the first use of the DSN (DSIF) as a sci-
entific instrument. Its immediate success marked the appearance of a significant
astronomical research tool at Goldstone, whose stature in the scientific community would
continue to rise as the years went by. The technical details of the Venus radar experi-
ment, and the history of what evolved into the Goldstone Solar System Radar, is described
in chapter 8, "The Deep Space Network as a Scientific Instrument."

By the time the Venus radar experiments ended in May 1961, it had become apparent
that JPL communications engineers needed a dedicated antenna research and develop-
ment facility at Goldstone, where new components used in the network could be test-
ed, measured, and evaluated. With the pressing needs of future missions in mind, and
the Ranger and Mariner missions about to start, it was simply not possible to devote
one of the two existing Goldstone antennas (Pioneer or Echo) solely to research work,
despite the vital importance to the technological future of the network. Predictions of
hture mission requirements mandated a need for two operational antennas, not only at
Goldstone, but also at the other two longitudes.

Furthermore, until a way could be found to simultaneously transmit and receive on a sin-
gle antenna without severely d+g the downlink signals, two antennas, one for m m i t t i n g
and the other for receiving, would be required for deep space communications.

O n this basis, the case was made for yet another antenna at Goldstone. The new anten-
na, which would be of the same type as the Pioneer and two overseas antennas, would
be erected at the Echo site where it could make use of the existing support facilities.
When that was completed, the original Echo Az-El antenna would be moved to anoth-
er site a few miles away, and established as the basis for a new facility dedicated to deep
space communications research and development (R&D).

The plan was rapidly approved by NASA and immediately put into effect at Goldstone.
Between February and May 1992, a new HA-Dec antenna was erected near the &-El
antenna at the Echo site. It was put into service two months later, just in time to sup-
port the first Mariner missions to Venus.

With a second operational deep space station now in place at Goldstone, the Az-El anten-
na could be released fiom network support and prepared for the complex task of relocation.7

In June 1962, the entire Az-El antenna-mounted on its pedestal-was jacked up, loaded
on a supporting undercarriage, and moved approximately four miles across the desert to
a new location that became known as the Venus site, after the now famous Venus radar
experiment. There, the antenna began a brilliant new life as a test facility for commu-
nications research and development, and over the next 40 years, proved vital to the
continued preservation of JPL pre-eminence in the field of deep space communications.

The research work that was conducted at that facility, known through the DSN
R&D Station 13, is described in chapter 7, "The Advance of Technolog in the
Deep Space Network."
WASHINGTON, DC

Before proceeding further with the history of the DSN, it will be instructive to review
the events occurring simultaneously in a related but quite different arena from that just
described. The focus of that activity was in Washington, DC.

Before NASA was created, ARPA had pIanned to develop a worldwide network of track-
ing stations to support its forthcoming programs for the exploration of deep space. The
network was to be called Tracking and Communications Extraterrestrial Network (TRACE).
Starting with the first three TRACE stations-Goldstone, Puerto Rico, and another small
station at Cape Canaveral-ARPA planned to expand TRACE into a full-scale world-
wide network by adding two additional overseas stations, similar to the one it had funded
for Goldstone. This expansion was the rationale for the AWA decision to purchase three
of the 85-foot-diameter antennas from Blaw-Knox in 1957.

Plans for the worldwide network were well under way in early 1958. ARPA had already
approved a JPL proposal for a three-station network which would provide optimum cov-
erage for tracking deep space probes. One of the sites would be sites at Goldstone,
California. The two overseas stations were to be located at Luzon, the Philippines, and
in Nigeria.8

Taking a broader view, however, the Department of Defense expressed some reservations
about the proposed sites in terms of their utility to all U.S. space vehicles, rather than
the specific, deep space probes considered by JPL. With this in mind, ARPA asked JPL
to reconsider its proposal for the station sites with the object of improving coverage for
a broader range of space vehicle trajectories. The resulting study showed that better
orbital coverage for all planned U.S. space missions would be possible if the Nigerian
site was moved north to southern Portugal or Spain, and the Philippines site moved
south to central Australia. The coverage of orbits with an inclination of 34 degrees to
51 degrees would be much improved. Future, piloted, Earth-orbiting flights were very
much in the minds of the planners at the time.9 In retrospect, their foresight was well
justified since these locations effectively served the needs of the major tracking stations
of the Deep Space Network for over forty years.

In the turmoil of space-related activity in that tumultuous year, 1958, ARPA never did
get the opportunity to deploy the overseas stations of its World Net. NASA was creat-
ed in October of that year, JPL was transferred to NASA in December, Pioneer 3 was
launched, and by the opening of the new year, 1959, it was NASA and not ARPA set-
ting the course for JPL and the future shape of the Deep Space Network. It was NASA
that ultimately acquired and erected the two remaining antennas procured by AHA.
To the nation's space program, and to the work at JPL, the formation of NASA was
transparent. It was mainly a transfer of authority and funding for the existing programs.
Instead of working under an Army Contract, JPL now worked under a NASA contract.
NASA inherited J P f i experienced personnel its facilities, including the new antenna at
Goldstone and, perhaps most important of all, the JPL vision of a worldwide network
of tracking stations for deep space probes. ARPA, representing military interests in space
applications, pressed forward with its own programs under the direction of DOD.

As soon as NASA began to make its long-range plans for a civilian space program known,
the need for completion of what ARPA had called the World Net immediately became
obvious. One station already existed at Goldstone and sites had been proposed for two
overseas stations. The antenna at Goldstone now belonged to NASA, but A H A still
owned the remaining two of the three originally ordered from Blaw-Knox. In January
1959, NASA and ARPA representatives decided each agency would get one antenna.
The NASA antenna would be shipped to Australia as part of a three-station NASA oper-
ated network with sites in Goldstone (existing), South Africa, and Australia. ARPA wodd
use its antenna as part of a deep space network for military purposes with stations in
Japan and Spain. However, the military interest in deep space soon faded and NASA
was eventually able to purchase the third antenna from ARPA to form the basis for its
station in South Africa.

In moving toward the ultimate creation of the worldwide network, NASA directed its
principal efforts in 1959 toward onsite surveys of the locations that had been proposed
for Australia and South Africa and the diplomatic negotiations that were necessary to
secure approval for their use.10 In both countries, NASA was able to secure the interest
of the government agencies-Australian Department of Supply (D0S)Meapons Research
Establishment (ETE) and che South Mican Council for Scientific and Industrial Research
(CS1R)INational Institute of Telecommunications Research (NITR), respectively-in coop-
erating in the operation of the proposed deep space stations. Both WRE and NITR
assisted the NASAIJPL survey team in identifymg sites for the stations which met the
three major criteria; a stable land area capable of supporting large antenna structures,
surrounding hilly terrain to provide natural shielding against electrical interference and
a surrounding area relatively free of radio interference in the frequency region useful for
space communications.
WOOMERA, AUSTRALIA

The favored Australian site lay on the southern edge of the great inland desert region
at a place called Woomera, about 350 kilometers north of the city of Adelaide, capital
of the state of South Australia. Approximately 110 degrees west of the longitude of
Goldstone, Woomera was already a missile and long-range rocket test center operated
by WRE. The local language was English, and the nearby rocket and missile test activ-
ities would provide a pool of technical expertise and facilities. Furthermore, in the same
area, the Australians had already installed, and were operating, a U.S. Navy minitrack
station and a Smithsonian Baker-Nunn tracking camera as part of its participation in
the IGY.

The signing of a "construction and operationyycontract for the antenna around early
April 1960 allowed both JPL and WRE to begin making major moves toward con-
struction of a NASA deep space tracking station at Woomera. WRE initiated the road,
buildings, power generation, and foundation work. JPL began shipping antenna com-
ponents and the electronics for the station.

The antenna was finally built at a site known locally as "Island Lagoon," so named for
the nearby dry lake which appeared to have an island at the center. Working under a
JPL contract with supervisor Floyd W Stoller, Blaw-Knox began assembling and erect-
ing the antenna in May 1960. By August, the antenna was complete and an electronics
team began installing the radio and tracking equipment, most of which had been sup-
plied by Collins Radio Company. 'When NASA built the second and third 26-m antennas
in Woomera, Australia, and Johannesburg, South Africa, the task of integrating the new
antennas with their electronics equipment and bringing the two new stations into oper-
ation fell to Richard "Dick" Mallis.

Richard Mallis was an outgoing inciividual, easy to work with, sociable, and much respect-
ed by his colleagues. He was an excellent manager with good communications and
technical skills and an appreciation for the different institutional environments at all
three antenna locations. When he went to JPL in 1955 to work on radio guidance sys-
tems for the Army's Sergeant missile program, native Californian Richard K. Mallis took
with him a degree in mechanical engineering from the University of Southern California
and a Navy background. Caught up in the changes that swept JPL into the space pro-
gram in 1958, he assisted with the construction of the first 26-m antenna at Goldstone,
and later implemented the down-range tracking station in Puerto Rico to cover the
launches of the Army's two Pioneer lunar probes.
Together with Goldstone, the three stations of the Deep Space Instrumentation Facility
finally formed a worldwide network. With these completed in time to support JW first
Ranger lunar missions, Mallis returned to JPL to take up a staff position in Renzerti's
new Communications Engineering and Operations Section. He was responsible for
Operations, regulating the way the Network carried out its day-to-day tracking func-
tions. In this role, he set-up a Network-wide logistics and repair program, a frequency
and timing standards program, a documentation system, and a training program for
operations and maintenance personnel. This essential idrastructure remained the basis
for all operations, maintenance, quality control, and configuration management process-
es as the worldwide Network expanded in size and capability through the years. He
integrated the first commercial contractor, Ben& Field Engineering Corporation, into
the DSIF as the operations and maintenance service provider for the Goldstone facili-
ty. In later years, as his responsibilities expanded to include the Space Flight Operations
Facility at JPL in addition to the DSN, he became Manager of the Operations Division.
Eventually he transferred elsewhere in JPL to further his professional career. However,
because of his unique experience with service contract management, he was frequently
called upon to assist the DSN in evaluating new contract proposals when existing serv-
ice contracts expired. He retired in 1993 after 37 years of service at JPL and later took
up residence in Australia.

In a find spectacular exercise on 3 November 1960, the Woomera station demonstrat-

ed its operational status by receiving voice and teletype messages transmitted fiom Goldstone
via reflection from the Moon. The JPL onsite manager, Richard K. Mdlis, departed
Woomera four days later, after turning the new facility over to WRE for its future man-
agement and operation. The Australian engineers soon demonstrated their ability to
handle the technical complexities of the new "space age" facility for which they had
accepted operationd responsibility. In a repeat of the "Moon bounce" experiment on 10
February 1961, during the official opening formalities, the station passed a congratula-
t o v message fiom NASA Deputy Administrator Hugh Dryden in Washington to Australian
Minister of Supply Alan Hume at Woomera over a "Moon bounce" communications
link.

The photograph of the completed Woomera Tracking Station in Figure 1-6 shows the
26-meter-diameter antenna, the electronics equipment building, and service and facility
structures. Island Lagoon is visible at the horizon to the left of the antenna structure.

The first overseas station of the Deep Space Instrumentation Facility (DSIF) was ready
to enter operational service. Designated DSIF 41, Woomera, it would see eleven years
of valuable service before being superseded by new stations at Tidbinbilla, near Canberra,
Figure 1-6. The 26-meter antenna and tracking station (DSIF 41), Woornera, Australia, 1961.

in southeast Australia. The Woornera station ceased operations on 22 December 1972,

as part of a NASA program to consolidate overseas station facilities. Initial proposals to
move the antenna to a new, more accessible location where it could be used for Australian
radio astronomy purposes were not successful because the Department of Supply deter-
mined that the cost of transporting the antenna was excessive. E v e n d y , it was dismantled
and sold for scrap despite the vigorous protests of several prominent members of the
Australian scientific community.
JOHANNESBURG, SOUTH AFRICA

The requirements for continuous tracking coverage of deep space probes outlined in the
JPL proposals for a worldwide network dictated that the third element in the network
should be located in the longitude band which included Spain and South Afiica. Although
a site in Spain was preferred by NASAJJPL, the political complexities associated with
existing international treaties with Spain caused NASAIJPL to look to South Africa for
a suitable site. South Africa also offered many advantages. The South African govern-
ment was anxious to participate with NASA in the new space venture, the language was
familiar, and a great deai of technical expertise was already available in that country.
Also, and most importantly, the flight path of all deep space probes launched from Cape
Canaveral would pass over or near the station within an hour or so of launch. This
meant that the downlink from the space robe could be first "acquired close to the
point at which the spacecrafi would be injected into its planetary trajectory. During this
period, mission controllers depended on telemetry and navigation data from the first
acquiring tracking station to make critical decisions very early in the mission, when cor-
rective action is necessary. Supplemented with the mobile tracking station that would
be moved from Puerto Rico, an 85-foot antenna at this site would provide the DSIF
with an excellent "initial acquisition" capability in addition to the normal tracking capa-
bility of the station.

The site that was chosen lay near the Hartebeestpoort Darn about 40 miles north of
Johannesburg, on government-owned property intended for use as a radio research facil-
ity. As had been the case at Woomera, a NASA Minitrack station and a Baker-Nunn
camera were already operating in the vicinity. An earlier site survey confirmed that the
Hartebeestpoort site also satisfied the antenna site-selection criteria.

Although site surveys conducted in mid-1958 had confirmed the suitabiIity of the site
for the antenna, it was not until September 1960 that an agreement between the gov-
ernments of the United States and the Union of South Africa permitted NASA to issue
a contract to the South African Council for Scientific and Industrial Research for the
construction, management, and operation of the station. NASA wanted the station to
be completed by July 1961 in time to support the first launch of new Ranger missions
to the Moon. Time was truly "of the essence." To prepare the site, erect the antenna,
install and test the equipment, and train personnel in its operation, under the pressure
of a high profile mission like Ranger, posed a formidable challenge for everyone involved.

Work began in earnest in mid-January 1961, as soon as the foundation for the anten-
na was ready. Antenna structural components travelling by ship, train, and truck from
Figure 1-7. The 26-meter antenna and tracking station (DSIF 51), Hartebeestpoort, South
Africa, 1961.

Philadelphia had arrived a few days earlier. The erection team arrived from JPL a f m
days later. Erection of the antenna finally began on 16 January and was completed 69
days later on 25 March. The JPL team-Don Meyer, Dick McKee, and Howard Olsen-
completed installation of the electronics in early June and spent the next few weeks
conducting the necessary tests, checkouts, and calibrations. The station was ready for
operation by 1 July 1961. Provided with teletype equipment and communication cir-
cuits by the South African Department of the Postmaster General, the new station was
able to participate in the first operational readiness test for the Ranger missions later in
July, as originally planned.

A photograph of the Johannesburg Tracking Station soon after its completion in 1961
is shown in Figure 1-7.
JPL and South African officials discussed the possibility of holding a public event to
mark the formd opening of the station as had been done for the Woomera station.
There was, however, some reluctance on the part of NASA Headquarters to publicly
recognize a cooperative project with South Afiica in light of the United States' attitude
toward the political situation then emerging in that country. Eventually, the South African
government held its own opening ceremony without formal NASA representation. Some
years later, this issue would intensify and ultimately lead to the closure of the station as
a part of the Deep Space Network.

Nevertheless, the new station, designated DSIF 51, Johannesburg, played a vital role in
the early NASA missions to the Moon and planets. In that role, it served the DSN with
distinction for 13 years until DSN operations at that site were terminated in June 1974.
Soon after, NASA transferred the antenna and equipment to the South Afirican National
Institute for Telecommunications Research (NITR) where it was used successfdIy for at
least another 25 years, as an instrument for radio astronomy research."

The completion of the Johannesburg station marked the fulfillment of the world net-
work initiatives proposed years earlier by JPL, carried forward under M A , and brought
to fruition under NASA. Together with a small launch monitoring facility at Cape
Canaveral, the three 85-foot antennas at Goldstone, Woomera, and Johannesburg became
known as the Deep Space Instrumentation FaciIity (DSIF). The DSIF had become a
separate facility of the NASA OfGce of Space Operations. It was managed, technically
directed, and operated by the Jet Propulsion Laboratory (JPL) for the California Institute
of Technology (Caltech), a prime contractor for NASAls Solar System Exploration pro-
gram, in Pasadena, California. Within a few years, the DSIF would change its name to
the Deep Space Network (DSN) and rapidly increase in size, complexity, and capabili-
ty to a level unimaginable to its founders. But that lay in the distant future. In the
immediate hture lay the first operational challenge for the new network-the Ranger
missions to the Moon and the first Mariner missions to Venus. It was to be a busy
hture. The Mariner Era was about to begin.
Endnotes

1. N. A. Renzetti et al., "A History of the Deep Space Network from Inception to
1 January 1969." JPLTechnical Report TR (September 1971): 32-1533.

2. William R. Corliss, '24History of the Deep Space Network" (Washington, DC:

National Aeronautics and Space Administration CR- 151915, 1976).

3. R. Cargill Hall, Lunar Impact: A History of Project Ranger (Washington, DC:

National Aeronautics and Space Administration, 1977).

4. Homer E. Newell, Beyond the Atmosphere; Early Ears of Space Science (Washington,
DC: National Aeronautics and Space Administration SP-42 11, 1980).

5. E. Rechrin and R. Jaffe, "The Design and Performance of Phase-Lock Circuits,"

IRE Trdnsactions on Information Tbeo (March 1 95 5).

6. Andrew J. Butrica, To See the Unseen, A History of PZanetay Rdddr Astronomy

(Washington, DC: National Aeronautics and Space Administration SP-42 18,
1996): 42.

7. Jet Propulsion Laboratory "Az-El Move and ReinstdIation," JPL Space Propms
Summary 37-13 (Vol. 111, 1 October 1962): 9-13.

8. Jet Propulsion Laboratory Description of World Networkfor Rddio Trdcking of Space

Kbicles (Pasadena, CA: JPL Publication No. 135, 1 July 1958).

9. E. Rechtin, H. L. Richter, fr., and W. K. Victor, "National Ground-Based Surveillance

Complex." JPL Technical Memorandum TM 39-9 (15 December 1958).

10. Support for NASA facilities in Australia, South Africa, and Spain was provided
under cooperative international agreements between the United States and those
sovereign countries. The diplomatic, political, and administrative complexities of
the negotiations that culminated in those agreements is beyond the scope of this
history of the Deep Space Network. Readers interested in pursuing those details will
find useful material in the JPL and NASA Archives.
11. A more detailed account of many of the topics discussed in this chapter is contained
in the unfinished and unpublished notes on the early (prior to 1962) history of the
Deep Space Network compiled by Craig W& at JPL in 1993. The Waf. notes are
heId in the JPL Archives.
THE MARINER ERA MISSION SET

The worldwide network, envisioned by Rechtin in 1958, became a reality between

July 1959 and July 1961. In that hectic two-year period, funding was proposed and
approved, international agreements were negotiated, suitable sites were found, equip-
ment was procured and shipped overseas, roads and facilities were built, and JPL
contractors erected antennas and installed equipment. The Deep Space Instrumentation
Facility now consisted of a real network of 26-rnerer-diameter antennas located at
intervals of approximately 120 degrees of longitude around the globe. Two such anren-
nas were located at Goldstone, California, and one each at Johannesburg, South Africa,
and Woomera, Australia. A world view of the DSIF as it existed at that time is shown
in the figure below.

Figure 2-1. World view of the Deep Space Instrumentation Facility, 1961. Although the loca-
tions of future sites, near Canberra, Australia, and near Madrid, Spain, are also shown on this
diagram, they had not been established at that time.
Teletype and telephone circuits linked the stations to a rudimentary flight operations
control center at JPL. In the 1961 photograph of the operations control center at
JPL shown in Figure 2-2, the presence of rotary telephones, mechanical calculating
machines, wall boards, and mechanical status displays is a poignant reminder of the
pre-digital age.

At JPL and at each of the uackng stations, highly motivated crews of operators and
engineers were trained to operate and maintain the complex, state-of-the-art equipment
that JPL had designed and tested, but not yet demonstrated, to communicate with dis-
tant spacecraft.

At the same time, DSIF planners knew that in the years ahead, the network would be
called upon to support much more complex missions to the Moon, missions to fly by
and possibly even orbit Venus and Mars, and heliocentric probes to measure fields and
particles and the solar wind in interplanetary space. At the extreme limit of possibility
lay spacecraft encounters with Jupiter and Saturn.

Figure 2-2.Flight operations control center, JPL, 1961.

To appreciate the magnitude of the challenge facing the embryonic DSIF at that time,
it is instructive to arrange this inventory of future missions into a mission set in the
following way:

Deea Soace Mission Set for the Mariner Era

As this table makes clear, all launches did not necessarily result in successful missions.
Failures were a common feature of the early programs but the failure rate greatly improved
as time went on. The Ranger program is an example. The launch span, the time between
the first and last launches of individual programs, is an indicator of the program's impact
on the DSIF. Many launches in a short span affected the level of tracking activity for
the DSIF but did not require significant new technology between launches. Lunar Orbiter
and Surveyor were such programs. The reverse was true for programs with a long launch
span like Pioneer and Mariner.

In terms of influence on the growth and technical development of the nenvork, the
Mariner program exceeded all of the other programs in both complexity and launch
span. It did not produce the longest duration missions, however. The Pioneer program
would still have viable missions in flight long after the last Mariner mission ended in
1974. Nevertheless, the Mariner program drove the early development of new capabil-
ity for the DSIF to a greater extent than any other program and, at a time before the
public and scientific palate had become jaded with a surfeit of "solar system science,"
it produced many of NASA's most spectacular scientific results. For that reason, the
period of DSIF history that began with the disastrous launch of the first Ranger in 1961
and ended with the hugely successful Mariner 10 spacecraft in orbit around Mercury
and Venus in 1974, is identified here as the "Mariner Era."

Of the five projects included in the Mariner Era mission set, three, the Ranger, Mariner,
and Surveyor projects, were l l l y managed by JPL. The Ames Research Center in Mountain
View, California, directed the Pioneer Program, while Lunar Orbiter was directed by the
Langley Research Center in Hampton, Virginia.

DSN participation in the Apollo program was directed by the Goddard Space Flight
Center, Greenbelt, Maryland. Because these later projects were not controlled by JPL,
the different management styles and demands for new standards of performance that
they brought to the DSIF led to fa-reaching changes in the DSIF arrangements for
supporting non-JPL or "outside" flight projects. It was the sum total of these ofeen
conflicting influences that shaped the character and capability of DSIF during the
Mariner Era.

In a real sense, the DSIF began its effective operational life as a complete operational
network with the Ranger program, an ambitious JPL mission to land a spacecraft on
the surface of the Moon, in August 1961.

While one stream of activity in JPL had, over the past several years, been focused on
the problems associated with approving, designing, and implementing the DSIF, the
principal effort had been devoted to developing the first spacecraft specifically designed
to reach the surface of the Moon. Designated the Ranger Program, this project had
been assigned to JPL by NASA in December 1959 and was to be completed within
36 months.

The Ranger spacecraft were much more complex than the earlier Pioneer-type probes.
In particular, the Rangers were to be attitude stabilized to keep their high-gain radio
antennas pointed toward Earth and the three antennas of the DSIF. This also allowed
the solar panels, which provided power for the spacecraft electronics, to remain point-
ed at the Sun during flight.
The bartery powered Pioneers had been simply spun like an arrillery shelt to maintain
their spatial orientation during flight. In addition to attitude stabilization and more
advanced radio communication systems, some of the spacecrafi would carry an ejectable
capsule. This capsule would descend to the lunar surface on a small retrorocket and
thereafier right itself to transmit data on local radio activity and moonquakes. Later
models were to carry high definition video systems to return detailed pictures of the
lunar surface prior to impact.

The original Ranger program comprised five spacecrafi, the first two of which were to
be test flights ejected into an elliptical orbit around Earth. These would be essentially
engineering flights to check out the new capabilities of attitude stabilization and solar
power for the spacecraft and, for the DSIF, the all important telecommunication design.
The remaining three spacecraft would consist of a basic spacecraft "bus" with an ejectable
capsule containing a science instrument package and the retrorocket to set it on the
lunar surface. During the flight to rhe Moon, these spacecrafi were to make fields and
particle science measurements and, during the final moments of approach before impact,
take close-up vidicon pictures of the lunar surface.

In retrospect, this appeared to be an ambitious plan indeed for a fledgling technology,

as later events proved. However, at the time, it was perceived as the appropriate major
step forward that was needed to reestablish the prestige and forefront position of American
scientific endeavor, so adversely affected by the Russian success with Sputnik and Lunik
some years previously.

Although all five of the first Ranger flights were unsuccessfd due to launch vehicle and
spacecrafi system failures, the excellent performance of the DSIF proved that the world-
wide network concept of tracking stations managed and controlled from a central location
was not only feasible but would be an essential element of hture space missions to other
planets and the Moon.

The Ranger flights demonstrated the value of continuous communication between

the spacecraft and the network stations. This provided the Mission Operation teams
back in the Flight Operations Control Center with immediate information about
the status and condition of spacecraft and operating systems during the entire flight.
This included engineering and scientific data. Ironically, this capability allowed the
spacecrafi controllers to estimate the reason for all of the Ranger failures attributed
to the spacecraft.
Following a soul-searching evaluation of the causes of failure of the first five Ranger
missions, and some consequent reorganization of technical responsibilities at JPL,
the program resumed in January 1964. The Ranger team redefined its objectives to
the sole task of securing close-up television pictures of the lunar surface, ostensibly
to aid in the design of two more sophisticated lunar missions then under study by
NASA, namely the unpiloted Surveyor landers and the piloted Apollo program.
Although the instrumented science capsule had been removed, the mission remained
very ambitious for those days, as it included a launch into Earth-parking orbit fol-
lowed by a second burn of the launch vehicle to transfer the spacecraft into a lunar
trajectory. Attitude control maneuvers then followed to orient the solar panels to
the Sun and the spacecraft antenna toward Earth. Further midcourse and terminal
maneuvers were required to be correctly executed before the Ranger vidicon driven
cameras could view the lunar surface for the brief period before final impact. During
that time, the precious video data would be transmitted to the DSN stations where
it would be processed and delivered to the Ranger team in the Operations Control
Center at JPL.

With the reputation of JPL on the line, the first of the final group of four Rangers was
launched in January 1964. Despite the political pressure to demonstrate success after the
poor results from the first group of five launches, Ranger 6 was also a failure. It per-
formed perfectly until the last 10 minutes before lunar impact, at which point the
spacecraft television system should have turned on to start transmitting pictures to the
Goldstone station. There was no indication that this happened, and no pictures were
sent. The cause of this failure was attributed to high voltage "corona discharge" in the
camera's electrical insulation and appropriate changes were made to the TV subsystem
on the remaining three spacecraft.

Six months later, the fortunes of the Ranger program changed dramatically. Ranger 7,
launched in July 1964, was a resounding success and radioed 4,300 high-resolution pic-
tures of the lunar surface back to Earth. Immediately after lunar impact terminated the
flighr, President Johnson called JPL and told Director William Pickering that Ranger 7
was "a magnificent achievement." O n behalf of the whole country he congratulated
NASA, JPL, and NASA contractors, saying, "This is a basic step forward in our order-
ly program to assemble the scientific knowledge necessary for man's trip to the Moon."
News conferences and presentations at the White House followed as soon as the films
arrived at JPL from Goldstone and could be processed. The mood at NASA was ela-
tion, the Ranger team at JPL was ecstatic.
The final flights of the Ranger program in February and March 1965 added to the suc-
cess of Ranger 7. Thousands more pictures of the lunar surface were returned by missions
8 and 9, and the reputation of JPL as a leader in the field of space exploration was
established to worldwide acclaim.

No less delighted than those who were directIy involved in the mission was the lunar
science community, which, for the first time, had over 17,000 close-up pictures of the
lunar surface for study and analysis. In addition, engineers and scientists designing the
Surveyor and Apollo missions, soon to follow Ranger to the Moon, had a greatly improved
basis for their designs.

In his History of Ranger project, Cargill Hall wrote, "No longer a liability, the Ranger
program had vindicated American space policies and presaged accomplishments yet to
come." It had indeed, and they would not be long in coming.

Important as it was, Ranger was not the only program demanding the attention of the
DSIF in 1962. By midyear, the first Mariner spacecraft, designed by JPL for planetary
missions as distinct from lunar missions, were on their way to the launch pads at Cape
Canaveral, Florida (later renamed Cape Kennedy).

Unlike lunar missions, planetary missions have only a very limited opportunity for launch,
or "launch window," which occurs every few years. For missions to Venus, 1962 was
one such year, and the next opportunity for launch would not occur until 1964. NASA
and JPL could not wait that long for a first planetary mission. The attempt to send a
spacecraf-t to Venus would have to be made in 1962.

Since the time between the NASA go-ahead and the opening of the Venus launch win-
dow was less than a year, much of the Ranger technology, including the radio subsystem,
was utilized for the Mariner spacecraft. For the DSIF, this meant a continuation of the
L-band uplink and downlink support and the possibility of competition with Ranger
for the attention of DSIF tracking stations when both Ranger and Mariner spacecrah
would be in view at the same time.

This situation, which first appeared in the RangerIMariner Era, was to become of immense
importance in the years ahead as the number of flight missions increased. Conflicts for
use of the DSN antennas when view periods of different spacecraf-t overlapped were a
frequent occurrence. In the onrush of preparing the Mariners for launch the problem
was deferred.

In addition to the existing 26-meter stations at Goldstone, Woomera, and Johannesburg,

the DSN had, at the time of the first Mariner launches, grown to include a Launch
Station at Cape Canaveral identified as DSIF 0, and a Mobile Tracking Station (DSIF
I), which was located near the Johannesburg station in South Africa.

It was the purpose of DSIF 0 to track the spacecrafi as it lifted off the launch pad atop
the launch vehicle, by manually pointing a small two-meter diameter dish antenna at
the rapidly disappearing launch vehicle/spacecraft combination. This was not a job for
the fainthearted, since the distance between the tracking antenna and the launch pad
was necessarily short to ensure that the spacecraft signal was received continuously for
as long as possible after liftoff. This period, seldom more than a minute or two, pro-
vided vital engineering information for the DSN regarding the status of the downlink
as the spacecrafi started its long journey around the world to reappear over South Africa
at the next tracking station, DSIF 5.

There, the small 3-meter antenna of the Mobile Tracking Station rapidly acquired the
spacecraft and used its pointing information to direct the large 26-meter antenna to the
spacecrafi. Once acquired, the spacecraft downlink was tracked and the data it carried
was processed, recorded, and returned to JPL by airmail, a process which at h a t time
could, and usually did, take seven days.

Mariner 1962

The first two Mariner spacecraft were essentially long-range versions of the Ranger space-
craft. They carried instruments to make scientific measurements in interplanetary space
while moving along its trajectory between Earth and Venus. On arrival, they would
measure the radiation and magnetic fields while the microwave and infrared radiome-
ters analyzed Venus' atmosphere. The spacecraft used the same type of radio system
(L-band) as had been carried by the Ranger spacecrafi.

Two complicated, inflight operations were to be carried out after launch. The first of
these was the midcourse maneuver to correct the trajectory to aim it more precisely
at Venus. The second maneuver would point the science instruments to scan the Venus
surface as the spacecraft flew by. The commands to carry out the maneuvers would
be sent from the tracking stations. The science data would be transmitted by the space-
cr&s three-watt transmitter to the DSN antennas, by then 58 million kilometers from
the spacecraft. Data would "trickle" across those millions of miles of interplanetary
space at just over eight data bits per second. Apart from the tremendous scientific
value of the science data itself, this was regarded as a major technological achievement
at that time.

Mariner 1 was launched in July 1962, but was destroyed by the Range Sdety when it
ran off course shortly after liftoff. However, Mariner 2 injected into a Venus trajectory
by an Atlas-Agena launch vehicle the following month, performed exactly as intended.
Following a perfect liftof& DSIF 0 followed the spacecraft until it disappeared over the
horizon. The spacecraft appeared over South Afi-ica some 28 minutes later and was
acquired by DSIF 1 with no problem. Within three minutes, the big, 26-meter
Johannesburg antenna had found the spacecraft and began receiving engineering data to
verify the spacecraft had survived the stresses of the launch environment.

The midcourse maneuver was carried out successfully and, after some further commanding
to correct a temperature problem, reached Venus in mid-December. After passing Venus
at a range of 40,000 miles, the spacecraft continued in orbit around the Sun until it
fell silent on 2 January 1963. The DSN stations searched the empty airwaves for sev-
eral days trying to find some sign of life from the first planetary spacecraft but none
was found. In its brief lifetime, Mariner had transmitted 11 million data bits of science
and engineering data back to Earth.

As a consequence of experience with the first Ranger and Mariner missions, the DSIF
began work to improve its capability so that engineering and science data could be
relayed back to JPL as it was received fiom the spacecraft, that is to say in real time.
Automatic monitoring of the DSIF stations would be added to permit the engineering
experts at JPL who had designed the station equipment originally to aid the remote sta-
tion staff in troubleshooting and problem identification. Even greater emphasis was to
be placed on good engineering practices, conservative design, and thorough testing to
minimize equipment failures in the field. Techniques to acquire rhe spacecraft downlink
as early in flight as possible were studied. Small wide-beam antennas for "initial acqui-
sition" were added to the network antennas and the accuracy of flight path predictions
was improved.

The experience gained from these early flights also led to a new approach to DSN
telecommunications link design, in terms of receiver sensitivity and transmission power
of the ground traclung station. Combined with similar considerations for the spacecraft
radio system, the basic concepts of uplink and downlink performance became firmly
established as the methodology for all future deep space communications engineering
design.

Mariner 1964

A "launch window" for Mars only opens every 25 months, and such an opportunity
occurred in late November 1964. To take advantage of this situation, NASA chose a
Mars-1964 project based on JPLls experience with the earlier Venus missions. Three
spacecraft were to be built, two for flight, with one as backup that operated at the then-
new S-band frequency.

The Mariner 3 and Mariner 4 missions were to be supported with the existing DSIF
26-m stations at Goldstone, Johannesburg, and Woomera, plus a new station in Australia
which had been built at Tidbinbilla near Canberra. The stations at Woomera and
Johannesburg that retained the old L-band receiving equipment were fitted with con-
version equipment, which enabled them to receive the S-band signals from the Mariner
spacecrafi. In addition, the R&D station at the Goldstone Venus site, which by then
had acquired a high-power, 100-kW transmitter, was to provide backup for the 10-kW
transmitters at the other stations in the event spacecraft problems required a more pow-
erful uplink for commanding purposes.

These first Mars flights also coincided with very significant improvements in the capa-
bility and operational management structure of the Ground Communications Facility
(GCF) and Space Flight Operations Facility (SFOF) at JPL. The SFOF was a large data-
handling machine, albeit housed in a barely finished new building at JPL, where data
arriving fi-om the tracking stations was processed in real time and distributed to the
DSN and Flight Project Operations Control Groups for action and to Science Groups
for analysis. The simple control room of the Ranger and early Pioneer days with a few
desks, teletype machines, and display beards was "history."

Although the first of the Mariner 1964 missions failed due to a launch vehicle prob-
lem, the second, Mariner 4, was launched successfdly on 8 November 1964. A little
more than seven months later, it reached Mars. On 14 July 1965, in response to com-
mands from the Johannesburg tracking station, the Mdriner 4 spacecraft turned on its
cameras, tape recorders, and scientific instruments in preparation for the first close-up
look at Mars. As the encounter point approached, the Mariner 4 cameras focused on
Mars and took a 25-minute sequence of pictures of the Martian surface before rhe space-
craft trajectory took it behind the planet. The pictures were stored on the spacecraft
tape recorder. Playback to Earth began when the spacecraft came into view of the track-
ing stations, about an hour later.

It must be pointed out here that although the many millions of data bits in the pic-
ture images could be recorded almost instantly by the spacecraft, it would take many
days to return the recorded data back to Earth, due to the limited capability of the
downlink. The first pictures of Mars took ten days to playback. This would change
radically in the years ahead, but at the time of Mariner 4, the downlink capability
between Mars and Earth for the 26-rn antennas of the DSIF was limited ro only
8.33 bits per second.

Figure 2-3.First close-up picture of Mars. Returned from Mariner 4 on 14 July 1965. The pho-
tograph shows the Phelgra region and Mars horizon from a distance of approximately 10,500 miles.
After a second playback of images and science data, Mariner 4 continued along its orbit
around the Sun, sending back more science and engineering data und the downlink
was inadvertently "lost" about two and a half months later.

Scientifically, the mission was judged to be a great success, affording humankind the
first close-up views of the Martian surface. Crude as they appear now, those first images,
transmitted laboriously at about 8 bits per second over millions of miles of deep space,
were hailed with tremendous enthusiasm by the public at large. A composite image of
the Phelgra region of Mars, taken by the Mariner 4 spacecraft, was proudly displayed
in the Main Lobby of the Engineering Building at JPL, where for many years, it served
both as a reminder of the humble beginnings of the planetary program and as the point
from which to gauge the astonishing progress that would take place over the next few
years. The original picture is reproduced in Figure 2-3.

For some rime after the downlink signal was lost by the DSIF stations, the R&D Venus
station at Goldstone continued to search for it. Eventually the signal was recovered and
used to check out and calibrate the new 64-meter antenna at Goldstone. It was also
planned to use Mariner 4 in conjunction with Mariner 5 to perform radio science exper-
iments in interplanetary space from three vantage points: two spacecraft and Earth.
However, the spacecraft supply of attitude control gas was exhausted during engineer-
ing tests in this extended phase of the Mariner 4 mission. After three years of operation
in space, and 1,119 passes by the DSN tracking stations, the mission was terminated.

Mariner 1967

NASA chose to use the 1967 planetary launch window for a second mission to Venus.
The spacecraft for this mission would be the one remaining spacecraft from the three
that had been built for the 1964 Mars missions. Fitted with a different set of science
instruments, including a photometer, radiometer, and plasma probe, the spacecrafi would
further explore the dense, optically opaque Venusian atmosphere.

The tracking and data acquisition requirements for support of Mariner 5 were compli-
cated by the DSN desire to involve Mariner 4, which had since been recovered, in
simultaneous radio science experiments in interplanetary space. In addition, new teleme-
try and command processing equipment added to the Network, between the Mariner
4 and Mariner 5 launches, further compiicated the technical and operational interfaces
between the DSN stations and the flight project organization at the SFOF. The space-
crafi was more complex, too. It carried, in addition to the standard lunar ranging system,
a newly developed system of ranging for its first flight evaluation. The Tau system, as
it was called, was designed to extend the DSN ranging capability beyond the limit of
the lunar distance.

At Goldstone, Mariner 5 would be supported by the two 26-m stations (Echo and
Pioneer) and the new 64-m Mars station. In Australia, it: would be Woomera and the
new 26-m station at Tidbinbilla that would be used. Johannesburg and another new
26-m station at Robledo, near Madrid, Spain, would complete the DSN support for
this mission. At the SFOF, a new 7044 computer system would see its first operational
use on Mariner 5 and demonstrate the capability to process Mariner 4 and Mariner 5
data simultaneously, even though the data streams came from different DSN stations.
To handle so many data streams from the increasing number of tracking stations, a
newly installed communications processor in the SFOF used computer switching to
direct all communications throughout the DSN.

Launched on 14 June, the spacecraft was first placed in an Earth parking orbit before
being injected into a final trajectory to Venus. Before reaching the limit of its capabil-
ity, the DSN lunar ranging system followed the spacecraft to a distance of 10 million
km, making Mainer 5 the first spacecraft to be ranged beyond lunar distance. The new
Tau ranging system was activated about two weeks before success in tracking the space-
craft to the Venus encounter and beyond, to about 75 million krn. Together with Doppler
and angle data, ranging data became a standard type of radiometric data generated by
the DSN and greatly enhanced the orbit determination process for the navigation of
future spacecraft. With the downlink running at 8.33 bits per second and encounter
activities proceeding smoothly, Mariner 5 made its closest approach to Venus on 18
October 1967, at a range of 3,946 km.The science return satisfied the mission objec-
tives and the DSN continued to track the spacecraft until it passed out of radio range
at about 160 million krn.

Mariner 1969

In discussing the background to the Mariner 1969 missions Corliss wrote, "Earlier probes
to Venus and Mars had indicated that Mars was the most likely planet in the solar sys-
tem to support life. The richly detailed and cratered surface of the planet revealed by
Mariner 4 had surprised planetologists and made them anxious for more photos. Mars
thus became NASA's prime planetary target. The Mariner Mars 1969 mission, therefore,
concentrated on TV imaging of the planet's surface, and experiments that might aid the
design of future missions, particularly those looking for life. Besides the TV cameras,
the two new Mariner-type spacecraft assigned to the mission carried an ultraviolet spec-
trometer, an infrared spectrometer, and an infrared radiometer."
For the DSN, the Mariner 1969 missions were of special importance because they includ-
ed the first flight demonstrations of the new DSN high rate telemetry system. Previously,
with downlinks limited to just over 8 bits per second, playback of telemetry data from
spacecrafi at the range of Mars or Venus was measured in days or even weeks. With the
DSN's new block-coded telemetry system running at 16,200 bits per second, roughly
two thousand times faster than the former system, low resolution pictures could be sent
in real time from Mars. The new high rate telemetry did not replace the existing low
rate system. It was an addition to the DSN's rapidly increasing capability for planetary
exploration. Corliss continued, "The DSN was ready for the much greater distances and
higher rates required (for planetary exploration). The various technical advances and new
facilities just described had begun years before the new missions left their launch pads.
As usual though, the new DSN capabilities, which had seemed perhaps unnecessarily
ambitious when proposed, were quickly absorbed by the new mission designers."

Tracking and data requirements for the Mariner 1969 missions (Mariners 6 and 7),
included not only the existing 26-m stations, but the new 64-m antenna at Goldstone
as well. The downlink would carry telemetry data at either 8.33 or 33.33 bits per sec-
ond for normal spacecrafi operations during the cruise period. When the spacecraft
reached Mars, the new high rate telemetry experiment would be turned on. Of course
telemetry reception at 16,200 bits per second would be possible only when the space-
crafi was in view of Goldstone, where the big 64-m antenna was required to enhance
the downlink sufficiently to carry data at this rate.

Orbit determination requirements for the very precise navigation needed to fly the space-
craft close enough for a Mars encounter would be based on standard radiometric data,
including an improved version of the Tau planetary ranging system.

Both Mariners 6 and 7 were launched successfully in February and March 1969. For
the first time, NASA had successfdly launched a pair of spacecraft. Both spacecrafi were
on target and, from a DSN viewpoint, both were in the same part of the sky. This
meant that both spacecrafi would be in view of a given DSN site, although not with-
in a single antenna beamwidth, simultaneously. Tracking both spacecraft simultaneously
required two antennas and two telemetry data processors, one for each downlink. At the
same time, there were the Pioneer spacecraft and backup support for the first Apollo
missions to be considered as well. The "loading" on the DSN antennas had begun to
reach the point of saturation, and the resource allocation system was hard pressed to
service dl of its customers.
As Mars began to draw near towards the end of July, encounter operations began with
Mariner 7 only five days behind Mariner 6. Corliss describes what happened next.
"All seemed to be going well until about six hours before the Mariner 6. encounter,
when Johannesburg reported that the signal from Mariner 7 had disappeared. It was
an emergency that came at the worst possible time. The Robledo, Spain[,] antenna
discontinued its tracking of Pioneer 8 and began to search for the lost spacecraft. When
Mars came into view for Goldstone, the Pioneer 26-m antenna joined the search,
while the Echo 26-m antenna continued tracking Mariner 6.. It was decided to send
a command to Mariner 7 to switch from the highly directional high-gain antenna to
its omnidirectional low-gain antenna. The spacecraf? responded correctly, and suddenly
both the Pioneer station and the Tidbinbilla station began receiving low-rate teleme-
try from the recovered spacecrafi. Something had happened to the spacecrafi but no
one knew just what."

While the DSN was committed to support one Mariner at a time in a mission-critical
phase, this situation presented one spacecraft approaching encounter and a second one
with a serious and unknown problem. To deal with it, the DSN applied its main effort
to the ongoing Mariner 6 encounter, while a special team at JPL studied the Mariner
7 anomaly.

Fortunately, the Mariner 6 encounter events executed without any problems. Many pic-
tures of Mars were taken and successfully returned to Earth using both the high-rate
and normal low-rate telemetry systems. The specid "Tiger Team" at JPL was able to
overcome the Mariner 7 attitude problem by using the real-time high-rate telemetry
sight, the TV cameras on Mars, in time to carry out a very successful encounter.

For both encounters, the new High-Rate Telemetry System (HRT) proved its worth,
not only in recovering from the Mariner 7 emergency, but also in providing a much
faster channel for playing back TV and other high-rate science from Mars to Earth. In
all, 202 pictures covering almost 20 times the area covered by Mariner 4 were returned
to Earth by the Mariner 6 and 7 missions.

Following the formal end of the mission in November 1969, both spacecraft, still oper-
ating perfectly, were used to perform radio science experiments in relativity, astronomical
constants, and electron densities in interplanetary space.
Mariner 1971

We turn again to Corliss for a view of these missions: "The logical follow-on mission
to Mariner-Mars 1969, using the lunar program analogy, would be picture-taking orbiters
around Mars. The next Mars opportunity was in the spring of 1971, and two Mariner-
class spacecraft were ~ r e ~ a r eaccordingly.
d The new Mariners drew heavily upon the
technology of the 1969 mission. Instrumentation was very similar, with emphasis again
on photography. The primary objecuves were the search for evidence of Iife and the
gathering of data that would aid Mars landers. After mapping as much of the surface
as possible, scientists wanted more data on the density and composition of the Martian
atmosphere. A 90-day orbital mission for each spacecraft was planned. These spacecraft
would be the first terrestrial satellites of another planet."

The requirements levied upon the DSN for these missions were even more extensive
than those of the 1969 missions. This was due to the precise navigation and maneuvers
needed to inject the spacecraft into orbit around Mars and to the DSN antenna cover-
age required for 90 days of orbital operations for two spacecraft simultaneously.

During the intervening two years between the Mars opportunities, DSN capabilities
expanded considerably. The Mulumission Telemetry System had been inaugurated on
the 1969 flights, and for the 1971 mission. The experimental HRT, which had been
demonstrated so successfully in 1969, was fLlly operational. Pictures and science could
be transmitted back from Mars to Goldstone much more rapidly. A 50,000 bit per sec-
ond wideband communications link was available to carry two 16,200 bits per second
(16.2 kbps) data streams simultaneously between Goldstone and JPL. In effect, the DSN
now had a downlink capable of delivering two simultaneous 16.2 kbps data streams
directly from Mars to JPL in real time. Of course the Mars station at Goldstone was
the only 64-m antenna in operation at that time, but the high speed readouts of the
Mariner 71 tape recorders were planned to take place only when DSS 14 had Mars in
its view. Another important capability of the DSN, which was planned for use on these
missions, was the ability of a single DSN antenna to simultaneously handle two space-
craft located in the same beamwidth.

The Mariner 71 flights began inauspiciously with a failure of the upper stage (Centaur)
of the Atlas-Centaur launch vehicle which resulted in the Loss of Mariner 8.
Investigation of the Centaur problem delayed the next launch to 30 May when
Mariner 9 lifted off on a direct ascent trajectory for Mars with an arrival in mid-
November 1971. AIthough the loss of one spacecraft placed the burden of
responsibility for the entire mission on Mariner 9, it reduced the burden of track-
ing for the DSN to a single spacecraft.

After a perfect midcourse maneuver in June and a nominal cruise to the planet, Mariner
9 was injected into Mars orbit on 14 November. Unfortunately, just at the time of
Mariner 9 s arrival, the planet was covered by a huge dust storm, which precluded any
useful photography. Mariner 9 continued to make its regular 12.567 hour orbits while
the scientists waited for the dust to sede. While the spacecraft waited in orbit, the other
instruments were busy and pictures of Mars's moons Phobos and Deimos were taken.

By 3 January 1972, the dust storm had cleared saciendy for the 90-day mission to
begin. Soon, the pictures came by the thousands to reveal, for the first time, deep flu-
vial channels, possibly cut by water, and evidence of ice action in the polar regions.
Detailed maps of the Martian surface, needed to plan the Viking landings to come a
few years later, were being drawn even as the Mariner 9 cameras covered more and more
of the surface of the planet.

Because the spacecraft was still operating so well at its nominal end date in April, NASA
extended the mission to re-examine some specially interesting areas and observe solar
occultations as the spacecraft moved through the shadow of Mars twice per day. Despite
its rigorous and longer than planned mission, Mariner 9 survived until 27 October 1972.

Mariner 1973

The launching of Mariner 10 in November 1973, on the first mission to Venus and
Mercury, marked the end of the Mariner spacecraft era. Immensely successful as it was,
the Mariner 1973 Venus-Mercury mission was the last to use the so-called Mark 111-73
network model which had supported the previous Mariner missions to Mars in 1969
and 1971 and the later Pioneer missions to Jupiter.

High-rate telemetry, wideband communication circuits connecting all complexes to JPL,

and 64-m antennas at all three sites formed the basis for the DSN support of the Mariner
10 mission. However, the spacecrafi carried a redesigned imaging system that required
the DSN to provide a downlink capability from Mercury of 117.6 kbps to M y exploit
the new imaging system capability. This data rate greatly exceeded the operational capa-
bility of the existing DSN high rate telemetry system of 16 kbps.

Responding to pressure from the Project, rhe DSN had proposed a substantial number
of improvements to the existing DSN downlink capabiIity to increase the mission data
return and enhance its science value. These improvements included 1 ) the installation
of a developmental supercooled maser and ultra cone at DSS 43; 2) the installation of
an SIX-band dichroic plate and feed cones at DSS 14; 3) the implementation of a spe-
cial 230-kbps wideband data transmission circuit from Goldstone to JPL for real-time
picture transmission and a 28.5-kbps circuit from Australia and Spain to JPL for trans-
mission of recorded data; 4) the redesign of the telemetry and command processor (TCP)
computer s o h a r e to handle the data rates as high as 117 kbps; and 5) the installation
of a Block IV receiver-exciter at DSS 14 for the SIX-band Radio Science Experiment.

As the mission extended into a second encounter with Mercury in 1974, the increasing
Earth-spacecraft range introduced hrther losses into the telecommunication path. To
compensate for this additional loss, the DSN implemented a new signal enhancement
technique, previously developed and demonstrated by Spanish engineers at the Madrid
Complex in 1969 and 1970. In this technique, the spacecraft signals from several DSN
stations were added rogether at a central point so that their combined signal strength
would be greater than that of a single station. The new "antenna arraying and signal
combining technique was implemented, this time at Goldstone, using stations DSS
14, DSS 12, and DSS 13. Tests with the spacecraft demonstrated an improvement in
downlink performance of 0.7 dB. In September 1974, the new technique was used for
the first time by the DSN, with good results, as a fully operational capability to sup-
port the second Mariner 10 Encounter of Mercury.

The Mariner 10 data return statistics were impressive. At its conclusion, it had returned
to Earth over 12,000 images of the planets Venus and Mercury. In terms of scientifi-
cally useful pictures, this exceeded the combined total of the Mariner 6 and 7 missions
to Mars by a factor of almost 15.

The time Mdriner I 0 had spent in deep space, short though it was, had been very event-
ful. Years later, Bruce Murray, who at the time was leader of the Mariner 10 Imaging
Team, expressed his view of the scientific worth of the mission, "The economy class
Mariner flyby of Venus and Mercury was one of the most productive space science exper-
imenrs ever carried

On 24 March 1975, shortly afier its third flyby of the planet Mercury and exactly 506
days after its Iaunch in November 1973, the last of the Mariner series of planetary space-
craft fell silent. With its attitude control gas supply exhausted, Mariner 10 started to
slowly tumble out of control, and controllers at the Jet Propulsion Laboratory in Pasadena,
California, sent a command to turn off its transminer to avoid contaminating the deep
space environment with an unwanted source of radio emissions.
From the DSN point of view, the Mariner 10 Mission to Venus and Mercury was asso-
ciated with a Iong list of "firsts."

The Mariner 10 mission was the

first multiplanet gravity assist mission,
first spacecraft to photograph Venus,
first spacecraft to fly by and photograph Mercury,
first spacecraft to have multiple encounters with a target planet,
* first JPL spacecraft to transmit fdl-resolution pictures in real time from planetary dis-
tances,
first mission to use dual-frequency radio transmission, and
first mission to use arrayed ground station antennas to improve signal-to-noise ratio.

For the DSN, the Mariner 10 mission to Venus-Mercury provided a fitting conclusion
to the Mariner Era. During that era, there had been many other important missions
sharing the attention of the DSN, which will be discussed later. Nevertheless, it was the
Mariners that influenced the future capabilities of the DSN for the longest time and in
the most significant way, as is evident in the summary of the Mariner missions con-
tained in the following table (on the following page).

Unlike the short missions of the five earlier Pioneer lunar probes, the second genera-
tion of Pioneer spacecraft consisted of long duration missions designed to measure the
interplanetary fields and particles environment along trajectories that placed the space-
craft in heliocentric orbits ranging between 0.7 and 1.2 AU. They were sometimes
described as interplanetary weather stations. The four spacecrk, Pioneers 6 9 , were each
launched approximately one year apart, starting in December 1965 with Pioneer 6 and
ending in November 1768 with Pioneer 7.

All four spacecraft transmitted continuous streams of scientific data from various
orbits around the Sun. This meant that the DSIF was required to provide almost
continuous tracking for first one, and eventually four Pioneers, to ensure that all of
the rapidly changing science data were captured. Furthermore, instead of dying after
the first year or so of operation as was expected, all four Pioneers continued to oper-
ate actively in deep space for many years. At the end of the Mariner Era they were
still going strong and had become a permanent feature of the DSIF routine track-
ing schedules.
Sufnmaxy of the Mariner Planetary Missions: 1962-73
....,.,...,.,.,..........
,>' , ..
,. Jd.. 2.. 2f.d .<<. A<,. i.<.../X ,&.
T<,.?:.::+:.
..>:.~.:.~:.-.:.:.:.-.>:.~.>~.>~:.~//.:<<<.:/.>~.>://,.:~.:<.-<.-/c-.~.:/.~/.~,z.-z/.-r/.-~.-<.-.-.-/.-.-.~.~.~<.~~.-.~.~~.-.-.-.~.-<.~.~.~.~*~,.~~<.~,.~<.-.~t.~.~.-.~.~.".~.~.~A+.~.~.~/.~.~.-t.~~.~.~.~

Mission; Objective Science Max. Data Rate

Launch Date Result Return at Encounter
1, Venus; Flyby, None None
Jul 62 Failure
2, Venus; Flyby, Scanned Venus surface 8.33 bitslsec
Aug 62 Success at encounter
3, Mars; Flyby, None None
Nov 64 Failure
4, Mars; Flyby, First close-up photos 8.33 bitslsec
Nov 64 Success of Mars surface
5, Venus; Flyby, Science data on Venus 16.33 or 8.33 bitslsec
Jun 67 Success surface environment
6, Mars; Flyby, High-res. photos of Mars' 16,200 bitslsec
Feb 69 Success equatorid region
7, Mars; Flyby, High-res. photos of Mars 16,200 bitslsec
Mar 69 Success southern hemisphere
8, Mars; Orbiter, None None
May 71 Failure
9, Mars; Orbiter, Mapped whole Mars surface 16,200 bitslsec
May 71 Success photos and data from Mercury
10, Venus1 Flyby, UV photos of Venus, 117,600 bitslsec
Mercury; Orbiter, close-up photos and
Nov 73 Success data from Mercury

The new Pioneer program was directed by the NASA Ames Research Center (ARC) at
Mountain View, California, and thus it became the first major non-JPL or "off-Lab"
flight project to require support from the DSIF. In addition, ARC wanted to exercise
mission control functions from Mountain View rather than from the mission control
center at JPL in Pasadena. These and other requirements, related to enhanced downlink
capability, presented a new situation for the DSN and forced it into broadening its man-
agement structure to deal with flight projects other than those directed by JPL. In due
course, special positions were created in the DSN Office to deal with the unique require-
ments of each new flight project. This unique position was given the title of "DSN
Manager for Project . . . ."
Like other flight projects in these early years, the Pioneer Project provided the track-
ing stations with a set of unique Ground Operational Equipment (GOE) for use
when the station was tracking its particuIar spacecraft. The stations were obliged to
take on the onerous tasks of accommodation, installation, maintenance and opera-
tion. The GOE provided real-time telemetry data processing and spacecraft
commanding functions, which could not be handled by the existing DSlF equip-
ment because the Pioneer spacecraft design differed from that of the current JPL
spacecraft. Eventually, the Madrid engineering staff developed new software for the
DSN's Multiple Mission Telemetry and Command Processor that emulated the GOE
functions and interface signals, thereby avoiding the need for the special Pioneer-
unique equipment in the Network.

For the most part, the DSIF used the network's 26-m antennas to support the Pioneer
missions. However, as the Pioneers reached the threshold of detection capability for the
26-m antennas, improvements in downlink sensitivity became necessary to the contin-
ued life of the missions. DSN engineers made systematic improvements in the S-band
maser sensitivity, microwave equipment, and receiver "tracking Ioop" performance and
advanced demodulation hardware, while Madrid engineering st& developed special con-
volutional decoding software for Pioneer 9. As a result of these improvements, the threshold
of performance for Pioneer on the 26-m antennas was extended from 0.4 AU in 1965,
to 1.5 AU by 1969. Although it was driven by the enduring performance of the Pioneers,
the improved downlink capability was, of course, of great benefit to later missions and
the to the DSN in general.

Inevitably, as the Pioneers moved steadily deeper into space, the strength of their down-
link signals dropped below the detection capability of the enhanced 26-m antennas.
Now, fully compatible with the Network's multiple mission capabilities, the Pioneer
spacecr& were transferred to the new 64-m antennas which were, fortuitously, just com-
ing into service in the network. W ~ t hthe great increase in downlink capability that the
64-m antennas brought to the Network, the communication range of the Pioneers was
extended significantly into, and even beyond, the next era.

Much smaller than the JPL-designed Mariner-class spacecrafi, the interplanetary

Pioneers were simple and rugged and, being spin-stabilized, avoided the complexities
of three-axis attitude control used by the Mariners. The eficacy of their design had
been amply demonstrated by four years of continuous deep space operation when, in
1969, NASA selected the "third generation" Pioneers for the first missions to Jupiter.
Alrhough the Jupiter Pioneers were technically more advanced than the previous space-
craft, the design philosophy was much the same. This project was directed as before,
by ARC.

The spacecraft were instrumented to carry out fields and particles measurements through-
out the entire mission, At Jupiter they were to explore the Jovian system, obtain the
first spin scan images of the planet, and investigate its enormous magnetosphere. To test
new technologies for future missions into this previously unexplored region of deep space,
where sunlight was too weak for satisfactory operation of solar cells, the spacecraft were
powered by Radioisotope Thermal Generators (RTGs). Instead of a mast-type of anten-
na, the new Pioneers carried 2.75-m parabolic dish antenna for communications with
the DSN. They were, for their time, very advanced spacecraft.

Pioneer 10 finally left the launch pad on 3 March 1972, and made its closest approach
to Jupiter 21 months later, on 4 December 1973, at a distance equal to 2.86 Jupiter
radii, 302,250 krn from the center of the planet. Science data from the encounter
instruments confirmed the complexity of the Jupiter system. For the first time, close-
up photos of the planet surface were obtained and an atmosphere was detected on the
satellite 10. The giant planet's magnetosphere was found to be disk shaped and bigger
than the Sun itself.

DSN support for the Pioneer 10 Jupiter mission was complicated by serious conflicts
for the 26-m and 64-m antennas by the Muriner 10 mission to Mercury and Venus and
by the need for Goldstone 64-m radar surveillance of possible Viking-Mars landing sites.
The multiplicity and complex nature of these overlapping view periods presented the
DSN with a problem it had not encountered up to this time, but which loomed omi-
nously in the years ahead. A Network Allocation Working Group was established to
resolve the conflicts by mutual compromise between the conflicted projects. The DSN
had moved from an age when it could assign antennas for the exclusive use of a single
flight project to an age when it would be necessary to assign missions to antennas on
a day to day or even hour to hour, basis. Thus was born the DSN scheduling system,
which soon became a permanent feature of daily DSN activity.

By the close of the Mariner Era, eleven DSN stations had provided 2000 tracking pass-
es for Pioneer 10 between April 1972 and January 1974, the equivalent of 21,000 hours
of tracking time. During this time the spacecraft distance from Earth varied from 22
million to 890 million kilometers. During the 60-day Jupiter encounter period, over
17,000 commands were sent to control the complex spacecraft sequences.
Except for a sling-shot (later called a gravity-assist) encounter with Jupiter, the Pioneer
11 mission was essentially the same as that of Pioneer 10. It was launched on 6 April
1973 and reached Jupiter at the end of the year.

Building on the success of the Pioneer 10 encounter with Jupiter, the ARC mission con-
trollers retargeted Pioneer 11 in flight in such a way that, at its encounter with Jupiter
in December 1973, trajectory was altered to eventually intercept the ringed planet Saturn.
Pioneer I1 went on to pass within 20,000 kilometers of Saturn's main outer ring sys-
tem on 1 September 1979, and became the first spacecraft to do so. The remarkable
success of Pioneer 11, discussed further in the chapter on the Voyager Era, paved the
way for the Voyager spacecraft to later pass Saturn.

Originally the objectives of the Surveyor Lunar Lander program, complemented by Lunar
Orbiters, were primarily scientific in nature. After Apollo became a national goal, the
objectives of the Surveyor program were redirected somewhat towards determining whether
a human could land safely on the surface of the Moon. The Surveyor program was man-
aged by JPL and controlled from the Space Flight Operations Facility (SFOF) which
had replaced the simple mission control center of the early days.

The DSN felt the impact of Surveyor in a number of ways. These stemmed from a new
(to the DSN) concept that Surveyor planned to use to control its spacecraft once they
landed. The spacecrafi had few automatic features and depended almost entirely on real-
time commands for its operation on the lunar surface. This concept of interactive control
of a distant spacecrafi was based on real-time video and other data displays of the cur-
rent state of the spacecraft. These data would be used by spacecraft controllers to make
decisions regarding the desired future state of the spacecraft. Commands transmitted
from the DSN stations would control the actions of the spacecr& accordingly. To make
this possible, video data streams from the tracking stations, 4,400 bits per second from
Goldstone, 1,100 bps from the overseas stations, were first transmitted via the Ground
Communications circuits to the SFOF. There the data was converted to television dis-
plays for operational decisions by the spacecrafi controllers, and to photographic images
for scientific interpretation by the project scientists.

Obviously this level of operational decision making was completely dependent upon the
reliability of the communications circuits between the DSN sites and JPL. To ensure
the integrity of the overseas communication circuits for Surveyor, all DSN traffic was,
for the first time, carried on a high-quality satellite channel provided by NASCOM.
As was the case for Pioneer, the Surveyor Project provided special mission-dependent
equipment in the form of a Command and Data Console (CDC), for installation at
each of the DSN stations, that would support the missions. The CDCs were operated
and maintained by project people who were resident at the stations, pending the trans-
fer of responsibility to the DSN. In the event of a loss of communications between a
DSN station and mission control at the SFOF, these personnel would be capable of
controlling the spacecraft to ensure its safety until the situation was corrected.

In the period prior to launch of the first Surveyor, lengthy and elaborate simulations
were carried out to prepare the DSN and Surveyor operations teams for the scenarios
they would likely face in interacting with the spacecraft on the lunar surface.

Using its new Atlas-Centaur launch vehicle, NASA launched seven Surveyors between
May 1966 and January 1968. The success of all seven launches demonstrated the reli-
ability of the new high-energy Centaur upper stage, a potent liquid-hydrogen-fuelled
booster rocket which had recently been developed at NASA's Lewis Research Center.
The Centaur subsequently came into general use as an upper stage for deep space mis-
sion launches, until the program was discontinued in 1986 as a consequence of the
Space Shuttle Challenger disaster.

Landing sites for the Surveyor spacecraft had been chosen primarily for their interest to
the future Apollo program and four of the spacecraft soft-landed successfully at, or near,
these sites and returned large quantities of engineering and scientific data. Two of the
spacecraft failed to reach the lunar surface. Surveyor 7, the find spacecrafi of the series,
was targeted to land in the more scientifically interesting lunar highlands. Afier a suc-
cessll landing it demonstrated the advantages of real-time control and manipulatory
capability by carrying out a number of experiments that involved digging trenches, mov-
ing rocks, and even recovering an alpha-scattering experiment that had failed to deploy
correctly after touchdown. The effort that went into the pre-launch simulation efforts
clearly proved to be worthwhile.

Despite the complexity of the Surveyor program, DSN support for all seven flights was
deemed to be excellent due in no small measure to the exhaustive testing and training
program that preceded the first launch. The table below, which summarizes the data
returned from each of the Surveyor missions, attests to the quality of the DSN data
retrieval support.
Summary of Surveyor Lunar Lander Missions

Mission; Launched Spacecraft Operations on the Lunar Surface

1; 1966; 30 Mar Sofi landing, operated for 2 lunar days. Returned 13,000 pictures,
I received 108 commands.
2; 1966; 20 Sep Spacecraft became unstable, destroyed itself in flight.
3; 1967; 17 Apr Sofi landing, operated for 1 lunar day. Returned 10,000 pictures,
I received 63,000 commands.
4; 1967; 14 Tul I Lost radio contact touchdown.
Soft landing, operated for 2 lunar days, 118 hours of alpha-
scatter experiments. Returned 27,000 pictures, received 123,000
commands.

16; 1967; 7 Nov Soft landing, operated for 1 lunar day, 30 hours of alpha-scatter
experiments. Returned 45,000 pictures, received 170,000 corn-
mands.
Soft landing- operated
- for I lunar day, 100 hours of alpha-
scatter, many surface sampler experiments. Returned 28,000 pic-
tures, received 150,000 comm&ds.

Note: One lunar day is equivalent to approximately 27 Earth days.

Lunar Orbiter Missions6

Although an orbital lunar mapping mission had been originally conceived by JPL as a n
adjunct to the Surveyor lander program, NASA eventually assigned the development and
management of the Lunar Orbiter project to the Langley Research Center in Hampton,
Virginia.

The main objective of the Lunar Orbiter project was to obtain high-resolution photo-
graphs of potentid Apollo landing sites on the Moon to complement those made by
Surveyor on the actual surface. From its continuous orbit about the Moon, it would
also make a detailed photographic survey of the entire lunar surface. To accomplish this
task, the spacecraft carried a complex wide-angle camera and film processing package
which employed facsimile transmission techniques to return the images to Earth via the
tracking starions of the Deep Space Network. There were five identical spacecrafr in the
Lunar Orbiter program, all launched on Atlas-Agena vehicles in one year, beginning
August 1966. All five missions were successful. Each of the missions orbited the Moon
at a different inclination to the lunar equator and was therefore able to scan a different
section of the lunar surface. The combined result of all five missions amounted to a sur-
vey of almost 99 percent of the entire lunar surface. By the time the project ended,
Lunar Orbiter had surveyed the entire front surface of the Moon and most of the back
surface, in some 2,000 individual high-definition, photographic images.

A standard S-band network of three deep space uacking stations was used to support
the five Lunar Orbiter missions. Like the Pioneer and Surveyor projects, the Lunar
Orbiter project provided these stations with special purpose mission-dependent gound
reconstruction equipment (GRE), as well as personnel to operate them. The Lunar Orbiter
missions were conducted concurrently with the Surveyor and Mariner-Venus missions.
This multiple-mission environment heavily taxed the resources of the DSN and, by
requiring precise schedding of rhe activities of all three projects and a high degree of
cooperation and coordination between them, presaged the later establishment of the
Network Resources Allocation Working Group.

The Lunar Orbiter missions provided the DSN with the first opportunity to evaluate
its new "turnaround" lunar ranging system in actual flight operations. Prior to this time,
Earth to spacecraft range had been estimated by electronically processing the Doppler
tracking data, a method that was cumbersome and prone to errors from "slipped cycles."
The Lunar Orbiter spacecraft carried a ranging transponder, which, h e r detecting a spe-
cial code carried on the uplink, would retransmit the code dong with telemetry data
on the downlink back to Earth. At the DSN tracking station, special ranging equip-
ment compared the downlink code with uplink code to calculate the time it took the
range code to travel out to the spacecrafi and back. M e r proper calibrations and cor-
rections were applied, spacecraft navigators used these data to obtain a very precise value
for the spacecrafi range (distance between tracking station and spacecraft). So success-
ful were these inflight demonstrations that turnaround ranging, in various more refined
forms, soon came into regular use along with Doppler data as the standard types of
radiometric data generated by the DSN for all lunar spacecraft navigation purposes.

In addition to refining the new turnaround ranging system, DSN engineers extended
its use to improve the time synchronization between DSN stations throughout the world.
This enabled a more accurate determination of the spacecrafi orbit about the Moon,
which in turn yielded more and better information about the geophysical propercies of
the Moon itself. In addition to purely scientific interest, these data also provided valu-
able background information for the design of the Apollo piloted missions to the Moon,
which were about to begin.

After the Lunar Orbiter program ended, scientists at JPL continued to analyze the wealth
of DSN ranging and other radiometric data that had been generated during the Lunar
Orbiter missions. The outcome of this work greatly improved our understanding of the
radius, orbit, and gravitational field of the Moon. It also led to the discovery of hith-
erto unknown anomalies in the lunar gravitational field due to unexpected concentrations
of mass, or "mascons" as they came to be called, lying beneath the lunar surface.

Just as NASA had assigned tracking and data acquisition responsibility for the planetary
program to JPL, it assigned tracking and data acquisition responsibility for the Apollo
program to the Goddard Space Flight Center (GSFC), Greenbelt, Maryland. And, just
as JPL built the DSN for its planetary program, GSFC built the Manned Space Flight
Network (MSFN) for supporting Apollo. The MSFN comprised two types of tracking
networks. The first consisted of a large number of 9-rn stations to handle the Apollo
spacecraft during Earth-orbit operations, while the second network consisted of three
26-m stations to handle the translunar and lunar phases of the flights. For obvious rea-
sons, the 26-m stations of the MSFN were located near the DSN stations at Goldstone,
Canberra, and Madrid.

Although the MSFN provided the prime network for tracking and data acquisition sup-
port for the Apollo flights, the DSN also contributed a great deal of technology and
facility support to Apollo. The DSN provided the MSFN with S-band receiving, trans-
mitting and ranging equipment and computer software for lunar trajectory orbit
determination purposes. Also, because the MSFN stations were (intentionally) very sim-
ilar in design to the DSN stations, it was decided to equip one DSN station at each
longitude (DSS 11, DSS 61 and DSS 42) with sufficient MSFN equipment to act as
backup for the prime MSFN 26-m stations at the same site, or one nearby These became
known as the "mutual" stations. Coriiss has described this aspect of the JPLlDSN role
in Apollo in his account of the history of the Piloted Space Flight Network (MSFN).

A second control room called the MSFN wing was added to each DSN station and
connected via a microwave link to the nearby prime Apollo station. The MSFN wing
housed the special Apollo transmitting and receiving equipment, the MSFN electronics
and recorders, and the switching connections that allowed the single DSN 26-rn anten-
na to be used for tracking either deep space or Apollo spacecraft.

By mutual agreement between JPL and GSFC, Apollo mission operations at both the
DSN mutual station and the MSFN prime station were directed by a single coordina-
tor provided by the MSFN at each complex. This operational arrangement proved to
be so effective that it was used without change for all the Apollo flights, beginning with
the Apollo 4 test flights in 1967 until the program concluded with Apollo 17 in 1972.

Following the success of the earIy test flights, DSN support for the Apollo program
expanded to include the Goldstone 64-m antenna as backup for special events and emer-
gencies. To make this possible, equipment was added to the receivers at the 64-m antenna
to allow the Apollo downlink signals to be transferred several miles via microwave link
to the prime MSFN station for processing and onward transmission to the Apollo Mission
Control Center in Houston. Furthermore, during Apollo tracks, the 64-m antenna would
have to be driven by computer generated pointing data derived from Apollo spacecrafi
trajectory information provided by the MSFN.

These new interfaces, implemented and tested in great haste, were completed just in time
for the first piloted lunar orbit flight on Apollo 8 in 1968. The then new 64-m anten-
na at Goldstone was called upon to provide backup support for the 26-m MSFN antennas
for receiving television signals for public broadcast while the spacecraft was en route to
and from the Moon and in orbit around the Moon. The arrangement worked perfectly,
and fiorn then on the Goldstone 64-rn antenna was used on a regular basis to provide
critical and, in some cases, emergency (Apollo 13) support for d l Apollo missions.

Apollo 13 m7as the third piloted mission with the Moon as its destination, and a landing
and return to Ear& as its objective. The mutual stations of the DSN, together with the
big Mars antenna (DSS 14) at Goldstone, had been scheduled to provide the backup cov-
erage for the MSFN prime stations as was normal pracrice by the time of Apollo 13.

After the Saturn V launch from Cape Kennedy on 11 April 1970, the mission pro-
ceeded "nominally." The DSN stations commenced to follow the detailed tracking schedule
developed to minimize impact to the Mariner 1969 Mars spacecraft, then in its extend-
ed mission. All went well until Apollo 13 was making its third pass over Goldstone on
14 April. At that point, about 55 hours into the mission, the unthinkable happenedzan
explosion in an onboard oxygen tank rendered the Lunar Service Module, with all its
redundant systems, useless. All mission objectives, except return to Earth, were aban-
doned. While the mission controllers at Houston set about trying to recover the crew,
all tracking stations were placed on high alert for the spacecrafi emergency. The Goldstone
Mars antenna was immediately requested to extend its tracking coverage of the current
and all subsequent Apollo 13 passes, to the maximurn possible, horizon to horizon.

The unique capabilities of the 64-meter antenna were about to be demonstrated once
again. To conserve electrical power, the spacecraft began transmitting with low power
through its low-gain omnidirectional antenna, an emergency procedure which resulted
in a much weakened downlink. Because of its great receiving capability, DSS 14 was
the only tracking station able to maintain contact with the disabled spacecrafi and its
crew during many parts of the aborted mission. Even when the 26-meter stations were
able to receive the downlink, the improved clarity of the astronauts' voice channels when
DSS 14 came into view was of great assistance and encouragement to those in danger
during the subsequent recovery efforts.

After the astronauts were sde and the mission concluded, the DSN received the fol-
lowing appreciative message from the director of flight operations at Houston. "We wish
to commend the entire Network for their superior performance in support of Apollo 13.
In the midst of this most difficult and critical mission, it was extremely reassuring to
have a Network with so few anomalies and one which provided us with urgently need-
ed voice and data to bring the crew back safely. We thank you for your outstanding
support." Cognizance of the mutual stations passed to the DSN at completion of the
Apollo program in 1972, and much of the electronics technology fiom the wing sta-
tions was subsequently incorporated into DSN systems.
THE DEEP SPACE INSTRUMENTATION FACILITY (DSIF)

When the Mariner Era opened in 1961, the network, then known as the Deep Space
Instrumentation Facility (DSIF), consisted of &tee stations--one at Goldstone, California;
another at Woomera, Australia, in the desert 320 kifometers north of Adelaide; and a
third near Johannesburg, South Mica. In addition to its 26-m diameter dish antenna,
each station had receive and transmit capabilities ar a fi-equency of 960 MHz (L-band),
and could process and record telemetry data at low data rates from the JPL-designed
space probes. Telephone and teletype circuits linked each site to a simple control and
coordination center at JPL. These circuits were carried between continents via undersea
cable, and by land line, or in the case of Johannesburg, by a high-frequency radio link,
to their ultimate destination.

At that time, the loose organizational structure of JPL allowed for a great deal of inter-
action between the technical groups responsible for the spacecraft and its operation, and
the groups responsible for the development, implementation, and operation of the DSIF.
This was of prime importance in the area of telecommunications, where it was essen-
tial that the spacecraft radio system was designed to be compatible with the radio systems
installed at the tracking stations. While the spacecraft and tracking station design groups
each focused on their individual design issues, important questions of "telecommunica-
tion compatibility" were responded to, informally, by the engineers involved. It was a
carryover from earlier times, when the spacecraft radio and DSIF radio systems had been
designed and built in the same technical division at JPL. This situation soon changed,
however, and the question of telecommunications compatibility, with aIl its ramifica-
tions, became a major concern at JPL as the NASA space program gathered momentum.

Mission Operations was another factor that dominated the planning processes for early
deep space missions. Determining and controlling the flight path of a spacecraft, dis-
playing and analyzing the engineering and scientific data received from it, selecting and
issuing commands to the spacecrafi, and coordinating the activities of the worldwide
stations of the DSIF and the communications links connecting them to JPL, are all part
of Mission Operations. This specialized activity is conducted in a central location called
the Flight Operations Control Center.

As early as 1960, JPL had completed studies for a facility that could handle mission
operations not only for the forthcoming Ranger mission, but also for all future space
missions. Plans called for a Space Flight Operations Facility consisting of the equipment,
computer programs, and groups of technical and operations personnel that would carry
out these and all other hnctions required to fly the Ranger missions. The facility would
Figure 2-4. Network Operations Control Center (NOCC) at JPL, 1969. Refurbished many times
since then, and organized in several different ways to meet the expanding needs of the DSN, the
NOCC continued to perform that vital function through t 997.

be located in a single building at JPL, and the same teams would work each of the mis-
sions to ensure continuity of experience and expertise.

In the short time available before the first Ranger launch, a temporary facility was estab-
lished next to the existing computer room. BIackboards and pinboards to post current
flight status lined the walls, desks, phones, calculating machines, and teletype machines
stood wherever space permitted. Later, a few console displays were added for closed cir-
cuit television display of printouts from the large mainframe IBM 7040 computers in
the adjacent computer room. A large static display board showed the status of the DSIF
stations and the communications links between them.

When it became obvious that the makeshift facility would be inadequate even for mis-
sions in the immediate future, planning and h d i n g for the much larger permanent
facility was accelerated. Designed specifically for spaceflight operations, the Space Flight
Operations Facility (SFOF) was completed and began supporting flight missions in 1964.
A separate control center for directing the expanded operational functions of the Network
occupied a large portion of the new SFOF as shown in Figure 2-4.
THEFACILITY
(DSIF) BECOMES (DSN)
THE NETWORK

In a memo to senior staff dated 24 December 1963, the Director of JPL, Wfiarn H.
Pickering, redefined the responsibilities of the DSIF to include <'all mission-independ-
ent portions of the Space Flight Operations Facility" in addition to the existing responsibility
for the tracking stations, the communications system linking them to JPL. The com-
bined organizational structure was named the Deep Space Network (DSN) and would
be directed by Eberhardt Rechtin, with the functional title of assistant laboratory direc-
tor for tracking and data acquisition.

At this time also, a most significant engineering change was being made in the DSN.
It concerned the change to a higher operating frequency for the uplinks and downlinks.
Mainly as a matter of convenience and expediency, the original receivers and transmit-
ters in the DSN had been designed in 1957 and 1958 to operate at a frequency of 960
MHz. In telecommunications terminology, 960 MHz lies within a narrow band of fre-
quencies identified as L-band. The Ranger spacecraft radio system had to match that
frequency to be compatible so that it too operated at L-band. However, there are sig-
nificant advantages to the uplink and downlink performance to be had from operating
at a much higher frequency of 2,200 MHz, or S-band.

By 1963, the availability of new radio frequency amplifiers and transmitters, which would
operate well at S-band, allowed the DSN to take advantage of the better uplink and
downlink ~erformanceat the higher frequencies. With an eye to the requirements of
future missions, the DSN set about converting a11 the tracking stations from L-band to
S-band. But where would that leave the Ranger program about to restart its lunar flights,
and how would it affect the recently approved Mariner missions to Mars?

The solution devised by DSN engineers was to install S-band to L-band conversion equip-
ment at the stations in parallel with the older L-band equipment. This would accommodate
the remaining four Rangers and the first two Mariner missions to Venus on L-band. It
could also provide for the later Mariner missions to Mars and the proposed Surveyor
missions, both of which would use the more efficient S-band uplinks and downlinks.
The LIS-band converter would remain in place until the end of the L-band missions, by
which time a new, M y S-band system would be in place. The LIS-band converters would
then be removed and the conversion to the more efficient S-band operation would be
complete. AL1 missions from then on would be on S-band. Of all the improvements in
the early years of the DSN, the move to S-band was probably the most significant.
The Mariner 1964 Mars missions were the first in which the DSN would use the newly
implemented LIS-band capability to adapt the spacecrafi S-band downlink to suit the
existing DSN L-band receivers. The capability to send commands to the spacecraft would
be provided by 10-kW transmitters recently installed at Goldstone-Pioneer,Johannesburg,
and Woomera stations.

In addition to the uplinkldownlink improvements accruing from the change to S-band,

significant improvements to navigation accuracy were expected from the introduction of
atomic clocks throughout the DSN to replace the less stable crystal controlled oscilla-
tors. These rubidium frequency standards improved the quality of the radio Doppler
data provided to the spacecraft navigators by the DSN stations, enabling them in turn
to improve the trajectory determination process necessary to deliver the spacecraft to a
small aim point in the vicinity of Mars.

Improvements in the NASA ground communications system (NASCOM), which con-

nected the far-flung stations of the DSN to the Flight Control Center at JPL, had
already taken place. Teletype links, voice, and high speed data circuits using a world-
wide network of microwave links, and undersea cables and radio circuits were connected
to a central communications center at the NASA Goddard Space Flight Center in
Greenbelt, Maryland. From there the circuits could be distributed to JPL and other
NASA Centers scattered throughout USA, as required. The DSN could call on
NASCOM to bring up the communications necessary to support a particular mis-
sion whenever needed. When the mission was completed, the circuits were turned
back to NASCOM to be used for some other NASA space mission, maybe at a dif-
ferent NASA Center.

By the end of 1964, the primitive Flight Operations Control Center used for the
first Ranger missions had been superseded by the newly completed Space Flight
Operations Facility (SFOF). In a dual string computer arrangement, new IBM 7094
computers performed the data processing for the Mariner 3 and 4 missions to Mars,
while the Ranger processing was still carried out on a later version of the original
7040 machines. This arrangement minimized the need for changes to the existing
Ranger software.

In the spacious facilities of the new SFOF building, a large high ceiling room in the
middle of the first floor housed the new Mission Control Center. An attractive entrance
lobby with reception desk and space-related displays occupied the front of the build-
ing, while offices and conference rooms for the various teams of flight project per-
sonnel surrounded a DSN Control Room on the other three sides. In the huge
basement below the Control Room, engineers in the DSN Communications Terminal
monitored, coordinated, and routed the flow of voices and data between the stations,
the DSN Controllers above them, and the data processing computers on the second
and third floors of the SFOF. An elaborate internal communications system enabled
all the users of the SFOF to access and direct data to and from sources and desti-
nations as their duties and authority required during the mission. The Facility was
designed to run 24 hours a day, and incorporated a generator-driven "uninterrupt-
ible power supply'' that would supply primary power if the commercial power fAled
for any reason. During critical parts of a mission, the generators would be turned
on anyway to ensure that the SFOF would not suffer any kind of a "gkitch" while
critical mission operations were in progress. The muffIed roar of the JPL generators
always informed the quiet neighborhood that something important was going on at
the Lab.

Over the course of thirty years, 1964 to 1994, the management responsibiliry for the
large data processing facilities in the SFOF passed back and forth several times between
the DSN and the Flight Project offices at JPL, depending on the current "financial cli-
mate" at NASA Headquarters.

However, responsibility for the two DSN control functions, Network Control and
Ground Communications Control, always remained in the hands of the DSN. The
consolidation of these vital functions, together with expansion of the Network itself,
benefited from a continuum of long-term planning and management. As a result, the
DSN was perceived from time to time to have a "monolithic structure" by other
organizations within the JPL and, indeed, even by NASA Headquarters. Good rela-
tions with NASA were essential to the well-being of continued JPL involvement in
the planetary space program. Nowhere was this recognized more acutely than in the
Tracking and Data Acquisition Ofice at JPL, where the encouragement of frequent
and open channels of communication between DSN personnel and NASA on DSN-
related matters did much to improve working relationships and dispel the "monolithic"
image of the DSN in chose years. With the beginning of the S-band missions and
the advent of the SFOF containing the DSN Network and Communications Control
Centers, as well as the Mission Operations Control Center and its data processing
facilities, the DSN began to assume the form, and much of the substance, of the
modern DSN.
To support the more sophisticated missions of the 1965 to 1968 period, the DSN rec-
ognized the need to expand and improve its communications, mission, and network
control capabilities. The two major lunar missions nearing launch readiness, Lunar Orbiter
and Surveyor, would pave the way for the start of the Apollo program and would trans-
mit data streams at thousands of data bits per second rather than the tens or hundreds
of bits per second received from the Mariners and Rangers. The increased complexity of
the spacecraft would require expanded and faster monitor, control, and display facilities.

For the first time, the DSN began to find that the simultaneous presence of several
spacecraft on missions to different destinations created new problems in network and
mission control. The vexing problem of DSN "antenna scheduling began to arise as
several spacecrafr began to demand tracking coverage from the single DSN antenna
available at each longitude. The difficulty of assigning priority among competing space-
craft whose view periods overlapped at a particular antenna site was to prove intractable
for many years. The problem was exacerbated by competition between flight projects
from NASA Centers other than JPL, each of which felt entitled to equal considera-
tion, for the limited DSN resources. The DSN was placed in the impossible situation
of arbitrating the claims for priority consideration. The regular "Network Scheduling"
meetings conducted by the DSN often resulted in the establishment of priorities that
were determined more by the dominant personalities in the group than by the real
needs of the projects.

With all of these imminent new requirements in mind, NASA decided to embark on
a program to construct a second network of DSN stations. Arguments as to where the
stations were to be located were complicated not only by technical considerations, but
by political and international considerations.

There were already two stations at Goldstone, one at the Pioneer site and a second at
the Echo site. Eventually, NASA decided to build two new stations, one at Robledo,
about 65 kilometers west of Madrid, Spain, and the other at Tidbinbilla, about 16 kilo-
meters from Canberra, Australia.

NASA looked to the Spanish Navy's Bureau of Yards and Docks to design and con-
struct the RobIedo station. For its new facilities in Australia, NASA dealt with the
Australian Government Department of Supply through its representative, Robert A. Leslie.
As an Australian foreign national, Robert A. Leslie played a major role in shaping
the relationship between NASA-JPL and the Australian government, on whose good
o&ces NASA depended for support of its several tracking stations in that country.
With family origins in the state of Victoria, Australia, and an honors degree in elec-
trical engineering from the University of Melbourne (1947), Leslie had worked on
radio controlled pilotless aircraft for the military in both England and Australia for
fifteen years before he encountered NASA. He was a high-ranking officer with the
Australian Public (Civil) Service (an &liarion that he retained throughout his career)
when, in 1963, he became the Australian government's representative for NASA's
new deep space tracking faciIity being built at Tidbinbilla, near Canberra in south-
eastern Australia.

As might be expected, the success of a NASA venture in a foreign country depend-

ed to a large extent on the personalities of the people who were directly involved on
each side of the international interface. The foundation for the success of what later
became the Canberra Deep Space Communications Complex (CDSCC) was, in no
small part, due to Bob Leslie's personal ability to "get along" with people at all lev-
els. In representing the Australian side of negotiations between NASA and JPL, Bob
Leslie was firm but gracious, capable, and friendly. His unassuming 'haternal" man-
ner endeared him alike to counterparts at NASA, his colleagues at JPL, and his staff
in Australia.

Along with a few key Australian technical st& members, Leslie spent a year at Goldstone
assembling and testing the electronic equipment that would subsequently be reassem-
bled at Tidbinbilla to complete the first 26-meter tracking station (DSS 42) at the
new site. He was the first director of the new Complex when it began service in the
Network in 1965. It was there that he established the procedures and protocols on
which all future DSN operational interactions between JPL and the Australian sta-
tions would be based.

A few years later, in 1969, Leslie lef? the "hands-on" environment of the deep space
tracking station to head the Austrdian Space Office, a branch of the Australian Government
that, under various names and government administrations, would guide future expan-
sion and consolidation of all NASA facilities in Australia. In that capacity, his charm,
experience, and wisdom served the DSN well.

Leslie's build was stocky and solid, his appearance craggy, his attitude "laid back."
Cheerful, sociable, easy to talk to, and blessed with a good sense of humor, he was
held in high regard by everyone he met, Australian or American. Tennis was his sport,
fishing his hobby, and "do-it-yourself" home building his passion. In his younger
days, he actually excavated the ground with shovel and wheelbarrow and single-hand-
edly built the family swimming pool at his home in Canberra. Many a JPL engineer
enjoyed a poolside barbecue at the Leslie home in the course of a technical visit to
the station.

Robert Leslie retired in 1983 and died in Canberra, Australia, in 1996.

By mid-1965, the two new stations were completed and declared operational. The DSN
then had two stations in Australia, (Woomera and Tidbinbilla), one in Spain, and one
in South Africa. In addition, a permanent spacecraft monitoring station had been built
at Cape Canaveral to replace the temporary facility with its hand-steered tracking anten-
na. Impressive as this growth was, still greater changes were in progress.

Consistent with the JPL vision for missions to more distant planets, and more power-
ful communications links to support them, the DSN had long recognized a need for
larger antennas, that is, larger than the existing 26-m antennas. Studies had shown that
a diameter of 64 meters was about the maximum practical limit to an antenna which
would have sufficient stability and structural integrity for DSN purposes. Interest in
building a giant new antenna of that size, employing radical new design and construc-
tion techniques, became a reality as early as 1962 at NASA and JPL. Feasibility study
contracts for an advanced antenna had already been issued to numerous U.S. corpora-
tions and, finally in January 1963, the Rohr Corporation was selected to build the
antenna at a suitable site a few miles from the existing Echo and Pioneer locations at
Goldstone. Since the advanced antenna was originally intended to support the first mis-
sions to Mars, the site was appropriately called "the Mars site."

Design work began immediately (1963), and procurement and fabrication of steel
components started a year later. Roads, concrete foundations, alidade structure, and
diesel generator buildings were completed in 1964, followed by the control room
and elevation bearings. The supporting framework for the antenna panels followed
next, and by mid-1966, the 64-meter Mars Antenna was ready for service. Although
the first signals from Mars-transmitted by Mariner 4---were detected in March
1966, a great deal of performance testing, personnel training, and calibration remained
to be completed before the Mars antenna was considered a fully operational addi-
tion to the DSN.
/ \

PRIMARY REFLECTOR STRUCTURE

FEED CONE HOIST

(IUADRIPOD SUPPORT
ELEVATION WHEEL AND
BACK UP STRUCTURE

BEARING ASSEMBLY

INSTRUMENT TOWER

25.3 (83)

ALIDADE BUILDING

PRIMARY REFLECTOR
SURFACE

NOTE: DIMENSIONS ARE IN METERS AND (FEET) CABLE WRAP-UP

Figure 2-5. Side elevation of the 64-meter azimuth-elevation antenna at Goldstone, 1966.
Figure 2-6. The 64-meter antenna at Goldstone, 1966. The pickup truck parked to the right
side of the pedestal gives an impression of the immense size of the antenna.

Designed to remain operating in wind speeds as high as 80 km per hour, the Mars
antenna was constructed of massive steel beams, which, together with its pedestal, weighed
about 33 million kilograms. At wind speeds exceeding 80 krn per hour, the antenna was
stowed in a fmed position to protect it from permanent damage. In that position' it
could withstand hurricane force winds. The dish, and its azimuth-elevation mounting
atop the pedestal, weighed nearly 13 million kg. The suucture rotated in azimuth on
three flat bearing surfaces that floated on a pressurized fdm of oil, about the thickness
of a sheet of paper. The reflecting area of the dish surface was 3,850 square meters. It
could be pointed to a given position in space with an accuracy of 0.006 degree. The
major structural components and dimensions of the 64-m antenna are evident in the
side elevation shown in Figure 2-5.

In a single, major step forward, the Mars antenna provided the DSN with more
than six times the transmitting power and receiving sensitivity of the 26-meter anten-
nas, and more than doubled their tracking range. This was a significant new capability
indeed. But, important as it was, it was only one in a series of stepwise improve-
ments that took place in the DSN in the mid-1960s. A photograph of the new
64-meter antenna at Goldstone, taken soon after its completion in mid-1966, is
shown in Figure 2-6.

Most of the credit for the conceptual design of the Goldstone 64-m antenna, the first of
its kind ever built for tracking planetary spacecraft, went to a brilliant electrical engineer
who had come to JPL originally in 1951 to work on missile tracking systems during the
Corporal and Sergeant tests at White Sands. His name was William D. Merrick. In 1958,
it was Bill Merrick who masterminded the design and construction of the first two 26-m
antennas at Goldstone before taking up the job of project manager for the Advanced
Antenna System. The "Hard Core" Design Team that Merrick formed to carry out the
complex design and analysis task for the new 64-m antenna contained the best engineer-
ing talent in the fields of servos, microwaves, mechanical, optical, structural, civil, electrical,
and hydraulic engineering, stress analysis and contract management that could be found
at JPL or elsewhere.

He managed his "Hard Core" team, and the contractors who eventually built the anten-
na, in a somewhat unorthodox manner, but the on-time, in-budget and to-specification
end result demonstrated the efficacy of his management style.

Although he insisted on high standards of performance from his people, his propensity
for practical jokes and his sense of humor endeared him to those who worked on his
team. Those who needed to direct or limit his activities found him difficult. He thought
of himself as a person who got things done regardless of obstacles such as budget con-
straints, administrative orders, or alternative opinions that stood in his path. He could
always find a way around them.
When the Goldstone 64-m antenna was completed in 1966, it was regarded by many
who appreciated the effort that he put into it as a testament to Bill Merrick. After the
64-m antennas in Spain and Australia were completed, Merrick moved fiom the "hands-
on" engineering, at which he excelled, to a staff position which offered little challenge
for his special talents.

Bill Merrick retired from JPL in 1984 to take up a position in industry, and died at
his home in Ventura, California, in 1997.

Important as it was, the new 64-m antenna at Goldstone was only one of many improve-
ments in technology and management that were added to the Network in the mid-1960s.
Like the 64-m antenna, each change represented an incremental increase in the overall
capability and performance of the Network. As Corliss put it, "The DSN did not sud-
denly change from a lunar to a planetary network, or a low-bit-rate network to a
high-bit-rate network, or a nenvork burdened with mission-dependent equipment to a
multi-mission nenvork. The DSN was always being upgraded, some of the steps were
small, some of the others were big." Although there were many changes in many areas,
some small, some large, among the most important at this time was the improvement
in the DSN ranging system.

The technique of measuring the range (distance) from a reference point on Earth, to a
spacecraft at lunar distance, to an accuracy of a few centimeters was one ding. To do
the same thing for a spacecraft at planetary distance required a significant new tech-
nique for measurement of range. A Lunar Ranging system, known as the Mark I, was
already in place in the mid-1960s. Although it was designed for the 1una.r missions, it
had successfully tracked Mariner 5 out to nearly ten times the distance of the Moon.

Planetary ranging, however, would require a hundredfold increase in capability. This could
be achieved with new and more complex range codes transmitted to the spacecraft on the
uplink, and new methods of detecting and decoding the range code retransmitted from
the spacecrafi on the downlink. More precise methods for calibrating the system, and for
measuring time, would also be required. Two planetary ranging systems were developed in
this period, one by Robert C. Tausworthe, the other by Warren L. Martin. The two sys-
tems were aptly named the "Tauyyand the "Mu." In Tausworrhe's design, the time delay
between a pseudo-random code transmitted on the uplink and returned on the downlink
was used to measure the range of the spacecrafi. Martin's design used a sequential binary
code, assisted by a Doppler rate-aided function to carry out the decoding, for range meas-
urement. This design also automatically produced a type of data called DRVID, for
Differenced Range versus Integrated Doppler, which allowed the range data to be cor-
rected for the effects of charged particles along the path between spacecraft and tracking
station. Together these improvements extended the performance of the Mu system to the
point where high-precision ranging measurements were possible out to a range of 2.6 AU.

The Mark I, and a later Mark LA version, was used for the Lunar Orbiter and Surveyor
missions in 1966 and 1967. The Tau planetary ranging system saw its first operational
use on Mariner 5 in 1967 and 1968, and an improved and more stable version of the
Tau system was used from 1969 to 1971 for Mariners 15and 7. Later srill, the Mu rang-
ing system appeared in the Network and was used for relativity experiments on the
extended missions of Mariners 6 and 7 in 1970.

For the accurate ranging required by the Lunar Orbiter mission, the electronic clocks
at the various tracking stations had to be synchronized to within 50 microseconds of
the master clock at Goldstone. In the early days, the DSIF had to rely on radio sig-
nals from 'WTWV for station time-keeping purposes. The most accurate measurement at
the time was in milliseconds rather than microseconds. Later, transportable cesium atom-
ic clocks from the National Bureau of Standards offered improved accuracy for station
time synchronization. However, this method proved to be too expensive and inconven-
ient for operational use. A more operationally convenient method of time synchronization
using the spacecraft itself was eventually implemented for Lunar Orbiter and improved
the accuracy to 20 microseconds.

As usual though, the DSN was looking for something better. It showed up in 1966
as the "Moon-bounce Time Sync" scheme. This depended upon the propagation delay
of a precision timed X-band signal which was transmitted from the Goldstone Venus
station to each of the overseas stations during their mutual lunar-view periods. The
Goldstone Venus station acted as master timekeeper and timing signals reached the
desired stations via the Moon. By 1968, the DSN was setting its clocks to an accu-
racy of 5 microseconds using this technique, which had been adopted for operational
use throughout the Network.8

Driven by the rapidly increasing data rates being transmitted by the new spacecrafi com-
ing into the DSN, the capability of the ground communications system had to be
expanded to permit the tracking stations to return data to JPL as rapidly as possible.
More stations needed more operational voice and teletype tr&c to control them.
ConsequentIy, in 1967, a new computer-based teletype communications switcher was
installed in the Com Center in the SFOF and the high-speed data circuits were upgrad-
ed to catry 2,400 bits per second.

The SFOF, too, felt the pressure to keep up with the ever-increasing requirements for high-
er data rates on the uplinks and downlinks for the new missions. The SFOF data processing
capability was expanded to three computer strings, each comprising an IBM 7044 and
large disk memory for input/output processor functions, followed by an IBM 7094. The
7094 was used as the primary processor for the complex calculations related to orbit deter-
mination; it generated antenna pointing and receiver tuning predictions used for initial
acquisition, calculated the parameters needed for spacecraft maneuvers, and manipulated
the spacecrafr tracking data generated by the DSN stations. In addition, a special system
was developed to display monitor and performance data for the DSN station controllers.

By 1967, the interactions between the DSN and new flight projects needing DSN sup-
port had become very formalized. No longer were all the missions managed by JPL, where
loose interdepartmental agreements would suffice to commit funds and resources to meet
a specific JPL objective. Now, other NASA Centers were responsible for the spacecraft
and its scientific payload, and the function of JPL was to provide the tracking and data
acquisition support using the resources of the DSN. Even within the JPL institution, the
changing management organization at NASA Headquarters was reflected in entirely sep-
arate funding channels for the flight project and the DSN organizations. This situation
resulted in a very formal process for the presentation of flight project requirements to the
DSN, and the acceptance (or rejecrion) of these requirements by the DSN.

NASA Headquarters insisted that these negotiations be set out in detail in two forma1
documents, the Support Instrumentation Requirements Document (SIRD) for the flight
projects, and the NASA Support Plan (NSP) for the DSN. Before any flight project
codd begin to effectively design its mission, it not only had to have a signed SIRD
from one office of NASA Headquarters but a signed NSP fi-om another. With these in
hand, a newly approved flight project could truly begin. In the years ahead, the rela-
tionship between the DSN and aU flight projects would always be a reflection, for better
or for worse, of the SIRDINSP negotiations.

Finally, the DSN was beginning to use computer assistance to resolve mounting conflicts
for sration time, not only for spacecraft tracking purposes but also for station maintenance,
implementation of new hardware and sohare, and testing time. Also, the scations were
becoming much more complex, and consequently required more time for calibration, con-
figuration and check-out prior to the start of each tracking pass. In addition, the flight proj-
ects were becoming less tolerant of station outages which depleted their hard won trackt.ng
time allocation. The persistent issue of tracking time allocations would never go away; it
was ameliorated somewhat as techniques to manage it improved, but it never went away.

This was the state of the DSN when, within a very short period it found itself deal-
ing with more spacecraft than any of its people could have imagined a few short
years earlier.

THEDSN BECOMESA MULTIMISSION

NETWORK

Referring to this period in the history of the DSN, William Corliss wrote, "The NASA
lunar exploration program absorbed the bulk of the DSN support capability during the
1966-1968 period. Surveyor, Lunar Orbiter, and the backup support provided to the
first Apollo flights combined to utilize the DSN almost fully. The lunar program at this
time consisted of short-lived missions, a few days long, and the DSN was able to divert
some of the support necessary for the Mariner 5 shot to Venus and also accord some
support for Pionee~~ GY, which kept on operating long after the ends of their nominal
missions. Mariner 4 was also picked up again and became another example of an extend-
ed mission. The new 64-meter Mars antenna was called upon to support almost all of
these missions, alrhough not always as a prime station. The DSN during this period
was not yet a multimission network, although it was supporting many missions simul-
taneously. The DSN stations were still crowded with mission-dependent equipment and
this situation was the very antithesis of a multimission philosophy."

These and other issues related to the general problem of data return from deep space are
discussed by Hall, Linnes, Mudgway, Siegmeth, and Thatcher9 and an excellent descrip-
tion of rhe growth of the DSN from inception through 1969 is given by Renzetti et d.10

To simplify the increasing problem and cost of accommodating the many different
kinds of special telemetry and command processing equipment appearing at the track-
ing stations to support the various missions, the DSN developed a "multi-mission"
philosophy.

Instead of each flight project bringing its own equipment (and people to operate it) to
the stations, the DSN would provide a generic set of equipment capable of operating
over a wide range of parameters at each station. Future projects would have to design
their uplinks and downlinks to fall within the capability of the equipment provided by
the DSN. That way, all flight projects could use the same set of ground support equip-
ment, and operations costs and complexity could be reduced. With some modification,
this concept has been preserved to the present time.

Its success has been largely due to rhe DSN policy of advanced development, in keep-
ing the DSN capability well ahead of the flight projects' demands to use it. The DSN
was able to do this by basing its long term planning on judicious forecasting of future
mission requirements.

By the early part of 1968, the Lunar Orbiter and Surveyor Lander missions had been
concluded. Spectacularly successful, they had returned an avalanche of lunar surface
and other science data, which was assimilated into the design studies for the Apollo
piloted missions.

It was not realized at the time that a DSN mission would not view the Moon again
for nearly rwenty-five years. In December 1992, a remarkable picture of the Moon in
orbit around Earth would be captured by a spacecraft called Gizhleu, outbound on a
mission to Jupiter and its satellites. By then, the DSN and the environment in which
it functioned would be very different indeed.

The missions to follow over the next six years (1968-1974) would focus the DSN's
attention outward to even more distant planets and to the Sun. During these years, the
DSN made enormous improvements in its operational capability to meet the require-
ments of the new planetary missions.

New telemetry and command systems with true multimission features were added through-
out the Nerwork. Not only were these additions multimission in nature but their ability
to run at higher data rates on the uplink and downlink had been increased to 16 kilo-
bits per second for telemetry and 32 bits per second for the command link. New forms
of coding the downlink data to improve the quality of the science data delivered to the
flight project had also been developed.

The speed of identification and correction of problems in the Network was improved
by giving station controllers berrer tools to monitor and control the configuration, sta-
tus, and performance of remote tracking stations.

The Doppler and ranging data generated by the tracking stations and used by the flight
projeccs for spacecraft navigation purposes was improved in accuracy and exrended in range.
Studies were made to better understand the disturbing effect of charged particles on the
radio path between the spacecraft and the DSN antennas. Thrown off by the Sun, these
particles introduce delays in the radio path and consequent inaccuracies in the navigation
data. Ways of calibrating out these undesirable effects were found and put into effect.

To create a realistic operational environment to train DSN operarors prior to the start
of a real mission, the DSN uses a Simulation System. Electrically generated signals can
be programmed to simulate a real spacecraft under a variety of flight conditions. Artificial
faults can be introduced and the operations personnel can be trained in the proper reac-
tions. With an increasing number of more complex spacecraft coming into the Network
in the early 1970s, a Simulation Center was established at JPL to carry out the simu-
lation tasks for all missions. Conversion assemblies at the stations allowed the individual
stations to interact with the "SIM Center" to suit their individual training needs.

The addition of "wideband circuirs that could handle data at 50 kilobits per second between
the Goldstone Mars station and the SFOF, and 28.5 kilobits per second to Spain and
Australia, gave the Ground Communications Facility (GCF) as it was now called, a capa-
bility to deliver the increased volume of data fiorn the now enhanced DSN to the SFOF.
Because the existing high speed circuits, operating at 4800 bits per second, could handle
much of the traffic formerly handled by teletype, the teIetype services were phased out.

It had become obvious that the SFOF itself would not be able to cope with the demands
of the hture flight projects for data processing and handling. The data streams planned
for the future Mariner and Viking missions to Mars would overwhelm the data-han-
dling capability of the old IBM 7040/7044 generation of computers. An Advanced Data
System study group recommended a completely new design for the new Mark I11 SFOF,
which involved, among other things, replacing the old machines with new IBM 360175s.

No longer an appendage to the DSN, the SFOF became an autonomous organization

with full responsibility for supporting the flight missions. DSN responsibility in the
SFOF was redefined to cover only Network control and the data processing necessary
to carry out that bnction.

The tremendous improvement (a factor of six times over a 26-meter station) in uplink
and downlink performance, af&ordedby the 64-meter antenna at Goldstone, was offset
somewhat by the limitation of a single antenna at only one longitude. To provide the
continuous coverage with high performance required by the new missions, similar anten-
nas in Spain and Australia would be needed. The viability of the 64-meter design had
already been established by the Mars station experience, and by June 1969, the Collins
Radio Company of Richardson, Texas, had been selected to build two more 64-meter
antennas. To simpIify logistics and support facilities, the new antennas would be buiIt
near the existing 26-meter sites at Canberra, Australia, and Madrid, Spain.

Construcrion of the Australian antenna began soon after and was completed in three
years, without significant delays, in mid-1972. Electronics were installed and tested and,
after a period of cdibration, operator training, and performance demonstration track-
ing, the station reached 111 operational status in April 1973.

Although work on the Spanish antenna encountered some problems along the way, it
followed a similar pattern and was declared operational in September 1973. By this time,
a second 26-meter antenna had been added to each of the complexes at Madrid and
Canberra.

The DSN now had two networks of 26-meter antennas plus a complete network of 64-
meter antennas. They would be numbered so as to indicate the areas in which the
antennas were located; 11 to 19 for Goldstone, 41 to 49 for Australia and 61 to 69 for
Spain. Thus the DSN would consist of Deep Space Station 14 (DSS 14) at Goldstone,
DSS 63 at Madrid, and DSS 43 at Canberra, plus the six 26-meter stations, which were
similarly identified.

At each location, the number and types of antennas continued to increase. Larger build-
ings were built to house the hundreds of racks of electronic equipment required to
control the antennas, and to transmit, receive, record, and process the spacecraft data
before it could be fed to the NASCOM communication circuits for transmission back
to JPL. Instead of a single antenna and a modest Control Room at each longitude, there
were now several antennas, a Signal Processing Center, power station, cafeteria, labora-
tory, workshops, offices, and supporting facilities. The entire establishment had become
a Deep Space Communications Complex (DSCC)-GDSCC for Goldstone, CDSCC
for Canberra, and MDSCC for Madrid.

The new complexes at Canberra and Madrid now provided the support formerly given
by Woomera and Johannesburg, both closed by NASA: Woomera for economic reasons,
Johannesburg for diplomatic reasons. The composition of the Nerwork as it existed in
1974 after these closings had taken place is tabulated in Figure 2-7.
 UPLINK-DOWNLINK: OF THE DEEPSPACENETWORK
A HISTORY

LOONNG BACK

Born out of the challenge to the prestige of American technology created by the appear-
ance of the Soviet satellite Sputnik in October of 1957, by 1974 rfie DSN had evolved
into the world's largest and most sensitive radio communications and navigation network
for unpiloted interplanetary spacecrafi engaged in the exploration of the solar system.

From a loose collection of remote transmitting and receiving stations maintained and oper-
ated by essentially dedicated engineering personnel born JPL, the DSN had matured into
a M y integrated global network of operational tracking stations, maintained and operat-
ed by competent nationals of the cooperating countries in which the sites were located.

It had become a worldwide organization of tracking stations, located on three conti-

nents and connected by high quality communications satellite and cable links to a Control
Center in Pasadena, California. From there, a web of phone, teletype, and modem cir-
cuits distributed scientific data in various forms to the science community in research
centers and universities throughout the U.S. and to several experimenters at locations
in other countries.

Fifteen or so years after inception, the Pioneer, Ranger, Surveyor, Lunar Orbiter, and
Mariner spacecraft had carried out scientific missions to the Moon, Sun, Mercury, Venus,
Mars, and Jupiter with great success. These spacecrd were spread out in a.U directions
and distances throughout the solar system. During these years, the DSN was fi-equent-
ly called upon to track as many as six such spacecrafi simultaneously, the limit to the
requirements being set by the number of antennas in the DSN itself.

Two big Voyager spacecraft were being built for 1977 missions to Jupiter. Two ambi-
tious V i n g missions to Mars, even though they would not be launched for another
year, were demanding and getting most of the DSN attention.

Yet, this was the era of the Mariners. As it drew to a close, rhe DSN was moving toward
a new phase of development. It would be simultaneously involved with several large
flight projects directed not by JPL, as had been the case in the Mariner program, but
by other NASA Centers or by foreign space agencies. We will refer to this period of the
DSN history as the Viking Era.
Endnotes

1. R. Cargill H d , Lunar Impact: A History of Project Ranger, NASA History Series

(Washington, DC: NASA SP-42 10, National Aeronautics and Space
Administration, Scientific and Technical Information Branch, 1977).

2. William R. Corliss, "A History of the Deep Space Nerwork," NASA CR-15 1915,
1976.

N. A. Renzetti, "Tracking a d Data Acquisition Support for the Mariner Venus 67

Mission," JPL Technical Memorandum TM 33-85, VoIs. 1-111, September 1969.

N. A. Renzetti, "Tracking and Data Acquisition Support for the Mariner Mars 1969
Mission," JPL Technical Memorandum TM 33-474, Vols. 1-111, September 1971.

R. E Laeser, et al., "Tracking and Data Acquisition Support for the Mariner Mars
1971 Mission," JPL Technical Memorandum TM 33-523, Vols. I-IV, March 1972.

E. K. Davis, "Mariner 10 Mission Support," Deep Space Network Progress Report

PR 42-27, MarchIApril 1975, 15 June 1975, pp. 5-9.

3. Bruce C. Murray, journey Into S p c e (New York: W. W. Norton, 1989).

4. Richard 0. Fimmel and Earl J. Montoya, "Space Pioneers and Where Are They
Now" (Washington, DC: NASA EP-264, Educational Affairs Division, National
Aeronautics and Space Administration, 1987).

N. A. Renzetti, "Tracking and Data Acquisition Support for Pioneer Project," JPL
Technical Memorandum TM 33-426, Vols. I-IV, November 1970.

A. J. Siegmeth, "Pioneer Mission Support," JPLTechnicd Report TR 13-1526, Vol.

111, 15 April 1972, pp. 8-15.

5. N. A. Renzetti, "Tracking and Data Acquisition Support for Surveyor Project," JPL
Technicd Memorandum TM 33-30 1, Vols. I-V, December 1969.

6. J. R. Hail, "Tracking and Data System Support for the Lunar Orbiter (Project)," JPL
Technical Memorandum TM 33-450, Vol. I, 15 April 1970.
7. William R. Corliss, "Histories of the Space Tracking and Data Network (STADAN),
Piloted Space Flight Network (MSFN) and NASA Communicarions Network
(NASCOM)," NASA CR-140390, June 1974.

I? S. Goodwin, ccApolloMission Support," The Deep Space Network, Space

Programs Summary SPS 37-56, Vol. 11, 31 March 1969, pp. 35-42.

R. B. Hartley, 'Xpollo Mission Support," JPL Technical Report TR 32-1526, Vol.

X, 15 August 1972, pp. 41-48.

R. B. Hartle~,"ApoIlo 13 Mission Support," The Deep Space Network, Space

Programs Summary, SPS 37-64, Vol. 11, 3 1 August 1971, pp. 7-1 1.

8. Jet Propulsion Laboratory, "The Deep Space Network," Space Program Summary,
Vol. 111, SPS 37-43, SPS 37-45, SPS 37-53, 1966-1968.

9. J. R. Hall, K. W. Linnes, D. J. Mudgway, A. J. Siegmeth, J. W. Thatcher, "The

General Problem of Data Return From Deep Space," Space Science Reviews 8 (1968):
595-664; Dordrechr-Holland: D. Reidel Publishing Co.

10. N. A. Renzetti, "A History of the Deep Space Network From Inception to January
1969," JPL Technical Report TR 32-1533, Jet Propulsion Laboratory, Pasadena,
California, September 1971.
 CHAPTER 3

THEVIKINGERA:1974-1978
THE VIKING ERA MISSION SET

The tracking and data acquisition requirements of the Viking Mission to Mars domi-
nated the engineering and operations resources of the DSN during the years 1974 through
1978. Obviously, there were also many other missions during this period, all of them
demanding and receiving the attention of the DSN as they moved through the various
phases of their respective missions. The Deep Space mission set for that period is shown
in Figure 3-1 with the relative importance of each noted.

Viking Era
Deep Space Network Mission Set

He1 Vkl
III]Viking Era Future Missions
Primary Missions Secondary Misions
L J
Figure 3-1. The Viking Era mission set.

Primary Missions are those in the prime mission phase of their project. Secondary Missions
are those that have completed their prime mission objectives and are in an extended
mission phase. Future Missions are those which, although not yet launched, have sig-
nificant prelaunch support demands on DSN attention and resources. Figure 3-1 also
shows the spacecrafi that were moving through these mission phases during this arbi-
trarily defined Viking Era. Marker 10, for example, was engaged in multiple encounters
with Venus and Mercury. Pioneers 6-9 were observing interplanetary fields and charged
particle phenomena from solar orbits inward and outward of Earth orbit. Pioneers I0
and 11 were making the first encounters with the planet Jupiter.
Also during this era, two Helios spacecraft f?om the Federal Republic of Germany (West
Germany), the first of the non-NASA spacecrafi to be supported by the DSN, were
launched, then traveIed on helio-centric orbits around the Sun. The long-awaited grand
tour of the solar system was also inaugurated with the launch of the two JPL Voyager
spacecraft. Vbyagers I and 2 would also become dominating influences on DSN opera-
tions and engineering and establish an era of their own. Toward the end of the Viking
Era, two other missions would revisit Venus: the Pioneer Venus Orbiter for remote sens-
ing, and the Pioneer Multiprobe to determine the composition and structure of the Venusian
aunosphere.

As the Viking Era opened, the Viking project was replanning its previously cancelled
1973 Mars Lander missions for two launches in 1976. Strongly managed by the h g l e y
Research Center in Hampton, Virginia, Viking was to dominate the DSN scene before,
during, and afcer the rather short Prime Mission, and for that reason it seems appro-
priate to consider these years as the Viking Era.

It would be the tracking and data acquisition requirements for the new missions in the
Viking Era Mission Set that would provide the rationale for DSN Operations and
Engineering activity in the years 1974 through 1978.
THE NETWORK

The DSN moved from the Mariner Era into the Viking Era with one complete sub-
network of 64-meter antennas and two complete subnetworks of 26-meter antennas.
The 64-meter antennas at DSS 14 Goldstone (Mars), DSS 43 Canberra (Tidbinbilla),
and DSS 63 Madrid (Robledo) had come into operational service in 1966, 1973, and
1974, respectively. Each station could transmit and receive on S-band. In addition to
S-band, DSS 14 could receive on X-band.

At the Canberra and Madrid Complexes, one of the 26-meter antennas was colocated with
the 64-meter antennas at the Tidbinbilla and Robledo sites, while the second 26-meter
antennas were located at the Honeysuckle and Cebreros sites. Goldstone 26-meter anten-
nas were at separate sites (Pioneer and Echo). The 26-meter antennas at Goldstone, Canberra,
and Madrid (DSS 11, 42, and 61) became operational on S-band in 1964, 1965, and
1965, respectively, while DSS 12, 44, and 62 became operational in 1965, 1974, and
1966, respectively. All 26-meter antennas had S-band receive-and-transmit capability

Operations control and signal processing for DSS 43142 and DSS 63/61 were conducted
from a single control hcility at each site. For this reason, these stations were called conjoint
stations. The remaining stations were operated and controlled by individual control centers.

By the end of the Viking Era in 1978, work was in progress to convert one 26-meter
subnet (DSS 12, 42, 61) to 34-meter, SIX-band capability. The first station in the series,
DSS 12, had been completed in October and was supporting inflight operations with
its new capabiIity in November. A block diagram of the DSN as it appeared toward the
end of 1978 is shown in Figure 3-2.

The Viking Era would see many significant changes in the structure, technical capabil-
ities, and operational management of the DSN.

The spacecraft and mission control functions for the inflight projects would be conducted
from a facility now called the Mission Control and Computing Center (MCCC). The
MCCC would be a completely separate entity from the DSN. Located in BuiIding 230
at JPL, the former Space Flight Operations Facility (SFOF), it would house the Mission
Director and his mission operations teams, as well as the computers and data processing
facilities needed to process, distribute, and display the real-time mission data as they
arrived from h e DSN tracking stations. A clear, manageable interface for the transfer of
data between the DSN and MCCC had been defined, and all DSN data were delivered
by the Ground Communications Facility (GCF) to that interface in accordance with pre-
established specifications regarding the quality, quantity, and validity of the data.
f \
JPL
r - - - - - - -1 -
I
I
I
I
I
- - - -----
- 7 -

I
I
- - -_ -_ -
--- - I
,I

------
I - - - _ _

- -- ----
-- -

COMM = GROUND COMMUNICATIONS NASCOM = NASA COMMUNICATIONS

DS = DEEP SPACE POCC = PROJECT OPERATIONS CONTROL CENTER
DSS = DEEP SPACE STATION SPC = SIGNAL PROCESSING CENTER
\ 1

Figure 3-2. Functional block diagram of the Deep Space Network, 1978.
As individual flight missions were completed, they moved out of the MCCC and new
missions took their place. Some flight projects, notably Pioneer and Helios, elected to
establish Project Operations Control Centers (POCCs) at their own institutions. In these
cases, the GCF routed the DSN data directly to the POCCs.

The operations control and management functions for the Network were conducted
from a facility now called the Network Operations Control Center (NOCC). It, too,
was located in the SFOF Building at JPL and housed the DSN Operations Teams and
the hardware and software necessary to generate and transmit the data products such as
predicts, sequence of events, configurations, and schedules needed by the stations to con-
duct real-time spacecraft tracking operations. At the NOCC, station controllers coordinated
activities with the remote site operations personnel. The controllers acted as a single
point of contact for the flight Project Mission Operauons people in the MCCC (or
POCC) and the DSN on matters concerning real-time flight operations.

All data passing to and from the Deep Space Stations (DSSs) and the MCCC also passed
in parallel through the NOCC. There it was monitored, validated for quality and com-
pleteness, recorded, and eventually transferred to the flight project in the form of a
digital tape called an Intermediate Data Record (IDR). A very extensive communica-
tions, monitoring, and display capability was provided in the NOCC to allow the DSN
real-time operations staff to perform all of these functions.

The operations control and communications circuit coordination h c t i o n s for the GCF
were carried out in the Central Communications Terminal (CCT), also located in the
SFOF Building at JPL. In the basement of the SFOF resided the complex terminal and
switching equipment for the worldwide network of communications circuits on which
the DSN depended for its voice, high-speed, and wide-band data communications with
the remote tracking stations of the DSN. Working closely with the N O C C real-time
operations teams, the GCF Operations Teams performed similar functions of monitor-
ing and coordinating the activities of the NASA Communications Division (NASCOM)
of the Goddard Space Flight Center, which provided the intercontinental circuits, the
local short-haul line carriers, and the communications supervisors at the DSSs.

All communications traffic between the DSSs and the NOCC/MCCC first passed through
the CCT, where it was routed to its prescribed destination. This iunction, originally car-
ried out by established line-switching methods, would be replaced by more flexible and
reliable message-switching techniques by the end of the Viking Era. By that time also,
new capabilities for automatic detection and correction of short line outages and data
block errors would ensure continuous and error free communications support for real-
time flight operations.
NETWORK OPERATIONS

As the urgency of the NASA space program of the mid-1970s carried the DSN forward
into the Viking Era of 1975-1977, it was not an American but a German spacecraft
that vied with the NASA missions for DSN attention.

Conceived in 1966 as a cooperative project between NASA and the Federal Republic
of Germany West Germany), the Helios Mission would eventually consist of two solar
probes launched toward the Sun--one in each of calendar years 1974 and 1975-to
achieve a perihelion distance (point of closest approach) from the Sun of approximate-
ly 0.3 AU (Astronomical Unit; the mean distance between Earth and Sun, approximately
150,000,000 kilometers or 93,000,000 miles).

Each spacecraft would carry ten scientific experiments to perform field and particle meas-
urements in the region between Earth and the Sun. The target perihelion distance of
0.3 AU was chosen because it bounded a previously unexplored region of interplane-
tary space and was within the estimated limits of the high-temperature capability of the
solar ceIls necessary to generate power for the spacecraft. As viewed from Earth, the
Helios trajectory took the form shown in Figure 3-3.

It would provide unique opportunities for high-value scientific observations during the
repeated periods of perihelion (closest approach), aphelion (farthest approach), and solar
conjunctions (spacecraft obscured by Sun).

The spacecraft carried two 20-watt S-band transponders, both of which operated on the
same uplink and downlink frequencies. The transponders provided a capability for coher-
ent or non-coherent operation, turnaround ranging, and gound command of the spacecraft.

The Helios spacecraft employed one telemetry channel to transmit both science and
engineering data back to Earth. Both data types were convolutionally encoded and mod-
ulated on to a single 32.768-kilohertz subcarrier, which in turn was phase-modulated
onto the S-band downlink. The combined science and engineering information data rate
could be varied from 8 bits per second to 4,096 bits per second in steps of two.

Ground command capability was provided by two actively redundant receiverlcommand

detector chains; one connected permanently to the low-gain (ornni) antenna, the other
to the medium-gain antenna system. Separate command subcarrier fi-equencies permit-
Figure 3-3. Heliocentric orbit of Helios I .

ted ground control to select either chain, regardless of spacecraft orientation. The com-
mand data rate was Manchester-coded 8 symbols/second, which phase-modulated either
the 448- or the 512-hertz subcarrier,

The first Helios spacecraft (Helios I) was launched on a Tiran-Centaur launch vehicle
from Cape Canaveral, Florida, on 10 December 1974.

On 15 March 1975, 96 days after launch, HeZios I reached its first perihelion at 0.31
AU (see Figure 3-3). Although the perihelion was the focal point of scientific interest
for the project, the period of 25 days about perihelion was also fascinating to investi-
gators. During this period, the spacecrafi would traverse hitherto unexplored regions of
the inner solar system and collect scientific data that would increase our understanding
of the Sun's influence on Earth and its surroundings.

Planning and scheduling the supporting stations (DSS 12, 14, 42, 44, and 62) to cover
this event were enormously complicated by the conflicting requirements for DSN sup-
port generated by the Mariner 1 0 mission (then making its third Mercury Encounter
on 16 March 1975) and support for Pioneer 11, which was completing encounter oper-
ations at Jupiter at the same time. To further complicate the problem, the Viking and
the Pioneer Venus projects were also demanding attention for pre-launch testing and
training activities. Because of this situation, Helios was unable to obtain some of the
prime 64-meter antenna support and much of the redundant 26-meter station coverage
it desired. Nevertheless, by a process of pass-by-pass negotiation, and a willingness by
a l l flight projects involved to adapt tracking schedules to real-time changes, the DSN
commitment to provide continuous tracking coverage throughout &is phase of the Helios
mission was largely satisfied.

Throughout 1975, the Helios I spacecrafi continued along its predetermined path around
the Sun, as shown in Figure 3-3. Perihelion, aphelion, and superior and inferior con-
junctions continued to afford unique opportunities for science data observations, which
in turn led to special requirements for DSN support by the 64-meter stations when the
Sun-Earth-Probe (SEP) angle was smaII (less than 7 degrees).

Plans for launching a second Helios spacecrafi in December 1975 were formally com-
pleted by the project at a Joint Working Group meeting in Munich in May 1975. The
second Helios mission (Helios 2) would closely parallel that of Hklios 1 and provide the
scientists with three high-interest, periheIion crossings during the course of the mission.
However, initial launch and early phases of the new mission would be controlled in a
markedly different manner.

The HeZios 2 launch phase would be controlled from the German Space Operations
Center (GSOC) at Oberpfaffenhofen near Munich rather than horn the Mission Operations
Control Center at JPL, as it was for Helios 1. This decision, which marked a significant
advance in the confidence and capability of the newly established German space agency
to manage deep space mission operations, would eventually create a resource of great
assistance to NASA in general and the DSN in particular. However, a backup spacecraft
operations team would be located at JPL during the critical launch period to deal with
any unforeseen emergency. The DSN would continue to provide standard tracking sta-
tion support, subject to availability of resources, throughout the mission.

Realizing that it would be unable to meet all of the demands for tracking station sup-
port in 1976, the DSN managers had, in 1975, negotiated an agreement with the
Spaceflight Tracking and Data Nenvork (STDN) for its stations in Goldstone and Madrid
to provide a limited receiving-and-analog tape recording capability for the Helios 1 and
2 spacecraft telemetry signals. These tapes were to be shipped to the STDN station at
Merritt Island, Florida (MIL 71), where they would be converted to digital recordings,
which could then be replayed to JPL via the NASCOM communications circuits. At
JPL, the digital telemetry data were merged into the Helios Master Data Record and
delivered to the Helios Mission Control Center in Germany in the same manner used
for a normal DSN telemetry record.

Beginning in January 1976, this rather cumbersome operational arrangement support-

ed Helios 1, while the DSN focused on the launch of Helios 2. Throughout 1976 and
1977, "STDN cross-support," as it was d e d , was used extensively to recover Helios
scientific data that would otherwise have been lost due to the over-subscription of DSN
tracking resources in those years.

The Helios 2 spacecraft was launched on 15 January 1976, with initial acquisition occur-
ring over DSSs 42 and 44. Primary control of the Helios 2 postlaunch activities was
carried out by the GSOC with a backup team at JPL. The spacecraft successfully car-
ried out the maneuvers necessary to establish its correct orientation for rhe life of the
mission and began its long journey around the Sun. During its perihelion passage in
mid-April 1976, it would pass three million kilometers closer to the Sun than did Helios
1 (0.29 AU versus 0.31 AU). Helios I and 2 traveled similar trajectories, and the obser-
vations from Helios 2 could be correlated with those of Helios 1 to improve the detail
and precision of the data.

The primary mission for the Helios 2 spacecraft ended in May 1976 as the first solar
conjunction approached. Despite the degraded downlink as the SEP angle fell below 5
degrees, the solar conjunction period continued to be of great interest to the Radio
Science Team. They continued to gather Faraday Rotation and Celestial Mechanics data
from the spacecraft downlink as the radio path fiom the spacecraft to the Earth approached
ever closer to the solar disk and eventually entered solar occultation. In this period (of
small SEP angles), the sequential ranging technique of the Mu-2 ranging machine proved
superior to the standard DSN planetary ranging system in generating good ranging data
in the presence of increasing solar noise.
In May 1976, the Helios project announced a plan to modify, by the end of the year,
the German Telecomrnand Station at Weifheim to include a telemetry data receiving
capability for Helios. The additional coverage aorded by a dedicated German station,
coupled with the spacecraft capability to store data for transmission during a later uack-
ing pass, lessened the advantages of the cross-support mode with its attendant losses and
time delays. The Helios requirement for STDN cross-support was therefore withdrawn
at the end of the Helios 2 prime mission.

The DSN managers decided, however, to continue engineering tests to evaluate the use
of a microwave link between a STDN and a DSN station for real-time telemetry pro-
cessing and commanding.

Helios I and HeZios 2 spacecraft, while in their extended mission phase, rated equal pri-
ority with Pioneer 10 and I I for DSN support. With Viking preparations for the first
landing on Mars approaching their climax, only limited DSN tracking support for
Helios could be expected. Unlike the JPL spacecraft whose lifetime depended critical-
ly on a supply of inert gas to power the attitude stabilization systems, the Helios
spacecraft were spin-stabilized like the early Pioneers and not dependent on a limited
gas supply. They could therefore continue to orbit the Sun and return science data as
long as the uplink and downlink with the Earth tracking stations (DSN and Weilheim)
remained operable.

Supported by the DSN at a low level of effort, the Helios mission continued to return
a wealth of science data relative to the Sun and its environment until the mission was
terminated in 1985.

Long before that point was reached, however, the focus of DSN attention had shified
to other missions.

It was June 1976, and the Viking Mars Lander Mission was about to move into the
spotlight and demand the f
dl resources of the DSN and a great deal of attention from
the media.

Any way you looked at it, the Viking Mars Lander project was big. Big in concept,
scale, price, success, and big in impact on the DSN.
Like the successhl Lunar Orbiter project some years earlier, the Viking project was
managed for NASA by the Langley Research Center (LaRC) in Hampton, Virginia.
The Project Manager was James S. Martin, Jr. However, JPL had two roles to play
in the Viking project. One was to design and build the Orbiter that would carry the
Lander into Mars Orbit and the other was to provide the tracking and data acqui-
sition support required by both Orbiters and both Landers, simultaneously. Langley
would provide the Lander. The launch and mission operations would be conducted
by a joint LaRClJPL mission operations team led by a Mission Director appointed
by LaRC.

The concept was to deliver an instrumented space vehicle to Mars orbit from where
it would be released to descend softly to a preselected landing site, there to search
for signs of extraterrestrial life, and make in situ observations of the Iocal surface fea-
tures and weather. Two such vehicles would be built and launched in the 1975 launch
opportunity. The cost of the Landers and Orbiters was estimated at about $620 mil-
lion (1977),1 an enormous amount for a single project in the 1970s. The heavy
combined OrbiterILander spacecraft weighed 3,500 kilograms and required the most
powerhl launch vehicle combination then available, the TitanlCentaur, to boost it
to the Red Planet.

Viking held the unique promise of finding an answer to a question that had tan-
talized the thoughts of humans throughout the ages: "Does life, in any form, exist
anywhere outside our own planet Earth?" At that time, the science community,
although not unanimous in its opinion, held that the most likely candidate was the
planet Mars.

Viking was a very high profile mission which demanded, and expected to get, d of the
DSN resources that were available. It also demanded significant new capability that the
DSN did not have, but was requested to implement specifically for Viking. The high-
er levels of DSN management that had for so long been used to accommodating JPL
missions, did not at first respond well to this off-lab method of doing business. They
were not ready to accept requirements for support, particularly those requiring new devel-
opments, with equanimity. They felt entitled to challenge the need for the stated support,
conduct an independent groundlspace telecorn link trade-off analysis and evaluate alter-
nate approaches. This was the very antithesis of the kind of response chat the Viking
Project Management Team could accept, and it took a long time and many harsh words,
with appeals to higher authority, before the two sides found a workable solution. Part
of the problem arose from the fact that the Viking Project Office was on the other side
of the country from California.
DSN representatives became involved in negotiations for DSN services and resources
far from their home office, with a project offlce that expected on-the-spot agree-
ments. The Viking people could not understand why this was not always possible.
The reason lay in the difference between the single-minded management style of the
Viking Project Office, where everything was focused on and controlled by the proj-
ect, and the Tracking and Data Acquisition Ofice at JPL, where the DSN was managed
somewhat along the lines of "the greatest good for the greatest number." At that
stage, despite a plethora of formal NASA documentation, the DSN process for work-
ing with off-lab projects had not matured to the point where there was well established
procedure for negotiating and approving new flight project requests for DSN sup-
port and resources.

Eventually, the DSN office produced a large, new document that specified all of its capa-
bilities and the technical details associated with them. It was titled the Deep Space Network
to Flight Project Inte'ce Design Handbook and was assigned the number 810-5.2 In the
future, new flight projects seeking DSN support for their missions would be expected
to use this technical interface document as a "shopping list" to identify their require-
ments on the DSN for support. To a large extent, this avoided the dificdt bargaining
experienced in the %king negotiations.

The scientific background to Viking stemmed from the Mariner missions to Mars in
1964-65, 1969, and 1971-73. In 1965, Mariner 4 was the pioneering flight to Mars,
providing a few television pictures of its cratered surface and defining the physical
parameters of its atmosphere in a fast flyby mission. In 1969, Mariners 6 and 7 con-
ducted more extensive approach and flyby imaging, acquired spectral data on
atmospheric composition, and did initial thermal mapping. Mariner 7, placed in orbit
about Mars in 1972, surveyed the entire surface at various image resolutions and per-
formed extensive spectral, thermal, and studies. A landing mission was
the next logical step.

The concept of the Viking project involved two Viking spacecrafi, each consisting of an
Orbiter and a Lander, boosted by a TitanICentaur launch Vehicle into a Mars trajecto-
ry at the 1975 launch opportunity. On arrival at the planet, each spacecraft was to be
placed in orbit. Then, following a mapping survey and landing-site verification, the
Lander capsule was to separate from its Orbiter, and descend through a complex land-
ing sequence to the Mars surface. The configurations of the Viking Orbiter and Viking
Lander are shown in Figure 3-4. Figure 3-5 depicts upIinMdownlink telecommunica-
tions among the DSN, the Viking Orbiter, and the Viking Lander on Mars.
\

Figure 3-4. Configurations of the Viking Orbiter (top) and Viking Lander (bottom).
fl \

Orbiter

lander to Earth
two-way radio link
* Lander science data
* Engineering telemetry
Doppler and range signals
Commands Lander to Orbiter
one-way relay radio link
Lander science data
* Engineering telemetry

Lander on Mars

\ /

Figure 3-5. Viking Orbiter, Lander, Earth Telecommunications System. This diagram, by Ezell
and Ezetl, shows the uplink and downlink between the Deep Space Network and the Viking Orbiter,
the uplink and downlink between the Deep Space Network and the Lander on Mars, and the
uplink data relay from the Lander to the Orbiter.
Although the primary science investigations would be carried out by the Lander on the
Mars surface, a significant science return would be obtained from the experiments con-
ducted by the Orbiter and by the Lander during its descent through the Mars atmosphere.

At its inception in the late 1960s, the Viking project was intended to be launched in
the 1973 opportunity, making use of the Mariner 69 and Mariner 71 Orbiter design
and operations experience, and their data on the planet. Eking 73 would complete the
Mariner-Mars orderly sequence of planetary exploration by soft landing a package of sci-
entific instruments on the surface of the planet. To increase mission reliability and scientific
scope, two launches were planned. Precise navigation data would be required to achieve
the desired landing accuracy, and large quantities of scientific and engineering data were
to be telemetered from the two orbiting and two Lander craft. Preliminary Tracking and
Data System planning was undertaken on this basis.

Construction of the 64-meter antenna subnet of the DSN was just then getting under-
way after a few years of successfd operation of the initial advanced antenna system at DSS
14 at Goldstone, California. A contractor had been selected to build and install the 64-
meter antennas at the overseas sites in Australia and Spain, and these facilities were scheduled
to be ready to support the 1973 Viking missions. In addition, during this period, other
new elements of the Tracking and Data Acquisition (TDA) system destined for Viking
support were being demonstrated, such as high-rate telemetry, high-power transmitters,
and the Differenced Range versus Integrated Doppler (DRVID) navigation technique.

At the beginning of 1970, hnding considerations created far-reaching changes in the

Viking project and resuited in an alteration of the flight schedule from a 1973 to a
1975 launch opportunity. The revised launch date permitted a more thorough assess-
ment of the TDA support requirements and capabilities, and an increased level of
commitment from what would have been possible for a 1973 mission.

By the end of 1973, the planning phase was essentially complete. The design had matured,
and key capabilities and interfaces had been defined. In addition, the network of 64-
meter antennas was operational, and the implementation of new capabilities needed for
Viking had begun. Initial design compatibility testing with breadboard models of the
Orbiter and Lander radio systems had begun in the Compatibility Test Area 21 (CTA21)
at JPL. Known as "DSN Compatibility Testing," this practice had become standard by
1977 for all spacecraft with which the DSN was involved, and was indeed part of the
agreement with the flight projects seeking tracking and data acquisition support from
the DSN. The DSN had established two compatibility test facilities, one at JPL called
Compatibility Test Area 21 (CTA 21), and the other on Merritt Island at Cape Canaveral
(MIL 71). Except for the actual large DSN antennas, both facilities were equipped with
hardware and sofrware representative of, and generally identical with, that which was
installed at the DSN tracking stations around the world. Radio frequency uplinks and
downlinks at S-band, X-band, and small (2-meter) calibrated antennas and radio fre-
quency (RF) attenuators were used to accurately simulate the signal paths berween the
inflight spacecraft antennas and Earth station antennas. Following these tests, simulated
spacecraft data were input to CTA 21 to verify the design of the end-to-end ground
data link, beginning with the radio frequency data from the spacecraft and ending with
data displays in the Mission Control Center. In progressive stages, these simulated oper-
ations were built to represent the maximum mission load conditions; system integration
and data compatibility were demonstrated during mid-1974.

A succeeding series of ground data system tests, using actual tracking stations and test
models of the Viking Orbiter and Lander provided training for the personnel of the
Flight Operations System who would actually conduct the mission, as well as the DSN
and Mission Control Center personnel who would support it.

Because there were to be two Viking launches close together, the DSN would be called
upon to track two Viking spacecraft en route to Mars simuItaneously. When they reached
Mars, each spacecraft would separate into an Orbiter and a Lander, so that then four
Viking spacecraft would be supported at various times by the DSN stations. Although
only one Lander at a time would require a direct link back to Earth, this multiple space-
craft task would refire all the tracking stations in the DSN to handle it.

One set of 26-meter stations was required to provide support for the launch and cruise
period of both Vikings until limiting uplink and downlink conditions necessitated the
use of the large 64-meter stations. The second set of 26-meter stations was required to
back up the first set if necessary, and to take over some of the tracking Ioad when the
Helios solar mission needed the services of the first set during its launch period later in
the year. All four uplinks and downlinks of the Viking Orbiters and Landers would use
the standard DSN S-band frequency channels, except that the Orbiters also carried X-
band downlinks. These were intended to demonstrate the advantages of the higher frequency
X-band channels for deep space communications, an advance in technology that the
DSN proposed to introduce into the Network in the years ahead. It was expected that
the increase in DSN capability would be even greater than that which resulted from the
change to S-band several years before.

The DSN capabilities required to support the 1975 Viking Missions to Mars are shown
in the following table.
Deep Space Station Capabilities for Viking (1975)

26-Meter Antenna Stations

Deep Space Stations (DSS) 11, 42, 61, 12, 44, and 62
One low-rate, plus one medium-rate telemetry data stream
S-band ranging, shared at DSS 11, 42, 43, 61, and 63
S-band Doppler at each station
One high-speed ground communications channel
Digital original data record with automatic total recall
Analog original data record with station replay

64-Meter Antenna Stations

Deep Space Stations (DSS) 14, 43, and 63
Single command uplink, 20 kilowatt prime, 100 kilowattt400 kilowatt
backup
Up to six Low-, medium-, and high-rate telemetry data streams
S- and X-band ranging
S- and X-band Doppler
One high-speed and one wideband ground communications channel
Digital original data record with automatic total recall
Analog original data record with station replay
S- and X-band occultation data, receive and record
S-band very long baseline interferometry data receive and record at DSS
14 and 42 only

Ground Communications Facility

50 kilobit per second wideband channel DSS 14 to JPL (GCF Central
Communications Terminal)
27.6 kilobit per second wideband channel DSS 43 and 63 to JPL
4800 bits per second high-speed channels tolfrom all stations and JPL
Circuit performance monitoring and error detection

Network Operations Control Center

Automatic total recall capabiliry
Intermediate data record production
Network system performance monitoring, display, and data system
validation
Tracking performance monitoring
Network operations control
 Mars Planetary Radar (Goldstone)
X-band transmit and receive
Up to 400-kilowan transmit power desired
Use of developmental (rather than operational) hardware and sohare

Individual uplinks and downlinks, each with its own particular parameters and critical
data path, would present an operational challenge at every level to all elements of the
DSN. The complexity of the mission is reflected in the downlink telemetry capabilities
of each of the two simultaneous Viking spacecrafi (see the following two tables).

Viking Orbiter Telemetry Channels

Channel Data Type Data Rate Subcarrier
(bits per second) (kilohertz)
Low Engineering 8 113 or 33 113 24
(uncoded)
High Science lk, 2k, 4k, 8k, 16k 240
(block-coded, 32, 6)
Science Ik, 2k, 4k
(uncoded)

Viking Lander Telemetry Channels

Channel Data Type Data Rate Subcarrier
(bits per second) (kilohertz)
Low Engineering 8 113 or 33 113 12
(uncoded)
High Science lk, 2k, 4k, 8k, 16k 72
(block-coded, 32, 6)
Science 250, 500, 1000 72
(uncoded)
Four Vikings (two Orbiters and two Landers), together with the other inflight mis-
sions being supported at the same time, were about to stress the DSN multimission
concept to its limits. At the outset, it was recognized by both the DSN and the proj-
ect that the most severe operational challenge lay in the successll establishment and
manipulation of multiple communication and data links with the four spacecrafi, two
Orbiters and two Landers. A very extensive series of engineering "compatibilityyytests
were planned to demonstrate the integrity of the uplinks and downlinks. Realistic mis-
sion sequences were simulated to train the operations personnel, and comprehensive
exercises of increasing complexity were used to measure the proficiency of the com-
bined DSN and Viking flight operations teams. At the same time, similar test and
training exercises were being carried out within the huge organization associated with
the actual Titan-Centaur launch itself.

Eventually, all the prelaunch readiness criteria were met, the final launch readiness
review was held and the approval for launch was given, and the countdown began.
Unfortunately, both launches were delayed by technical problems that could not be
corrected on the launch pad and that necessitated removal of the spacecraft from the
Titan-Centaur launch vehicle. Reassembly and testing further delayed rhe lifioff. Finally,
Eking I was launched on 20 August, followed by Viking 2 on 9 September 1975.
Both launch vehicles performed perfectly to inject their payloads into Mars trajecto-
ries with arrival times at Mars of 19 June and 7 August 1976, respectively.

In both cases, the Australian 26-meter stations performed the initial acquisition sequences
precisely as planned-a tribute to the careful planning and intensive preflight testing
and personnel training that preceded it. Both Viking spacecraft survived the stresshl
launch conditions without incident, and the "cruiseyyphase of the Viking mission to
Mars began.

During the next few months, each spacecrafi would move continuously along its indi-
vidual trajectory, gradually crossing the region of deep space between the injection
point on the orbit of Earth (around the Sun) to its final injection point on the orbit
of Mars (around the Sun). It would be performing maneuvers to correct its trajecto-
ry for small errors, carrying out scientific experiments in deep space, checking and
calibrating the science instruments that the Orbiter would use to make observations
from Mars orbit, and checking the condition of the all important Lander and the
instruments it would use to carry out experiments on the Mars surface.

The DSN would maintain continuous uplinks and downlinks to communicate with
both Vikings, generate radiometric tracking data for spacecrafi navigation purposes,
and evaluate the performance of the new X-band downlink. The DSN would also be
completing implementation of the new and extended capabilities required for the Mars
"Orbiter operations" phase and "Lander operations" phase. Expansion was required in
almost every area. The quality of the Doppler data used for spacecraft navigation had
to be improved, and the data rates for the Orbiter telemetry had to be increased, as
did the capability of the tape recorders used to record it at the Complexes. New hard-
ware and sofnvare was needed to provide monitor and control functions for the new
equipment. Earlier, the Viking project determined that in the event of adverse land-
ing attitude conditions, the standard DSN 20-kilowatt transmitter power would not
be sufficient to uplink commands to the lander to work around the situation. It
demanded the newly installed 100-kilowatt transmitters which had not been proven
for support of regular network operations, let alone critical support of the type sug-
gested. Furthermore, because of greatly increased emphasis on the evaluation of alternate
Mars landing sites, a request was made for the assistance of an experimental 400-kilo-
watt X-band transmitter that was being used at Goldstone for the study of planetary
surfaces using radar techniques. At one stage, over 400 major change items were work-
ing their way through the DSN change control system to meet the Viking schedule.
Engineering and operations staff were augmented to cope with the work and ongo-
ing operations-everyone seemed to be working on Viking.

Nevertheless, the changes and additions were made and tested, and the DSN was ready
when the first Viking arrived at Mars.

On 19 June 1976, Viking mission controllers started the 28-minute retro-engine burn
that slowed Viking I down from its approach speed of 14,400 kilometers per hour to
10,400 kilometers per hour to allow it to be captured into Mars orbit. Later, after
evaluation of the navigation data, the retro-engine was burned again, this time for
132 seconds, to adjust the orbit slightly to bring it over the prime landing site. The
spacecraft was now in a 24.6-hour orbit, corresponding to the length of a Martian
day. The landing site would be viewed by the Orbitei- on each subsequent revolution
near the lowest part of the orbit. With the completion of orbit insertion and trim,
the "cruise" phase of the Eking I mission had been completed.

In the days that followed, the andysis team evaluated the Orbiter pictures and the
Goldstone radar data to the last possible detail to arrive at consensus regarding the
best (or rather, least risky) place to Iand consistent with meeting the principal science
objectives of the Mission. The range of possible landing sites was constrained not only
by the surface features, which could pose a threat to the safety of the lander as it
descended, but also by the need for a biologically promising area, which would be
indicated by the possible presence of water in some form.

The science teams searched long and hard for an agreement, and findy selected "alpha
site" in the large central basin, at 22.5 degrees north latitude and 47.5 degrees west lon-
gitude. Once the decision to land had been made, the necessary planning followed
rapidly. At 10:30 P.M. on 19 July, &er calling for readiness reports from each of the
major elements of the project, the project manager announced the "Go" for separation
of the lander from the Orbiter. The separation and descent sequences were initiated and
all events followed exactly as planned.

'With the Orbiter low on its radio horizon, DSS 43 transmitted the Lander separation
"Go" command to the Orbiter at 07:47:18. The Orbiter released the Lander about an
hour later, and began sending relay data from the Lander as it descended through the
Mars atmosphere to the surface. By 09:20, the Orbiter was beginning to set over the
Australian Complex, and control was passed from Canberra to Madrid where DSS 63
continued to receive the Lander descent data and forward it to JPL. Back in the Mission
Control Center at JPL, the Viking flight team anxiously watched the Lander telemetry
data as the tension mounted. The Lander Team Leader called off the radar-altimeter
readings and the critical events as they were executed in perfect sequence by the Lander.
Finally, at 05: 11:43 Greenwich Mean Time (GMT), came the announcement, "We have
touchdown." Knowing that several critical housekeeping chores remained to be carried
out before the first picture taking could begin, the jubilation in the Mission Control
areas was rather restrained. The Lander appeared to be in a nearly nominal attitude,
positioned in a stable, level landed configuration. The shutdown and reconfiguration
tasks were accomplished without incident, and the first real-time imaging sequence began.
The first picture (Figure 3-6) appeared on the monitors in Mission Control half an hour
later. It showed the circular footpad at the end of one of the three legs of the Lander,
neatly settled on the Martian soil. A few light-colored deeply pitted rocks lay in the
foreground.

A second picture (Figure 3-7) soon followed. It showed a 300-degree panorama of what
appeared to be a sandy plain covered with rocks of various sizes.

The quality of the first pictures horn the surface of Mars astonished the flight team, the
hundreds of guests gathered at JPL to observe the event, and the national teIevision audi-
ence that saw the pictures a short while later. These images were humankind's first close-up
view of Mars. Congratulations from the project manager, the NASA Administrator, and
President Gerald Ford soon followed.
Figure 3-6. First Viking close-up picture of Mars surface. Viking Lander 1 took this close-up pho-
tograph of the Martian surface just minutes after its successful touchdown on the Plain of Chryse on
20 July 1976. The center of the image is about 1.4 meters (5 feet) from the spacecraft camera. A
portion of the Lander footpad is visible in the lower right corner of the picture. The spacecraft con-
tinued to observe surface features and caw out numerous other scientific experiments on the Martian
surface and environment until it ceased operating on 13 November 1983.

Figure 3-7. First panoramic picture of the surface of Mars.

With one Lander on the surface, one Orbiter in orbit, and the Viking 2 spacecraft rap-
idly approaching its Mars orbit insertion point, the tempo of activity by the Viking
flight team and in the DSN reached an unprecedented level. While some of the sci-
entists were anxious to initiate the biology experiments related to the search for life,
others were faced with the task of selecting the second landing site. The spacecrafi con-
trollers were involved with managing Eking 2, then in Mars orbit, as well as operat-
ing Lander I and Orbiter 1. The DSN was doing its best to keep up with the demands
for more and more tracking time, as well as with the avalanche of data from both
Orbiters, now running at 16,000 bits per second.

Responding to commands from its controllers at JPL, Lander 1 scooped up the first
samples of Mars soil on 28 July and carefully transferred them to its various science
instruments for biological, chemical, and molecular analyses.

By mid-August, Orbiter I had been in orbit around Mars for 55 days making a com-
prehensive photo reconnaissance of possible landing sites for Lander 2. Lander 1 had
been on the surface for 24 days obtaining striking pictures of the Martian landscape,
and making continuous observations of Mars weather, surface features, and compo-
sition, as well as analysis of the Mars atmosphere. For the first time, the DSN could
establish a direct radio link between its antennas on Earth and an antenna on the
surface of another planet. Radio scientists used this capability as a highly precise
instrument to refine existing knowledge about the physical properties of Mars and
its atmosphere.

The data from the Lander I biology instrument, which was analyzing the Martian soil
samples, was indicating the possibility of a biologic+response. However, other possible
causes of the response were being considered and tested as the experiment continued.
It was still much too soon to reach any conclusions regarding the existence (or nonex-
istence) of Martian Me.

Late in August 1976, a decision was made regarding the second landing site. Lander 2
would land in the eastern end of the Plains of Utopia at latitude 48 degrees north, lon-
gitude 226 degrees west. Once that decision was made, events followed a similar course
to those of Viking 1. Touchdown would occur on 3 September. At 9 A.M. that day, the
Viking Flight team met for the "Go/no go" for separation decision. All systems report-
ed "Go" for separation. The commands were sent from DSS 14, and the separation
sequence began to execute.

But there were some heart stopping moments to come before Viking mission control
would know that Ldnder 2 was on rhe surface. Shortly after separation, 0rbit.er 2 ceased
sending the telemetry data it was relaying fiom the descending Lander. Swifi action by
the spacecraft controllers at Mission Control and the station controllers at Goldstone
determined that the spacecrafi attitude stabilization system, on which it depended to
keep its main antenna pointed toward Earth, had failed. Without the main antenna, the
Orbiter would not be able to transmit the Lander images to Earth.

In anticipation of such an eventuality, the Orbiter had been programmed to switch over
to a small backup antenna and begin sending basic engineering data to aid in uou-
bleshooting. Immediately, commands were sent to turn on the Orbiter tape recorder so
&at the pictures from the Lander would be saved while the engineers figured out how
to reestablish the main Orbiter to Earth radio link.

Using the downlink fiom the small backup antenna on the Orbiter to transmit a star
field map for reference, the spacecraft was commanded to slowly roll back to its origi-
nal position and lock on to the celestial reference star, Vega. The main antenna was
activated, and to the great relief of all concerned, the downlink signal returned to nor-
mal. All subsequent sequences were executed as planned, and the first pictures from
Lander 2 were received early in the morning of 4 September, nine hours later than orig-
inally scheduled.

The first two pictures of the Utopia site were essentially similar to those taken by Lander
1 in Planitia six weeks earlier. The first showed a footpad slightly embedded in sandy
material with some debris flung around from the exhaust of the Lander retro-engine.
The second picture revealed a barren rock-strewn plain, somewhat less rocky than the
corresponding picture fiom Lander I. Following the landing, the science instruments
began to gather data to complement the similar observations being made by Lander I
at Planitia.

For the next two months planetary operations continued at a furious pace. The flight
teams and DSN personnel and resources were hlly extended. Problems arose, and were
corrected. Failures occurred and were worked around or backup systems were activated.
Some data were lost, but very little. A few mistakes were made, but none which harmed
any of the four active spacecraft. Within the Network, new operational procedures were
developed to overcome the difficulties of working with multiple spacecraft simultane-
ously, and the multimission hardware and software held up to its arduous task of running
continuously at top speed.

Although the other experiments delivered a wealth of fascinating new scientific data
about Mars, the biology experiments continued to provide inconcIusive results.

On 16 November 1976, the primary scientific objectives of the Viking Mission to Mars
had been achieved and, as planned, the "Prime Mission" was declared complete.
The scientific community began the task of analyzing the new Mars data and pub-
lishing findings in science journals around the world. The findings of the biology
experiment and its potential to reveal the presence of some form of life on Mars gen-
erated particular public interest. The investigators needed to exercise extreme caution
in reaching a conclusion on these results. One of the investigators told a press confer-
ence, "Nobody wants to be wrong on a question as important as that of life on Mars."
Efforts to explain the inconclusive results of the biology experiments continued, and
hopes of a nice crisp answer faded. An enormous amount of scientific debate ensued
over the following years, but eventually it was conceded that the rhree biology exper-
iments did not provide conclusive evidence for the presence of living organisms on
Mars. Neither did it rule out the possibility. A NASA spokesperson declared, "No clear,
unambiguous evidence of life was detected in the soil near the landing sites." And
added, "The question of life on Mars, according to the Viking biologists, remains an
open one." So the question remained unanswered.

Although the Prime Mission had officially ended, NASA decided to extend Mars plan-
etary operations with the four active spacecraft for 18 months. By that same time,
interest and activity gradually moved to the events surrounding the rapidly approach-
ing solar conjunction, the period of time when Mars moves behind the Sun as viewed
from Earth. Radio scientists would observe changes to the S-band and X-band down-
links as they slowly became intercepted by the solar disk.These effects would be interpreted
ro yield new information about the structure and composition of the solar corona and
the solar wind.

As Mars, with its four attendant spacecraft emerged from behind the Sun in late November
1976, the spacecraft-to-Earth telecommunications links began to return to normal, mis-
sion operations activity increased, and the Viking "extended mission" began.

No one realized it at the time, but the Viking mission would continue in various forms
for the next six years. Although the Landers were designed to operate for three months
in the Prime Mission, one of them, Lander I would remain active until its last contact
with DSS 14 in mid-November 1982. The rwo Orbiters eventually depleted their atti-
tude stabilization gas supply and, without their antennas pointed toward Earth, were
unable to maintain an uplink or downlink. Lander 2 ceased operation in February 1980
when its battery failed.

The extension of the Viking mission into 1977 continued to place heavy demands on
the DSN, not only because of the addition of radio science investigations to the origi-
nal science experiments (except for biology, which had been completed in the Prime
Mission), but also because the DSN was now engaged in replacing all its obsolete SDS
computers with new MODCOMP minicomputers. Each of the nine tracking stations,
in turn,would have its old telemetry and command processors replaced with new machines
and sobare, and then be requalified to resume inflight support. It would be called the
Mark-I11 Data System (MDS) upgrade, and was to be accomplished without impact to
any of the inflight missions, including Viking (then running at its peak).

It was under these di&cult circumstances that the focus of DSN attention moved to
another new flight project that would extend the capability of the DSN uplinks and
downlinks to even further limits. This new flight project was called Voyager.

In its time, the Voyager project was one of the most ambitious planetary missions ever
undertaken. Vqyager I, which would encounter Jupiter in March 1979, was to investi-
gate the planet and its large satellites; later, it would explore Saturn, its rings, and several
of its satellites. Vqyager 2 would arrive at Jupiter in July 1979. This spacecraft would
also examine Jupiter and Saturn and their satellites, then it would be hurled on toward
an encounter with Uranus, which it would reach seven years later, in 1986. En route
to these encounters, both spacecrah would make measurements of the interplanetary
medium and its interactions with the solar wind.4

Each Voyager spacecraft follows a billion-kilometerpath to

Jupter. Except for minor thruster firings to achieve small
trajectory corrections, each Voyager coasts from Earth to
Jupiter, guided by the gravitational pull of the Sun. At
Jupiter, the powerful tug of the giant planet deflects the
spacecraft and speeds them up, imparting an extra kick to
send them on their way toward Saturn.

\ 1

Figure 3-8. Voyager trajectories to Jupiter, Saturn, and Uranus.

Except for minor adjustments to the trajectories made by the spacecraft thrusters, the
spacecraft would ini-jaliy coast from Earth to Jupiter, guided by the powerfd gravita-
tiond pull of the Sun. At Jupiter, the gravitational force of the planet itself would be
used to redirect the spacecraft toward Saturn.

Due to a problem that required the removal of Vqyager 1 from the launch pad for repairs,
bydger 2 was launched first. The launch from Cape Canaveral, Florida, in August 1977
utilized the most powerful launch vehicle then available, the mighty TitanICentaur. Viager
1, launched sixteen days later on 5 September, on a shorter, faster trajectory, soon over-
took Vqager 2 and eventually reached Jupiter four months earlier than Wyager 2.

To inject these heavy spacecrafi into an Earth escape trajectory to Jupiter would require
a five-stage rocket-motor burn sequence, two burns of the Titan, two burns of the
Centaur and a final single burn of the solid rocket motor in the spacecraft's own propul-
sion module. All these complex events were executed flawlessly on both occasions. Sixty
minutes after liftoff, the Voyager spacecraft were free of the influence of Earth's gravity,
coasting on their way to Jupiter at a speed of more than 10 kilometers per second.

The first DSN involvement with the Voyager spacecraft had taken place in the early months
of 1977 when the spacecraft uplinks and downlinks had been subjected to a rigorous series
of radio "compatibility" tests to ensure that spacecraft and DSN would be able to com-
municate without problems after launch. For the Voyager tests, CTA 21 at JPL and MIL
71 at Cape Canaveral were each configured to simulate a 64-meter antenna station, and
each included both Block-I11 and Block-IV Receiver-Exciter Subsystems and the new MDS
for telemetry, command, and radiometric data. The early tests at CTA 21 in March and
April were run with the spacecraft in the space simulator at JPL and established the valid-
ity of the radio system design as represented by the spacecraft components being tested.
The tests with the actual flight configuration spacecraft took place at MIL-71 in July and
August at Cape Canaveral. This was the last chance before launch to confirm beyond all
reasonable doubt that the Voyager-to-DSN uplink and downlink would function in space
"as designed." Both spacecraft passed the tests without reservation.

Sixty minutes afier liftoff, separated from the Centaur and beginning its long jour-
ney to Jupiter, Voyager 2 approached its second involvement with the DSN "initial
acquisition."

"Initial acqtdsition" is the term used by the DSN to describe the chain of events that
take place at the first tracking station to observe, lock onto, and track the downlink sig-
nal after the spacecraft has separated from the launch vehicle and leaves Earth orbit to
begin its planetary trajectory. A successfd, rwo-way, initial acquisition is crucial to the
safety of the spacecrafi and subsequent mission events. When completed, the data it
provides gives the mission controllers knowledge of the spacecraft condition via teleme-
try, the ability to control the spacecraft via the command uplink, and the capability to
determine the spacecrafi orbit from the Doppler radiometric data generated by the DSN
from the interaction of the coherent uplink and downlink.

Based on long experience of many launches of diierent spacecraft and different launch
vehicles, the DSN had refined the initial acquisition procedures by the time of the
Voyager launches to the point where the probability of success was related to two main
factors. The first of these, dispersion errors in the powered flight part of the launch
sequence, affects exactly where and when the spacecraft will rise at the first viewing sta-
tion and, therefore, whether or not it will rise within the beamwidth of the DSN antenna
set to the predicted pointing angles. The second factor is the uncertainty in the pre-
dicted downlink frequency at the time the spacecraft appears above the radio horizon
of the first acquiring station. This determines whether the spacecrafi downlink will appear
in the bandwidth of the DSN receiver set to the predicted downlink frequency. Time
is of the essence. Because all the uncertainties grow with time, the longer it takes to
acquire the spacecraft, the more difficult it becomes to accomplish.

When a spacecraft first appears over the radio horizon, its angular movement across the
sky is too great for the large 64-meter antennas to track. For this reason, the spacecrafe
is first acquired by the smaller, more agile, wide-beam 26-meter antenna. Much later,
when the trajectory has been well established and more accurate predicts become avail-
able, the spacecraft can be transferred to the large 64-meter antenna. To hrther increase
the 26-meter antenna's ability to view the spacecraft in the event of significant trajecto-
ry dispersions, a small, wide-beam acquisition antenna is attached to the outer edge of
the 26-meter antenna and fed to a separate receiving and display device. Station oper-
ators can use the acquisition antenna data to help point the 26-meter antenna to "lock
on" to the rapidly moving spacecraft.

The designated initial acquisition station for T/oyager 2 was DSS 12 at Goldstone. The
station had been configured according to the "Initial Acquisition Plan," the necessary
prediction data for antenna pointing angles and rates had been loaded into the anten-
na pointing system (APS) and the receiver frequencies and Doppler offsets had been
updated based on the latest spacecraft frequencies measured just before liftoff, Similar
preparations had been made on board the NASA tracking ship Vdngtldrd,which was on
station in the Indian Ocean, and the Goddard STDN tracking stations on Guam, Kwajalein,
and Hawaii.
AU of the Titan and Centaur powered flight events observed by the downrange track-
ing stations appeared to be "nominal." About sixty minutes &er liftoff, the spacecraft
separated from the Centaur, burned its propulsion module engine for 45 seconds, and
was finally injected into a direct orbit for Jupiter. Within the next several minutes the
spacecrafi carried out several pre-programmed maneuvers. After turning to point the roli
axis toward Earth, the science and RTG booms were deployed, and the spent propul-
sion module was jettisoned. Then, with its S-band downlink transmitter turned ''on"
and its low-gain wide-beamwidth antenna pointed to Earth, it moved rapidly away from
Earth orbit, passing high across the Indian and Pacific oceans toward the west coast of
the United States. In rapid succession, each of the antennas on Vdnguard,Guam, Kwajalein,
and Hawaii briefly acquired the downlink signal as the spacecraft rapidly passed through
their field of view.

Finally, approximately 1 hour 14 minutes after launch, the spacecrafi signal appeared in
the acquisition receiver at DSS 12. The receiver was immediately "locked up one-way"
on the spacecraft signal and the 26-meter antenna put into the auto-track mode. A few
minutes later, the DSS 12 transmitter was turned "on'' and the uplink frequency swept
across the spacecraft receiver rest frequency Immediately, the spacecrafi receiver locked
up on the DSS transmitter signal and, after a small readjustment of the downlink fre-
quency, the DSS established two-way acquisition of Voyager for the first time.

Once the first tracking station had established an uplink and d o d i n k with the Ihyuger
spacecrafi, the telemetry, command, and radiometric data capabilities of the DSN became
available to the Voyager Mission controllers. As the spacecraft continued to move away
from Earth, its apparent eastward motion slowed and evenrually reversed to the normal
westerly motion as Earth rotated beneath it. A few hours later, Voyager set at Goldstone
and the spacecraft was transferred to the Canberra stations without incident. DSN sup-
port for Voyager had begun.

The powered flight and initial acquisition sequences for the V y g e r 1 launch on
5 Seprember 1977, followed essentially the same pattern of events as described above
for Vqydger 2, so that by early September 1977, the DSN had two additional planetary
spacecraft to deal with, both of them bound for Jupiter.

DSN support on the 26-meter Network for both Wyagers I and 2 during the last quar-
ter of 1977 was fairly routine. The priority enjoyed by the project at that time was not
curtailed by scheduling conflicts with other inflight missions. The mission controllers were
involved with trajectory correction maneuvers, optical navigation imaging, analysis of space-
craft problems, and the calibration of various science instruments (the High-Gain Antenna
and the S- and X-band communication links. During this time, the telemetry data rate
was running at 2,560 bits per second, and both spacecraft were operating well.

In November 1977, both spacecraft crossed the orbit of Mars to begin a nine-month
crossing of the asteroid belt. By the end of the year on 15 December 1977, both space-
craft were more than 1 AU from Earth, with Voyger 1 having overtaken Voydgeer 2.
One-way radio communication time was 9 minutes 49 seconds for V~yageerI , and 9
minutes 40 seconds for Voydger 2. Kyager 1 continued to increase its lead and would
arrive at Jupiter in March 1979, four months ahead of Vyageer 2.

In 1978, routine cruise operations continued normally on both spacecraft until early
April. In the course of switching from primary to backup receivers on 5 April, it was
discovered that the backup receiver in Vyagger 2 had developed a fadt in its carrier track-
ing loop. Analysis by the project and DSN teIecommunications engineers determined
that a capacitor in the receiver carrier tracking loop had probably shorted. However,
soon after switching back to the prime receiver, it was found that it also had developed
a fault that proved to be permanent in nature. Since the spacecraft was unable to receive
commands fi-om Earth on the now-permanently failed primary receiver, intense efforts
were made by the DSN to develop a strategy that would permit commanding the space-
crafi using the partially disabled secondary receiver, despite its failed tracking loop5

The carrier tracking loop enables the spacecraft receiver to follow changes in the uplink
carrier frequency caused by the relative movement of the ground transmitter as Earth
rotates. This is the radio telecommunications equivalent of the well known Doppler
effect, and knowledge of its precise value is an essential element in the technique of
spacecraft interplanetary navigation. When no uplink is present, the spacecrafi receiver
remains active at a frequency called its "rest frequency." To activate an uplink and send
commands, the uplink transmission must first be tuned to this rest frequency, after which
the spacecraft receiver tracking loop will lock-on and follow the uplink frequency changes
due to Doppler throughout the tracking period.

The strategy that finally evolved from these efforts, to compensate for the inoperative
tracking loop, required the DSN to slowly change rhe transmitter uplink frequency dur-
ing a spacecraft tracking period in such a way that the frequency changes due to Doppler,
as seen by the spacecraft receiver, would be cancelled out. The receiver could then lock
on to the DSN uplink at its rest frequency. No tracking loop action would be neces-
sary and commands would be received by the spacecraft in the normal way.
On 13 April, the Voyager 2 recovery sequence, which was based on this strategy, was
initiated by DSS 63. In less than an hour it was confirmed that the first commands
had been received by the spacecraft receiver. Spacecraft controllers immediately sent a
large sequence of commands to suspend all but essential spacecraft activity; these com-
mands would minimize any risk of losing the vital uplink wMe they set about evaluating
the spacecraft condition and determining how to deal with it in the future.

Throughout the emergency, TJoydg 2 had been tracked continuously by the 64-meter
antennas with the cooperation of the Viking, Pioneer, and Helios projects, which were
also dependent on those antennas for their mission support. Only the 64-meter anten-
nas could receive the Iow-power S-band signals that Eyageer 2 had been transmitting
when the emergency occurred. As soon as possible, commands were sent to turn the S-
band transmitter to high power so that the spacecraft could be tracked by the 26-meter
antennas, thus relieving the pressure for 64-meter antenna support.

The Eyuger 2 emergency placed an additional burden on the DSN operations organi-
zation in terms of planning, tracking analysis, and real-time operations. It occurred during
a period when there was a large number of very significant changes in progress through-
out the Network. These changes were associated with the MDS implementation, discussed
in more detail later. Suffice it to say here that in 1978 new telemetry, command, track-
ing, radio science, and data delivery systems were installed at all sites. DSS 12 was
withdrawn from operations and upgraded from a 26-meter, S-band station to a 34-meter
S- and X-band station.

Apart from the actual installation and testing, all of these changes necessitated new pro-
cedures and retraining of the Network operating personnel, while maintaining a high
level of support for all of the inflight projects. NevertheIess, by using innovative opera-
tional procedures to economize on station preparation time, and detailed negotiation
and scheduling of the available resources, the DSN was able to maintain satisfactory
tracking and data acquisition support for all ten interplanetary missions in 1978.

Over the next few months, the Tracking-Loop-Capacitor (TLC) tuning sequence was
greatly refined and would eventually become a routine feature of DSN operational sup-
port for I/ayger 2.

As a result of this timely recovery action, the Voygeer 2 mission was able to resume
throughout 1978 with the high expectation of achieving all of its Jupiter mission
objectives.
Both spacecrafi continued to move through the 1978 cruise phase of the mission with-
out incident. Voyager support in the last few months of 1978 was reduced somewhat
to allow the DSN to divert resources to support the Pioneer Venus and Multiprobe mis-
sions. Vyager I emerged from its crossing of the Asteroid belt in September followed
in October by Voyaer 2. No events of any significance were reported. By the end of
the year, the project attention was focused on a new phase of the mission, due to begin
on 4 January 1979. It was called the "Observatory Phase" and would be the prelude to
the Jupiter encounters.

Pioneers 6 1 1

Since the start of the second-generation Pioneer program beginning with Pioneer 6 in
1965, the DSN had introduced many modifications toward improving the threshold
capability of the Network for Pioneer telemetry support. Seven 26-meter stations were
equipped with improved masers, more efficient microwave equipment, linear antenna
polarizers, 3-Hertz receiver carrier tracking loops, advanced demodulation hardware, and
sequential decoding s o h a r e (for support of Pioneer 9).By 1969, these systematic improve-
ments had resulted in an extension of the Pioneer telemetry threshold on the 26-meter
antennas from 0.4 AU (1965) to 1.5 AU (1969).' The availability of the 64-meter sta-
tions increased the telecommunications link margins for the Pioneers so much that they
continued to be a significant presence in the DSN mission set throughout the Viking
Era (Figure 3-1).

The Pioneer 1 0 and I1 missions to Jupiter were designed to investigate the interplane-
tary medium beyond the orbit of Mars, the asteroid belt, and the near-Jupiter
environment, hitherto unexplored regions of interplanetary space at that time. The space-
craft incorporated many technological advances over those in the interplanetary Pioneers
6 9 , although the design philosophy was the same and the same design team (NASA's
Ames Research Center and TRW Systems) developed and built the spacecraft.' The
spacecraft were launched in 1972 and 1973, respectively, to solar system escape trajec-
tories. Pioneer 10 went on to become the first spacecraft ever to penetrate the asteroid
belt and, in December 1973, to obtain close-up images of Jupiter. Pioneer 11 reached
Jupiter one year later and was successfully retargeted for an encounter with Saturn in
September 1979.

The DSN supported both flights with the 26- and 64-meter stations, but there were
serious conflicts with the overlapping Mariner 1 0 mission to Mercury and Venus, dur-
ing 1973 and 1974, and the Viking requirements for prelaunch test time, and for Mars
radar observations from DSS 14 at Goldstone in 1975.
In March 1776, the Pioneer project was estimating a useful spacecraft life to 1983 for
Pioneer Wirh the then-existing 64-meter station, the downlink teleco~munications
limit at minimum bit rate (16 bits per second) would be reached in mid-1780 at a
range of about 22 AU, which is 2 AU beyond the orbit of Uranus.

Pioneer was already pushing the outer limits of long range deep space communication
links, but there was a strong motivation to push the uplinks and downlinks even fur-
ther, hopefidly to the point of matching the projected end of useful spacecraft life.
Downlink performance improvements were also of interest for the Pioneer 11 Saturn
Encounter, where the highest data rate possible was desired to maximize the imaging
science data return. The predicted downlink telecommunications performance for Pioneers
10 and 11 with both 26- and 64-meter antennas is shown in Figure 3-9.

At the time, there were a number of possibilities for improving the existing downlink per-
formance of the DSN antennas; these are listed in the table below. The table also gives the
factor by which the resulting telecommunications downlink margin would be increased (in
decibels) and the multiplying fictor by which the maximum range would be increased.

Possible Downlink Performance Improvements for Pionem 10 and I I

Downlink
Margin Increase Range Increase
Item Improvement (decibel) Factor

(1) Reduce receiver tracking 1.0

loop (BW = 3 Hertz)

(2) Improved low-noise 0.7

feed-cone (18.5 kelvin)

(3) Reshape the 64-meter 0.5

subreflector

(4) Increase 64-meter antenna 0.8 1.10

diameter to 70 meters

With the 1976 configuration of the 64-meter antenna, Piunee~10 would reach the limit
of its telecommunications downlink at about 22 AU in mid-1980. This limit could be
extended to 24.6 AU, which the spacecraft would reach in early 1981, by using the 3-
Hertz receiver tracking loop bandwidth then available in the new Block IV receivers.
The improved S-band feed-cone would reduce the downlink system noise temperature
Figure 3-9. Downlink performance estimates for Pioneers 10 and 71.
to 18.5 kelvin and extend the potential range even further to 26.8 AU, which Pioneer
10 would reach at the end of 1981.

Although the last two items would have extended the maximum range to 31.3 AU in
early 1983, coincident with the projected life of the spacecraft, it was not thought at
the time that they could benefit Pioneer 10, not planned for Network impIementation
until h e r 1983.

Over the course of time, these estimates turned out to be grossly pessimistic and, as
subsequent chapters will show, Pioneer 10 continued to maintain its downlink with the
DSN until it finally vanished on 31 March 1997. Pioneer 11 obtained some benefit
from the low-noise feed-cone but not enough to enable it to achieve its desired down-
link telemetry rate of 1,024 bits per second.

Following their successful Jupiter encounters in 1973 and 1974, the Pioneer 10 and 1 I
spacecraft moved through the Viking Era, with two essentially trouble-free spacecraft on
solar system escape trajectories. Along the way, Pioneer 11 would return imaging data
from an encounter with Saturn in September 1979.

The Pioneer 10 and 11 missions would utilize every available enhancement to the DSN
downlink capabilities to extend the life of the missions and to enhance their interplan-
etary science return. The longevity of the spacecrafi and this approach to mission design
and planning would yield far-reaching results into the future, much further than could
have possibly been foreseen in that era.

Investigation of the planet Venus continued in 1978 with the launch of two spacecraft,
one of them a finw Orbiter, the other a finus Multiprobe. Managed by Ames Research
Center as part of the very successful ongoing Pioneer program, the Wnus Orbiter was
launched by Atlas-Centaur in May 1978, and the Venus Multiprobe in August 1978.
These rwo launches would also be noteworthy because they were last expendable launch
vehicles to be used for planetary launches before the newly developed Space Shuttle took
over all deep space launches in the 1980s.

The Orbiter reached Venus on 4 December, followed five days later by the Multiprobe
on 9 December. Afier insertion into Venus orbit, the Orbiter mapped the Venus surface
by radar, imaged its cloud systems, explored its magnetic environment, and observed
interactions of the solar wind. Advantage was taken of its many occultations to conduct
Earth-based radio science experiments to determine the refractivity, turbulence, and ionos-
pheric structure of the atmosphere.

O n arrival at the planet, the Mult+robe spacecraft (bus) released four heat-resistant probes
to penetrate the Venus atmosphere at widely separated locations. Each probe contained
an S-band radio transmitter that continued to return telemetry measurements of atmos-
pheric temperature, pressure, and density until it impacted the planet surface.

Both Pioneer Venus 1978 missions presented significant challenges to DSN engineering
because they required a substantial amount of the new equipment and new technology
then being installed as part of the MDS implementation program. In addition, the
Multiprobe mission presented unique operational problems because of the need to rap-
idly acquire and simultaneously track all four probes as they were released into the Venus
atmosphere from the bus.

Shortly afier arrival and insertion into Venus orbit, the Orbiter began a long series of
Venus occultations, during which the DSN conducted ninety S- and X-band radio sci-
ence observations as the spacecraft alternately entered and exited obscuration by the
planet. Open-loop and closed-loop receivers together with wide- and narrow-band dig-
ital recorders were used to capture the S- and X-band radio science data on each occasion.
Between occultations, normal telemetry data capture continued.

The initial occultations occurred only over DSS 14. The spacecrafi periapsis passage was
allowed to move later each day until it reached the DSS 14lDSS 43 mutual view peri-
od in January 1979. Thereafter radio science was conducted at both stations until passing
completely to DSS 43 later in 1979.9

Occultation radio science observations on the Orbiter were suspended for two days
on 8 and 7 December to allow the stations to give their full attention to the probe
entry event.

DSS 14 and DSS 43 had prime responsibility for the supporting probe entry events in
real time. They were to provide telemetry and Doppler data, and analog recordings of
the open-loop receiver outputs, and to establish and maintain an uplink to the large
probe during the entry and descent period. After the probes had impacted the surface,
they were to support the bus until it, too, entered the atmosphere and was destroyed.
In conjunction with the stations of the STDN at Santiago, Chile, and Guam, they
would also provide recording for the Differential Long Baseline Interferometry (DLBI)
experiment.lo
A special configuration had been implemented as part of the MDS at DSS 14 and DSS
43 for the Pioneer Multiprobe mission to allow simultaneous reception of five carriers,
and for the processing of four carriers in real time. For backup purposes, four open-
loop receivers with bandpass filters and analog recorders allowed for recording a wideband
spectrum around the anticipated carrier frequencies of the four entry probes. These data
could be played back in non-real time at CTA 21, and converted back up to S-band
for reception and processing the Mark 111-75 telemetry equipment.

DSS 11 and DSS 44 would support the bus during the entry period with command,
telemetry, and Doppler data while the 64-meter stations were tracking the four probes.
DSS 42 would provide equipment for generating Doppler data from the third small
probe. At DSS 14 and DSS 43, the operational complexiry of dealing with ten signal
flow paths for the four probes and the bus gave cause for concern. R e c a n g that these
paths were all active simultaneously and that the time available for activating them once
the probes were released was about five minutes, it is not surprising h a t the operator
crews at the stations approached the Multiprobe entry event with some trepidation.

The fact that the entire event was supported perfectly, and resulted in the capture of 90
percent of the possible data from all four probes during the brief period (50-60 min-
utes) of their descent to the surface of Venus, attests to the care in planning, thoroughness ,

of the training, and the skill and dedication of all the DSN people involved.
NETWORK ENGINEENNG AND IMPLEMENTATION

THEDSN M k u c 111 SYSTEM

DESIGN

The enduring design of the DSN Mark I11 Data System originated with a presentation
by Wdter K. Victor, N. A. Renzetri, J. R. Hall, and Douglas J. Mudgway to the Tracking
and Data Acquisition Organization and the Tecknical Divisions in October 1969. At
that presentation, Victor first proposed a radical new approach to the DSN architec-
tural structure." It would be called the DSN Mark 111-73 and would include the DSN,
NOCC, GCF, and the MCCC. Due to institutional restructuring, the MCCC would
be separated from the DSN in 1972, and thereafter was excluded from the scope of the
Mark I11 Data System.

Walter K. Victor deserved to be numbered among the founders of the DSN. Bringing
with him an engineering degree from the University of Texas (1942) and a brilliant
mind, Victor came to JPL in 1953 to work on radio guidance systems for the Corporal
and Sergeant missiles then being developed for the Army He was soon collaborating
with a colleague, Eberhardt Rechtin, in developing more efficient receivers for tracking
very weak radio signals from distant missiles or space probes. Based on ccphase-lock"
tracking principles developed earlier by Rechtin and R. Jaffe of JPL, the ccphase-lock
receiver" eventually became a key element of all deep space tracking stations. Walt Victor's
association with the "phase-lock receiver" became legendary throughout the DSN.

When JPL made the transition to NASA in 1958, Victor became Manager of
Communications Systems Research in Rechtin's Telecommunications Division. In the
years that followed, he moved to successively higher positions of authority within the
tracking and data acquisition organization. During that time, he directed many of the
major technical advancements that led to improvements in the performance, reliability,
and productivity of the Network. He Iefi the Tracking and Data Acquisition Office in
1973 for higher responsibility elsewhere at JPL.

Td, autocratic, and conservative in approach, Victor would not "suffer fools gladly." As
one of his engineers described his meetings with Victor, "With Walt Victor, there was
no middle ground, no gray area. You were either right or wrong; you either knew, or
you did not know, the answers to his always detailed questions." His penetrating mind,
breadth of experience, piercing gaze, and abrupt manner brought terror to many an
inadequately prepared engineer in the course of a technical presentation.
A very private man, not much given to socializing, Victor elicited great respect from his
coworkers, at both JPL and NASA Headquarters, by virtue of his unequivocal com-
mitment to excellence in every problem, task, or project he addressed. He retired fiom
JPL in 1987.

As visualized by Victor, the DSN Mark 111-73 Data System would comprise one sub-
net of 64-meter diameter antennas and one subnet of 26-meter diameter antennas
supplemented by an additional 26-meter subnet shared with the Manned Spaceflight
Network. Two launch support stations, one at Cape Canaveral (MIL 71) and one at
Johannesburg, South Africa (DSS 5 I), would provide near-Earth tracking coverage for
planetary missions, and a spacecraft-to-DSN compatibility test facility at JPL (CTA 21)
would be used to verify the integrity of uplinks and downlinks before a spacecraft moved
to the prelaunch phase of activity. The station in South Africa was eventually dropped
from the plan for political reasons. Network control was to be exercised from a central
facility at JPL (NOCC), which would have the necessary capability for the processing,
display, and storage of network control data. (In its original form, the plan also includ-
ed a data processing facility for mission control, but this became irrelevant when the
MCCC was separated from the DSN.) The addition of a "simulation center" to the
DSN would provide simulated spacecraft data streams for use in end-to-end testing of
new installations at the tracking stations over real communications circuits, and train-
ing station personnel in new operationd procedures.

Two important, new, and lasting basic concepts were introduced to form the basis for
future design of the Mark I11 data system: closed-loop control of major data systems to
minimize loss of data, and specific validation by the DSN of all spacecraft data deliv-
ered to the flight projects.

The first of these, "closed-loop control", implied the development of a new capability
that would continuously monitor the status of the Network and the real-time data streams,
compare these to preestablished standards, and generate alarms when these values exceed-
ed specified limits. Immediate corrective action would follow. The new capability would
become the DSN Monitor and Control System.

The "data validation" concept implied no dteration of, or delay to, the incoming space-
craf? data by the validation technique. The validation process also included the ability
to determine the status of the DSN in real time, and to make these data available to
the flight prcjects.
Finally, as a basic framework on which to design and manage the future Mark I11 Data
System, Victor identified a hierarchy of "systems and subsystems" for the Network and
its (then) three facilities: the SFOF, GCF, and DSIF. (As mentioned previously, the SFOF
was later deleted from the plan and the DSIF became known as the DSS.)

The definition of the DSN Systems was logically based on the h c t i o n s to be per-
formed, and became formalized as listed below:

DSN Tracking System

DSN Telemetry System
DSN Command System
DSN Monitor System
DSN Simulation System
DSN Operations Control System

A similar hierarchy was identified for each of the facilities based on its specific hnc-
tions as an eIement of the Network.

Because of its intuitive simplicity and inherent ability to create cleat manageable inter-
faces between them, the system hierarchy for the DSN, GCF, and DSS continues in
use today throughout the Network.

In a form substantially the same as originally proposed, the Mark 111-73 Data System
was impIemented in time to support missions in the closing years of the Mariner Era,
and was soon followed by an upgrade to a Mark-111-74 model to support the major
new missions of Pioneer and Helios. Then came the Viking mission to Mars.

The process of DSN implementation to support missions in the Viking Era was essen-
tially one of addition and modification. Except for the 64-meter stations in Ausualia
and Spain, which had come into service in 1973 and 1974, respectively, and the NOCC,
all the facilities had been in existence and had supported many previous flight projects
in more or less the same form for several years? Further, the network configuration
required for Viking launch and early cruise (up to February 1976) was substantially that
which already existed. The major increase in capability was not required for fid plane-
tary testing with the project until February 1976, well before the planetary encounter
in June 1976. Therefore, the actual implementation was carried out in phases corre-
sponding to the phase of the mission activity for which it was required.
Prior to March 1972, Network implementation plans had included the SFOF, since it
was at that rime a part of the DSN. Following reorganization of the Offices of Tracking
and Data Acquisition and Computing and Information Systems in March 1972, the
SFOF was separated from the DSN, and the plans for a Network upgrade were rearranged
to encompass only the deep space tracking stations, compatibiliry test stations, ground
communications, and the NOCC. The former SFOF then became h o w n as the Mission
Control and Computing Center (MCCC) to distinguish it from the NOCC, which
resided in the same building (230) at JPL. The DSN and SFOF functions had origi-
nally been conducted as a single entity under the control of the DSN. They were separated
in 1961 only to be recombined in 1963 and separared again in 1972.'

During 1972 and 1973, implementation activity related to Viking was relatively srnd,
partly because the Viking mission was being replanned from the cancelled launch in
1973 to a 1975 launch, and also because DSN attention was focused principally on the
Mariner Venus-Mercury mission.

By September 1973, however, implementation for the new CTA 21 had begun to move
forward, and the longer range Viking implementation was findiy planned. CTA 21 was
urgently needed to allow the all-important Viking multiple spacecraft testing to get
underway.

The implementation of CTA 21 made such good progress that a year later (August
1974) the Viking radio frequency and data system compatibility tests had been com-
pleted using two Orbiter streams and one Lander stream simultaneously between CTA
21 and the Mission Control Center. In the course of these tests, up to six data streams
had been passed across the interface and processed correctly, simultaneously with the
generation of two command links. This completed the implementation of CTA 21.
Implementation at the STDN-MIL 71 site at Cape Kennedy and DSS 14 at Goldstone
for support later in rhe year was in progress.

The end of 1974 saw the Block TV receiver instalIed at MIL 71 and the new station
integrated with the NOCC. Block N receivers for DSS 43 and DSS 63 were in pro-
duction and expected to be installed before Viking launch, then projected for July 1975.
Construction of facilities for the overseas high-power transmitter instdarions were about
complete at the Madrid site while an Ausrralian contractor was about to begin work in
Canberra.

Despite the interruption for most of July to recodigure for the Helios launch, and a
further diversion to cover the Pioneer I 1 Jupiter Encounter in November 1974, the
installation of the Viking planetary configuration was complete at Goldstone stations
DSS 14 and DSS 11 by January 1975. Testing of the new capabilities began immedi-
ately and soon indicated areas in which improvements were needed, particularly with
regard to the analog and digital recording of original data records. Implementation of
the Viking planetary configuration at overseas stations in Madrid and Canberra were
slowed somewhat by the restrictions on configuration changes imposed during the Viking
prelaunch test period.

Nevertheless, all implementation required for launch was completed by the time of the
first DSN-Viking Launch Readiness Review on 9 July 1975, and the DSN was declared
fully prepared to support both Viking launch and cruise operations.

DSN MARK 111-75 MODEL

With the advent of full planetary operations on 20 July 1976, following the successful
landing of Eking Lander I, the entire resources of the DSN were required to provide
simultaneous support for the two Orbiters and the Viking Lander.

In addition to Viking, the DSN was also being called upon to provide support for
Pioneers G I 1 and the Helios I and 2 solar missions. The extent of the coverage involved
can be adjudged from Figure 3-1, which shows all the missions in flight at the time.

The support required by all of these users of the Network encompassed telemetry, com-
mand and tracking support, error-free data transmission services fiom Madrid, Spain;
Canberra, Australia; and Goldstone, Cdifornia; as well as Network control and moni-
toring and production of complete data records.

The resources included three 64-meter stations, six 26-meter stations, DSN and
NASCOM ground communications facilities, and the NOCC. The configuration of the
Nerwork was then called the DSN Mark 111-75.

The DSN Mark 111-75 was, at this time, the latest in a progressive succession of con-
figurations or models based on the original DSN Mark 111 Data System design beginning
with the Mark 111-73. Each Model represented a new increment of DSN capability,
driven by the tracking and data acquisition requirements of the future missions (see
Figure 3-1). The key telemetry characteristics of each of the models through the peri-
od 1973-76 reflect the steady development of DSN capability in this period. As can
be seen in the following table, additional telemetry characteristics were added to the
Mark 111-73 Model and the result was the Mark 111-74 Model, to which other char-
acteristics were added to create the Mark 111-75 Model.13

Key Telemetry Characteristics for Mark I11 System Models 73 Through 75

Deep Space Network Mark 111-73

* Multimission capability at each station for receiving and formatting
uncoded and block-coded telemetry.

Special s o b a r e for Pioneer convolutional coded telemetry.

Centralized monitoring of Telemetry System performance by observa-

tion of project displays in the MCCC.

* Capability at each station for a single carrier, dual subcarrier, and for-
matting for high-speed and wideband communications.

Recording of pre- and post-detection analog records with nonreal-time

playback.

Production of digital OriginaI Data Record (ODR) at each DSS and

playback via manual control or automatic response to project inputs.

* Real-time reporting of telemetry status at the station to the Monitor

and Control Subsystem.

Deep Space Network Mark 111-74

Multimission capability for sequential decoding of long constraint con-
volutionally coded telemetry data up to 4,000 bits per second using
the Fano algorithm, for both Pioneer and Helios.

* Real-time monitoring of Telemetry System performance at the NOCC.

Deep Space Network Mark 111-75

Capability at the 64-meter DSS to handle multiple carriers (up to
four) and multiple subcarriers (up to six) with decoding, ODR, and
formatting for communications circuits.
 * Centralized monitoring of Telemetry System performance at the
NOCC, and reporring via the DSN Monitor and Control System.

* Central logging (on a Network Data Log) of all data received at the
DSS, with gap accounting, and automated recall of missing data from
the DSS ODRs.

* Generation of a time-ordered, gap-fiee record of all received data

called an Intermediate Data Record (IDR) for delivery to the flight
project.

Generation of Telemetry System predicts and configurations for trans-

mission to each DSS for use in the manual control of system
elements.
d2UJJJ.!2
~ 2 X P X X < ~ ? I . r C C C < . d 2 ~ " l I C ~ / . ~ . ' C X ~ C C C i . i C I ' / ? J r 7 r 7 I I I ~ Y I I . ~ I ~ C 2 ~ ~ f Z # ~ ~ ~ ~ ?"/f~R
CXC~
2Z~2~/ r2?
~ ICZ ~ ~
iC22C2Z~XZ~~<d1?~FZIX/~r7IPr'ddY~I~~~>2CI'J2~2I~XZC~rY2~fl.'d.'~2d?UXdd~2II~CX~If~iIfIirrlI~~~

The ground communications capability associated with the Mark 111-75 model provid-
ed for transmission of formatted telemetry data fiom each DSS to the NOCC and
MCCC. The communication circuits were furnished by NASCOM, and consisted of
one 4,800-bit-per-second high-speed line from each DSS, and one additional 28.5-
kilobit-per-second wideband line from each 64-meter DSS in Spain and Australia, with
50 kilobits per second from Goldstone.

A functional block diagram of the DSN Mark 111-75 configuration for the 64-meter
stations (DSS 14 at Goldstone, DSS 43 at Canberra, and DSS 63 at Madrid) is shown
in Figure 3-10.

Two Block 111 receivers and two Block IV receivers provided a redundant capability for
up to three radio frequency carriers simultaneously; e.g., two Viking Orbiters and one
Viking Lander. Block I11 receivers were capable of only S-band reception, while Block
IV receivers could receive at either S-band or X-band. Two open-loop receivers provid-
ed for radio science occultation and solar corona experiments.

Six subcarrier demodulators were connected by an extremely flexible switching matrix

to associated block decoders symbol synchronizers, and data decoders which, together
with the telemetry a d command processor, provided the capability to handle six simul-
taneous data streams, or two subcarrier data streams per spacecraft.
The block decoders were used to decode the block-coded data streams fiom the Viking
Orbiters at 16,000 bits per second and 1,000 bits per second from the Landers.
Convolutionally coded data fiom Pioneer or Helios were handed by the data decoders
at 512 bits per second and 4,096 bits per second, respectively. Uncoded data, up to
about 33 bits per second, could be handled directly by the telemetry and command
processor in addition to its other command and telemetry data handling tasks.

The Data Decoder Assembly (DDA) and the Telemetry and Command Processor (TCP)
delivered the processed telemetry streams to the ground communications subsystems for
transmission via global NASCOM high-speed and wideband communication circuits to
the GCF terminal at JPL. From that point the data could be directed to the appropri-
ate Mission Control Center for flight project use and to the NOCC for monitoring and
validation and recording purposes.

Under TCP control, two Command Modulation Assemblies and dual transmitter excirers
afforded redundant paths to either a 20-kilowatt or 100-kilowatt transmitter for uplink
command purposes, one uplink at a time. At Goldstone only, a 400-kilowatt transmit-
ter could be used in place of the 100-kilowatt transmitter, if necessary.

Radiometric (Doppler and ranging) data were generated by the station tracking subsys-
tem, which consisted of two Doppler counters and two Range Demodulation Assemblies.
This arrangement, together with a Planetary Ranging Assembly, provided S- or X-band
ranging simultaneousIy with S- or X-band Doppler. When tracking a spacecraft close to
the Sun, the Planetary Ranging Assembly could be replaced by the Mu-I1 Ranging
Machine, which embodied a different technology to give better ranging performance in
the presence of solar noise.

Digital data records were made by pairs of %track high-density tape recorders attached
to the DDA and analog records of baseband, and detected data were made by Ampex
FR1400 analog tape recorders. Two high-performance Honeywell machines were avail-
able for baseband playback at the stations when necessary.

Rubidium frequency standards were the basis for all station tracking and frequency ref-
erences. Interstation time synchronization to 20 microseconds was accomplished by means
of an X-band Moon bounce link from the Madrid and Canberra stations to the Network
master clock at Goldstone.

The Mark 111-75 Data System was designed to permit the maximum flexibility in switch-
ing and interchange of assemblies in the telemetry, command, and tracking subsystems.
 O P E N LOOP
RECEIVER

---------- -!
----

J
Figure 3-10. Functional block diagram of the DSN Mark 111-75 configuration for 64-meter stations.
 ~G&~GcSUBSYSTEM---------

------------

I----

-----.

- - - - -------
A N D P O S T D E T E C T I O N R E C O R D I N G SUBSYSTEM

i
?terstations.
This system also provided some degree of redundancy when stations were called upon
to support planetary operations for three spacecraft simultaneously This approach enabled
the stations to be rapidly reconfigured to support the different needs of the Viking,
Pioneer, and Helios inflight missions.

The specific configurations needed for each mission were separately defined and coded
into a Network Operations Plan for use by the station operators in manually setting up
the station equipment.

A hnctional block diagram of the DSN Mark 111-75 configuration for the six 26-meter
stations (DSS 11 and 12 at Goldstone, DSS 42 and 44 at Canberra, and DSS 61 and
62 at Madrid) is shown in Figure 3-1 1.

Because DSS 42 and 61 shared the same control room and some of the same equip-
ment as the 64-meter stations (DSS 43 and DSS 63), they were known as "conjoint"
stations in the Viking Era.

Two Block I11 receivers provided a capability for receiving two S-band downlink carri-
ers if necessary. Generally, they were used for only one carrier with a backup receiver,
since the two subcarrier demodulators that followed could accommodate only one low-
rate (33 bits per second) and one medium-rate (2,000 bits per second) data channel.
Block coding at 2,000 bits per second or less was accomplished in the data decoders
feeding the symbol synchronizers.

Two XDS 920 computers, known as TCPs, performed the data processing functions for
two telemetry channels and two command channels. Each TCP could format and out-
put telemetry data to a single GCF high-speed communications line for transmission to
JPL. Each TCP could also deliver command data to a command modulation assembly
for modulating a single S-band exciter and 20-kilowatt transmitter for transmission of
commands to the spacecrafi.

S-band ranging and differenced range versus integrated Doppler (DRVID), data gen-
erated by the planetary ranging assembly was available only at DSSs 12, 42, and 61.
At the conjoint stations, the ranging capability was shared with the 64-meter sta-
tions, so that either, but not both stations, could provide ranging support at any
time. Using their Block 111 Doppler extractors, all 26 stations could also generate S-
band Doppler. The Tracking Data Handling (TDH) assembly was used to sample
and condition the radiometric data for transmission to JPL via the GCF high-speed
communication circuits.

Digital data were recorded at the output of the TCE Baseband and detected analog
records were made at the input and output of the Subcarrier Demodulator Assembly.

At the conjoint stations, a single crew was used to operate both stations from the one
control room. To reduce the demands upon these crews during operations, the station
monitor and control hnctions were exercised from a newly developed Station Monitor
and Control Assembly known as the SMC 113. The SMC provided remote control of
some functions, such as receiver tuning, and, by bringing other monitoring functions
to a single console, permitted the station to be operated during periods of high activi-
ty with fewer operators than would otherwise have been required.

To meet an early Viking requirement for simultaneous commanding to two spacecraft

(Orbiterlorbiter or OrbiterlLander) from a single antenna, the DSN planned to provide
dual 10-kilowatt uplinks from a single 100-kilowatt transmitter.'*, l5 In early 1973, the
dual carrier single transmitter scheme presented technical difficulties that were not resolv-
able at the time and a recommendation was made that separate antennas with command
capability at the same longitude be used to meet the Viking requirements. However, fix-
ther evaluation of the uplink performance of the Viking Lander under worst-case conditions
revealed that, irrespective of the dual-carrier requirement, there remained a need for a
100-kilowatt transmit capability at the two 64-meter stations (DSS 43 and DSS 63), and
in addition, a 400-kilowatt transmit capability at DSS 14 at Goldstone. These latter
enhancements would "avoid unacceptable risks andlor constraints to the Viking Lander
operations, particularly Mission B." Accordingly, these big new transmitters now became
part of the Viking planetary configuration required at a l l the 64-meter stations.

Originally, the DSN stations included 10-kilowatt to 20-kilowatt S-band transmitters to

provide the command and tracking data uplink to the spacecraft. However, in 1970, a 100-
kilowatt high-power S-band transmitter had been installed at the new 64-meter antenna at
Goldstone, and planning was initiated to provide similar transmitters at the other 64-meter
antennas in Spain and Australia. Driven by the Viking requirement however, the plans were
accelerated to have the DSS 63 subsystem installed and transferred to operations for V i i g
support by July 1975, followed by the DSS 43 subsystem in mid-October 1975.16
To save time, vendors shipped many of the components directly to the overseas sites.
Other components were sent to the high-power transmitter test facility at (DSS 13)
Goldstone for testing before overseas shipment.

By June 1974, elements of the DSS 63 transmitter had been assembled into a test fm-
ture to simulate other parts of the transmitter; they were run up to fuU. power (1 megawatt)
for 24 hours. These tests assured sufficient warrage from the power supply group for
either a 100-kilowatt klystron (Virian X-3060) or a 400-kilowatt klystron, which was
being developed for the DSN at Varian but was not available at that time. In July, four
technical personnel from Spain arrived at Goldstone for a short period of training prior
to dismantling and shipping the subsystem to Spain for installation on the 64-meter
antenna. After following a similar procedure, technicians shipped the DSS 43 transmit-
ter to Goldstone in December 1974.

The construction of facilities for the new transmitter subsystem began in August 1974 at
DSS 63. The facilities consisted of a building to contain the transformer/rect5er, a filter
choke and control junction box, and a set of concrete pads for mounting the 1,750-horse-
power (1,300-kilovoltampere), 400-cycle-per-second motor generator set. Housing was also
required for the motor control center, auxiliary heat exchanger, the distilled water replen-
ishing unit, and underground water tanks. The high-power Mystron would be mounted
in the tri-cone on the antenna itself and connected via waveguide couplings to the anten-
na microwave system. High-voltage power supplies and control equipment were to be
mounted in the antenna pedestal, and a heat exchanger mounted in the alidade (non-
tipping part of the antenna structure) would supply the Mystron with distilled water coolant.

As the facilities became available, equipment was assembled and installed, and check-
outs proceeded rapidly. Final overall and load testing was completed without incident
and the 100-kilowatt high-power transmitter at DSS 63 became operational and avail-
able to support Viking in July 1975. Events at DSS 43 followed a similar pattern and
the 100-kilowatt transmitter became operational at DSS 43 in October 1975.

In April 1974, the high-power transmitter at DSS 14 was shut down for approximate-
ly four weeks to i n s d an experimental 500-kilowatt klystron that would raise the maximum
transmit power to 400 kilowatts as required for Viking.

Early in 1976, the klystron tube developed a short circuit in its filament assembly and had
to be returned to the mandacturer, Varian Associates, for repair. By that time, however, it
had been determined that the real advantages of the 400-kilowatt capability as compared
to the 100-kilowatt capability were slight, and the 500-kilowatt klystron was held as back-
up in the event of some unforeseen emergency.17 Fortunately neither of the Landers
for whose emergency support the backup was intended, ever had need to use it. Even as
the last Viking Lander fell silent in February 1983, an anaIysis of the telecommunications
link showed that at a transmitter power of 100 kilowatts nominal (actual power of 80 kilo-
watts), there was a 10- to 14-decibel margin for command with the Earth antenna
pointing-angle passing through the side lobes of the lander spacecrafi antenna.18

In contrast to the 1971 and 1973 Mars radar opportunities, which were conducted at
S-band frequencies, the 1975 Mars observations of the Viking landing sites were con-
ducted at X-band. This was due in part to the availability of an experimental high-power
(400-kilowatt) X-band transmitter at DSS 14, and in part to the increased emphasis on
an evaluation of the Mars landing site characteristics, particularly surface roughness.

The observations were due to commence in December 1975, while the Viking space-
craft were in the cruise phase of their missions.'"20

At that time,21.22 the experimental X-band transmitters included two, 250-kilowatt, Varian
VA949J klystrons in a four-port hybrid arrangement, which would combine the two
individual outputs into a single waveguide to deliver a nominal output power of 400
kilowatts. Klystron fGlures and other hardware problems plagued the early attempts to
make successful radar observations, but in the last days of December 1975, two good
observations were made ar a power level of 165 kilowatts.

As the Earth-Mars range increased in the first months of 1976, intense efforts were made
to increase the Mars radar transmitter power to its full 400-kilowatt rating. Varian Associates
contributed significantly to this effort by responding rapidly to DSN engineers' calls for
modification or refabrication of klystron components. While this work was in progress,
Mars radar data gathering continued as opportunity allowed. By the end of February
1976, problems in the crow-bar protective circuits, high-voltage power supply and kly-
stron body current had been identified and corrected, and the radar was delivering its
full rated output of 400 kilowatts at X-band. From that point on, aIl Mars radar tracks
were conducted at full-power output.

By the end of April 1976, the strength of the radar signal returns from the planet had
f d e n to about one-eightieth of their strength in December 1975, when Mars was clos-
est to Earth. A a consequence, the data became too noisy for practical use, and the
data processing technique was changed from the range-gated technique to a continuous
spectrum technique. This change resulted in less noise, but there was some loss in res-
olution compared to the former technique.

During May and June 1976, the Mars radar facility at Goldstone continued to support
an intensified program of Mars observations. Also during this time, additional observa-
tions of overlapping areas of coverage were provided by the Cornell Radio Astronomy
Observatory at Arecibo, Puerto Rico.

By mid-June, the data had become too noisy for analysis and the X-band radar support
for Viking was concluded.

The planetary radar work continued to expand and improve in the years ahead to become
a unique part of the DSN Science program. Eventually it became known as the Goldstone
Solar System Radar (GSSR) and is discussed in more detail in later chapters.

The improvement in uplink and downlink performance that could result from opera-
tion at frequencies higher than S-band (approximately 2,000 to 3,000 MHz) had long
been recognized in DSN engineering and planning. However, such improvements could
not be realized without corresponding improvements in the microwave components,
which together comprise the antennalmicrowavelreceiver subsystem. Simply changing
the operating frequency alone would not be effective.

A considerable effort in research and development had been in progress during the early
1970s to develop feed-cones,23.2**25 antenna surfaces, masers, and waveguide compo-
nents that would operate effectively at X-band frequencies (approximately 8,000 to 9,000
MHz) in the operational Network environment.

By 1973, this work progressed to the point where the first inflight demonstration of an
X-band deep space telecommunications link could be carried out using the Mariner 10
(Mariner Venus-Mercury [MVM]) spacecrafi to carry an SIX-band transp~nder.~~ In addi-
tion to the standard S-band uplink and downlink, the spacecraft transponder would
transmit an X-band downlink carrier signal that would be coherent with the normal S-
band carrier. The X-band carrier would not carry telemetry but could be switched to
carry a range modulated signal. The demonstration was officially called an SIX-band
experiment to distinguish it from the prime MVM mission objectives, and because the
outcome of the effort was far from certain.
There remained the problem of how to adapt the DSN 64-meter antennas to receive
X-band signals in addition to their existing transmit-and-receive capabiliry on S-band.
The DSN's Advanced Technology program provided the answer. Known as the "Dichroic
Feed," rkis complex microwave device is discussed in more detail in chapter 6 In con-
cept, this would be accomplished by adding an ellipsoid reflector over the S-band feed
and a dichroic mirror over the X-band feed, with each feed-horn mounted in a sepa-
rate cone structure. The dichroic mirror would pass an X-band signal directly into the
X-band feed, but would direct an S-band signal on to the ellipsoid reflector. The ellip-
soid reflector would then refocus the S-band signal to enter the S-band feedhorn.

The tri-cone on the 64-meter antenna at DSS 14 would be the first to be modified.
When this very complex installation was completed in early 1973, performance meas-
urements attested to the excellence of the design. No signal degradation was observed
at S-band, and measured loss at X-band was less than 0.2 decibels. There was no warp-
ing or distortion of the structures, sometimes caused by thermal effects, when the S-band
uplink was transmitting at full power.

The other components required for the SIX-band experiment, the new Block IV receiv-
er, the Block IV exciter, and a new S/X-band ranging system, were available by the
middle of 1973, and verified for compatibility with the onboard flight radio system.

During the early portion of the MVM mission, the SIX-band equipment experienced
many problems with noise, instability, and Doppler cycle slips. Gradually, the prob-
lems were identified and corrected, and by the Mariner 10 superior conjunction
during May and June 1974, the DSN was able to provide dual frequency SIX-band
Doppler, range, and open-loop receiver data for radio science support. This support
provided the first opportunity for spacecraft dual frequency analysis of solar corona
and gravity effects, and marked the first demonstration of X-band downlink capa-
bility in the Network.

The MVM SIX experiment had demonstrated the viability of an X-band downlink and
the considerable advantages of an SIX-band for radio science; this experience was to be
carried a step hrther in the Viking mission.

Like Mariner 10 (MVM), both V i n g Orbiter spacecraft were fitted with SIX-band
transponders; i.e., they would receive an S-band uplink and would uansmit an S-band
downlink together with a coherent X-band downlink.
For Viking support, both DSS 43 and DSS 63 were fitted with the dichroic plates and
the Block IV receivers and exciters necessary for S- and X-band receiving capability, as
had been done at DSS 14. Based on the results from the MVM mission, the Viking
radio science and ranging requirements were much more stringent in terms of noise,
stability, and cycle slips, than had been the case for Mariner 10.

Because of this new capability, the Radio Science experiments became an important part
of the DSN support for Viking. Earth occultation, solar corona, and general relativity
experiments were carried out successfiJly using the DSN SIX facilities.

In late 1975, at the end of the Viking Prime Mission, the Viking Radio Science Team
reported the following:

Since November 1975 when the X-band transponders on both Orbiters

were rurned on, several hundred S- and X-band Doppler and ranging passes
of excellent data quality have been acquired during Mars planerary opera-
tions. These data were ememely useful in evaluating the new X-band system
performance, monitoring solar flares and solar corona noise, and for the
charged particle calibration of Doppler and ranging data. The data afforded
the first practical demonstration of the validity of the dual frequency calibra-
tion of charged particle effects in the interplanetary medium. Also, the
achievement of a 0.1 percent accuracy in the general relativity experi-
ment depended entirely upon the near simultaneity of the S-band and
X-band tracking of an Orbiter and the S-band tracking of a Lander.27

The DSN SIX-band capability had arrived and would form the basis for all radio sci-
ence observations for aII flight missions from this time forward.

In mid-August 1976, the 64-meter antenna at DSS 14 (Goldstone) developed a poten-

tially serious problem when it became difficult to maintain a satisfactory oil film height
between the hydrostatic bearing and the azimuth bearing pads. The area of concern
lay between the 128- and 140-degree azimuth positions of the antenna. This partic-
ular area of the hydrostatic bearing had been a problem since the antenna first came
into service in 1966. The previous history predicted a failure condition arising about
every three months, although the film height had been stable for several months at
that time. The last maintenance that had been carried out about three months earli-
er (in June) used a shimming technique that had reached its limits and could no
longer be used for further maintenance. Further delay in correcting the problem would
only increase the chance of a catastrophic failure of the antenna's azimuth rotation
bearing.

A new technique, which required an 18-hour downtime on seven successive days, had
been proposed but never tried out, and therefore carried with it some question as to
how effective the proposed "fix" might be. Occurring as ir did at such a critical time in
the Viking mission, just before the second Mars landing, the situation and issues involved
were cause for great concern to the DSN.

Rather than proceeding with the "fix," a plan was put into effect which required
daily monitoring and reporting of the oil film height against an established ser of
safety criteria. In this way some warning of a deteriorating condition would become
visible, permitting a reevaluation of the risk involved in continuing operations. It was
also decided that a firm work plan for refurbishment would be scheduled for 15
November 1976, or sooner if the bearing showed any sign of deterioration prior to
that time. This plan worked very well. The oil film height remained stable through
5 November, at which time the operational usage of DSS 14 was such that the work
was carried out on alternate weekdays and brought to a successfu1 conclusion on 19
November 1976.

A further incipient problem had been known for some considerable time in the ball and
socket joint on the antennas at DSS 43 and DSS 63. An inspection carried out in late
1975 revealed that the ball joint that permitted the pad to rotate as it followed the con-
tours of the hydrostatic bearing runner was frozen because of improper lubrication. A
cross-sectional diagram of the hydrostatic bearing and the ball joint assembly is given
in Figure 3-12.

Prolonged attempts to force grease into the joint under high pressure were unsuc-
cessful, so a spare unit was shipped to Canberra in January 1976. However, prior to
installation on site at Canberra, this unit was found to be defective and was rerurned
to the United States. New units were ordered for both DSS 43 and DSS 63 only,
since the DSS 14 units were satishctory. Installation was postponed to coincide with
the Viking solar conjunction period in November and December, and the station
downtime for the MDS work in Australia. The installation was successhlly compler-
ed during that time.
r \

ALIDADE STRUCTURE

CORNER WELDMENT
WT = 1088 Mg

LUBRICATED
. I

SURFACE BOLTED
BALL AND INTERFACE

-
SOCKET JOINT
WT 2700 kg / /
FILM HEIGHT
HYDROSTATIC BEARING PAD
WT = 5600 kg
-- ------- ,- --
HYDROSTATIC BEARING RUNNER
4";
-
[
LAMINATED SHIMS

GROUND SHIMS
-
I _

-I---------------------

., 4 0
-

- - -

-
o a QGR&T.
( 0 * 0 0 -

SOLE PLATE
PEDESTAL

i J

Figure 3-12. Hydrostatic bearing and ball joint assembly.

DSN MARKI11 DATASYSTEM

PROJECT

The concepts of the DSN Mark I11 Data System established by Victor in 1969 were
updated, refined, and developed into a Mark I11 Project Plan by Easterling in 1974.28

Easterling's fundamental objective in planning the Mark I11 project was to provide sup-
port to flight projects in the 1973-1978 period and establish a standard, easily managed
interface with the MCCC at JPL and with other remote POCCs. A key b c o r in accom-
plishing this objective was seen to be the development of a Network Control System that
would be entirely separate fiom and independent of the hardware and software used by
the flight projects. The upgraded Network would include new technology in radiometric
data generation using two station radio-interferometry techniques, increased use of automa-
tion throug!nout the Network, and real-time transmission of spacecraft video data from the
DSSs to the MCCC using a global network of wideband communication circuits.

The plan advocated a phased design approach to hardware implementation with formal
reviews at the completion of each phase, culminating in acceptance tests and a formal
/ \

DSN MARK Ill NETWORK

t
NETWORK SYSTEMS (5)
0 DSN TRACKING
0 DSN TELEMETRY
a DSN COMMAND
DSN MONITOR & CONTROL
s DSN TEST &TRAINING

I
NETWORK CONTROL GROUND COMMUNICATIONS DSS SUBSYSTEMS (12)
SUBSYSTEMS (7) SUBSYSTEMS (5)
9 TRACKERS HIGH SPEED DATA ANTENNA MECHANICAL
* MICROWAVE
* TELEMETRY 0 WIDEBAND DATA RECEIVERIEXGITER
* COMMAND e VOICE
* TRANSMllTER
0 FREQUENCY&TIMING
MONITOR & CONTROL @ TELETYPE a TECHNICAL FACILITIES
PRVPOST DElECTlON

I I:E%Zy 1
* TEST &TRAINING MONITOR & CONTROL

1 /I
RECORDING
TRACKING
0 SUPPORT

D l s p
0 MONITOR & CONTROL
TEST & TRAINING
L
Figure 3-13.Mark 111-77 Network systems and subsystems.

transfer to operations status. An efficient, cost-effective s o h a r e design philosophy devel-

oped by a s o h a r e technology group would employ structured, top-down programming
techniques to provide for simple interfaces, well-controlled development, and easily mod-
ified programs. A phased design approach culminating in acceptance tests and transfer
to operational status would be used in producing all DSN s o h a r e programs. After
transfer, DSN software would be subject to a rigorous program of configuration con-
trol and managemenr, maintained by a centralized DSN S o h a r e Library.

The Network was defined to indude five Network systems: Tracking, Telemetry, Command,
Monitor and Control, and Test and Training. Lower tier subsystems were defined for each
of the three elements of the Network: Deep Space Stations, Ground Communications,
and Network Control. The Network systems and subsystems are shown in Figure 3-13.

The upgraded Network, which no longer included the SFOF, would be identified as
DSN Mark 111-77 and is shown in Figure 3-14.

Working within this hierarchical framework, engineers developed key characteristics for
all systems and subsystems which reflected the known and anticipated requirements of
the NASA Mission Set through 1981.
f
GOLDSTONE

" OPERATED BY STON + CONVERTED TO 64-MER

" INERFACES WITH OSS ANTENNA POST-1977
'" INTERFACESWITH MCCC

i I

Figure 3-14. General configuration of the DSN Mark 111-77 Network.

The telemetry system would take advantage of improved telecommunications capabil-

ity to handle data rates up to 115 kilobits per second at Jupiter distance. It would
accommodate multiple data streams and perform the block decoding, sequential decod-
ing, and maximum-likelihood convoIutional decoding functions, using standard
techniques. The command range and command data rate would be extended and an
improved "store and forward" command transmission process would be introduced.
Radiometric data types would then include long baseline interferometer data, as well
as planetary ranging and Doppler observables. The introduction of Hydrogen Maser
frequency standards in place of the older Rubidium Standards would improve the accu-
racy and stability of all radiometric data types to improve the uncertainties in spacecrafi
navigation targeting parameters.
Network operations would be conducted from an NOCC using dedicated facilities to
serve multiple missions at several locations simultaneously. It would also include a for-
malized DSN scheduling procedure based on a 7-day, 8-week, 52-week, and 5-year
Network allocation plan.

Improved management of the operational interfaces between the flight projects and the
DSN would result from the newly defined DSN Monitor and Control, and Test and
Training capabilities.

The GCF capability for duplex high-speed data transmission would be raised from 4,800
bits per second to 7,200 bits per second, while the wideband capability would increase fiom
28.5 kilobits per second to 50 kilobits per second &om all sites to JPL. Up to 230 kilobits
per second would be available fiom Goldstone to JPL. The GCF would create a GCF log
of all flight project data generated by the DSN and deiiver it on a negotiated schedule.

The DSSs would be upgraded with S- and X-band downlink capability for tracking and
telemetry and for acquiring up to five spacecraft in one antenna S-band beam width.

Finally, the plan proposed a management approach to carry out the Mark 111-77 proj-
ect. It included an overall Project Manager, as well as an Implementation Manager for
each of the Network Data Systems, and for Quality Assurance and Reliability. The man-
agers were to be supported by cognizant system and subsystem engineers. Guidelines
were also given for scheduling, and progress and problem reporting.

It was early 1974, and the window of opportunity for implementation of these major
changes in the Network was approaching. The window would open at the end of the
Viking Prime Mission (last quarter of 1976) and close just before the Pioneer Venus
Prime Mission started (second quarter of 1978).There was no time to lose. Implementation
Task Managers were appointed for the data systems, and began to develop technical
requirements derived from the key characteristics contained in the plan, and schedules
based on the rapidly approaching window.

The largest of the tasks included several of the Network systems, and for that reason
was called the Mark 111 Data Systems Implementation Task; it was established in February
1975 with Paul T. Westmoreland as Task Manager.

Paul T. Westmoreland started at JPL in February 1960 as one of the original engineers
in Renzetti's Communications Engineering and Operations Section. Except for a two-
year lateral transfer to the Mission Control and Computing department in 1977, he
rose through the DSN to become the assistant laboratory director for tracking and data
acquisition in 1997. Born and educated in California, Westmoreland received a B.Sc.
in eiecuid engineering fiom USC in 1960, just prior to joining JPL, and a master's
degree in the UCLA Engineering Executive Program in 1968.

Westmoreland's early work led to the introduction of digital computing techniques for
operational control of the Network in the mid-1960s. Based initially on Scientific Data
Systems (SDS) 910- and 920-type computers, and subsequently on Modcomp machines,
and later still, on distributed computing techniques, this approach provided the opera-
tional flexibility that was the key to the future success of the Network control process. As
the applications for digital technology proliferated in the Network, so did the requirements
for s o b a r e to drive the digital processes. In collaboration with Walt Victor, Westmoreland
introduced new and innovative s o b a r e standards that provided management with insight
and control of the cost and status of formerly runaway software development processes.

Later in his career, he directed the DSN Systems Engineering Office, where he was ulti-
mately responsible for the overall architectural design and potential of the Network.
In 1997, he was appointed director of the Telecommunications and Mission Operations
Directorate.

Paul Westmoreland was a pleasant though serious man for whom pride in his appear-
ance complemented obvious pride in his achievements. His well-cut suits with fashionabIe
ties were notable among his more casually dressed coworkers. He was an expert ball-
room dancer, and he cast gold costume jewelry as a hobby. He was highly respected at
JPL and at NASA Headquarters for both his technicd judgment and his fiscal astute-
ness. His management style owed much to his early years at JPL when Rechtin and
Victor held the positions that he would later hold. In a real sense, the Network's abil-
iry to support many of the major spacecraft-related projects described in this history
owes much to the technical foresight and budgetary awareness of Paul Westmoreland.

He retired horn JPL in 1998.

(MDS) IMPLEMENTATION
MARK I11 DATASYSTEM

A plan for implementing the portion of the Mark I11 project that involved the pro-
cessing and transport of all data within the Network was issued by I). T. Wesunoreland
in February 1975.29The "Westmoreland plan" was intended to embody the principal
features of project management, system and hardware design, and software design phi-
losophy enunciated by Easterling in the Project Plan.30
 TWO-MONIHS OOWNIIME FOLLOWED BY

Figure 3-15. DSN Mark Ill Data System implementation schedule, 1977.

This major Network-wide reconfiguration was called h e DSN Mark I11 Data System
(MDS) implementation task and involved all stations of the Network in sequence, through-
out 1977 until completion, on schedule and under budget, in mid-1978.

In carrying out this task, the DSN would replace all its obsolete SDS-920 computers
with new MODCOMP 11-25 minicomputers. Each of the nine tracking stations in turn
would have its old telemetry and command processors replaced with new machines and
software, and then be requalified to resume inflight support. The implementation task
schedule, which was carefully planned ro avoid impact with the inflight mission sup-
port activity and the launch of the new Voyager mission to Jupiter and beyond, is shown
in Figure 3-15.

Because of the size of the effort involved in this task, the basic implementation was
accomplished by an integration contract with Univac Corporation to provide for the
fabrication, test, and integration of the Modcomp and other assemblies to form sta-
tion-level configurations for onsite installation and test by traveling teams of DSN
engineering personnel. Three basic station-level configurations were designated, one
for the 26-meter stand-alone stations (DSS 11, 12, 44 and 62), one for the 26/64-
meter conjoint stations (DSS 42/43 and DSS 61/63), and a 64-meter configuration
for DSS 14 and DSS 63, as well as for CTA 21 at JPL and MIL 71 at Merritt
Island, Florida. Each configuration contained up to three of each of the following
major assemblies:

Telemetry Processor Assembly (TPA)-Assumed the telemeuy hnctions

of the existing DSS Telemetry and Command Processor (TCP) by
replacing the older XDS-920 computers with new Modcomp mini-
computer to form a separate Telemetry Subsystem. The existing
Symbol Synchronizer remained in place.

Command Processor Assembly (CPA)-Assumed the command hnc-

tions of the existing DSS Telemetry and Command Processor (TCP)
by replacing the older XDS-920 computers with new Modcomp mini-
computer to form a separate Command Subsystem. The existing
Command Modulation Assembly remained in place.

Metric Data Assembly (MDA)-Replaced the existing DSS Tracking

Data Handling Subsystenl (TDH) and functioned as part of the DSS
Tracking System.

Communications Monitor and Formatter (CMF)-Functioned as part of

the existing GCF to provide for inputloutput formatting of high-speed
data and wideband data between the MCCC and the CPA, MDA,
and TPA assemblies. It also carried out the function of making
Original Data Records (ODR).

Digital Instrumentation Subsystem (DIS) to Mark I11 Interface

(DM1)-A non-minicomputer assembly that provided for the intercon-
nection of the CMF, MDA, and TPA minicomputer assemblies, and
interfaced them with the existing DIS.

* Time Format Assembly (TFA)-A non-minicomputer assembly that

functioned as an extension of the existing Frequency and Timing
Subsystem (FTS) to provide Greenwich Mean Time (GMT) inputs
and various interrupt pulses to the TPA, CPA, MDA, and CMF.

Data Systems Terminal (DST)-A non-minicomputer assembly that pro-

vided a central means of connecting associated Megadata terminals
 (computer monitor displays) and G. E. Terrninets (computer input ter-
minds) to any two of up to seven minicomputer assemblies.

Each station-level configuration was designated an "MDS" and was based on the arrange-
ment shown in Figure 3-16.

In total, the MDS assemblies and peripherals occupied 214 DSN standard size cabinets
assembled into 87 individual modification kits, tested and shipped to the stations for
installation during the 2-year period of implementation.

As Viking conducted important relativity experiments with DSSs 14, 42, and 63, the
MDS implementation activity started at DSSs 12, 44, and 62. As each of these srations
was withdrawn in turn from Viking, Helios, and Pioneer support, the MDS reconfig-
uration was carried out and the station was returned to operational status following an
appropriate period of testing and crew retraining.

Figure 3-16. Basic organization of the Mark Ill Data System.

By mid-July 1977, DSSs 12, 14, and 62 had completed system performance testing
in their new configurations and resumed full operational support. Stations 42 and
43 were released from tracking support on July 15, began a ten-week period of recon-
figuration and retest, and were returned to operation in mid-October. By the end of
January 1978, the work had been completed at DSSs 61 and 63 and the stations
returned to operations support. The last of the six stations, DSS 11, returned to serv-
ice on 26 April 1978.

The accomplishment of this major reconfiguration throughout the entire Network,

in the midst of a heavy and demanding tracking support schedule that included two
major launches (Voydgers 1 and 2), presented a formidable and exacting task. That
it was carried out on schedule, without any significant problems, and without impact-
ing the ongoing flight support was a credit to all the engineering and operations
personnel involved.

DSN MARK 111-77 MODEL

At the completion of the MDS implementation task in April 1978, the new DSN model
Mark 111-77 would have all the capabilities needed for support of the remaining mis-
sions in the Viking Era Mission Set (see Figure 3-I), plus future missions including the
(then-planned) 1983 Galileo mission to Jupiter. These missions include Pioneers G-ll,
the Pioneer Venus and Pioneer Multiprobe, Helios, Viking, and Voyager.

The continued growth of the DSN capabilities, in response to flight project require-
ments, is obvious to anyone who examines the key characteristics of the telemetry system.
The additional telemetry capabiIities resulting from the MDS transformation of the Mark
111-75 model to the Mark 111-77 model are listed be10w:~'

1) High-rate X-band telemetry capability (up to 250 kilosymbols per sec-

ond) at both a 34- and a 64-meter subnet in addition to S-band
telemetry at all subnets.

2) Maximum likelihood decoding (Viterbi algorithm) of short-constraint-

length convolutional codes at all DSSs. Deletion of block decoding
after completion of the Viking mission.

3) Replacement of the TCP with a dedicated processor, the TPA, for

telemetry.
 4) Precise measurement of received signal level and system noise temperature.

5) Simultaneous reception of five carriers at a selected DSS for Pioneer

Venus Multiprobe (to be deleted after &e mission).

6) Replacement of the Data Decoder Assembly by incorporating its func-

tions into the TPA.

7) Formatting of all decoded data for high-speed or wideband uansmis-

sion to the NOCC and MCCC via the GCF.

8) Real-time arraying of signals received from two stations at the same

longitude.

DSN MARK 111-77 TELEMETRY

SYSTEM

A simplified block diagram of the telemetry system for the DSN Mark 111-75 model
is shown in Figure 3-17.

Prediction messages are initially generated at the NOCC for transmission by a high-
speed data line (HSDL) to the DSSs for the purpose of establishing the proper
telemetry modes and configurations. Such messages consist of predicted frequencies,
data rates, signal levels, and tolerances.

At the DSS, the radio frequency spacecraft signal is collected by the antenna, ampli-
fied by the Antenna Microwave Subsystem and passed to the Receiver-Exciter subsystem.
The radio frequency (RF) carrier is tracked by the receiver, and the telemetry spec-
trum is routed to the Subcarrier Demodulation Assembly (SDA) where the subcarrier
is regenerated, and the symbol stream is demodulated. The resulting demodulated
symbol stream is passed to the Symbol Synchronizer Assembly (SSA) where it is dig-
itized. The digitized stream for convolutional encoded data is then routed to the
Maximum Likelihood Convolutional Decoder (MCD) for decoding of short-con-
straint-length convolutional codes, or to the TPA for decoding of long-constraint-length
codes, or uncoded data. The digitized symbol stream for block encoded data is sent
to either the TPA or the Block Decoder Assembly (BDA) depending on the data
rate. All the data are formatted by the TPA for transmission by high-speed or wide-
band data line to the NOCC and MCCC.
At each DSS, an ODR of the decoded data is written in GCF blocks by either rhe
TPA for high-rate data or the Communications Monitor and Formatter (CMF) for
low-rate data. The data are passed to a high-speed or wideband buffer depending
on the data rate, and then transmitted to the MCCC at JPL or in some cases to a
Remote Mission Operations Control Center (RMOC), as in the case of Pioneer
(ARC) or Helios (GSOC) and, in parallel to the NOCC for network control and
monitoring purposes.

At the NOCC, a limited amount of decommutatian of engineering telemetry data

is performed to analyze system performance and to obtain certain spacecraft param-
eters usehl in controlling the Network. The NOCC also receives and displays DSN
Telemetry System parameters.

AI1 the data received by the NOCC either in real time or by recall from the ODR
are written to the Network Data Log by the Network Log Processor. The Data
Records Assembly provides for the recall of any missing data by activating the O D R
after each station pass to complete the NDL. It uses the NDL as a source to cre-
ate the IDR for delivery to the flight projecr. The IDR is a complete, time-ordered
record of all spacecraft data received at each DSS during a spacecraft tracking pass.

DSN Telemetry System performance is displayed and validated in the NOCC by a

comparison of the telemetry system monitor data, which are transmitted from the
DSS, with preset standards and limits for the particular configuration being used.

DSN MARK111-77 COMMAND

SYSTEM

In the DSN Mark 111-75 Command System model, the DSS command processing
function shared the same computer (XDS-920) as the telemetry processing function.
In the Mark 111-77 model, the command function resided in a new dedicated com-
puter (MODCOMP 11-25) to be known as the CPA. The CPA provided higher
reliability, greater processing speed, and more core memory for the command func-
tion. The increased processing speed allowed the DSN to support higher spacecraft
command data rates. In che TCP XDS-920, the 8-bits-per-second command bit rate
used by Helios was near the upper limit of the Mark 111-75 Command System.
Voyager required a command rate of 16 bits per second, and the new CPA would
accommodate this and even higher bit rates anticipated for future mission^.^^^ 33

Later, in 1978, using the increased data storage capability of the CPA, a new com-
mand technique known as "store and forward was introduced. This technique allowed
f -l
GCF C E W
DEEPSPACE COMMUNlCATlONS NFlWORK OPUIAnONS
STATIONS (DSS) rmMIN# {CCT) CONTROL CENTER (NOCC)

t \ f I

MNN4 MICROWAM TRANSMITER RECENER- DSS C O M W D

MECHAN- SUBSYSTEM SUBSYSTEM EXCITER SUBSYSTEM (OCD)
IW SUBSYSTEM
SUBSYSTEM
-+-r *
TRANSMIITER,
EXCITER SWCH
-
EXCITER
ASSEMBLY
- COMMAND COMMAND
WKCH MODUMOR
COMMAND
s
PROCESSOR
STAR
SWITCH
C
CATIONS
~ HIGH-SPEED
~
DATA UNES
~
GW HIGH-SPEED DATA SUBSYSTEM

ERROR ~
DETECTION
HIGH-SPEED
~
SWiTCH
~ NmYORK
LOG
~ - NmVORK
MMMUNI-
* STAR
SWCH
ASSEMBLY ASSEMBLY ASSEMBLY ASSEMBLY CON- MONITOR L (HSDL) CORREC- (HSW) PROCESSOR CAllDNS CDN-
(CSA) (CW (CP4 TROLLER FORMlTER nON (NLP) EQUIPMENT TROLLER
@SO (WO EOC) (NCQ (SSCI

----------------------------------
DSS 11,12,44,62PDSS12IS3dm)

-------

JPL MISSIONCOWIAOL AND

COMrnNG CENTER (MCCC)

DSS 42.61 ( 7 0 BECOME 34-m IN 1980)

FFIEQUENCY DSS MONiTOR

B TIMIMINO b CONlROL
SUBSYSTEM SUBSYSTEM
(W [DMC)
\ /

Figure 3-18. DSN Mark Ill Command System, 1978.

the mission operations controllers to prepare large files of spacecrafi commands in
advance and then f o m d them to the DSS at the beginning of a spacecraft track-
ing period. At the DSS, the files were stored on the CPA disk and placed in a queue
for radiation in the specified order at the designated time.

The DSN Command System as it existed after the MDS reconfiguration and sub-
sequent enhancements is shown in Figure 3-18.

In addition to these improvements, the overall Network command capability was

further enhanced by an increase in the uplink transmitter power at DSS 12 and
DSS 62 from 10 to 20 kilowatts to meet the Network standard.

DSN MARK111-77 TRACKING

SYSTEM

The art of spacecrafr navigation is based upon the three types of radiometric data
generated by the DSN Tracking System. The radiometric data types are S- and X-
band Doppler, S- and X-band range, and antenna tracking angles. Doppler and
range in various combinations are the most powerful data types, and are used direct-
ly in the spacecraft orbit determination process. The Mark 111-75 Tracking System
had provided these data types for Pioneer, Helios, and Viking. To support Venus
and Voyager, additional new capabilities were needed, and these were provided in
the Mark 111-77

By improving the DSS frequency stability by two orders of magnitude, the new
Hydrogen maser frequency standards that repIaced the older Rubidium frequency
standards extended the precision range and Doppler capability of the DSN Tracking
System to beyond 20 AU to meet the long-range navigation requirements for Voyager.
Radiometric data could now be time tagged to the 10-nanosecond level relative to
the DSN master clock.

Calibration of the DSN radiometric data for the effects of the tropospheric and
ionospheric media along the transmission path between spacecraft and Earth was
made possible by a new Meteorological Monitor Assembly that measured these data
at the stations and transmitted them via the new GCF to the Network users.

At the DSSs, all of the old hardware and software associated with the Tracking
System was replaced with new units based on the MODCOMP computers. The
planetary ranging capability was improved for operation when the line of sight to
the spacecraft passes close to the Sun, and a new unit called the MDA carried out
all metric data handling functions and controlled the interface with the upgraded
GCF.

At the NOCC, improvements were made in the generation of tracking data pre-
dicts based on a long span of trajectory data suppIied by the Bight project. The
display, real-time monitoring, and data validation Eunctions were upgraded to sup-
plement the DSS and GCF improvements in the Mark 111-77 model.

DSN MARK111-77 GROUNDCOMMUNICATIONS

For the Mark 111-77 model, the GCF capability that provided for transmission of
formatted telemetry data from each DSS to the NOCC and MCCC was also increased.
The high-speed data line rate from all stations was increased from 4,800 to 7,200
bits per second. The wideband capacity from each 64-meter DSS in Spain and
Australia was raised from 28.5 to 56 kilobits per second with a capacity of 230 hlo-
bits per second being available from Goldstone.

End-to-end encoding with new 22-bit error polynomials provided automatic error
detection and correction to significantly reduce the possibility of undetected high-
speed transmission errors. With this correction scheme the DSN was able to approach
nearly 100 percent delivery of real-time data, and cut down on the station time and
personnel effort previously required for post-track replay of missed telemetry data.
These capabilities as they were set up for Viking and the later missions of the Viking
Era, are shown in Figure 3-19.

In mid- 1978, the GCF high-speed-data subsystem was converted from the older line
switching method of routing messages that had been used for Helios, Viking, and
earlier missions, to programmable message switching capability. The upgrade also
included changes to the NOCC and MCCC interfaces with the GCF. However, the
line switching interface for Viking was maintained until March 1979. By then, mes-
sage switching was standard for all

CAPABILITY
DSN MARK 111-77 ARRAYING

As part of the Mark 111-77 capability for Voyager, a two-station arraying capability was
to be provided at each complex. The combining function was to be carried out at the
subcarrier level using the subcarrier spectrum outputs from a 64-meter station and a 34-
meter station to provide a single output for telemetry processing. The arraying was to
accommodate subcarrier frequencies of 24 to 500 kilohertz and symbol rates from 2 to
250 kilosymbols per second. The combining process was to introduce less than 0.2 deci-
bel degradation and to operate at a received signal-to-noise ratio as low as 0.8 decibel
from a 64-meter DSS and -5.2 decibels from a 34-meter DSS. This capability would
find its first operational application as the DSN entered the Voyager Era and the Voyager
spacecrafi approached its first encounter with Jupiter.36

DSN MARK111-77 DECODING

CAPABILITY

To accommodate the decoding requirements of the Viking Era mission set the DSN Mark
111-77 model provided for the following three types of telemetry channel decoding:"

Viking-Block Codes 32:6; biorthogonal codes handled by the BDA at

the 64-meter DSS at data rates up to 16,000 bits per second. At other
DSS, the block decoding is performed by the TPA software at rates
limited to 2,000 bits per second.

Pioneers 6 - 1 1 and Helios-Long-Constraint Length Convolutional

Codes (32:112); Sequential decoding by the Fano algorithm using TPA
s o h a r e at output rates limited to 4,000 bits per second.

Pioneer Venus, Voyager, and future missions-Short-Constraint-Length

Convolutional Codes (7:112); maximum likelihood decoding by the
Viterbi algorithm using the MCD at output rates up to 250,000 bits
per second.

The block decoding functions for Viking were not substanrially &ected by the transi-
tion to the Mark 111-75 model. The long-constraint-length, convolutional decoding
functions for Pioneer and Helios were also essentially unaltered except that they were
now carried out within the TPA by a separate processor that interfaced to the new
MODCOMP 11-25 computers.

However, the short-constraint-length, convolutional decoding function required by

Pioneer 78, Voyager, and fbture missions was an entirely new capabiliry.This new decoder
was called a Maximum Likelihood Convolutional Decoder (MCD) and utilized the
Viterbi decoding algorithm operating on short-constraint-length codes (K=7) to provide
a gain of about 5 decibels over uncoded data at data rates from 10 to 250,000 bits per
second. The obvious bit rate advantage of maximum likelihood decoding is due to the
fact that a large part of the computations can be performed in parallel, rather than in
sequence, as in sequential decoding.
The MCD was a special purpose digital computing device that received coded, quan-
tized symbols from an SSA and output decoded data to the TPA for formatting and
transmitting to the Mission Control Center at JPL. The h c u o n a l requirements for the
MCD and the factors used in their selection had been the subject of a great deal of
study development, test, and evaluation within the DSN for several years prior to their
implementation in the MCD.

The salient features of the MCD are described below:

1) Short-constraint-length (K=7) provided simplified encoder design, ade-

quate decoding performance, and reasonable computational load.

2) Minimum node synchronization time (grouping of incoming symbols

into pairs for rate 112 are triplets for rate 113).

3) Code rates of 112 and 113 are provided to satisfy Voyager and provide
growth capability for future requirements. (Note: For a given symbol
rate, rate 113 provides about 0.3-decibel improvement over rate 112,
but at the expense of reduced data rate.)

4) Maximum data rate of 250,000 bits per second was chosen to be

about double the maximum SSA capability to allow for future increase
in SSA capability,

5 ) Decoder bir-error rate performance represented a state-of-the-art capa-

bility and is given as a function of the ratio of energy-per-bit (Eb) to
noise-spectral-density (No) as shown below:

Energy-per-bitiNoise-spectralDensity

Bit-Error Rate Code (7:1/2) Code (7:1/3)

10(-3) 3.0 2.7

lo(-4) 3.8 3.5

lo(-5) 4.5 4.2

10 (-6) 5.2 4.9

For a comprehensive discussion of the theory and application of telemetry channel encod-
ing and decoding by Joseph Yuen, and other experts in that field, the reader is referred
to the book "Deep Space Telecommunications Systems Engineering."38

In consonance with the Mark 111-77 project guidelines, a study group was formed
early in 1975 to prepare an implementation plan and budget estimates for upgrad-
ing the 26-meter anrennas ro larger diameter and adding an X-band downlink
capability. It was called the 26-Meter S-X Conversion Project. Trade-off studies using
telecommunication link design parameters for the Pioneer Venus S-band mission
and the Voyager SIX-band mission showed that the optimum diameter for the new
antennas would be 34 meters. Following a review at NASA Headquarters in December,
it was decided that the stations to be modified would be DSS 12, DSS 44, and
DSS 62 and that they would be completed on the following schedule: DSS 12
(Goldstone), 1 December 1978; DSS 44 (Canberra), I April 1980; and DSS 62
(Madrid), 1 October 1980.39

The implementation plan issued by N. A. Renzetti on 1 July 1976, called for exten-
sive changes in the existing Antenna Mechanical Subsystem, the servo and antenna
pointing control mechanisms, the Antenna Microwave Subsystem, and the Receiver-
Exciter Subsystem. These changes were to be made on a negotiated schedule to
minimize conflict with the ongoing flight missions.40

The aperture increase from 26-meters to 34-meters required extensive modification

of the existing antenna structure. These changes, together with changes to the
microwave elements of the antenna, are shown in Figure 3-20.

The new X-band masers for the 34-meter SIX-band stations would have the same
basic amplifying portion and cryogenics as the existing operational masers on the
64-meter antennas, but would be provided with superconducting magnets for improved
gain and stability These X-band masers were also repackaged to make a more com-
pact and easily maintained unit, and they were supported by much improved
instrumentation. In particular, the microwave switching control system for the new
SIX-band waveguide assembly was completely redesigned to be compatible with the
centralized station control and monitoring facilities, and included a graphics display
of each selectable microwave config~ration.~'
f \

REFLECTOR PANELS - (RESETTING)

REFLECTOR EXTENSION

REFLECTOR BACKUP STRUCTURE

REINFORCEMENT (WELD JOINTS)

COUNTERWElGHTS
S-X CONE SHELL

REFLECTOR PANELS NEW

COUNTERWEIGHTS

CABLES, TRAY AND

HYDRAULIC UNES
(RAISE AND ELONGATE) RAISED PEDESTAL

IADDER - EXTENSION

AREAS OF CHANGE FROM 26-meter

i 2

Figure 3-20. Structural changes to the 26-meter antenna for extension to 34-meter diameter.

The addition of the X-band receive capability at the new 34-meter stations also
required the addition of a fourth harmonic filter to the transmitter subsystem. The
purpose of rhe filter was to suppress the fourth harmonic output from the S-band
transmitter klystron amplifier, which otherwise could cause interference with the X-
band signals in the X-band maser receiving pass band.

In order to provide sufficient long-term frequenry stability for outer planet navigation
and potential interferometry applications, an upgrade to the existing 26-meter frequen-
cy standards was required. For this purpose, two high-grade Cesium standards were
provided, which could be referenced to the hydrogen maser frequency standards at the
nearby 64-meter stations.
The most cost-effective way of providing receivers to complement the X-band maser front
end was to first down-convert the X-band maser output to S-band, and then feed the S-
band signals to the existing Block I11 S-band receivers. To maintain uplink and downlink
frequency coherence, the reference frequency for the downconverters was obtained through
appropriate frequency multipliers and dividers from the transmitter-exciter frequency source.

When completed, the 34-meter SIX-band antennas were required to have the following
RF performance:

S-band: Transmit: 2,110 (210) megahertz

Gain: 55.3 (k0.7) decibels
S-band: Receive: 2,285 (-+15) megahertz
Gain: 56.1 (+0.3/-0.9) decibeIs
System Noise Temperature:
Diplexed, 27.5 (k2.5) degrees kelvin
Listen only, 21.5 (k2.5) degrees kelvin
X-band: Receive: 8,420 (k20) megahertz
Gain: 66.9 (+0.3/-0.9) decibels
System Noise Temperature:
Diplexed, 25 (k3.0) degrees kelvin

The first station in the series, the Goldstone 26-meter antenna at DSS 12, was decom-
mitted from flight project support in June 1978 to undergo the upgrade to 34-meter,
SIX-band. The work was carried out in the period June through September 1978. System
and subsystem tests were conducted at the station, in October 1978, to verify that the
antennas met the design specifications. A period of extensive crew training was followed
by ground data system tests and demonstration tracks on Voyager before being declared
available for full operational support in November 1978. A photograph of DSS 12 after
its conversion to the 34-meter S/X-band configuration is shown in Figure 3-21.

The large concrete blocks on which the antenna pedestd stands were required to pro-
vide sufficient ground clearance for the increased antenna diameter. At the 34-meter sites
in Australia and Spain, large, concrete-lined trenches were formed in front of the anten-
nas as an alternative method of providing the necessary ground clearance.

Following the completion of DSS 12, effort on the 26-meter S-X conversion project
was transferred to the two overseas sites; they were completed in the Voyager Era, but
Figure 3-21. Goldstone Station, DSS 12 antenna after conversion to 34-meter, S-X-band
Configuration, October 1978.

by that time the original plan had been changed. Rather than convert the 26-meter sta-
tions at Cebreros and Honeysuckle as originally planned, the conjoint stations, DSS 42
and DSS 61, were upgraded to 34-meter diameter. DSS 44 and DSS 62 faced an uncer-
tain future and were eventually retired.
Dominated by Viking and driven by the requirements for tracking and data acquisition
support for the other major new planetary missions of Helios, Voyager, and the later
Pioneers, DSN capabilities expanded in the years of the Viking Era, 1974 through 1978.
Competition between the flight projects for antenna tracking time led to a much more
formalized approach to forecasting and allocation of these finite DSN resources. While
the initial approach lek much to be desired, it was continually refined on the basis of
experience and became more effective as time went on. The introduction of a formal-
ized architecture for the individual stations of the Network in the form of an integrated
assemblage of interactive systems did much to enhance the manageability of fither
improvements and new implementation in the Network as a whole. The Mark I11 Data
Systems Project that preceded the Mark 111-77 model of the Network is a prime exam-
ple. New technology in low-noise X-band receivers, arraying of multiple antennas, and
advanced coding techniques were combined to push the limits of downlink sensitivity
to new levels to match the demands for higher data rates at ever increasing distances.
Advances such as these were augmented with improvements to existing resources such
as the conversion of the existing 26-meter antennas to 34-meter diameter and retro-
fitting them with X-band receivers.

Although it was not apparent at the time, the launch of the Pioneer multiprobe space-
craft to Venus in December 1978 marked the beginning of the longest hiatus in planetary
spacecraft launchings in the history of NASA. For the next ten years, until the launch-
es of Magellan and Galileo in 1989, the DSN would be involved in supporting the
deep space missions already in flight and supporting new and existing Earth-orbiting
missions. It is to these events that we now turn in that period of the history of the
DSN dominated by the remarkable Voyager spacecraft-the Voyager Era.
Endnotes

1. Edward C. and Linda N. Ezell, On Man: Eqloration of the Red Planet, 1958-1978,
NASA SP: 4212 (Washington, DC: NASA 19841, p. 452.

2. Jet Propulsion Laboratory, "Deep Space Network to Flight Project Interface Design
Handbook," DSN Document Number 810-5, Latest Revision, 1995.

3. D. J. Mudgway and M. R. Traxler, 'Tracking and Data Acquisition Support for the
Viking 1975 Mission to Mars," JPL Technical Memorandum TM 33-783, Vol. I (15
January 1977), pp. 5, 35, 123.

4. David Morrison and Jane Sam, Vqqe to Jupiter, NASA SP439 (Washingcon, DC:
NASA, 1980), p. 25.

5. J. E. M e n and H. E. Nance, "Voyager Mission Support," DSN Progress Report PR

4 2 4 9 , NovemberIDecember 1978 (15 February 1979), pp. 29-33.

6. N. A. Renzetti, "A History of the Deep Space Network From Inception to January
1969," JPL Technical Report TR 32-1533 (September 1971), p. 41.

7. William R. Corliss, "A History of the Deep Space Network," NASA CR- 151915;
(Washington, DC: NASA, 1976), pp. 92, 194.

8. R. B. Miller, "Pioneer 10 and 11 Mission Support," DSN Progress Report PR 42-

33, MarchIApril 1976 (15 June 1976), pp. 21-25.

9. T W. Howe, "Pioneer Venus 1978 Mission Support," DSN Progress Report PR 42-
51, MarchIApril 1979 (15 June I979), pp. 19-30.

10. R. B. Miller, "Pioneer 1978 Venus Mission Support," DSN Progress Report PR 42-
3 1, NovemberIDecember 1975 (15 February 1976), pp. 11-14.

11. W; K. Victor, N. A. Renzetti, J. R. Hall, and 0.J. Mudgway, "DSN Mark I11 Data
System Development Plan, Oct-Nov 1969," presentation given at JPL,
21 October 1969, Deep Space Network Records, Jet Propulsion Laboratory
Archives.
12. R. Cargill Hall, Lunar Impact: A History of Projen Ranger, NASA History Series;
NASA SP-4210 (Washington, DC: NASA, 1977), pp. 71, 23.

13. M. L. Yeater, "DSN System Requirements: DSN Telemetry System 1977 Through
1982," DSN Document 821-2: Rev. B (15 September 1778).

14. D. A. Bathker and D. W. Brown, "Dual Carrier Preparations for Viking," JPL
Technical Report TR 32- 1526, Vol. XIV: January/February 1973, (15 April 1973),
pp. 178-179.

15. D. A. Bathker and D. W. Brown, "Dual Carrier Preparations for Viking," JPL
Technical Report TR 32-1526, Vol. XD(: NovemberIDecember 1973, (15 F e b r u q
1974), pp. 186-72.

16. J. R. Paluka, "100Kw X-band Transmitter for FTS," JPL Technical Report TR 32-
1526, Vol. VIII: JanuaryIFebruary 1972 (15 April 1772), pp. 9 6 9 8 .

17. D. J. Mudgway and M. R. TraxIer, "Tracking and Data Acquisition Support for the
Viking 1975 Mission to Mars," JPL Technical Memorandum TM 33-783, Vol. I1
(15 March 1777), pp. 94, 125, 137, 208.

18. D. J. Mudgway, "Tracking and Data Acquisition Support for the Viking 1975
Mission to Mas; May 1980 to March 1983," JPL Publication 82-107 (15 May
1983), pp. 3-13.

19. G. S. Downs, R. R. Green, and PE. Reichley, "A Radar Study of the Backup Martian
Landing Sites," DSN Progress Report PR 42-36, SeptemberIOctober 1976 (15
December 1976), pp. 49-52.

20. M. A. Gregg and R. B. Kolby, "X-band h d a r Development," DSN Progress Report

PR 42-20, JanuaryIFebruary 1774 (I 5 April 1974), pp. 44-48.

2 1. C. I? Wiggins, "X-band Radar Development," JPL Technical Report TR 32- 1526,

Vol. XII, SeptemberIOctober 1972 (15 December 1972), pp. 19-21.

22. R. L. Leu, "X-band Radar System," JPL Technical Report TR 32-1526, Vol. XM,
NovemberIDecember 1973 (15 October 1973), pp. 77-8 1.
23. R. W. Hartop, "X-band Antenna Feed-Cone Assembly," JPL Technical Report TR
32-1526, Vol. XTX: NovemberlDecernber 1973 (15 February 1973), pp. 173-175.

24. P. D. Potter, "Improved Dichroic Reflector Design for the 64-m Antenna S-and X-
band," JPL Technical Report TR 32-1526, Vol. XIX: NovemberlDecember 1973
(15 February 1973), pp. 55-62.

25. D. L. Trowbridge, "X-band Travelling Wave Maser Amplifier," JPL Technical Report
TR 32-1526, Vol. XVII: JulyIAugust 1973 (15 October 1973), pp. 123-30.

26. E. K. Davis, "Mariner VenusIMercury 1973 Mission Support," JPL Technical

Report TR 32-1526 Vol. XIV, January/February 1973 (15 April 1973), pp. 5-13.

27. D. J. Mudgway and M. R. Trader, "Tracking and Data Acquisition Support for the
Viking 1975 Mission ro Mars," JPL Technical Memorandum TM 33-783, Vol. I11
(1 September 1977), p. 105.

28. M. F. Easterling, "DSN Data System Development Plan, Mark I11 Project," DSN
Document 803-1, Rev. A (15 March 1974).

29. I? T. Westmoreland, "Technical Requirements Document for DSN Mark 111 Data
Systems Implementation," JPL Technical Requirements Document, TRD 338-300,
Rev. B (7 February 1975).

30. D. C. Preska, "DSN Standard Practices for S o h a r e Implementation," DSN

Progress Report PR 42-29, July/August 1975 (15 October 1975), pp. 119-30.

3 1. E. C. Gatz, "DSN Telemetry System Mark 111-77," DSN Progress Report PR 42-42,
SeptemberIOctober 1977 (15 December 1977), pp. 4-12. ,

32. W. G. Stinnett, "Mark 111-77 DSN Command System," DSN Progress Report PR
42-37, November/December 1976, (15 February 1977), pp. 4-1 1.

33. Thorman, H. C., "DSN Command System Mark 11-78,'' DSN Progress Report PR
42-49, NovemberlDecember 1978 (15 February 1979), pp. 11-1 8.

34. W. D. Chaney, "DSN Tracking System Mark 111-77," DSN Progress Report PR 42-
40, MayIJune 1977 (15 August 1977), pp. 4-13.
35. M. S. Glen, "DSN Ground Communications Facility," DSN Progress Report PR
42-36, SeptemberIOctober 1976 (15 December 1976), pp. 4-12.

36. E. C. Gatz, "DSN Telemetry System Mark 111-79," DSN Progress Report PR 42-49,
NovemberlDecember 1978 (15 February 1979)) pp. 4-1 0.

37. J. H. Wilcher, "A New Sequential Decoder for the DSN Telemetry Subsystem,"
DSN Progress Report PR 42-34, MayIJune 1976 (15 August 1976))pp. 84-87.

M. E. Alberda, "ImpIementation of a Maximum Likelihood Decoder in the DSN,"

DSN Progress Report PR 42-37, NovemberlDecember 1976 (15 February 1977),
pp. 176-83.

J. W. Layland, "Convolutional Coding Results for the MVM 1973 X-band

Telemetry Experiment," DSN Progress Report PR 42-48, SeptemberIOctober 1978
(15 December 1978), pp. 18-21.

38. Joseph H. Yuen, "Deep Space Telecommunications Systems Engineering," JPL

Publication 82-76, Jet Propulsion Laboratory, California Institute of Technology,
(July 1982), pp. 220-47.

39. V. B. Lobb, "26-meter S-X Conversion Project," DSN Progress Report PR 42-39,
MarchIApril 1977 (15 June 1977)) pp. 157-67.

40. N. A. Renzetti, "DSN Data System Development Plan, 26-meter S-X Conversion
Project," DSN Document 803-7 (1 July 1976).

4 1. H. R. Buchanan, "S-X 34-meter Conversion Receiver and Microwave Performance,"

DSN Progress Report PR 42-50, JanuarylFebruary 1979 (15 April 1979), pp.
219-25.
CHAPTER 4
THE VOYAGER ERA

From the overall DSN point of view, the transition from the Viking Era to the Voyager
Era was not marked by a clearly defined change in activity or objectives. Rather, driv-
en by the press of real-time mission events, the focus of DSN attention merged gradually
from Viking-related matters to those of Voyager.

It began with the successfd launches of two large planetary spacecraft in 1977 and 1978.
For almost the next decade, the two Voyager spacecrafi repeatedly astonished the world
with a flood of dazzling science data and images as they transmitted their data from
Jupiter, Saturn, and Uranus. At the same time, many other inflight missions vied with
Voyager for the support of the tracking stations of the Deep Space Network. The appear-
ance of a new class of mission, the Earth Orbiters, to the DSN "inventoryyyfurther
complicated the situation in this period.

DSN resources were stretched to the limit, not only by the number of simultaneous
missions but by the requirements for new and enhanced capabilities to support the rnul-
tipliciry of missions and the ever-extending uplinks and downlinks from the two Vyagers.

For the purpose of giving an account of DSN history between the Voyager launches in
1977 and the first appearance of Galileo in 1986, this period may be identified as the
Voyager Era, and it is to this era that we now turn our attention.

The scale of Network activity during the Voyager Era is best appreciated in the context
of the Deep Space and Earth-orbiting missions which were then of concern to the DSN.
The chart shown in Figure 4-1 depicts the situation.

This workload, together with engineering and implementation of upgrades to the capa-
bilities of the Network, is the subject of &e discussion that follows.

Except for the activity associated with the receiver problem on I/ayager 2, the Voyager
launches and early cruise events in 1977 and 1978 were "nominal." DSN Operations
resources were dominated more by the multiple spacecraft requirements of the Viking
mission and the complexities of the Pioneer Venus missions during that period. The
Mark 111 Data System (MDS) implementation effort, together with the start of the 26-
m to 34-112 antenna upgrade and the SIX-band conversion tasks, Mly occupied DSN
engineering and implementation personnel.
f

Voyager Era 7
Deep Space Network Mission Set

P6-9 PI0 He1 Vkl Vk2 He2 Vg2 Vgl PI2 Pmlp UIys Gll HE0 VB-P
0Voyager Era Future Missions
Primary Missions EZl Secondary Missions
\ J

Figure 4-1. Voyager Era Deep Space mission set, 1977-86. Each of the vertical bars in this
chart represents a deep space or Earth Orbiter mission which was in the DSN "inventory" during
the Voyager Era. During the period shown on the vertical scale, all missions passed through vari-
ous phases of their lives. Future missions were in the planning stage, primary missions were in
flight, and secondary missions had already completed their primary missions and were being tracked
by the DSN stations on a lower priority basis as "extended missions." The number and phase of
the missions in any year represents the relative level of activity in the Network at that particular
time. For example, from mid-1977 through 1980, the Network handled five, and for a short time
six, primary missions simultaneously with four secondary missions and three missions, including
Earth Orbiters, in the planning stages.

However, by the beginning of 1979, the focus of DSN attention was rapidly changing.
The Mark I11 Data System implementation task had been completed, the new capabil-
ities required for the Voyager Jupiter Encounter were in operation throughout the Network,
and operations teams were trained in their use. The Viking mission had been extended
through May 1978 and then further continued through February 1979. Both Pioneer
Venus missions had been successllly completed in 1978, and b o h Voyager spacecraft
were approaching Jupiter and were expected to carry out a full program of science exper-
 UPLINK-DOWNLINK:
A HISTORY NETWORK
OF THE DEEPSPACE

iments and imaging sequences during their brief encounters with the planet in March
and July 1979.

For approximately two years, the Voyagers moved along separate trajectories to equally
spectacular and scientifically rewarding encounters with Saturn in 1980 and 1981. From
there, Vayaer 2 would be retargeted to Uranus for an encounter in 1986, leaving kyager
I to continue along an inevitable path toward the outer reaches of the solar system. At
Uranus, Vyager 2, because of its remarkable longevity, would be retargeted yet again-
this time for an encounter with Neptune in 1989.

For their success, each of the Voyager encounters depended not only upon operable
spacecraft, but also upon ever greater significant enhancement of the uplink and down-
link DSN capabilities of the Network. At the same time, a heavy expenditure of DSN
operational resources in personnel, training, and facilities was required simply to main-
tain a viable science data return from existing missions.

As the mission set in Figure 4-1 clearly shows, there were as many as ten inflight, deep
space missions throughout the solar system, supported at various rimes during this period.

A closer look at the chart also shows that there were no new planetary launches after
the Pioneer Venus missions in 1978. To a large extent this hiatus was due to debates
within NASA about the objectives and funding for the Lunar spaceflight program
using the Space Shuttle, and the robotic space exploration programs using expendable
launch vehicles. 1

Galileo was the new flight project most affected by this sirnation. As the vertical bar in
Figure 4-1 shows, Galileo remained in "new mission" status for ten years. Between 1978,
when the original plan for a 1982 launch was approved, and 1987, when the final plan
for a launch in 1989 was approved, the Galileo launch vehicle was changed four times.
Each change to the launch vehicle necessitated a complete redesign of the mission with
corresponding changes to the requirements for tracking and data acquisition support.
With each change, the DSN responded with a plan for new or modified capabilities
and schedules. Change was the name of the Galileo game in those years.2

The Ulysses mission also remained in a state of suspension during the Voyager Era.
Ulysses was a joint NASAIESA mission to study the poles of the Sun and interplane-
tary space both above and below. As part of the joint effort, NASA had committed a
Shurtle launch to ESA for the purpose of deploying the Ulysses spacecraft into Earth
orbit. From there, Ulysses would use a solid motor Inertial Upper Stage (IUS) to inject
it into a trajectory toward Jupiter. The gravitational field of Jupiter would redirect Ulysses
out of the plane of the ecliptic and into a heliocentric orbit passing over the poles of
the Sun. First planned for launch in 1983, Ulysses finally made it to the launch pad in
October 1990 afier two postponements, one in 1983 and one in 1986. The design of
the mission and its requirements on the DSN were not affected by the delays in the
Shude program.

Toward the end of the Voyager Era, a truly international cooperative mission made its
appearance. The Venus-Bdoon mission in mid-1985 involved the Soviet, French, and
United States space agencies. Although of very short duration, it presented a complex
engineering and operational challenge for the DSN. Its successful completion established
a basis for future relationships between these agencies in the area of tracking and data
acquisition support for deep space missions.

Earth-Orbiting Missions

The DSN began the Networks Consolidation Program, which expanded the capa-
biliry of DSN tracking and data acquisition support to include support for high-Earth
orbiting (HEO) missions as distinct from Deep Space (DS) missions in 1980. H E 0
missions were defined to be those which were beyond the capability of the newly
developed NASA space network called the Tracking and Data Relay Satellite System
(TDRSS). The H E 0 class of missions would have orbits with an apogee in excess
of 12,000 km, and would be supported by the 26-m and 9-m networks added to
the DSN as part of the Networks Consolidation Program. Mainly through the per-
sistent efforts of W. R. Martin, this criterion was eventually modified to allow the
DSN to provide support, in the near-Earth region below 12,000 km, for a whole
new class of non-NASA missions known as "reimbursables." Under the reimbursable
arrangement with NASA, foreign space agencies were able to employ the services of
the DSN for part or all of their missions, subject to compliance with DSN techni-
cal and resource conditions. NASA would be reimbursed at a standard hourly rate
for the antenna time provided to the foreign space agency. Within a few years, the
term "HEO" fell out of common usage and became simply "Earth Orbiter (EO),"
to include these latter missions.

DSN support for Earth-orbiting missions began in 1984 with ICE (International Cometary
Explorer). Under a different name, ICE had been placed into a heliocentric orbit by a
Delta launch vehicle in 1978. As a cooperative mission with ESA, it was ro monitor
solar phenomena. Later, the name was changed to ICE and its orbit was changed to
encounter the Comet Giacobini-Zinner in September 1985. It was soon joined by the
Active Magnetospheric Particle Explorer (AMPTE) series of missions. Designed to study
the transfer of mass from che solar wind to the magnetosphere from an Earth orbit
above 100,000 km, these missions were also a cooperative effort between NASA, the
United Kingdom, and the Federal Republic of Germany.

DSN support for ICE and AMPTE continued through the entire Voyager Era, sup-
porred a number of short duration reimbursable missions, and reached a peak of
thirteen in 1985.

Since 1984, support for Earth-orbiting missions in the form of cooperative, reimbursable,
or NASA projects has become a permanent feature of the DSN purpose in the United
States space program.
NETWORK OPERATIONS

Voyager at Jupiter

The Voyager Project was the ultimate realization of a much earlier NASA plan to send
two robotic spacecraft to visit all of the outer planets of the solar system in the later
years of the 1970s. As early as 1969, even as Congress was approving funds for the first
exploratory missions to Jupiter (Pioneers 10, l l ) , NASA was making grand plans for
Euture planetary exploration.

Based on a report from the United States National Academy of Sciences, which found
that "exceptionally favorable astronomical opportunities occur in the late 1970s for mul-
tiplanet missions," this "mother of d missions" came to be known as the Grand Tour.
An additional report, published in 1971, concluded with a specific recommendation
"that Mariner-class spacecraft be developed and used in Grand Tour missions for the
exploration of the outer planets in a series of four launches in the late 1970s." At least
in theory, this unique juxtaposition of the planets would allow a passing spacecrafi to
use the gravitational pull of one planet to alter its trajectory in such a way as to redi-
rect it toward a flyby of the next planet. The process could be repeated, as required, to
make a complete tour of all the outer planets. The technical challenges of such a mis-
sion were enormous. Amongst them were precise celestial navigation and deep space
communications, both primary functions of the DSN.

In its original form, the NASA plan for the Grand Tour encompassed dual launches to
Jupiter, Saturn, and Pluto in 1976 and 1977, and dud launches to Jupiter, Uranus, and
Neptune in 1979. A total cost, over the decade, was about $750 million.3 Later, how-
ever, political and budgetary constraints forced NASA to scale back the original plan to
two missions to Jupiter and Saturn, with an option for an encounter with Uranus. The
totd cost of the new missions was to be $250 million, a more acceptable figure in the
fiscal climate of the early 1970s.

Congressional approval was soon forthcoming and the official start of the Voyager mis-
sion was set for 1 July 1972. NASA designated JPL as the Lead Center and Edward C.
Stone, a distinguished expert on magnetophysics fiom Caltech, as Project Scientist. Because
the new mission was based on the proven, JPL-designed Mariner spacecraft, it was ini-
tially named MJS, for Mariner Jupiter-Saturn; the name became Voyager in 1977 but
most of the early documentation retained the original name. At JPL, Raymond L. Heacock
became Project Manager with Richard I? Laeser as his Mission Director, and Esker K.
Davis as Tracking and Data Systems Manager, representing the DSN. Although these
names changed several times as the mission progressed over the next twenty-five years,
the functions always remained the same.

In both concept and execution, the Voyager Project was one of the most ambitious plan-
etary space endeavors ever undertaken. The Tlqyager I spacecraft was to investigate Jupiter
and several of its large satellires, and S a w n and its rings and large satellite, Titan. Vqyager
2 was also to observe Jupiter and Saturn and several of their sateuites after which it was
to be redirected toward an encounter with Uranus in 1986. After their final encoun-
ters, both spacecrafi would eventually cross the boundary of the solar system into interstellar
space. Each spacecrafi. would carry instrumentation for conducting elwen scientific inves-
tigations in the fields of imaging science, infrared radiation, ultraviolet spectroscopy,
photopolarimetry, planetary radio astronomy, magnetic fields, plasma particles, plasma
waves, low energy charged particles, cosmic ray particles, and radio science. A new onboard
computer system gave the Voyagers greater independence from ground controllers and
more versatility in carrying out complex sequences of engineering and scientific opera-
tions than on the Mariners earlier.

On each spacecraft, uplink communications with the DSN were provided by two S-
band radio receivers while downlink communications used four 25-wan transmitters,
two of which operated at S-band and two at X-band. Each spacecraft carried a large,
3.7-meter diameter, high-gain antenna (HGA) in addition to a smaller, low-gain anten-
na intended for use as backup. Combined with the HGA, the X-band downlink was
designed to deliver telemetry data up to 115.4 kilobits per second to the DSN 64--m
antennas from the distance of Jupiter.

That was the plan and the basis for h e Voyager project requirements for DSN track-
ing and data acquisition support for the first part of the overall mission-namely the
launch phase, cruise phase, and with the Jupiter Encounter.* The Network responded
to these requirements with the DSN MDS. The MDS and the events associared with
the Voyager launch and cruise phases marked the closing stages of the V i n g Era. Events
associated with the approach of the two Voyager spacecraft to Jupiter, toward the end
of 1978, soon showed that transition to rhe Voyager Era had already begun.

As the year 1979 opened, the space drama unfoIding at JPL and featuring Eyagev I at
Jupiter had begun to attract the attention of space scientists and observers throughout
the world. Excerpts from "Voyage to Jupiter"5 describe the mounting interest. "In mid-
January, ~ h o t o sof Jupiter were already being praised for 'showing exceptional details of
the planet's multicolored bands of clouds."' "By early February, Jupiter loomed too large
in the narrow-angle camera to be photographed in one piece." Sets of pictures called
mosaics were necessary to cover the entire planet body. A spectacular movie covering
ten Jupiter days and displaying the swirling vortices of the upper atmosphere would be
assembled from thousands of images transmitted fiom the spacecraft as it approached
the planet. These images, transmitted at the rate of about one image per 90 seconds
over the X-band downlink at 117 kilobits per second, required 100 hours of continu-
ous DSN coverage to complete.

The Voyager I spacecraft reached its point of closest approach to Jupiter at 4:42 A.M.
PST, on 5 March 1979, at a distance of 350,000 km from the planet center and 660
million km fiom Earth. The governor of California was present at JPL, and a special
TV monitor was set up in the White House for President Carter ro witness this his-
toric event. A short time before the closest approach to Jupiter, Voyager had begun an
intensive sequence of observations of 10. Much of the data, taken during the Canberra
pass, was stored on the spacecrafi tape recorder and played back later over the Goldstone
stations so that it could immediately be transmitted to JPL.

Typical examples of the many images of Jupiter and 10 taken by Vyager I are shown
in Figures 4-2 and 4-3.

The passage of Vyager I through the Jupiter system is shown in Figure 4-4.

The entire event, from the time it crossed the orbit of Callisto inbound to the time it
recrossed it outbound, took only 48 hours. In that short time, equal to two complete
passes around the DSN, Vyageer made observations on a major planet, Jupiter, and five
of its satellites-Amalthea, 10, Europa, Ganymede, and Callisto. Never before had a
planetary encounter yielded such a wealth of new and unique scientific data. It was truly
a major milestone in the history of planetary exploration. In his essay, "Voyager: The
Grand Tour of Big Science,"G Andrew Butrica noted, "Voyager is planetary exploration
on a grand scale." The mission was only just beginning.

As Voyager I receded from Jupiter, it continued to work through the carehlly planned
post-encounter mission sequences, and accelerated on to a new trajectory that took it
to an encounter with Saturn in November 1980.

Voyager 2, four months behind Voyder 1, moved into the Jupiter observatory phase and
took imaging sequences for another time-lapse movie and activated its W and fields
and particles instruments.
Figure 4-2. Voyager 7 image of Jupiter, 13 February 1979, at a distance of 20 million kilome-
ters, showing lo (left) and Europa (right)against the Jupiter cloud tops. One of the most spectacular
planetaty photographs ever taken was obtained on 13 February 1979, as Voyager 1 continued to
approach Jupiter. By that time, at a range of 20 million kilometers, Jupiter loomed too large to fit
within a single narrow-angle imaging frame. Passing in front of the planet are the two Galilean satel-
lites. lo, on the left, already shows brightly colored patterns on its surface, while Europa, on the right,
is a bland, ice-covered world. The scale of all of these objects is huge by terrestrial standards. lo and
Europa are each the size of our Moon, and the Great Red Spot is larger than Earth.

For the next several months, DSN operations activity would be dominated by these two
high profile planetary missions. lhyager I post-encounter activities were kept at a low
level so that the majority of the support facilities could be devoted to preparation for
the Voyger 2 encounter. The Doppler tracking loop problem in the tijyageer 2 radio
receiver complicated each DSN tracking pass by requiring a continuous change in the
uplink frequency to compensate for the Doppler shift at the spacecrafi. This change
could now be predicted and included in rhe computer driven predicts, but it was also
necessary to monitor the spacecraft telemetry to detect any drift in receiver rest fre-
quency and modify these predicts in real time to maintain the spacecraft receiver in lock.
Figure 4-3. Voyager I image of lo, 4 March 1979. Taken at a distance of 400,000 kilometers,
the picture shows the giant volcano Pele and strangely colored deposits of surface material.

Despite these diEculties, all transmitted command sequences were accepted by the space-
craft, and uplink and downlink communications were maintained successfdly.

As Vyager 2 closed with Jupiter, the returning images showed clear evidence of the great
changes that had taken place in the Jovian atmosphere since the I/oyger 1 encounter in
March. In passing through the Jupiter system, Voyager 2 would be able to fly by the
same satellites as Vyager I. It would "see" different faces of Callisto and Ganyrnede and
pass much closer to Europa and Ganymede. Unf-brtunarely, it would pass much further
away from the volcanic satellite 10. Although it was not realized at the time, the data
from the closer pass to Europa and the h t h e r pass from 10 would be of great signifi-
cance when the Galileo spacecrafi returned to Jupiter seventeen years later. Some of the
Voyager sequences were changed as a result of the Vlager I experience, but in genera1
they followed the same pattern.

The Jupiter Encounter occurred at 3:29 P.M. PDT, on 9 July 1979, amid great excite-
ment as new data on the satellites and the planet itself poured in and were released to
f

-- - Satellite closest approach

L

Figure 4-4. Voyager I trajectory through the Jupiter system, 5 March 1979.

the pubIic in the daily press conferences. A new satellite of Jupiter, Andasrra, was dis-
covered and a ring quite unlike those of Saturn was observed for the first time. The
NASA Associate Administrator for Science, Thomas A. Mutch, was moved to remark
that "such events are dearly read into the record. And I submit to you that when the
history books are written a hundred years from now . . . the historians are going to cite
this particular period of exploration as a turning point in our cultural, our scientific,
our intellectual development."7

During and after the encounter, there was some concern about the effects of the Jupiter
radiation environment on the spacecraft radio receiver. This caused more rapid drift in
frequency than expected and actually resulted in loss of the uplink connection with rhe
spacecraft on the day after encounter. By sending repeated commands at various fre-
quencies, the DSN operators were finally able to find a frequency at which the spacecraft
receiver would accept the commands, just in time for the commands needed to fire the
thrusters for the trajectory correction maneuver that would enable the spacecrafr to make
the turn for Saturn.
While Vqyager 2 added considerably to the volume of science delivered by Vqyuger 1, it
also added to the number of unanswered questions about the "King of Planets." The
answers would be a goal for Galileo, the next Jupiter mission, which lay many years in
the hnire. At that time, however, the next exciting goal for both Voyagers lay Iess than
two years away. It was the planet Saturn.

The demand for contiguous support for Voyager and the Pioneer Venus Orbiter in 1979
had forced the DSN to consider economies in the use of tracking station time for pre-
track preparation, i.e., the configuration and calibration of the station hardware and
software needed for an upcoming spacecraft tracking pass. Prior to this time, the DSN
had generally been able to assign a complete station to a flight project in a critical
encounter phase, to the exclusion of all other users. In this situation pretrack prepara-
tion times of as much as six to eight hours were normally used. Time needed for essential
station maintenance consumed additional time. Early in 1979 this was no longer pos-
sible because of end to end Voyager and Pioneer view periods of 8 to 10 hours each at
the same 64-m sites. To contend with this situation, DSN operations devised a new
strategy for pretrack preparation. Pretrack and post-track activities for both spacecraft
were drastically shortened and carried out during a single allocated time period.8 "Quick
turnaround reduced the time allowed between the end of one track and the start of
the next to the time required to reposition the antenna and to mount new tapes. In
addition, time for routine station maintenance was based on a new formula which ensured
an essential, monthly minimum number of hours for each type of antenna.

From this time on, the DSN was under constant pressure to increase the amount of
tracking time available for flight project use, often at the expense of reducing or even
eliminating, essential maintenance rime with consequent risk to station reliability. Over
the years, the basic rules devised in 1979 for allocation of pretrack and post-track prepa-
ration time have served the DSN well in striking a balance between conflicting demands
for maximum antenna tracking time and maximum station reliability.

Voyager at Saturn

In terms of uplink and downlink capability, the requirements of the Voyager missions
to Jupiter had been mer by the DSN Mark 111-77 Data System described in detail in
the previous chapter. In fact, several years earlier, the Voyager requirements for teleme-
try, command, and radiometric data had been one of the principal drivers for that major
upgrade to the Network capability and the schedule on which it was carried out (see
chapter 2). When zompleted in April 1978, it provided the Voyager missions with an
X-band, high-rate (up to 115 kilobits per second) telemetry downlink at Jupiter range
from all three 64-m stations. Radiometric products, in the form of two-way Doppler
and ranging data, provided for precise navigation and the S-band, 20-kilowatt, trans-
mitters were more than adequate for command purposes at Jupiter range.

However, to support the Voyager spacecraft at Sat-urn, double the range of Jupiter, and
possibly even Uranus, four times the Jupiter range, substantial enhancements to the Mark
111-77 capability, particularly in the detection of weak signals, were required. Because
the downlink signal strength diminishes as the square of the spacecraft distance from
Earth, the signals reaching the DSN antennas when the spacecraft reached Saturn would
be only one-fourth of those received from Jupiter, one-fourth less again when the space-
craft reached Uranus.9

These and other issues related to DSN operational support for Voyager were routine
topics on the agenda of the weekly Voyager project meetings at JPL. Led by the proj-
ect manager or the mission director, project meetings were the established JPL forum
for negotiation of all requirements and interactions between the DSN and all flight proj-
ect organizations. Once the actual mission began, the agenda of the regular project
meetings was expanded in scope (and duration) to include status and progress report-
ing, as well as future requirements on the DSN. The meetings comprised representatives
from each of the institutional organizations involved with that particular flight project,
of which the DSN was one. Marvin R Traxler represented the DSN to the Voyager
project. The DSN appointed similar representatives to each of the flight projects to speak
on DSN matters.

It should be pointed out that negotiations between the flight projects and the DSN,
which involved the design and implementation of new DSN capabilities, were com-
pleted as much as five years prior to the "time of need," due to the long lead time
required for the DSN approval, funding, contractual, implementation, testing, and oper-
ational training processes. In some cases, even five years proved to be insufficient time.
That is the reason for the long planning periods represented by future missions in the
charts of the DSN mission sets for the various eras (see Figure 4-1 for the Voyager Era).
A formal set of top level documentation conveyed the project "requirements" and the
DSN "commitments" to the two Program Ofices at NASA Headquarters, the Office of
Space Science (OSS) and Tracking and Data Acquisition (OTDA), for formal, top-level
approval. Lower tier documents within the project and the DSN disseminated the nec-
essary technical and operational detail to the implementing organizations. These steps
infrequently occurred in serial fashion. Rather, driven by the always-pressing flight proj-
ect schedules, work proceeded on the assumption that the necessary formalities would
be eventually completed. Interactions between the flight projects and the DSN were iter-
ative in nature; as space communications technology advanced, the technical require-
ments of deep space missions advanced to jusufy the implementation of the new technology
in the Network, and vice versa.

Such was the case for Voyager. Even as the two Voyager spacecraft left their launch pads
bound for Jupiter in 1977, engineers in Robertson Stevens's Telecommunications Division
at JPL had turned their attention and considerable talents to the enhancements that
would be needed for the existing Network capabilities to support the Voyager spacecraft,
if and when they reached Saturn, Uranus, or even Neptune. Of course these enhance-
ments would benefit all deep space missions; NASA insisted that they be multimission
in nature, and so they were, although they were generally attributed to the first mission
to make use of them.

Ultimately, the flight project requirements that had been approved for implementation
in the Network were published in the form of a plan, jointly signed by project and
DSN representatives. The referenced DSN Preparation Plan for Voyager is rypical of
those prepared for all flight projects which used the Network at various times.10

The final plan to meet the Voyager project requirements for DSN support of the two
Saturn encounters included the addition of major new or improved capabilities, to the
existing Network, in the following technical areas:

1. S-band and X-band antennas: Capability to receive both S-band and X-band down-
links at three 34-m stations and three 64-m srations. All 64-m antennas were optimized
with X-band, low-noise masers for improved downlink performance, and with spe-
cial microwave feeds for radio science experiments during encounter.

2. Downlink signal enhancement: Two-staxion arraying at each Complex using the 34-m
and 64-m antennas and the Real-Time Combiners to improve the signal margin by
about 1.0 dB (approximately 25 percent) compared to the 64-m antenna alone. This
would allow the downlink telemetry rate to run as high as 44.8 kilobits per second.

3. Precision navigation: Up to six different kinds of radiometric data to enhance space-

crafi navigation and radio science experiments by allowing for the removal of charged
particle effects in the interplanetary media. These data types consisted of various
combinations of S-band and X-band Doppler and ranging data. Besides basic irnprove-
ments in the accuracy of the ranging system, the DSN Tracking System also included
automatic uplink frequency tuning to compensate for the failed frequency tracking
loop in the Vqageer 2 transponder.
4. Radio science augmentation: New precision powered monitors, spectrum signd indi-
cators, open-loop receivers and multiple-channel, wideband, digital recorders together
with appropriate s o h a r e were installed at DSS 63 Madrid, the prime radio sci-
ence station designated by Voyager for covering the occultation and Saturn rings
experiments during the Voyager I encounter. Later, some of this equipment was
moved to the Canberra site and installed at DSS 43 to cover the Voyager2 encounter.
This equipment measured changes to the inherent radio frequency characteristics
(polarization, phase delay, spectral spreading, scintillation, etc.) of the Voyager radio
signal as it grazed the Saturn atmosphere or passed through the rings. Later analy-
sis of wideband recordings of these data provided significant new scientific data on
the composition of the atmosphere and the structure of the rings.

The manner in which this work, together with other project-related activity, was carried
out during the Voyager Era is described in a later section on Network engineering and
implementation.

Following their highly successful encounters with Jupiter in 1979, both Voyager space-
craft commenced the Jupiter-Saturn cruise phase of the mission. The Saturn trajectories
had been estabIished as the spacecraft passed Jupiter and, in the months that foliowed,
both spacecraft carried out routine engineering, science, test, and calibration activities.
While these routine activities were in progress, the spacecraft carried out a number of
special activities.

In February 1980, a delicate cruise science maneuver, which involved turning the space-
craft away from the Earth pointing direction, was performed. The maneuver went well
and the spacecraft reacquired the uplink after the antenna came back to Earth point.
The scan platform was calibrated in March, and the Canopus star tracker sensitivity was
checked in April. Numerous navigation cycles were carried out to refine the radiomet-
ric data used for spacecraft orbit determination. A navigation cycle consists of one
continuous uninterrupted pass around the Network. Radiometric data, consisting of
Doppler, ranging, and a new data type called DeIta Differential One-way Ranging (Delta-
DOR), are generated by each of the participating stations in turn, as the spacecrafi passes
through each longitude.

While the real-time operations elements of the Network were supporting all of these
Voyager related activities, other parts of the DSN operations organization were also busy,
preparing for the upcoming Saturn Encounter.
In March 1980, operations and engineering representatives fiom the stations in Spain
and Australia arrived at Goldstone for training in the installation and operation of a
new device called a Real-Time Combiner (RTC). The RTC enabled the signals fiom
two or more separate antennas to be combined electronically to produce a single out-
put of considerably greater strength than either of the input signals alone. When used
in this manner, the antennas were said to be arrayed. The RTC had been used earlier
in a two-station array at Goldstone during the Voyager 2 Jupiter Encounter, with encour-
aging results. The X-band downlink signals from the DSS 14, 64-m antenna had been
combined with the signals from the new DSS 12, 34-m antenna to give a telemetry sig-
nal gain of 1.1 dB relative to the 64-m antenna alone. This result, obtained under
real-time operational conditions, agreed well with the theoretical, predicted value of 1.1
(+0.2) dB. There was great interest in adding a similar capability at the two other 64-
m sites where the 26-m to 34-m antenna upgrades were just being completed. After
completing the classroom courses and receiving operations experience, the trainees would
return to their home sites to replicate the Goldstone installations.

The overseas sites, Deep Space Station 42 (DSS 42) in Australia and DSS 61 in Spain,
were being requalified for operational support following the upgrade of their antennas
from 26-meter to 34-meter diameter. In May, DSS 14 (Goldstone) and DSS 62 (Spain)
were returned to operational status after requalification following antenna downtime for
repIacement of the subreflector at DSS 14 and repair of the antenna drive gear boxes
at DSS 62.

ALl of the DSN upgrades and modifications to hardware planned for the Voyager Saturn
Encounter had been completed and declared operational by mid-1980. In addition, seven
new software packages needed for the antenna, communications, command, radiomet-
ric, meteorological, occultation, and planetary ranging systems were installed throughout
the Network and certified for operational use. The DSN was ready for the first Voyager
Saturn Encounter.11

Voyager 1 began its concentrated observations of Saturn on 22 August 1980, just 82 days
before its closest approach to the ringed planet. At that time, the spacecraft was travel-
ling with a heliocentric velocity of 45,675 miles per hour, at a distance of 67.6 million
miles from the planet. The radio signals from the DSN antennas were taking 80 min-
utes to travel the distance of 901 million miles from Earth to the Vqager spacecraft.12

The DSN began a series of navigation cycles around the Network to provide precise
orbit determination data on which the spacecrafi Navigation Team would base the param-
eters for the final trajectory correction maneuver to fly close by, but not impact, Titan.
In addition, the unique geometry of the Saturn Encounter, zero declination, required
highly accurate ranging measurements from two stations simultaneously to provide radio-
metric data from which the declination of the spacecraft orbit could be determined.

Earlier, the spacecraft had experienced some minor hardware problems in the Canopus
star tracker and the scan platform supporting the cameras, but neither was expected to
pose a serious problem to the planned Saturn Encounter activities.

The Canopus star tracker helped stabilize the spacecrafi and keep it properly oriented
by tracking Earth, the Sun, and a reference star, normally Canopus. A backup star track-
er was available for use if needed. The scan platform supported the imaging cameras
and other science instruments, and under certain conditions of operation had shown a
slow drift which could be compensated for, if necessary.

By making use of the 28 percent gain in downlink signal power that would result from
using the DSN 34-m and 64-m antennas in the arrayed mode at all three complexes,
the Voyager mission controllers planned to run the spacecraft downlink at a telemetry
data rate of 44.8 kilobits per second. 'Without arraying, the maximum data rate possi-
ble from Saturn would have been only 29.9 kilobits per second. By contrast, the data
rate at the Jupiter Encounter had been 115.2 kilobits per second due to the much shorr-
er spacecraft to Earth range of 413 million miles.

The onboard planetary radio astronomy experiment had been used to determine the rota-
tion rate of Saturn with greater precision than was possible with Earth-based measurements,
and cyclical bursts of nonthermal radio noise had been detected. Daily scans of the plan-
et by the ultraviolet spectroscopy instrument were searching for hydrogen sources, and
fields and particles instruments were constantly monitoring the interplanetary medium
near Saturn. Hundreds of photographs were being taken by the imaging system to com-
pile a full color 'time-lapse" movie as the Voyager spacecraft zoomed in on the planet.
One of the spectacular pictures from this period is shown in Figure 4-5. At the time this
picture was taken, Eyager I was 80 days from Saturn Encounter at a distance of 66 mil-
lion miles. In addition to the planet itself, three satellites could be seen, and the inner
and outer rings, separated by Cassini's division, were clearly visible. The imaging data was
transmitted from the spacecraft over DSS 43 at a data rate of 44.8 kilobits per second.

Before the spacecraft reached Saturn, it passed behind the Sun as viewed from Earth.
During this period of solar conjunction, from 3 September through 6 October, the angle
defined by the Sun, Earth, and Voyager became 15 degrees or less, and the radio noise
emitted by the Sun gradually degraded the radio downlink to the DSN tracking sta-
Figure 4-5. Voyager 7 image of Saturn and its rings, August 1980. Taken on 24 August 1980,
80 days before closest approach at a distance of 66 million miles, this image shows the Saturnian
satellites Enceladus, Dione, and Tethys.

tions. However, these conditions provided a unique opportunity for rhe radio astrono-
my observations of the Sun and heliosphere as the radio signals from the spacecraft
passed through the solar corona. At each of the Deep Space Communications Complexes,
rhe radio science equipment had been upgraded with new and improved hardware and
s o h a r e in anticipation of these imminent events.

Following the solar conjunction period, the uplink and downlink performance returned
to normal as the spacecraft continued to rapidly approach the planet.

By 24 Ocrober, 19 days before encounter, the field of view of the narrow angle camera
could no longer cover Saturn in one frame. It took four pictures to image the entire
planet. Ten days lam, the Saturn image was larger still and more mosaic pictures were
needed to cover it. Attention was then focussed on more detailed examination of the
Figure 4-6. Voyager I image of Saturn and its rings, November 1980, shows approximately
95 individual concentric rings.

planetary features. The extraordinarily complex structure of the Saturn ring system was
shown in a mosaic of two images taken on 6 November from a distance of 4.9 million
miles. (See Figure 4-6.)

O n 11 November 2980, twenty-six hours before reaching the point of closest approach
co the planet, the spacecrafi began executing the encounter sequences which had been
preprogrammed into its onboard computers months earlier. Downlink telemetry rates
varied between 19.2 kbps and 44.8 kbps as the spacecraft automatically switched through
various formats to return imaging, general science, and playback data in rapid succes-
sion. Taking high-resolution pictures as it wenc, the spacecraft would first pass close
(7,000 km) by the haze-covered satellite Titan before dipping below rhe ring plane as
it accelerated toward the point of closest approach. Eighteen hours later, on 12 November
1980, Voyager I reached its closest: point to Sawn, 184,000 krn from the cenrer and
124,000 km above the cloudtops of the shadowed southern hemisphere.
Some 100 minutes after closest approach, the spacecrafi passed behind Saturn and remained
obscured fiom Earth for 90 minutes. During this time, and at entrance and exit from
occultation, DSS 43 and DSS 63 carried out valuable radio science measurements relat-
ed to an understanding of the structure and composition of the Saturn atmosphere and
ionosphere. At this point the radio signal from the spacecraft was taking four and one-
half hours to reach the Earth-bound antennas of the DSN.

Vqydger 1 continued to observe the planet and its satellites through December 15, at
which time the Saturn post-encounter phase ended and a new phase of scientific data
collection began. For as long as the DSN could maintain an uplink and a downlink
with the spacecraft it would continue to observe the planetary medium, participating in
celestial mechanics and solar experiments with other spacecraft travelling through the
solar system. It was now on a solar system escape trajectory that would take it out of
the plane of the ecliptic, the plane of Earth's orbit around the Sun. Someday, Trayger
1 and other planetary spacecraft will reach the heliopause, the outer edge of the solar
system. The DSN will continue to track these spacecraf?, each going in a different direc-
uon to uy to determine the size of this invisible region of our solar system. It was
expected that Vyager 1 would cross the heliopause in about ten years, 1990, at a dis-
tance of 40 AU.13

Vyager 1 had completed its primary mission of planetary exploration. There were no
other planets along its new path nor could its path be changed even if there were.
Eventually, it would exit the solar system altogether, climb above the ecliptic towards
the constellation Ophiacus, and chase but never catch it for all time.

The attention of the Voyager project returned to Vyageer 2, now seven months away
from its encounter with Saturn in August 1981.

During the encounter phase of the Vyager 1 mission, the mission operations activity
on Eyager 2 had been relatively quiet, with most of the DSN tracking support being
provided by the 26-m network, while Voydger I dominated the 64-m and 34-m net-
works. Beginning in February 1981, however, the pace on Vqydger 2 picked up as the
DSN raced to complete several major new capabilities needed to support the Vayager 2
Saturn Encounter.

The equipment used for the Vyager 1 radio science observations had to be moved from
DSS 63 in Spain to DSS 43 in Austrafia, the prime viewing site for Eyager 2 solar con-
junction. This equipment, which included the four-channel narrow band and wide band
receivers and associated digital recording assemblies, had to be integrated with new sofi-
ware in the Occultation Data Assembly and tested while operational procedures were
developed and crew training progressed in parallel.

The receiver problem on Voyager 2, described earlier, had now become a "fact of life"
with which the DSN had to deal. To this end, the DSN had developed new s o h a r e
to automatically control the Digitally Controlled Oscillator (DCO) which drives the fre-
quency of the uplink transmitter. The DCO would be programmed to compensate for
the inability of the Vqager 2 spacecrafi receiver to acquire and track the uplink fre-
quency transmitted by the DSSs, because of its failed tracking loop. It would reside in
the Metric Data Assembly (MDA).

There was, however, a complication introduced by spacecraft internal temperature change.

It had been determined that various spacecrafi activities would cause compartment tem-
perature changes, which would in turn cause the center frequency of the spacecraft
receiver to drift in an unpredictable manner. This would make it very difficult or impos-
sible for the DSN to set the transmitter frequency to the exact Best Lock Frequency
(BLF) needed to track the spacecrafi receiver. To provide background data on the fre-
quency offset and drifi caused by spacecrafi temperature variations, the DSN had been
supporting a special tracking procedure known as "adaptive tracking." In an "adaptive
tracking sequence, the 34-164-m station carries out a series of uplink frequency ramps
estimated to pass through the BLE By observing the spacecraft receiver reaction to the
uplink frequency ramps, a real-time determination of the correct value of the BLF and
its drift can be made. These data can then be used to program the DCO to keep the
uplink fiequency centered in the slowly drifting receiver pass band.

The DSN would need to use "adaptive tracking during the Voyager 2 Near Encounter
period to ensure rapid and reliable acquisition of the spacecraft receiver for uplink com-
mand purposes. The key to this new capability was the new MDA s o h a r e which drove
the DCO.

This sofiware was installed at all sites by midyear. After the operations crews were trained
in its use, it was verified for encounter operations by conducting demonstration tracks
with the live spacecraft.

Early in 1981, the DSS 12 antenna, upgraded to 34-m in time for the Jupiter Encounter,
was decommitted from support operations for six weeks for further upgrade work to
improve its radio efficiency. The two outer rows of antenna surface panels were replaced
and reset, and the subreflector surface was replaced with one designed for better iflu-
mination of the primary antenna. The controller for the subreflector was also upgraded.
The net result of this work was an improvement of 0.7 dB in antenna gain, demon-
strated on the first operational pass when the antenna returned to service in April. The
DSS 12114 array would also benefit during encounter operations.

On 5 June 1981, the observation phase of the Vqager 2 Sarurn Encounter began. The
first activity was the movie sequence, which started over the Madrid Complex with Deep
Space Stations 61 and 63 (DSS 61 and DSS 63) tracking the spacecraft, and was con-
cluded on 7 June over the Canberra Complex with tracking support by DSS 42 and
DSS 43. During the movie phase, the arrayed 34-164-m cofiguration was used at all
complexes to enhance the received imaging telemetry data. The performance of the
arrayed stations was well within the predicted tolerances and resulted in excellent: pic-
ture data quality.

While the spacecrafi team conducted a sequence of scientific observations somewhat sim-
ilar to those conducted by Vyageer I , the DSN completed the remaining hardware and
s o h a r e items needed for encounter, and enhanced the proficiency of the DSS crews
with Operational Verification Tests.

Beginning with the passes over DSS 61 and DSS 63 on 13 August, and continuing
around the Network for several days, high rate imaging data were obtained by the DSN.
The 34-m and 64-m stations were arrayed for this support and the real-time data were
received at 44.8 kbps with playback data at 29.8 kbps. These images, which were taken
under better lighting conditions and at a better approach angle than had been possible
on Vyageer 1, were later used by the project to compile a Saturn rings movie.

A final pre-encounter trajectory correction maneuver on 18 August was supported over

DSS 12 and DSS 14. During the maneuver sequence, the spacecraft antenna was placed
off Earth point for over an hour resulting in loss of the downlink for that time. DSS
14 reacquired at the proper time and the telemetry data indicated that the trajectory
correction had been performed correctly.

In preparation for the extensive radio science activities which would occur over DSS 43
during this encounter, a final radio science Operational Readiness Test was conducted
with the Canberra Complex on 19 and 20 August. AIl of the encounter equipment was
operational, and the acrual operauonal sequence was used for the test. w t h the suc-
cessful completion of this test, the DSN was declared ready to support the second
Voyager encounter of Saturn.14
The Near Encounter Mission phase started on 25 August 1981. Recording of celestial
mechanics data in the form of closed-loop Doppler and ranging data had begun earli-
er and was continued through the encounter period. High-rate imaging data of the
closest approach sequences from the narrow-angle camera was obtained at 44.8 kbps in
the arrayed mode. Image reception in the arrayed configuration was excellent and no
images were lost due to DSN operations. With most of its Saturn observations com-
pleted, Vqydger 2 passed within 63,000 miles (101,000 krn) of the Saturn cloudtops, at
9.50 P.M. (Earth Received Time), 25 August 1981. Still to come were observations of
the dark side of the planet and southern hemisphere, the underside of the rings, and
several satellites.

Shortly after 10 P.M. PDT, the imaging sequence ended and the spacecraft began to dis-
appear behind the disk of Saturn. For the next 90 minutes, it would remain occulted
by the planet. As it entered occultation, and again as it exited, the DSN radio science
System at DSS 43 recorded open-loop and closed-loop receiver data. The data would
yield information regarding the Saturn atmosphere and ionosphere and the microwave-
scattering properties of the rings.

When the spacecraft exited occultation and the DSS 43 downlink was reestablished,
telemetry data indicated that the scan platform had not carried out its programmed
pointing sequences while it was behind the planet. Only black sky image frames were
being received. Playback of the tape recorder data from the spacecraft indicated that
the scan platform had functioned properly while the spacecraft was occulted and the
fault had occurred just prior to egress. Investigation of the problem by the project
resulted in restoration of the scan platform capability some days later and the Saturn
imaging sequences resumed. Throughout dl of this action, the DSN continued to pro-
vide Network support and accommodate numerous schedule and sequence changes as
the situation required.

A trajectory correction maneuver on 29 September refined the Vyager 2 flight path to

Uranus with a swing-by assist toward Neptune.

Preliminary science results born Vyager 2 revealed that subtle changes had taken place
in the Saturn atmosphere since the byager I visit nine months earlier. Vyager 2 saw
more detail in the atmosphere and much more detail in the rings, which codd be num-
bered in the thousands. Voyager 2's trajectory took it on a wide arc through Saturn's
magnetic field, expIoring different regions and adding to the data obtained by Pioneer
I I and Vqager 1. On its passage through the Saturn system, Vyagu 2 passed closer to
some of the satellites, hrther away from others, than did Vyageer I, and it returned a
magnificent set of images of the planet and all its major satellites.

DSN support of the Vyager 2 Near Encounter activities at Sarurn was accomplished
without any significant problems. Radio science played a major part of the encounter
operations at DSS 43. During the closest approach period, DSS 43 generated 10 medi-
um-band and 40 wide-band Digital Original Data Records (DODRs) for radio science.
These were used to produce 484 radio science intermediate data records (IDRs). AU
DSN Near Encounter operations were conducted in the arrayed mode. The quality of
telemetry imaging data was excellent and no images were lost.

Even with the scan platform problem, it was considered that the Voyager Sarurn mis-
sion objectives had been met.

The DSN contributed to this success in two most significant ways. First, the successfd
operational use of "adaptive tracking" mode enabled the DSN to accommodate the
uplink problems created by the failed spacecraft receiver tracking circuit, and second,
the excellent performance of the Network in using the "arrayed configuration to enhance
the downlink performance to the point where 44.8 kbps real-time telemetry data at
Saturn range became a reality for the first time ever.

By the end of August 1981, Vaydgeer 2 had completed nearly half' of its three-billion-
mile journey to Uranus, measured from the launch in August 1977 to Uranus Encounter
in January 1986. Then four years old, the spacecraft was in good condition except for
the radio receiver problem discussed above and some difficulty with the scan platform
pointing sequences which stuck, shortly after closest approach to Saturn on 25 August.
Since then, the anomaly had been thoroughly analyzed and understood, and the scan
pladorm had been maneuvered successfully several times. Prospects for a successfd Uranus
Encounter appeared to be good.

Nevertheless, at the final press conference for the Saturn 2 Encounter, several of the
speakers, including Ed Stone, the Imaging Team Leader, and Bruce Murray, the JPL
Director, reminded those present of the long hiatus in deep space missions that lay
ahead. It would be five years before the DSN would see the launch of another deep
space mission, and that would be Galileo--or so they thought.
Venus BdoonlPathfinder

In the period between 1983 and 1985, as international relations between the Soviet
Union and the United States began to improve, the first signs of scientific collaboration
in space-related endeavors appeared in the form of a cooperative project involving sci-
entists of the Soviet, French, and American space agencies. It would be called called
Venus BalloonIPathfinder. As Bruce Murray observed, "This technical partnership
between the United States, Europe, and Russia came about despite the absence of any
formal relations between NASA and the Soviet Union. The original U.S.-USSR bilat-
eral space agreements of 1972 (which facilitated, among other endeavors, the Apollo-Soyuz
handshake in space in 1975) expired in 1982. Renewal became a casualty of U.S. hos-
tility to the USSR, triggered by the Soviets' suppression of the Solidarity movement in
Poland and their invasion of Afghanistan."l5

On 11 and 15 June 1985, the Soviet WGA 1 and W G A 2 spacecraft released two
instrumented balloons into the atmosphere of Venus. The VEGA spacecraft continued
past the planet on their way to a rendezvous with Comet Halley in March 1986. Drifcing
with the Venus winds at an altitude of about 54 krn, the balloons travelled one-third
of the way around the planet during their 46-hour lifetimes. Sensors carried by the bal-
loons made periodic measurements of atmospheric pressure and temperature, vertical
wind velociry, cloud particle density, ambient light level, and frequency of lightning. The
data were transmitted to Earth and received at the DSN 64-m antennas and at severd
large antennas in the USSR Approximately 95 percent of the telemetry data were suc-
cessfdly decoded at the DSN complexes and in the Soviet Union, and were provided
to the international science community for analysis.lG These data would supplement cur-
rent knowledge of the Venus atmosphere obtained by earlier Soviet Venera spacecraft
and the NASA Pioneer Venus Probes.

Ground-based tracking support for the Venus balloon experiment involved an interna-
tional network of about a dozen radio astronomy antennas organized by the French
space agency, CNES, a more limited internal Soviet Network and the three 64-m anten-
nas of NASA's Deep Space Nenvork.17

Consequent upon the negotiation of appropriate diplomatic and technical agreements

in Moscow, Paris, and Washington, the DSN managers determined the trajectories of
both the VEGA 1 and 2 bus spacecraft during the Venus flyby phase to recover eeleme-
try from the balloon signal and, as part of the international network, to acquire VLBI
data from each balloonlbus pair while the two were within the same antenna beamwidth.
To meet these commitments, the DSN would employ, as far as possible, existing capa-
bilities normally used for planetary spacecrafi navigation, radio science, and radio astronomy.
However, there were some special requiremenrs connected with the Soviet spacecraft's L-
band downlink and telemetry system.

To deal with the downlink, JPL engineers designed, built, and installed L-band feed-
horns and low noise amplifiers at all three 64-m antennas. An L-band to S-band
frequency-up converter provided an S-band signal spectrum to the radio science and
receiver-exciter subsystems for subsequent extraction and recording of telemetry spectra,
one-way Doppler and the essential DDOR and very long baseline interferometry (VLBI)
data. The telemetry system was not so straightforward. The peculiarities of the Venus
Balloon telemetry system required the development of special s o b a r e to extract the
telemetry data burst from an open-loop recording of the L-band spectrum transmitted
by the balloon. Owing to the somewhat unusual nature of this mission, the normal JPL
s o b a r e development resources were not available to the DSN which had, therefore, to
seek help elsewhere. The necessary help and expertise needed to produce the s o h a r e
came from Spain in the form of a s o b a r e development team led by Jose M. Urech
and the engineering st& of the Madrid c0mplex.~8

Jose M. Urech was director of the Madrid Deep Space Communications Complex at
Robledo from 1981 until his retirement in 1999. Prior to becoming Director, he had
served for fifteen years as servo engineer and station analyst for NASAs 26-m station at
Cerebros near Robledo. At retirement, he had been associated with NASA tracking sta-
tions as a foreign national for thirty-three years.

Dr. Urech had roots in Madrid, Spain, and was educated in both Spain and France prior
to earning a doctorate in engineering {?om the Polytechnic Univexsiry of Spain in 1969.

Quiet-spoken, courteous, and low-key in manner, with a partiality for good food and
wine, he engendered confidence in all who dealt with him, at whatever Ievel, in NASA
business. His social graces, technical ability, and leadership skills were of great help in
integrzting his team of Spanish engineers into the American-oriented methodology of
the space program, and in minimizing the inevitable effect of cultural differences in
resolving issues that occasionally arose between the rwo.

In addition to carrying out his management responsibilities, Jose Urech made time to
pursue his technical interests. In 1969 and 1970, he first developed and demonstrated
the concept of combining the output of two antennas to improve telemetry reception.
This experiment, the first of its kind in the DSN, arrayed two antennas, 20 krn apart,
to enhance the downlink signal fiom Pioneer 8. In 1985, Jose Urech led a team of engi-
neers from the Madrid station in the development of the special s o h a r e needed to
process the telemetry signals &om the Venus Balloon mission. He also contributed to
that mission by coordinating the various technical efforts of participants from JPLINASA,
CNES (France), and IKI (USSR).

Throughout his career, Urech was highly regarded by JPL as a valuable consultant and
additional resource that could be, and frequently was, called upon to address technical
questions related to the performance and productivity of the tracking stations.

He retired from NASAtINTA in 1999 to pursue his interests in active outdoor activi-
ries, music, and science, and to assist his wife with her business ventures.

The success of the Venus Balloon experiment depended on two new, very precise orbit
determination techniques-Delta differential one-way ranging (DDOR), which was used
for the Voyager planetary spacecraft navigation by the DSN, and the VLBI, which was
used by radio astronomers for determining the position of extra-galactic radio sources.

The DSN would use DDOR and one-way Doppler techniques to determine the main
spacecrafi (bus) trajectory, for a two-week period, around the time of Venus flyby Balloon
position and velocity would be obtained from VLBI measurements between the main
spacecraft and the balloon, taken by an international network of ground-based radio
astronomy antennas, which included the three 64-m antennas of the DSN, over a peri-
od of approximately two days near each Venus Encounter, when both bus and balloon
were in the same DSN antenna beamwidth.

The DDOR navigation technique employed by the DSN for very precise determina-
tion of spacecraft orbits required a measurement on a nearby extra-galactic radio source
(EGRS). To find EGRS suitable for VEGA orbit determination, the DSN first had to
make measurements of the L-band correlated flux densiry of 44 potential EGRS taken
from the existing JPL radio source catalog of 2.3 GHz and 8.4 GHz sources.19

During the actual encounter period, the DSN performance in all three areas of support
was satisfactory. Balloon telemetry data obtained by the DSN were provided to the sci-
ence teams on computer-compatible magnetic tape in the form of original recorded
spectra and in the form of demodulated and decoded data streams. Most of the decod-
ed telemetry data was provided from MDSCC where Spanish engineers succeeded in
adaptively adjusting their data processing software to compensate for the totally unex-
pected wind conditions on Venus that created Doppler rates up to fZiy times ;\eater
than the predicted values on which the software designs were based.

Only seven of the ninety-two balloon telemetry transmissions were missed, mainly due
to downlink signal variations caused by the balloon itseK20

During the fifieen-day period of Venus Encounter observations, the DSN succeeded in
obtaining good DDOR data on 85 percent of the attempts. These data formed the basis
for the VEGA flyby orbit determination. The DSN also obtained good VLBI data on
sixty-seven of the sixry-nine balloon transmissions. These data, in conjunction with the
VEGA orbit data, were used to estimate balloon position and velocity.

The end of the Venus Balloon experiment did not, however, spell the end of DSN
involvement with the two VEGA spacecrafi. Nine months later, in early 1986, the DSN
would again be providing D D O R data from the VEGA spacecraft in support of Patl&nder
operations for the Giotto mission to Comet HaUey.

DSN support for these missions took place as the DSN moved into the Galileo Era
described in the next chapter.

The Downlink Problem

As the excitement associated with the successful Saturn encounters in 1980 and 1981
subsided and the two Voyager spacecrafi began to move along their new trajectories,
Vyageer I toward the edge of the solar system and Voydgeer 2 toward Uranus, the Voyager
Project and the DSN began negotiations for even more enhanced tracking and data
acquisition support than that provided for Saturn. At a disrance of 20 AU from Earth,
double the range of Saturn, with the downlink signal correspondingly reduced to 25
percent of its strength, the Voyager requirements for Uranus posed a further challenge
to the ingenuity, innovation, and expertise of JPCs Telecommunication Division, which
provided engineering support to the DSN. Basically, the problem was that of compen-
sating for the loss of downlink due to increased spacecrafi range (at Uranus) with increased
signal gathering capability on the ground.

In the early 1980s, the DSN had adequate technical capability to support the planetary
missions then in the mission set, at distances of 5 to 10 times the Earth-Sun distance,
5 to 10 Astronomical Units (AU). It was not until the successful Saturn Encounter in
August 1981 that a Uranus Encounter at 20 AU in 1986 became a real possibility. The
DSN antennas did not then have the additional 4 to 6 dB of downlink capability that
would be required to support the desired imaging data rate fiom Voyagm 2 at the dis-
tance of Uranus.21

The NASA planetary program was entering the hiatus period of the 1980s and, other than
Voyager 2 at Uranus and possibly Neptune, there were no future missions in view that
would jusufy additional, large, new antennas. However, earlier DSN studies had drawn
attention to the sigrdicant improvement in downlink performance that could be obtained
kom an array of existing antennas.22 Furthermore, it was (theoretically) possible, by the
use of data compression techniques in the spacecrafr, to increase the efficiency of the
telemetry data stream itself. By combining the improvements derived from DSN arraying
with the improvements resulting from spacecrafr data compression, the downlink capa-
bility required to obtain the desired science data return at Uranus could be realized.

Voyager 2 at Uranus in 1986 could not be regarded as merely a five-year extension of

the 1981 Voyager Jupiter-Saturn mission, for it was indeed a new mission from a DSN
point of view. It had been renamed the Voyager Uranus Interstellar Mission.

To this end, a comprehensive study of DSN options for employing arraying techniques
to enhance DSN downlink performance for Voyager encounters of Uranus and possi-
bly Neptune was undertaken in 1982. The study was not limited to DSN antennas and
therefore included a survey of all h o w n large antenna facilities, including those of
Australia, England, Germany, Japan, Italy, and Russia. The "Interagency Array Study"
Team was led by J. W. Layland and published its recommendations in April 1983.23

In considering arrangements for interagency arraying, the study recommended that per-
manent ties should be sought with other space agencies such as Japan's Institute for
Space and Aeronautical Science (ISAS), but that for shorter term goals, such as support
of Voyager at Uranus, a radio astronomy observatory seemed more appropriate. In this
context, the array of the Canberra Deep Space Communications Complex (CDSCC)
with the 64-m antenna of the Australian Commonwealth Scientific and Industrial Research
Organisation (CSIRO) radio astronomy observatory at Parkes, New South Wdes, was
recommended as most viable DSN configuration for support of the Voyager at Uranus.
A photograph of the 64-m antenna of the CSIRO radio astronomy observatory at Parkes,
New South Wiles, Australia, is shown in Figure 4-7.
Figure 4-7. Parkes 64-m radio astronomy antenna.

The downlink performance of a l l the DSN arrayed complexes and the enhancement
that would be provided by the addition of the Parkes antenna to the CDSCC array is
shown in Figure 4-8, which is taken from this study.

In addressing the specific needs of Voyager at Uranus, the study had started with the
Voyager project requirement for an imaging science data return of 330 111 hame images
per day with about 30 to 50 images per day in the few months preceding encounter.
Coupled with this requirement was a possible improvement in the efficiency of the space-
craft telemetry data transmission system resulting from the use of new coding and data
compression techniques in rhe spacecraft Flight Data Systems. The new spacecrafi capa-
bility would be implemented by inflight reprogramming of the rwo redundant Flight
Data System processors to hnction as a dual parallel data processor. This enabled the
spacecraft to transmit the science telemetry at lower data rates than would otherwise
have been possible.
f
I
29.9

21.6

~DIN~,
-------- 19.2
DB HZ DATA RATE
KILOBITS -
I SEC
14,4 90% CONFIDENCE

8.4

7.2

19 20 00 04 08 12
GMT HOURS
i J

Figure 4-8. Enhanced link performance for Voyager 2 at Uranus. The figure predicts how the
telemetry data downlink between the Voyager spacecraft at Uranus and the DSN antennas on Earth
would vary as Earth rotated. The vertical scales depict the maximum sustainable rates at which
science data could be transmitted over the downlink (right) with their equivalent signal-to-noise
ratios (left). The rotation of Earth is represented on the horizontal scale in terms of GMT. The lower
curves represent the situation for arrayed antennas at each of the DSN sites in Goldstone, California;
Canberra, Australia; and Madrid, Spain. The upper-center curve shows the greatly enhanced per-
formance that would result from coupling the DSN arrayed antennas at Canberra with the Australian
radio astronomy antenna at Parkes.

An assessment of the imaging science data return resulting from this new spacecraft capa-
bility and the link performance shown in Figure 4-19 concluded that the combined
capability of all three DSN complexes, supplemented at CDSCC with the Parkes Antenna,
would be able to return 320 full frame images per day under optimum conditions.

Al&ough there were some qualifications to that estimate relating to spacecraft tape recorder
strategy and telecommunications link uncertainties, these were understood and accept-
ed. The requirement was very close to being satisfied in a feasible and cost-effective way
and planning to support Voyager at Uranus with each Complex M y arrayed with one
64-m and two 34-m antennas (only one 34-113 at Madrid) supplemented with the 64-
m antenna at Parkes, went forward.

This would be the first time that interagency support had been provided for a major
planetary encounter, and as such the Parkes-Canberra Telemetry Array (PCTA) would
be a "pathfinder" for hrther applications of this technology. And much sooner than was
foreseen at the t i e , interagency arraying would become established as an alternate resource
for DSN support of many NASA and non-NASA deep space missions.

It would be a busy time for Director Tom Reid and his staff engineers a s implemen-
tation of the PCTA and preparations for the Voyager Uranus Encounter played out
simultaneously at the Canberra Complex. Nevertheless, it was a situation Reid had expe-
rienced on several occasions in the past. In fact, most of the major expansion of the
tracking facilities at the Canberra Deep Space Communication Complex (CDSCC), in
Tidbinbilla near Canberra, took place during Tom Reid's directorship of the Complex.
In that eighteen-year period, the 64-m antenna was built and later increased in size to
70 m; the 26-m antenna was enhanced to 34 m; a new, high-efficiency, 34-m antenna
was built; and X-band uplinks and downlinks were added. Their successhl integration
into the Network and subsequent record of outstanding service to NASA spaceflight
programs owed much to his cooperation.

A native of Scotland and an electrical engineering graduate of the University of Glasgow,

Reid served in both the Royal Navy and the Australian Navy before taking charge of
telemetry services for the Australian Weapons Research Establishment's rocket test range
at Woomera in 1957. His association with NASA began in 1963 with radar support for
the Gemini program, at the Red Lake site near Woomera. In 1964 he took charge of
the NASA tracking station at Orroral Valley, moved to Honeysuckle Creek in 1967, and
in 1970 he was appointed to succeed R. A. Leslie as Director of the Tidbinbilla Station.

His crisp management style and penchant for clear lines of authority, particularly in his
relations with JPL and NASA personnel, made a visit to "his" Complex a memorable
experience for many Americans. He ran the station in a disciplined, formally organized
way that attracted and retained the best technical staff available. As a direct result of
their teamwork and his leadership, the CDSCC played a critical role in all of NASA
deep space missions in the years 1970 to 1988.

At the time of his retirement in 1988, his wife, Margaret, represented the Australian
Capital Territory as a senator in the Australian Parliament.
The design and operational use of the arraying system wodd be based on the success-
ful demonstration of arraying techniques in a DSN operational environment that had
been observed during the Jupiter and Saturn encounters, and the additional new DSN
capabilities that were becoming available as part of the Mark IVA implementation effort.

Unlike the DSN antennas, which were colocated at each Complex, the Parkes antenna
was some 280 kilometers CDSCC. In addition, it required completely new receiving
and telemetry equipment to make it compatible with the Voyager downlink. The design
and implementation of the PCTA was a joint effort of JPL, CSIRO, and the European
Space Agency (ESA).Z* ESA became involved because of an existing arrangement with
CSIRO to support an ESA mission to Comet Halley called Giotto, in 1987, shortly
after the Voyager Uranus Encounter.

In summary, CSIRO was to provide a new feedhorn and antenna with upgraded sur-
face for X-band operation, ESA was to provide the X-band maser and down converter,
and JPL was to provide an interfacing microwave assembly, 300-MHz receiver, format-
ting and recording equipment and a baseband data interface to the 280-km intersite
microwave link supplied by the Australian Department of Science.

Together, these facilities provided the critical elements of a real-time combining system
similar to that used at Goldstone for the S a w n Encounter with DSS 14 and DSS 12.
However, in this case a special, long-baseline, baseband combiner would be required at
CDSCC to compensate for the much longer delay on the baseband signal path between
Parkes and CDSCC. At CDSCC, baseband signals from DSS 14, DSS 42, and DSS
45 would be combined first, before being combined with the Parkes baseband signal in
the long-baseline combiner. In addition, the new Mark IVA monitor and control facil-
ities provided greatly improved conditions for correctly operating these very complex
configurations in a critical real-time environment. A functional block diagram of the
Parkes-CDSCC telemetry array configuration is shown in Figure 4-9.

Because of the critical role of CDSCC and Parkes in the Uranus Encounter strategy,
both sites were provided with pairs of Mark N A digital recorders to back-up the real-
time system with baseband recordings, at both sites.

Arrangements at the Goldstone and Madrid complexes were similar, but simpler. At
Goldstone, the baseband signals from DSS 14 and the two 34-m antennas, DSS 12 and
DSS 15, were combined in the new Mark IVA Baseband Assembly (BBA) which, in
addition to the baseband combining function, carried out the functions of subcarrier
demodulation and symbol synchronization. These two latter functions had been carried
 ------ MARK IVA DEN
ESA PARKES

--------
ESA PARKES
CSlRD PARKES

-4 4
FREQUENCY AND TIMING I
A
FACtUTlES

I P A S 4- PCTA PARKES TELEm

FREQUENCY AND TIMING

F'CTA cDscc

I
Figure 4-9. Functional block diagram of the Parkes-CDSCC telemetry array for Voyager Uranus Encounter, January 1986.
out by separate assemblies in the former Mark I11 system. At Madrid the array consist-
ed of just the 64-m antenna and the 34-m srandard antenna at DSS 61.

In addition to its downlink capability a hrther item of great concern in connection

with the Uranus Encounter was the pointing capability of the DSN antennas.25 To avoid
degrading the downlink signal from the spacecraft significantly, it was determined that
the DSN antennas must be pointed to less than 0.16 of the anrenna half-power beamwidth.
This corresponded to less than six one-thousandths (611,000) of a degree or 6 millide-
grees (mdeg) for the 64-m antennas, and 11 millidegrees for the 34-m antennas. The
new Mark IVA antenna pointing system contained new computers and new equipment,
all of which required new calibrations and procedures for making the calibrations.

There were two ways of pointing the DSN antennas. In "blind pointing," an antenna
pointing prediction program derived from the spacecrafi ephemeris, the antenna com-
puters are fed data for necessary corrections before sending pointing signals ro the servos
to drive the antenna. This method did not rely on any downlink signal from the space-
craft, and was used during radio astronomy occultation observations and spacecraft
acquisitions. In "conical scanning" (conscan) the antenna was moved in a conical scan-
ning motion about the approximate direction of the spacecraft. Error signals, derived
from a comparison of the downlink signal strength at opposite parts of the scan pat-
tern, were used to drive the antenna to a "null" position. The ccconscan"mode &us
enabled the antenna to sense the apparent direction of the spacecraft radio signal. This
method was useful in the absence of accurate predictions, or for searching for a "lost
spacecraft" signal.

During the pre-encounter period, a process for initial pointing calibration of the Mark
IVA antenna system was developed at the Madrid Deep Space Communications Complex
(MDSCC), using radio stars as distant target points of known position. After a series
of radio star observations, the data were reduced to pointing angle offsets and used to
produce a systematic pointing error model. These data then provided the corrections
needed by the antenna predicts program for "blind pointing." Typical accuracies obtained
fiom these observations were less than 4 mdeg for the 64-m antennas and less than 6
mdeg for the 34-m antennas.

However, rado stars near the 23-degree south declination of the Vyagm 2 spacecrafi were
not available. For that declination, "conscan" offset data from live Voyager tracking passes
were used. Comparison of the apparent direction of the spacecraft signal given by conscan,
with the direction given by the corrected predict program, was the conscan offset. To the
extent that the conscan properly sensed the apparent direction of the spacecraft signal, the
conscan offsets determined the total system pointing errors. Usiing this method on a large
number of tracks at Goldstone, it was determined that the conscan defined axis and the
a c d antenna beam axis were coincident within 1 to 2 idlidegrees. Subsequent tests at
Goldstone verified that the downlink signal level was not significandy affected by the method
used to point the receiving antenna. W1th this basic information in hand, the CDSCC car-
ried out an extensive program of blind pointing exercises in preparation for support of the
radio science events connected with the Uranus encounter.

These investigations substantially improved DSN confidence in the ability of its anten-
nas to support the outer planet missions without incurring significant downlink
degradation due to pointing errors, and laid to rest the initial concerns about antenna
pointing errors reIative to the imminent Voyager Uranus Encounter.

The regular weekly Voyager project meetings provided a forum for discussion and
resolution of the issues and concerns described above. Based on these discussions,
Voyager and DSN representatives gradually developed a mutually acceptable plan for
DSN support of what was by then called the Voyager Uranus Interstellar Mission.26
Marvin R. Traxler continued to represent the DSN in these negotiations, while the
project was represented by George I?. Textor, the newly appointed Mission Director
for the Voyager project.

The DSN plan for Uranus met the project requirements for downlink telemetry at a
maximum data rate of 29.9 lsilobits per second, improved the quality of radiometric
data for spacecraft navigation purposes, and M e r improved and expanded the radio
science data gathering capability of the Network. Antenna arraying techniques, based on
past experience with the Jupiter and Saturn encounters, and including the new, 34-meter
high efficiency antennas and the Parkes radio astronomy antenna in Australia, were used
to enhance the downlink signal power received by the DSN stations, while the Mark
IVA- 1985 model of the Network provided the new capabilities required to provide teleme-
try, command, radiometric, and radio science services for Vqager 2 at the Uranus encounter.

Both of these initiatives represented major increments in Network capability and were
subsequently used to great advantage not only by Vyager 1 and 2, but by all of the
other ongoing missions. The engineering and implementation particulars of these con-
current streams of activity and the sensitive task of bringing them into service to meet
the Voyager need date (mid-1985), without disruption to the routine operational sup-
port of the ongoing missions, are discussed later this chapter.
f- \

M518T
SATURN CIA 719n9

1124186
URANUS CIA
107.000 krn
OCCULTATION
EXPERIMENTS

INTERSTELLAR

LEGEND

CIA CLOSEST APPROACH

FE FAR ENCOUNTER
1214- 14/85 BOW SHOCK ME NEAR ENCOUNTER
CROSSING
OB OBSERVATORY
PE POST ENCOUNTER
TCM TRAJECTORY
CORRECflON
MANEUVER
\ I

Figure 4-1 0. Pictorial diagram of Voyager 2 mission.

O n 4 November 1986, 81 days before Voydgeer 2 was due to flash by the outer planet
Uranus, the Voyager flight team began continuous, extended observations of the Uranian
system at better resolution than possible from Earth. The one-ton spacecraft was then
travelling at nearly 15 krn per second (relative to Uranus), and radio signals travelling
at the speed of light were taking 2 hours and 25 minutes to reach the DSN antennas
on Earth, 2.88 billion krn away. Uranus lay 103 million krn ahead. This was the start
of the observatory phase during which the spacecraft and DSN would be fine-tuned to
prepare for the close encounter observations. A pictorial diagram of the Eyager 2 mis-
sion including all its planetary encounters is given in Figure 4-10.

Two weeks earlier, the Voyager Flight Team had conducted a Near Encounter Test (NET)
to validate the readiness of all elements of the project, including the DSN Flight Team
and Science Team, to commence the Uranus Encounter operations depicted in Figure
4-22. DSN participation in the NET demonstrated that while the DSN had the basic
capability to support the Near- Encounter phase, additional operational proficiency would
be needed to reach the level deemed appropriate for those critical operations. This was
in large part attributable to the schedule slips that had occurred in the Mark IVA imple-
mentation project. Hampered by budget changes and s o h a r e problems, completion of
the implementation tasks had consumed the time that was originally planned for oper-
ational training and proficiency activity.

O n 5 November 1986, the DSN reviewed its status and responded to the deficien-
cies that had shown up in the NET. Arraying, radio science, and operator training
were the principal areas of concern. The review found that, with the help of special-
ist engineering personnel supplementing the station operations crews during critical
periods, the DSN was ready to support the observatory phase. Further, the observa-
tory phase would provide sufficient additional operational experience to enable the
stations to deal with the Far- and Near-Encounter phases. With this proviso and sev-
eral hardware and software liens against various elements of the new Mark IVA system,
the DSN prepared to meet what was possibly its greatest challenge yet, the Vyager 2
encounter of Uranus.

The spacecraft made its closest approach to the planet on 24 January 1986. Despite
the earlier anxiety about its scan platform, onboard computers, and one remaining
active radio receiver, the Voyager spacecraft performed perfectly. A final trajectory cor-
rection maneuver scheduled for 19 January was canceled since the flight path was
deemed satisfactory without further refinement. After travelling an arc of nearly 5 bil-
lion km to Uranus, the Navigation Team estimated that the spacecrafi passed within
20 km of the aim point. This astonishing navigational accuracy had been achieved by
the use of spacecraft optical navigation to complement the Doppler, ranging and Delta
differential one-way ranging (DDOR) data types provided by the DSN. Navigation
accuracy is critical to the success of crucial science observations that depend not only
on knowledge of the spacecraft position but also on proper, accurate pointing of the
instruments aboard the steerable scan platform. The spacecraft executed these com-
plex sequences without incident. The science data return quickly grew to avalanche
proportions as each of the science instruments carried out its preprogrammed obser-
vations. Previously unknown facts about the planet, its atmosphere, rings, magnetosphere,
winds, and satellites were soon being presented to the "happiIy bewildered" scientists.
Typical of the many beautiful imaging science results is the striking picture of Miranda,
one of the inner moons of Uranus, shown in Figure 4-1 1.
Figure 4-11. Voyager 2 image of the Uranus satellite Miranda.

At a distance from Earth of two billion km (1.212 billion mi) and traveling at 65,000
krn (40,000 mi) per hour, Vyager 2 passed within 31,000 km (19,000 mi) of Miranda's
surface. Image motion compensation prevented smearing of the image during the long
exposure times necessitated by the low lighting conditions at Uranus. The clarity of this
image amply demonstrates the capability of the Voyager imaging system combined with
the benefits of image motion compensation and the excellent performance of the DSN
downlink communications channel.

The new Mark IVA systems all worked well throughout the encounter period. At all
sites, array performance, including the Parkes element at Canberra, was close to pre-
dicted values. The complex and lengthy radio science sequences at CDSCC were executed
without significant loss of data. The new Mark WA Monitor and Control System cre-
ated some difficulty at first, while the operations crews adjusted to the concepts of
centralized control in a critical operational environment.
For the DSN, the Voyager Uranus Encounter in January 1986 had been a challenge in
deep space communications and operations complexity. To meet the Voyager require-
ments for science data return, the DSN had designed, implemented, and put into operation
a complex system of arraying all the DSN antennas at each longitude. At Canberra, the
array had been supplemented with a non-DSN 64-m antenna, 280-km distance, belong-
ing to a foreign agency. O n the project side, a risky but successful inflight reprogramming
of onboard compurers had increased the efficiency of the telemetry data stream to com-
plement the DSN improvements in the downlink performance. Special uplink tuning
sequences were required to compensate for the Voyager spacecraft receiver problem.

At the same time, the DSN had completed a' Network-wide upgrade of the entire data
system to increase its overall capabilities and decrease operations costs by introducing
centralized control to station operations.

Together, the new technology, new methodology and limited time, created a difficult
position for the DSN in the weeks prior to encounter. Responding to this situation,
DSN operations and engineering personnel in Pasadena, Goldstone, Canberra, and Madrid
were able to meet the challenge and the science returned from the Voyager Uranus
Encounter attests to their success.

In the Voyager Uranus Encounter, three major advances in DSN capability, represent-
ed by the Mark WA Data System, antenna arraying, and centralized operations control,
had been successfully demonstrated under the most critical operational conditions. These
capabilities would carry the DSN forward into the Galileo Era, where once again the
Voyager mission would stretch the DSN downlink capability to its limits-the next
time at Neptune.
EXTENDED MISSIONS

H~~ros

The two Helios spacecraft successfully completed their prime missions in mid-1976 and,
with both spacecraft in good operating condition, continued to orbit the Sun and return
prime scientific data during repeated perihelion and superior conjunction periods for
many years. The DSN continued to support Helios as an extended mission by provid-
ing tracking support as resources became available, mainly on the 26-m Network. This
support was supplemented by the German tracking station, DSS 68 at Weilheim. Under
these conditions, the supportable downlink data rate was 64 bps.

In March 1980, the downlink transmitter on the Helios 2 spacecraft failed, and despite
numerous attempts by the DSN to restore communications with the spacecraft, no use-
ful data could be recovered. O n 8 January 1981, a command was sent to turn the Helios
2 spacecraft transmitter off to prevent it from becoming an uncontrolled source of radio
interference.

Helios I , however, continued to operate normally and, supported primarily by a net-

work consisting of the two DSN 26-m stations, DSS 11 (Goldstone) and DSS 44
(Honeysuckle), and the German station DSS 68 (Weilheim), it participated in impor-
tant collaborative experiments with Voyager, Pioneer, and ISEE-3 to study scientific
phenomena in the inner solar system.

When time became available on the 64-m subnetwork in December 1981, Helios 1 was
able to obtain continuous 64-m coverage for a six-day period coincident with the solar
conjunction period following the fourteenth perihelion passage. During this time, the
DSN radio science equipment was used to make measurements of Faraday rotation from
a Sun-Earth-Probe (SEP) angle of 4.1 degrees on entry, to an SEP on exit of 3.8 degrees.
To carry out this type of experiment, the 64-m stations had been equipped with rotat-
able, microwave, linearly polarized feeds and closed loop polarirneters to automatically
and precisely measure the orientation of the linearly polarized downlink signal.
Measurement of the Faraday rotation caused by passage of a radio signal through the
solar corona provides important scientific information about electron density and the
solar magnetic field.

Helios I continued to perform functions of this kind for several more years into the
Voyager Era, supported mainly by the 34-m antennas at low data rates (8-256 bps). As
the power from the solar generators weakened, the spacecraft could be operated only for
a few months in each orbit. Eventually, both spacecraft receivers ceased to operate, and
without an uplink command capability to send correcting signals, the high-gain anten-
na gradually drified off Earth "point." The downlink from Helios I was last seen at DSS
43 Canberra on 10 February 1986, nearly 12 years afier its launch in 1974. It was pre-
dicted that, with diminished power available, the spacecrafi would automatically shut
down its transmitters soon afier.27

In June 1978, NASA gave approval to continue the Viking Extended Mission for a fix-
ther Martian year (1.88 Earth years). This period of the Viking mission was called the
Viking Continuation Mission (VCM). Its purpose was to replace Orbiter imaging data
that had been lost or degraded during Sun occultation periods, to gather radio science
data including near simultaneous Lander Orbiter ranging measurements, and to con-
tinue to make occultation and gravity fields observations. Both of the Viking Landers
were placed in an automatic mode of operation to reduce the work load needed to man-
age Lander operations.

At the beginning of the VCM, management of the Viking project was transferred from
the Langley Research Center to the Jet Propulsion Laboratory. The original flight team
was reduced from about 800 personnel to 150, and DSN support for Viking mission
operations was cut from 168 to 40 hours per week. The operational workload was fur-
ther reduced by placing both Landers in an automatic mode of operation to eliminate
the need for daily attention by ground controllers. These economies in resources allowed
DSN personnel to redirect their efforts to support the more critical mission phases of
the Voyager and Pioneer Venus flight projects.

The depletion of expendable attitude control gas had started to take its toll on the
Viking Orbiters. Although Eking Orbiter I (VO-1) was still performing normally, Viking
Orbiter 2 (VO-2) continued to be plagued by intermittent gas leaks in the yaw-axis con-
trol jets. O n 25 July 1978, the transmitter on VO-2 ceased operating and the spacecraft
was presumed to have finally lost attitude control capability. By the end of rhe year,
both of the S-band transmitters on VL-2 had also failed, leaving the Lander dependent
on the relay link via VO-1 for returning its data to Earth. Nevertheless, both Landers
continued to collect meteorology and imaging data from the Mars surface and return it
to Earth either directly or via the relay through VO-1. Throughout the year, VO-I
remained in good condition and returned excellent radio science, imaging, and general
relativity data.
DSN support for the two Landers and the one remaining Viking Orbiter was curtailed
even hrther in 1979 to make resources available for the Voyager Jupiter encounters.
VO-1 was placed in a housekeeping mode in March and transmined only spacecrafi
engineering data. The science systems were deactivated and new operating and protec-
tion instructions were stored on the spacecrafi to make it as self-sufficient as possible.
Daily contact was not required, but the spacecraft was interrogated every two weeks to
check on its status.

In late 1979, VL-1 became a separate mission called the Viking Lander Monitor Mission,
operated autonomously by command sequences stored in the Lander onboard comput-
er. At the same time, VO-1 was designated as the Orbiter Completion Mission and
DSN support for it was increased.

Resuming its normal operating mode, VO-1 began returning high-resolution photo-
graphic coverage of areas of the Martian surface that had not previously been obtained
or adequately covered in the earlier phases of the Viking mission. Included in this cov-
erage were possible landing sites for future Landers. (The Mars Pathfinder mission in
1997 used these images.) Frequent occultation events provided opportunities for radio
science observations of Mars atmospheric fluctuations, correlation of ionospheric plasma
temperature with solar activity, and improved knowledge of Mars topographic features
and gravity anomalies. Both VL-1 and VL-2 remained in the automatic mode, per-
forming repetitive observations on 37-day cycles. VL-I ranging data obtained directly
from the surface of Mars was collected to aid in improving the existing models of che
ephemeris of the planet. Meteorological data and imaging data were also returned peri-
odically, as well as data from the inorganic analysis of the final sample of Mars soil.

In July 1980, the Ofice of Public Information at JPL issued the foltowing press release:
"After more than four years exploring Mars, NASA's Eking Orbiter 1 has almost reached
the end of its mission. The Orbiter has used most of its attitude-control gas, which keeps
its solar panels pointed to the Sun and the antenna aimed at Earth. When the gas was
exhausted, about 23 July, controllers at JPL sent commands to turn off Viking Orbiter I
to end its long and productive mission.

"Meanwhile, Viking Lander-1 is programmed to operate unattended on Mars into 1990,

perhaps even into 1994 . . . . As long as it survives, Viking Lander-1 will continue to
collect photos and weather data from Mars and, on command from Earth, transmit
them on approximately a weekly basis."
And so it was that at 20:15 GMT on 7 August 1980, a command from DSS 61 in
Spain turned off the S-band and X-band transmitters on VO-I. This would prevent
them from becoming an uncontrolled source of radio interference. Eking Orbiter I
would continue to silenrly orbit Mars for many decades until its decaying orbit would
lead it to destruction in the upper reaches of the Mars atmosphere.

The Viking Orbiter Project Scientist compiled the following statistics of general inter-
est for the Viking Orbiter missions.

V i g Orbiter Statistics of General Interest

Item VO-1 VO-2

Days from launch to end of mission 1,814 1,050
Orbits of Mars 1,488 706
Pictures recorded in orbit 36,622 16,041
Data bits played back fi-om the tape recorders 358x109 161x109
Distance of tape travel across recorder heads (krn) 2,955 1,397
Total number of commands sent by DSN 269,500
Tracking passes provided by DSN 7,380
Hours of tracking time provided by DSN 56,500

The Wking Orbiter 1 mission to Mars ended h e r five years of continuous DSN oper-
ational support. The demise of the last Orbiter also deprived Eking LanAr 2 of its
Earth relay capability and its mission, too, was ended.

The last of the Vikings, Lander I, which was located in Chryse Planetia, had been oper-
ating in an automatic mode since March 1979 transmitting stored engineering, imaging,
and meteorological data on command from one of the DSSs. During the first extend-
ed mission, one of the two spacecrafi radio receivers failed, leaving the other receiver
permanently connected to the Lander high-gain antenna. Since that time, the Lander
had been operating quite successfully but was completely dependent for its uplink and
downlink communications on a long-term antenna pointing program, previously stored
on the spacecraft, and was valid through 1994.

Managed and controlled by fewer than thirty people at JPL, this was to be the Viking
Monitor Mission and hopes were high for a low-cost, long-lived science observation post
on Mars. Radio science, meteorology, and comparative time-lapse imaging were to be
the principal mission objectives. These goals were achieved, over the next two years,
using whatever time became available in the busy DSN tracking schedules.

Despite the steady flow of good science data in 1982, all was not well with Lander 1.
Early in the year, three of the four nickel-cadmium storage batteries showed signs of
losing their energy storage capacity. For several months, the normal recharging sequences
failed. Finally, in an attempt to extend the life of the batteries, advice was sought from
experts in this field all over the country. These reviews led to recommendations for a
new battery charging strategy which would be incorporated into new Lander sequences
and uplinked to the Lander on 17 November 1982 from DSS 43.

After the uplink was transmitted to the Lander, the station searched for the expected
downlink, but could find no sign of an S-band carrier signal. Shortly thereafter, the
spacecraft controllers realized with horror that the new battery-charging sequence had
been inadvertently written to the memory locations occupied by the Lander high-gain
antenna pointing program.

It followed that the high-gain antenna was no longer pointing to Earth and that until
the situation could be corrected by giving the high-gain antenna the correct Earth point-
ing instructions, communicarion with the spacecraft was no longer possible. In the months
that followed, strenuous efforts were made ro get commands into Lander I to retrieve
control of the high-gain antenna but to no avail. Telecommunicationslink analyses showed
that with the 100-kW DSN transmitters sending commands there was sufficient signal
strength at the Lander to reach the receiver even through a side lobe of the high-gain
antenna. Other failure modes were proposed, tried, and failed to produce a downlink.

Finally the Mission Director decided that no further effort was warranted and, in March
1983, the Viking Lander Monitor Mission was terminated.

During the course of the Viking mission, a substantial contribution to the field of radio
science was made by investigators using the Orbiters, or Landers, or both. Based on
these results, scientific papers were published in the areas of general theory of relativity,
solar-wind scintillation, dynamics shape and gravity of Mars, gravity waves, solar gravi-
ty, solar-wind electron density, and improvement in the Mars ephemeris.

Observations from the VL-1 meteorology sensors provided major insight into many Mars
atmospheric processes such as frontal systems, annual climate variations, and dust storms.
The VL-1 cameras were programmed to view each Martian scene periodically with the
same camera parameters, including pointing. In this way, comparative imaging could be
used to observe for dust, condensation, or erosion over a period of time. Until the mis-
sion was terminated, several imaging frames showing these processes were obtained.

The Eking Lander 1was eventually renamed the "Thomas A. Mutch Memorial Station."
Dr. Thomas A. Mutch was the Viking Project Scientist who disappeared while on a
climbing trip in the Himalayas in 1980.

The world press noted the passing of the last Lander with many tributes and accolades.
For nearly twenty years, the Viking presence on Mars had provided a topic for a con-
tinuous stream of articles and reports of fact and speculation. In May 1983, the New
York City Tribune published an article by international science writer J. Antonio Huneeus,
which recognized the Viking endeavor quite succinctly: "Together the two Landers took
more than 4,000 photos of the Martian surface, while the Orbiters took over 50,000
photos. They mapped 97 percent of the Mars' surface with a resolution of 25 meters.
More than three million Martian weather reports were received between the two Landers
and a number of other experiments were conducted.

"However, what received most publicity at the time were the experiments designed to
search for 'evidence of living organisms and organic molecules on the Martian surface.'
Again, it was the first, and so far only, time that such an experiment had been con-
ducted on another planet. Although some scientists had high hopes that the experiment
would reveal important principles about extraterrestrial life processes, the results were
inconclusive.

"In NASA's own words, 'no clear, unambiguous evidence of life was detected in the soil
near the landing sites.' But the space agency adds that, 'the question of life on Mars,
according to Viking biologists, remains an open one.' "

And that is where it remained for the next twenty years until the new Mars initiative
opened in 1997, with the spectacular appearance of Mars Pathfinder on the world's space
scene. It, too, was renamed to memorialize one of its strongest advocates. The new sta-
tion became the "Carl Sagan Memorial Station."28

Pioneers 6, 7, 8,7

The Pioneer 6, 7, and 8 spacecraft remained in an orbit around the Sun similar to that
of Earth, varying between 0.8 and 1.2 AU. All three spacecraft remained in good oper-
ating condition as they continued to be supported as extended missions, through the
Voyager Era. They provided field and particle data individually and in conjunction with
other solar wind measuring spacecraft when conditions of alignment were favorable.

The last successful DSN acquisition of Pioneer 9 took place on 18 May 1983. Until
that time, the spacecrafi had been tracked periodically as schedules permitted, but on
the last scheduled track in October, a downlink could not be detected. Over a span of
several years, commands were sent to configure the spacecraft into all possible combi-
nations of receivers, decoders, transmitters, and antennas in an effort to reestablish the
downlink. The SET1 rnultispectral analyzer a1 Stanford was used to try to detect a weak
signal, but with no success. In all, eighteen separate attempts were made to revive the
spacecraft to no avail. The Pioneer Project Office declared the end of the Pioneer 9 mis-
sion on 5 March 1987.

Pioneers 10, 11

The Pioneer 10 spacecraft, launched in March 1972, reached Jupiter in December 1973
and continued on a path 2.9 degrees above the ecliptic that would eventually take it
out of the solar system. Throughout the Voyager Era, it continued to maintain an uplink
and a downlink with the DSN while returning a continuous stream of particle and field
science data, and participating in several collaborative experiments with other spacecraft
in different parts of the solar system.

The Pioneer 11 spacecraft was launched in April 1973 and reached Jupiter in December
1974. Using the strong gravitationa1 field of Jupiter to redirect its trajectory, Pioneer 11
reached Saturn in September 1979. Throughout the long cruise time between Jupiter
and Saturn, the spacecraft remained in excellent condition with few system problems.
During this period, the DSN provided routine tracking and data acquisition support
while the mission was controlled from the Pioneer Mission Operations Control Center
at Ames Research Center (ARC) in Mountain View, California. The heliocentric geom-
etry of the Pioneer 10 and Pioneer I1 trajectories are shown in Figure 4-12.

After its Saturn Encounter, Pioneer II escaped the solar system travelling a direction
opposite that of Pioneer 10, which was already on a solar escape trajectory. It was expect-
ed that Pioneer II could be tracked by the DSN through 1987.

The DSN prepared to support the Pioneer 11Saturn Encounter with new S-band masers
in all three 64-m stations. Additional downlink performance would be obtained for a
few days, during the closest approach, by operating the 64-m stations in the receive-
f \

\
u
URANUS

P - 11 OUTBOUND
ASYMPTOTE 12.6 deg
ABOVE ECLIPTIC
1. P - 10 JUPITER ENCOUNTER
DECEMBER 4,1973
2. P - 11 JUPITER ENCOUNTER
DECEMBER 3,1974
3. P - 11 SATURN ENCOUNTER
SEPTEMBER 1,1979

Figure 4-1 2. Pioneer I0 and 1I heliocentric trajectories, September 1979.

onIy mode to gain an extra 0.7 dB. Because of the extensive uplink command activity
1

required, the uplink would be provided by the 26-m stations consisting of DSS 44 and
DSS 62 and the recently completed 34-m antenna at DSS 12.

To further enhance the downlink capability during the 14-day encounter period, an exper-
imental. arraying technique would be used at Goldstone to combine the signals from the
34-m antenna at DSS 12 with those from the 64-m antenna at DSS 14 in a real-time
combiner. An improvement in signal to noise ratio of 0.4 dB to 0.5 dB was predicted
at antenna elevations above 13 degrees. This would be verified by demonstration tracks
using live spacecrafi data at 1,024 b/s (coded) over DSS 12 and DSS 14 prior to encounter.

Because the closest approach distance would be less than two Saturn radii from the cen-
ter of the planet, the rate of change of the S-band Doppler frequency at encounter would
reach as high as 70 Hz per second. This would be too great for the DSN receivers to
track unaided. This fact, coupled with the low downlink signal level resulting from the
extreme Saturn range, necessitated the use of preprogrammed receiver tuning at the sta-
tions that would observe the closest approach event of the encounter. The receivers at DSS
12, DSS 14, and DSS 63 would be automatically tuned to ramp out the rapid change in
Doppler as the spacecraft swung by the planet. In the absence of an automatic tuning
capability, relays of operators were employed to manually tune the receivers at DSS 62.
To assist the spacecrdi receiver in tracking the uplink frequency changes, similar arrange-
ments at DSS 12 and DSS 62 would control the DSN transmitter exciter frequency.

During the seven-day period on either side of closest approach, the DSN efforts to sus-
tain a 1,024 bps data rate to support the imaging polarimeter met with various degrees
of success. On occasions, problems with the Goldstone array, with telemetry decoder
performance, Canberra weather, and some radio interference from a Russian satellite
forced the project to reduce the data rate to 512 bps. The net result was a return of
approximately twenty pictures better than Earth-based resolution instead of the forty
pictures that would have been possible had the 1,024 bps data rate been sustained for
the entire period. Even so, the science data return was still very extensive, leading the
Pioneer Imaging Team to estimate that the scientific data return from the Pioneer 11
Saturn Encounter exceeded that from both the Pioneer 1 0 and Pioneer I1 encounters
of Jupiter.

In addition to imaging and field and particle science, several radio science experiments
were carried out during the encounter period. The ring plane crossing on 1 September,
the Saturn occultation shortly afterwards, and superior conjunction beginning on I I
September afforded opportunities for radio science observations.

Following its Saturn encounter on 1 September 1979, Pioneer 11, too, began to move
along an endless path above and out of the ecliptic in the opposite direction to Pioneer
1 0 and at a considerably steeper angle. All systems aboard the spacecraft were perform-
ing normally. L i e Pioneer 10, it would remain a M y operating spacecraft, collecting
fields and particle data and participating in collaborative experiments, as its signal slow-
ly weakened with increasing range.

Managed from the ARC Mission Control Center in Sunnyvale, California, both space-
craft would remain a presence on the DSN tracking support schedules throughout the
entire Voyager Era.

Performance estimates, based on measured ~erformancein 1980, indicated that the 64-
m stations using the 18.5-kelvin low-noise S-band receivers would be able to support
the Pioneer 1 0 spacecraft at a data rate of 16 bps through 1989, 4 years and 13 AU
greater than previous estimates. By 1990, the distance from Earth would be 7 billion
Ism, the DSN radio transmissions would be taking over 13 hours to travel to Pioneer
10 and back, and the signal level received at the 64-m DSN antennas would be 172
dB below 1 milliwatt (-172 dBm). At the same time, the Pioneer 11 distance would be
4.5 billion km and its data rate would be 32 bps. The life of both spacecraft would
ultimately be determined by the life expectancy of their Radioisotope Thermoelectric
Generator (RTGs) power sources, rather than the downlink telecommunication limits
of the DSN antennas.30

Pioneer 12 (Venus Orbiter)

Pionee~12, the orbiting spacecraft of the two Pioneer Venus missions of 1978 (the other
was a Venus atmospheric entry multiprobe) continued to operate normally through the
Voyager Era, collecting a wealth of new information about Venus. In 1980, the first
detailed radar maps of Venus were compiled using data returned from Pioneer 12. As
the shape of its 24-hour orbit gradually changed, Pioneer 12 observed the bow shock
and the ionospheric tail of Venus in 1983 and 1985. Making use of its ultraviolet spec-
trometer, Pioneer 12 observed Comets Encke in 1984, Giacobini-Zimer in 1985, and
Hdley in 1987, as well as several other comets.

Although the spacecraft used an S-band uplink and downlink for regular communica-
tions with the DSN, it also carried a 750-&watt, X-band transmitter intended specifically
for carrying out radio science experiments during occultation periods. The DSN sup-
ported numerous radio science experiments with Pioneer 12 using this capability Celestial
mechanics experiments determined irregularities in the Venus gravitational field and dual
frequency (S- and X-band) occultation experiments provided new data about the Venus
atmosphere.

Still active in its prime mission phase, Pioneer 12 was supported regulariy on the 64-m
network through the entire Voyager Era.
HIGHLY-ELLIPTICAL EARTH ORBITERS

INTERNATIONAL EXPLORER
COMETARY (ICE)

The ICE spacecraft started life as the htemational Sun Earth Explorer (ISEE) and spent
the first five years of its life in heliocentric orbit observing solar wind effects in the
region between Earth and the Sun. During this period, tracking support was provided
by the 26-m stations of the Ground Spaceflight Tracking and Data Network (GSTDN).
In 1983, ISEE-3 was retargeted to intercept Comet Giacobini-Zinner on 11 September
1985, and renamed the International Cometary Explorer (ICE). The Spacecraft to Earth
distance at comet encounter would be 0.47 AU. On its new trajectory, ICE soon exceed-
ed the tracking capability of the 26-m antennas and, in January 1984, became the first
of the highly elliptical Earth-orbiting satellites to depend on DSN 64-m antennas for
tracking and data acquisition support.

Designed for use in a near-Earth orbit of about 0.01 AU, the ISEE-3 downlink would
be reduced in strength by a factor of 2,500 when the spacecraft, as ICE, encountered
the comet. Nor only would the DSN 64-m antennas be required, but significant, inno-
vative enhancements to the existing DSN receiving and telemetry data processing capabilities
would also be necessary to support a telemetry data rate of 1,024 bps from ICE at
comet encounter.

The ICE spacecraft carried two identical, S-band, 5-watt transmitters that could be used
interchangeably to carry telemetry or ranging data. One transmitter operated at 2,270
MHz, the other at 2,217 MHz. To overcome the problem of downlink signal margin,
the DSN proposed to transmit telemetry on both downlinks simultaneously and to com-
bine the separate signals at the DSSs before feeding the combined signal to the
demodulation, synchronizing, and decoding processors. In addition to the two down-
links from a single 64-rn station, single downlinks from nearby 34-m stations could be
added to rhe combiner, to further enhance the downlink prior to data processing.

Earlier studies had evaluated the feasibility of this approach and estimated that, under
optimum conditions, this arrangement could provide an improvement of 3 to 4 dB
compared to the use of a single 64-m antenna and a single downlink.

Implementation of the scheme, however, proved to be very difficult. Simultaneous recep-

tion of the two downlink carriers on a single 64-m antenna required the addition of
new wide-bandwidth masers, which initially experienced problems with gain and band-
width stability. Special multiport resistive combiners were built and installed at each of
the DSCCs.

Several hundred hours of link performance on the ICE data were accumulated and evd-
uated in mid-1985 as the encounter time approached. The marginal signal power available
from the spacecraft at this range made it essential to account for every 0.1 dB of teleme-
try performance that could be extracted from the combined downlinks.

The large radio astronomy antenna at Arecibo, Puerto &co, was also scheduled to sup-
port the ICE for a short time during the crossing of the comet tail. For this purpose,
new receiving and telemetry equipment for the 2,270 Mhz downlink was provided by
GSTDN.

Additional support near the Australian longitude was to be provided by the Japanese
64-m antenna at Usuda, Japan. Under a joint NASA-ISAS agreement, the station would
be implemented with tbe necessary low-noise maser amplifier and telemetry processing
equipment by the DSN. The high aperture efficiency of the Usuda antenna, together
with the low-noise maser configuration, resulted in a downlink performance almost equiv-
alent to the DSN 64-m dual channel configuration.32 A block diagram of the DSN
configuration for support of the ICE encounter with comet Giacobini-Zinner is shown
in Figure 4-13.

Configuration testing, resolution of problems, and intensive crew training continued at

all sites right up to the actual encounter time. ICE passed through the tail of Comet
Giacobini-Zinner on 11 September 1985. The three sites having view of the event,
Madrid, Goldstone, and Arecibo, successfully acquired the spacecraft and reported good
downlink performance at or above the predicted level. Telemetry data was sent via high-
speed data line to the ICE Principal Investigators at GSFC from all three sites
simultaneously. The Madrid Complex sent arrayed data from the 64-m antenna and one
34-m antenna, while the Goldstone Complex sent data from its 64-m antenna and two
34-m antennas. The Arecibo site sent data from one downlink during the overlap of
the Madrid and Goldstone view periods. The multiple data streams allowed &e mission
controllers at GSFC to choose the best data stream for delivery to the investigators for
scientific analysis.

The return of telemetry data at 1,024 bps, from this historic first encounter w i h a
comet, was made possible to a large degree by the application of DSN analysis, engi-
neering, and operations expertise to refinement of the ICE downlink.
f 'r
COSCC
I GDSCC MDSCC

NOCC ECS

Figure 4-1 3. DSN configuration for ICE encounter of Comet Giacobini-Zinner, 11 September 1985.
For the rest of the Voyager Era, the DSN would track ICE as schedules permitted while
it continued to return science data, generally at a lower data rate, as it pursued its high-
ly elliptical Earth orbit around the Sun.31

ACTIVEMAGNETOSPHERIC
PARTICLE EXPLORER
TRACER (AMPTE)

The AMPTE spacecraft consisted of three separate satellites: Charge Composition Explorer
(CCE) provided by the U.S., Ion Release Module (IRM) provided by the Federal Republic
of Germany, and the United Kingdom Satellite (UKS) provided by the U.K. From high-
ly elliptical Earth orbits, the three satellites would engage in an international, cooperative
study of the interaction of the solar wind with Earth's magnetosphere. The prime mis-
sions would last about one year for the IRMIUKS, and four years for the CCE. Project
management was exercised by Goddard Space Flight Center (GSFC), for NASA, in
cooperation with DFVLR.

Following launch from Cape Kennedy by a single Delta 3924 launch vehicle in August
1984, dl three satellites were placed in highly elliptical Earth orbits by onboard propul-
sion units. The final orbit for CCE was 1,148 kxn by 7.78 Earth radii (Er), while that
of the IRM and UKS was 407 krn by 17.7 Er.

All S-band uplinks and downlinks for telemetry, command, and ranging communica-
tions provided by the spacecraft were compatible with the capabilities of both ESA and
DSN ground tracking stations.

The AMPTE would be the first Highly-elliptical. Earth Orbiters (HEO) to be support-
ed by the DSN following completion of the Network consolidation project. Beginning
in August 1984, the three 26-m antennas recently transferred fiom GSFC (DSSs 16,
46, 66), would supplement the DSN 34-m standard antennas to provide coverage for
the CCE. The 34-m standard antennas would support all three spacecraft. The three
64-m antennas would be required to support critical downlinks only. The ESA tracking
stations provided additional support for the IRMIUKS satellites.

Following a critical period from August 1984 through April 1985, when daily 8-hour
passes were required, tracking coverage was reduced to two-hour tracks, twice daily, for
CCE spacecraft tape recorder data dumps and radiometric data support.

Downlink telemetry data rates for the CCE varied from 3.3 kbps (real time) to 105.6
kbps (playback). Uplink command data rate was 125 bps. Radiometric data consisted
of standard DSN ranging data only. Telemetry data was provided to the project at GSFC
in the form of digitd Original Data Records (ODRs). Telemeuy data from the 26-m
stations were provided in the form of analog recordings.

The AMPTE missions introduced h e DSN to the new H E 0 mission types which were
to become a very significant part of the total DSN effort in ensuing years. Although
the IRMIUKS satellites ceased to be of concern to the DSN in mid-1985, the CCE
continued to require support as an extended mission through 1989.
REIMBURSABLE MISSIONS

By 1984, the reputation, maturity, and unique capabilities of the Deep Space Network,
as well as the international value of the dollar, had made the DSN attractive to many
non-NASA and foreign space agencies as a viable alternative to their own facilities for
tracking and data acquisition support of Earth-orbiting missions. In this environment,
NASA had begun to explore the possibilities of providing DSN support for non-NASA
programs on a reimbursable basis, in addition to that which it already provided to non-
NASA missions on a cooperative basis. Representatives of the NASA Office of Tracking
and Data Acquisition (OTDA), supported by the technicd st& from the DSN, began
to establish working relationships with several other national space centers and foreign
space agencies to explore the possible use of DSN resources for non-NASA, Earth-orbit-
ing missions, on a reimbursable basis.

As a result of these actions, requirements for support of non-NASA missions began to

appear, along with those for NASA missions, in the standard DSN documentation of
the period.33 Suddenly, the DSN was involved, not only with a new class of Earth-
orbiting missions, known generically as "reimbursables," but also with many new foreign
space agencies, some of which are listed below.

1. Europe: European Space Agency (ESA)

2. Italy: Agencia Spaciale Italiana (ASI)

3. Germany: Deutsche Forschungs-und Versuchsanstalt her Luft und Raurnfahrt (DFVLR)

4. Germany: Bundesministerium h e r Forschung und Technologic (BMFT)

5. Japan: National Space Development Agency (NASDA)

6. Japan: Institute of Space and Aeronautical Science (ISAS)

7. Spain: Insrituto Nacional de Technica Aeroespacial (INTA)

8. France: Centre Nationale &Etudes Speciale (CNES)

9. Russia: (Former Soviet Space Agency) Astro Space Center (ASC)

Prior to the formalization of the reimbursable process in 1991, interactions between
those taking part in the interagency negotiations were rather informal. Joint technical
Working Group Meetings initiated by the DSN usually alternated between the various
foreign agencies and JPL, and were used to plan support for upcoming missions, dis-
seminate information of mutual interest regarding DSN current status and future plans,
and assess potential DSN support for hture reimbursable missions. Science considera-
tions for these missions were not a topic at these meetings. Where technical and operational
support by the DSN was the subject of discussion, Robert M. Hornstein, or another
associate, represented NASA Headquarters, while R J. Amorose, or one of his man-
agers, spoke for the DSN.

It was the joint Working Group Meetings that afforded a medium for the definition,
discussion, and approval of the detailed technical and operational interfaces between the
DSN and the non-NASA user agency. Later, these interfaces would be published in the
appropriate interface control documents. Quite frequently these activities were taking
place while the formal approvals were still in progress. It, too, was an iterative process
which, since NASA looked to the DSN for an estimate of the costs involved, necessi-
tated a thorough understanding of the prospective user's requirements before the desired
support could be formally approved. These meetings were a necessary precursor to the
formal request to NASA for reimbursable support.

Generally, the DSN appointed a Tracking and Data Systems Manager (TDSM) to con-
vene the meetings, represent it in the negotiations, and be responsible for all aspects of
DSN support once the mission began. The TDSM for the reimbursable missions, of
which there were several in progress at any one time, was also accountable for the DSN
antenna time and other resources which were expended in support of the mission, and
for reconciling with the original estimates on which the support had been based.

In a directive issued in 1991,34 the Director of Ground Networks, Robert M. Hornstein,

set out a "NASA policy covering the use and reimbursement of the NASA ground net-
works by non-NASA users." This document applied to the Deep Space Network, in
addition to the two other NASA ground networks, the Spaceflight Tracking and Data
Network (STDN), and the Wdops Tracking and Data Facility. It identified the avail-
able, standard services, which included tracking, radiometric data for orbit determination,
acquisition of telemetry data, transmission of commands, radio science data recording,
prelaunch support planning, and documentation and delivery of the data to a NASA
Communications Network (NASCOM) gateway. Other special services, unique to par-
ticular missions, were also identified. It was, in short, a "111service" policy.
Requests for service were to be provided "36 months before the requested service date
for NASA's assessment and determination if the service could be provided." Users were
required to "submit a letter of intent, together with an earnest money deposit to initi-
ate planning activities leading to the provision of services." Work could not begin until
a reimbursable arrangement describing, "the support to be provided, the schedule for
the support activities, the cost estimate, the progress schedule and any necessary terms
and conditions for the support," had been completed.

The costs for reimbursable services were based on 'the number of actual support hours pro-
vided by each antenna multiplied by an appropriate rate schedule," which was revised by
NASA each fiscal yea. To this were added miscellaneous costs for Headquarters and DSN
technical s&, travel and administration, software modification, DSN navigation support,
documentation, testing, and support from other elements of the ground network

A NASA Management Instruction (NMI), issued in 1991, conveyed the essence of this
policy to all NASA Field Centers, including JPL. The NMI included specific conditions
and procedures by which non-NASA organizations could request DSN support. Within
the DSN, these instructions were translated to the working level for the guidance of the
technical and administrative staff that would implement reimbursable mission operations
in the Nenvork.35

The costs and number of reimbursable missions in the DSN Mission Set varied con-
siderably over the years following their first appearance in 1985. The hourly cost for a
26-m or 34-m antenna from 1985 through 1990 was $920 per hour. By 1992 this had
risen to $1,100 per hour. The hourly rate for the 64-m antennas was double those rates.
In 1993, this rate, which was then adjusted annually to reflect the current operating
costs of the Network rose to $2,825 per hour for the 26-m and 34-m antennas, and
$5,650 per hour for the 70-m antennas.

Reimbursable missions made their first appearance in the Network toward the end of
the Voyager Era, with four television broadcast satellites, TV-SAT (German), TDF-1
(French), and BS-2B and MS-T5 (Japanese). DSN support for each of them was of rel-
atively short duration (days to weeks) and involved only the 26-m and 34-m antennas
with existing capabilities. From this modest beginning in 1985, the number of "reim-
bursables" in the total DSN mission set would steadily increase to become a major effort,
as the DSN moved into the Galileo Era.
NETWORK ENGINEERING AND IMPLEMENTATION

NEW34-METER HIGH-EPPICIENCY
ANTENNAS

With the success of the DSN 26-m to 34-m antenna conversion program fresh in
mind, the earliest design for the NCP/MK IV program was predicated on upgrad-
ing the 26-m GSTDN antennas which had been moved to the DSN sites as part of
the Consolidation program, to 34-m. These antennas were, however, X-Y mount
design and not at all suitable for DSN, deep space tracking applications. Further
reconsideration and cost versus performance studies led to a decision to build a set
of completely new antennas, based on a design study on larger antennas versus mul-
tiple antenna arrays, which had been made for the DSN by Ford Aerospace in the
mid-1970s. The new antennas were to be built as 34-m array elements having receive-
only capability. There would be no transmitter and the common aperture feed would
be optimized for X-band at the expense of S-band performance. DSS 15 and DSS
45 were funded in FY 1981 to meet the immediate need of Voyager at Uranus, and
funding for DSS 65 was deferred until FY 1984. By that time, the future Magellan
requirements for an X-band uplink were in part responsible for the completion of
DSS 65.36

The high-efficiency antennas, so-called because the aperture efficiency at X-band was
relatively high (approximately 70 percent), were intended to supplement the 34-m
standard antennas and the 64-m antennas in an arrayed ensemble of antennas. They
differed from the standard 34-m antennas mainly in having an azimuth-elevation
(Az-El) wheel and track mount rather than an Hour-Angle-Declination (HA-DEC)
mount. They had a single, common feedhorn for both S-band and X-band and no
transmitter. A Block I1 maser, transferred from the 64-m stations, provided for X-
band reception and a cooled Field Effect Transistor (FET) was used for S-band
reception.37

The antenna structure was designed to maximize stiffness, minimize distortion, and pro-
vide accessibility for maintenance. It was mounted on a four wheel azimuth turntable
consisting of a square pladbrm with self-aligning wheels, rotating on a circular, machined
track. The Az-El wheel and track mounting structure can be clearly seen in this photo
(Figure 4-14) of the first HEF antenna to be completed, DSS 15 at Goldstone.

The main reflector surface was shaped by the contiguous positioning of individual pan-
els to present a homologous paraboloid for improved microwave efficiency. The quadripod
and apex structure was designed to limit aperture blockage to five percent. The alu-
Figure 4-14. 34-meter high-efficiency (HEF) antenna: DSS 15, Goldstone.

minum subreflector was constructed from six individual panels overlaid with aluminum
tape to form a continuous conducting surface. The subreflector three-axis positioning
mechanism served to minimize microwave performance loss due to gravity deformation
of the antenna primary structure, quadripod, and primary reflector when the antenna
was operating at various elevation positions.

The antenna drive system was integrated with the Mark IVA configuration in the
SPC via interconnecting microprocessor computers. Antenna position control signals,
trajectory prediction data, systematic error correction tables, transformation of coor-
dinates, and monitor data were handled by drive and control equipment located on
the antenna, and by the pointing computers in the SPC. Microwave subsystem con-
figuration control was also exercised from the SPC. No operators were required at
the antenna site.

When completed, the high-efficiency antenna cost about $4,800,000 (1986 U.S. dol-
lars) each.38

Typical microwave performance parameters are given below:

Parameters X-band (8,400 GHz) S-band (2,300 GHz)

Antenna Gain (dB) 68 56
System Noise Temp.
(kelvin) 20 40

The parameters given in the table above are intended for illustration only Under oper-
ational conditions these values changed significantly, depending on the microwave
configuration in use, the specific low-noise front-end amplifier, the antenna elevation,
weather, etc. These effects were important factors to be taken into account in designing
uplinks and downlinks, and were given, explicidy for that purpose, in the "Deep Space
Network to Flight Projecr Interface Design Handbook."39

The first HEF antenna at Goldstone, DSS 15, was brought into service with the
Mark IVA and SPC 10 in October 1984, followed by DSS 45 at SPC 40, Canberra,
in December 1984. The HEF antenna at the Madrid Complex, DSS 65, became
operational with SPC 60 in April 1987 with the additional capability of a 20 k W
X-band transmitter to support the Galileo and Magellan missions. The added X-
band uplink transmit capability was retrofitted to DSS 45 and DSS 15 in 1987 and
1988.

Nowhere in the Deep Space Network was the ardent pursuit of refinement in the uplink
and downlink more apparent than in the 70-meter antennas. They were the single most
obvious structure to draw and hold the attention of the thousands of people who vis-
ited the complexes in Australia, Spain, and California each year. Technical and
nontechnical visitors alike were fascinated by the huge, gleaming white structures that
readily conveyed the sense of deep space communications links with spacecraft in dis-
tant parts of the solar system. These were not fragile laboratory instruments continuously
nurtured by white-suited laboratory technicians. These instruments were rugged engi-
neering marvels that were designed to run continuously, exposed to all kinds of
environmental conditions, from the searing heat of a Mojave Desert summer, to the
driving winds and fieezing rain of a Canberra winter, while maintaining a mechanical
and electrical performance equivalent to the most sophisticated of laboratory instruments.
Their size was deceiving. Seen alone against the natural contours of the surrounding
country, the size of the antennas did not appear to be their most striking feature. It was
the intricate complexity of the structure that impressed the viewer. But seen against a
human figure at the base or on the antenna, or a truck or building, the visitor was con-
founded by the size of the structure. Taller than a 20-story building and over one acre
in surface area, the antenna itself was impressive even when stationary. When it was in
silent motion, slewing from the horizon position of a setting spacecraft to acquire the
downlink from another spacecraft, rising like a planet from the opposite horizon, the
sense of awe was complete.

Impressive though the external appearance of the 70-meter antenna was, the exquisite
design and engineering that allows it to function as a major element in the uplink and
downlink between Earth and the distant spacecraft could not be seen from outside.
Indeed most of it could not be seen at all, for it lay embedded in the technical details
of the electrical and mechanical performance specifications that were associated with an
operational 70-m antenna of the Deep Space Nenvork.40

It was no accident that this "jewel in the crownyyof the DSN looked and performed
as well as it did, for it represented the culmination of a process of continuous devel-
opment, refinement, and operational experience extending over twenty years since the
first 64-meter antenna became operational at Goldstone, in 1966. From that time
until 1987, the network of three 64-m antennas served the DSN well in supporting
the mission sets of the Mariner, and Viking, and Voyager Eras described in this and
earlier chapters.

By 1982, however, interest in enhancing the capability of the 64-m antennas had reached
the point where the chief engineer of the Tracking and Data Acquisition O&ce, Robertson
Stevens, would recommend "that we embark on a program to design and implement
the full technical performance potential of the 64-meter antennas." He went on to give
performance and cost predictions for several options.

For Robertson Stevens, JPL was his first, and only, employer. Hired straight after col-
lege graduation (1949) with a master's in electrical engineering from the University of
California at Berkeley, Robertson (Bob) Stevens remained with rhe DSN throughout his
working life. With a family background associated with Naval aviation, Bob Stevens's
education path moved through San Diego, Sacramento, and the Naval Academy, gath-
ering qualifications before reaching Berkeley.

Stevens's career began with antennas and communications systems for the Corporal and
Sergeant missiles, during JPCs Army period. By 1960, the start of the Ranger period,
Stevens had become head of the Communications Elements Research Section in Rechtin's
Telecommunications Division. While Stevens concerned himself with developing the ele-
ments of communications systems; antennas and amplifiers, masers and microwaves,
transmitters and timing systems, Walt Victor's Communications Systems Research Section
focussed on research at the protoype systems level. In due course, Stevens followed Victor
to become manager of the Telecommunications Division. Most of the significant advances
in communications technology that eventually found their way into the operational Network
were made in that division, under his direction.

A great believer in physical fitness, Bob Stevens rode his bicycle five miles to and from
JPL each day across the busy Pasadena city streets. O n the job, he always seemed to be
around to encourage or help his engineers when the going was rough, sleepless like
everyone else during all-night tests at Goldstone, or calm and confident in the Network
Control Center at JPL, wearing his good-luck red vest during a launch or critical plan-
etary encounter. Together with Mahlon Easterling, who introduced the DSN to
block-coding as an enhancemenr for the Surveyor downlink, this author and Bob Stevens
appeared regularly with his flute or guitar to play carols at midday Christmas parties
around the "Lab" in the freewheeling days when that was permitted. Tousled dark hair,
a ruddy complexion, a ready smile, and a hearty laugh were typical of the man. Stevens's
casual, slightly ruffled appearance and somewhat shy manner belied the intensity of the
attention that he brought to bear on matters that concerned him. He had a wonderfd
ability to resolve issues between upper management, who often did not understand the
technicality of the issues before them, and engineers, who did not appreciate the con-
straints those technicalities an issue. Later, as chief engineer for the Telecommunications
and Data Acquisition Ofice, he applied his many talents to the role of "troubleshoot-
er" for the office.

He retired from JPL in 1991. As Stevens saw it, the "full potential of the 64-m anten-
nas" would only be realized by the "Ultimate" option described by the performance and
cost parameters shown below. And that, he said, was his recommendation. It was a pow-
erful statement, and on that basis the project went forward.
Options for Upgrading the 64-Meter Antenna

Cost per
Upgrade Option X-band Perf. S-band Perf. Antenna X-band Value
Increase Increase (FY 1982) in $M/dB4
Ultimate' 1.9 dB 1.4 dB $6,000,000 3.2

Gravity3 Compn. 0.3 dB 0.0 $330,000 1.1

1. Ultimate: Extend 64-m to 70-m; shaped precision reflectors; gravity

compensation.
2. Intermediate: Extend 64-m to 70-m; shaped reflectors; gravity
compensation.
3. Gravity compensation: Braces and Y-axis correction to 64-m or 70-rn
tipping structure.
4. For comparison, the arrayed addition of one new 34-m antenna to a
standard 64-m antenna provides approximately 1.3 dB performance
increase for a cost of $6.OM, as installed and equipped for X-band
reception only. The equivalent value is $4.6MldB. However, the added
34-m antenna would provide a capability for simuItaneous multiuser
support which is not given by the single, upgraded 64-m antenna
option.41

The 64-meter Antenna Rehabilitation and Performance Upgrade Project was established
in September 1982 with D. H. McClure as project manager. The guidelines to the proj-
ect manager were to "increase the technical performance of the 64-m antennas in terms
of gain, efficiency, and reduced (radio) noise," and to "carry out the rehabilitation of
the 64-m subnet."

The following account of the execution and successfd accomplishment of this task is
an abridged version of an unpublished report on the project by D. H. McClure and
F. D. McLaughlin.42

Overall program management was under the cognizance of the NASA Office of Space
Operations. Responsibility for project management was assigned to the Office of
Telecommunications and Data Acquisition at JPL. Implementation responsibility was
assigned by that ofice to rhe manager for the project, D. H. McClure. The personnel
required to carry out the project consisted primarily of JPL employees with support from
contractor personnel for some engineering, administrative, and documentation functions.
Major technical areas were monitored by cognizant specialist engineers assigned to the
project.

The project consisted of two main initiatives, rehabilitation and upgrade. The rehabili-
tation task included repair of the hydrostatic bearing and replacement of the radial
bearing, at the Goldstone 64-meter antenna, and investigation and determination of the
cause of the tilt on the 64-m antenna at Canberra. The performance upgrade task involved
increasing the existing 64-m antenna diameter to 70-m, improving the stiffness of the
structure by modifying the structural braces, and improving the subreflector focus capa-
bility by adding automatic Y-axis focussing. This latter task was eventually carried out
as part of the Mark TVA task.

In summary, the project costs were $3.8 million for the rehabilitation task and $4.1 mil-
lion for the upgrade task.

The rehabilitation task was completed at Goldstone by June 1984 and the DSS 43 tilt
investigation was completed in early 1985.

The upgrade task was completed and the (now) 70-m antennas returned to mission
operations service with 70-m performance as follows: DSS 63, July 1987; DSS 43,
October 1987; DSS 14, May 1988.

Rehabilitation of DSS 14 Pedestal

The "hydrostatic bearing" assembly, which forms part of the azimuth rotation system,
was a key feature of the 64-m antenna design. In this design, the 13.2-million-kilogram
rotating structure was supported by a film of high pressure oil approximately 0.25 mil-
limeter thick. The oil film formed the load-bearing medium between each of three "pads"
that carried the entire rotating structure, and a large, horizontal circular steel "runner"
bearing on which the shoes rotated in azimuth. The runner bearing had a finely machined
upper surface to support the pads and was itself supported by a massive circular con-
crete pedestal. Between the upper surface of the concrete and the lower surface of the
runner lay a thick layer of grout which was supposed to provide a stable, impervious,
interface between the steel bearing and the concrete pedestal. The dimensions of the
components of the hydrostatic bearing were impressive. Each pad was approximately 1
meter wide, 1.5 meter long, and 0.5 meter deep. The runner bearing was 24.4 meters
Figure 4-1 5. Rehabilitation of the DSS 14 hydrostatic bearing and pedestal concrete, June 198%
June 1984.

in outside diameter, 1.12 meters wide, and 12.7 cm thick. The walls of the pedestal
were more than one meter thick and were topped by a massive "haunch" 2.1 meters
high and 1.8 meters thick. The haunch provided the foundation for the grout and the
runner. A cross-section showing the hydrostatic bearing components, and the portion of
the concrete pedestal that was removed and replaced during the year-long rehabilitation
task at Goldstone, is shown in Figure 4-15.

Throughout the life of the 64-m antennas, deterioration of the grout and the concrete
had resulted in variations in the clearances between the pad and runner surfaces as the
pads moved around the runner during azimuth rotation of the antenna. On numerous
occasions, the film height reached minimum value and the antenna was automatically
halted, bringing mission operations to a stop. The antenna was then out of service until
the faulty section of the runner could be repaired by jacking it up and inserting long
tapered shims between the grout and the runner, or even replacing the grout itseK
During the twenty-month period between January 1981 and August 1982, 12,500 hours
of effort were expended in maintaining the grout. By this time, the antenna was oper-
ating with film heights at or near the alarm level (0.005 inch).

During the Voyager 2 Encounter with Saturn in August 1981, the unevenness in the
runner surface reached the safety limit. It was only by posting an engineer on the anten-
na to personally monitor the film height that DSS 14 was able to complete its final
tracking pass with fiyageer. It was a very close call. After that, the antenna was out of
service for several weeks while temporary repairs were made. Had this problem occurred
a few weeks earlier, the Saturn Encounter would have been significantly affected. Reliable
operation of the antenna was essential to support not only the ongoing flight projects
but the also the T/oygeer 2 Encounter with Uranus in January 1986. It was clear that
drastic remedial measures were called for.

Over the years since the hydrostatic bearing problem first appeared, a great deal of study
and analysis of the problem had been carried out in an effort to determine the h d a -
mental reasons for the failure of the load-bearing grout. This work led to the discovery
in 1983 chat the principal cause of the problem was not the grout but the concrete in
the pedestal region beneath the grout. Briefly, it was determined that the pedestal con-
crete had deteriorated due to a chemical reaction between the aggregate and the cement.
The aggregate reacted with the alkali in the cement to form silica gel, which in turn
absorbed moisture, expanded, and caused microcracks in the concrete. The net result
was a decrease in the stiffness of the concrete to the point where it was no longer able
to provide the proper support for the hydrostatic bearing. Based on this and previous
engineering analyses, the decision was made to remove all the concrete in the haunch
at DSS 14 and replace it with concrete of a new type of mix which would not be sub-
ject to the former aggregate problem.

In June 1983, the antenna was removed from service and repair of the pedestal and
refurbishing of the hydrostatic bearing had begun. The six-million-pound rotating struc-
ture was raised and placed on temporary supporting columns to allow the hydrostatic
bearing and the runner to be removed for rework. All the concrete in the pedestal haunch
was then removed one section at a time by a four-person crew with jackhammer and
drills. Because of the enormous amount of reinforcing steel embedded in the concrete,
it took the crew about five days to remove a 40-foot-long section. As each section was
removed, the new concrete was poured and alIowed to cure. When the new "haunch"
was completed, the radial bearing was replaced and the hydrostatic bearing components
inscalled and aligned. Final tests on the new type of concrete showed that the original
specifications for stiffness had been met or exceeded. Various phases of this work as it
progressed are shown in Figure 4-15.

The DSS 14 antenna was returned to service almost exactly one year afier it was taken
down for the pedestal rework. Raising the six-million-pound antenna, placing it on a
temporary support structure, removing the hydrostatic bearing and runner and remov-
ing and replacing 450 cubic yards of high-strength concrete had been a monumental
task. The task was completed on schedule, within budget, and without a lost-time acci-
dent. As a result of this work, the DSS 14 pedestal was expected to provide excellent
support for the hydrostatic bearing for many years.43

DSS Azimuth Radial Bearing

While the work described above was in progress, another key element of the antenna's
azimuth rotation system was also being repaired. This was the radial bearing.

The radial bearing consisted of a 30-foot-diameter steel ring surrounding a concrete col-
lar at the top of the pedestal. A vertical wearstrip was anached to the runner to provide
a track for three wheel assemblies equally spaced and attached to the rotating alidade
structure. The wheel assemblies, bearing on the vertical wearstrip and runner, maintained
the correct vertical axis of rotation for the entire antenna.

In the twenty years afier the antenna was built, the grout behind the radial bearing run-
ner and wearstrip had deteriorated to the point where the risk of failure in the entire
radial bearing assembly was unacceptably high. Failure of this bearing would result in
many months of downtime for repair. Accordingly, the decision was made to take advan-
tage of the current downtime to repair the radial bearing as well as the hydrostatic
bearing. A new runner and wearstrip were fabricated and installed. It took several weeks
to adjust and align the runner and wearstrip to be concentric with the center of the
antenna foundation. New grout was placed and cured for two weeks for maximum
strength. The wheel assemblies were removed and shipped to a contractor for refur-
bishing. Despite some problems with correct realignment of the wheel assemblies during
the reinstallation process, the work was completed successfully, on time, to support the
firsr rotation of the antenna on the new pedestal.

DSS 43 Pedestal

The final item in the rehabilitation task concerned the pedestal tilt at DSS 43 in Canberra.
This had been discovered in 1973 and later attributed to nonuniform loading on the
pedestal. When it was decided to expand the 64-m antennas to 70-meter diameter, a
special analysis was made with new data to determine the cause of the tilt and the poten-
tial impact of the increased weight on the foundation. The ensuing report concluded
that the tilt was due to the presence of softer supporting soil and bedrock at the south
side of the pedestal, compared to that which supports the pedestal on the north side.
The report further concluded that the soil had compressed to its maximum amount,
and that the proposed additional weight would not increase the tilt enough to be cause
for concern.

Antenna Performance Upgrade

The concept of enhancing the performance of the 64-m antennas had been studied for
many years. There were a number of options and alternatives available, each with its
own advantages in terms of performance, cost, and viability. Chief amongst the sup-
porters of this initiative was Robertson Stevens, a significant partner in the development
and implementation of the original 64-m antenna at Goldstone in the mid-1960s. His
report on the evolution of large antennas in the DSN traces the course of these stud-
ies from 1958 to the 1782 recommendation to proceed with the 70-m extension pr0~ratn.44

By 1783, with a solution to the problem of supporting Kyager 2 at Uranus (20 AU)
in 1986, the DSN addressed the next question of how to support the mission at even
greater distance, nameIy at Neptune (30 AU) in 1989. Using a metric called an aper-
ture unit (Ap.U.), equivalent to the effective aperture of a DSN 64-m antenna with a
system noise temperature of 25 kelvin at X-band, Stevens showed that for the Neptune
Encounter, 10.92 Ap.U. would be required. The Uranus configuration had provided
onIy 5.25 Ap.U., 4.45 Ap.U. from an array of DSN antennas, and 0.8 Ap.U. from the
Parkes antenna. Where would the shortfall in downlink performance come from?

Drawing on a background of previous engineering studies, Stevens ~ r o ~ o s etod increase

the DSN contribution from 4.45 Ap.U. to 6.45 Ap.U. by upgrading the existing 64-m
antennas for a gain of 1.65 Ap.U., and by adding a new high-efficiency antenna at
Madrid for an additional 0.35 Ap.U. The baIance of 4.47 Ap.U. would be contributed
by non-NASA facilities consisting of the Parkes, Australia, 64-m radio astronomy anten-
na, the very large array antennas of the U.S. National Radio Astronomy Observatory in
New Mexico, and the 64-m antenna of the Japanese Space Agency in Usuda, Japan.
This approach was considered viable and cost-effective, and with the Voyager Neptune
encounter as the driving motive, pIanning for upgrading the three existing 64-meter
antennas to 70-meter diameter, on that basis, went forward.
Performance analysis studies had predicted that in terms of gain improvement relative
to the 64-m antennas, 0.6 dB would come from the microwave feed system, 0.5 dB
from the precision surface of the subreflector and main reflector, and 0.8 dB from the
increased diameter, for a total of 1.9 dB or about 55 percent. Overall performance at
S-band would be about 0.5 dB less due to the fact that the reflector surface refinement
had little effect at the lower frequency.

Notwithstanding its importance to the Voyager Neptune Encounter, upgrading the

64-m antenna had many other technical advantages. The additional antenna gain applied
to the uplink as well as the downlink, &ording a considerable enhancement of com-
mand capability under adverse conditions. In a new navigation technique called very
long baseline interferometry (VLBI), which involved two 70-m antennas alternately
tracking a spacecraft and a natural radio source such as a quasar, the increased sensi-
tivity of the 70-m antenna allowed a much wider choice in the selection of radio
sources that were suitable for this purpose. In arraying with smaller antennas, the sig-
nal from the larger antenna could be used to aid the carrier trachng of the signal
from the smaller antenna at or near signal threshold. Furthermore, while the cost of
extending the 64-m antennas to 70-m diameter was about the same as building a new
34-m antenna, the cost per aperture unit was only about 60 percent as much. Finally,
estimates of the maintenance and operations cost for an additional 34-m antenna
amounted to $200,000 per year, while the costs for maintenance and operation of the
existing 64-m antennas were not expected to increase significantly by their enlarge-
ment to 70-m diameter.

When the project began, it was estimated that modification of the 64-m antennas to
70-m would require about 12 months for each antenna, or three years if they were mod-
ified in sequence. However, the demand for the antennas to support ongoing missions
was such that the time period for the entire job had to be reduced to fit the 18-month
period following the Galileo launch scheduled for mid-1986, and the Galileo Probe
release scheduled for the end of 1987. This 18-month window was the only period in
the foreseeable future when the antennas could be taken out of service without signifi-
cant impact to in-flight missions. The challenge became one of finding a way to complete
the implementation in half the time it would take under normal circumstances.

Normally, an implementation task of this type would be done by a single contractor

who would be responsible for the design, fabrication, erection, and testing. The lead
time for this approach would run into many months. Instead, to save time, the design
was done by JPL engineers with assistance in detail design horn TIW Systems of SunnyvaIe,
California. Except for the Australian antenna, JPL acted as the prime contractor for the
major items on a build-to-print basis. In Australia, the fabrication of the structural ele-
ments and the quadripod and the erection work were contracted locally. In addition to
this time saving strategy, it was planned to return each 70-m antenna to service with a
gain performance at least equal to the 64-m antenna, and to perform the final fine-tun-
ing to reach predicted 70-m performance later as the opportunity provided.

For various reasons associated with the support of ongoing missions, particularly Voyager
and the upcoming GaIileo launch, urgency of the work at Goldstone, the seasonal weath-
er conditions at each site, and availability of subcontractors and equipment, it was decided
to carry out the antenna expansion work in this order: Spain, Australia, and Goldstone.

The way the plan eventually worked out is shown in the following table.

summary of 64-m to 70-m Antenna Expansion Start and Finish Dates

Location Start Date Return to Operations with 70-m Performance

Madrid 1 Aug 1986 29 July 1987

Canberra 1 Feb 1987 21 Oct 1987

Goldstone 1 Aug 1987 1 June 1988

The loss of the Shuttle Challenger in January 1986 caused a delay in the launch of Galileo,
and as a result the eighteen-month downtime constraint was removed. This left the Voyager
Neptune Encounter in August 1989 as the next scheduled mission constraint. However,
the project elected to hold to its June 1988 completion date for the last antenna at the
expense of some overlapping downtime between the Madrid and Canberra antennas.

Microwave Design Considerations

Working within the constraints of antenna diameter (70 m), existing feed location, and
existing 64-m antenna rib shape, DSN engineers designed a microwave optics system
that would optimize each of these constraining parameters. An asymmetrically shaped
subreflector of approximately 25 feet in diameter would optimally illuminate the new,
high-precision, shaped surface of the 70-rn main reflector. Microwave energy, focussed
by the subreflector, would enter a single hybrid-mode horn, to feed the existing low-
noise X-band maser. The choice of microwave feed was based on an existing, proven
design, and offered a performance and cost advantage over either the SIX-band feed-
horn designed for the 34-m high-efficiency antennas or the dual hybrid-mode horn
which had been used on the 64-m antennas.

The single hybrid-mode horn, mounted on rhe X-band Receive Only (XRO) cone
would provide the ultimate in X-band receive performance. S-band receive and trans-
mit capability would be provided by the existing dichroic plate and reflex feed
arrangement on the S-band Polarization Diversity (SPD) cone, while the third cone
carried an X-band/K-band Receive (XKR) feed, used principally for radio and radar
astronomy purposes.

Measurements of RE efficiency and system rernperature were made at each 64-meter

antenna before it was taken out of service and again before it was returned to service
after upgrade to 70 meters. That design objectives were met is plainly obvious from the
results given in the table below.

Summary of Measured Antenna Gain Before and After Upgrade

Gain DSS 14 DSS 43 DSS 63 All Ants.

X-band X-band X-band S-band
64-m 72.33 71.94 71.9 1 61.40
70-m 74.17 74.10 74.28 63.34

The reasons for the differences between stations would be understood and corrected as
time progressed. But, as the project: ended, the gain improvement of 1.9 dB had been
realized and all three antennas were back in service in time to meet the immediate
requirements of the Voydgeer 2 Neptune Encounter in August 1989.45

Kicker Braces

Before the 70-m expansion work on the two overseas 64-rn antennas could begin, it
was necessary to install additional structural bracing or "kicker.braces" on each anten-
na. This additional bracing was required to mitigate the effects of gravitational deflection,
with consequent radio performance loss, as the antenna was moved in elevation. The
modification that involved replacement of two short, existing knee braces with very sub-
stantial steel members, had proven very effective on the Goldstone 64-m antenna.

The modification to the Australian anrenna was started in October 1984 and completed
six weeks later. S-band performance measurements on the antenna in January 1984 showed
a gain versus elevation curve which was similar in shape but 0.4 dB lower than the cor-
responding Goldstone data. The reason for the discrepancywas not clear. When the Spanish
antenna was modified some monrhs later, its gain versus elevation curve also Wered fiom
those of both DSS 43 and DSS 63. It was decided to defer any fi.uther investigation of
these discrepancies until after the 70-meter extension work had been completed.

Subreflector

The shaped subreflector for the 70-m antenna was almost 8 meters in diameter, about two
meters deep, and 1.6 cm thick. It was constructed of six, cast aluminum panels, welded
together to form two large pieces, and then bolted together for machining. At the time, it
was probably the largest cast duminurn suucnu-e that had ever been built. It required state-
of-the-art casting, welding, and machining technology to achieve the desired asymmetrical
surface shape, with an accuracy of 2.54 millimeters (rnm). The three subreflectors were cast
and machined by specialty contractors in Spain. However, the Spanish contractors soon ran
into technical problems with the casting and machining processes, and the resulting delays
to the work were compounded by transportation, economic, language, and cultural diff.i-
culties. Early estimates of subreflector fabrication time changed fiom 27 weeks to 40 weeks
and directly threatened the completion schedule for the Australian antenna.

A contingency plan involving fabrication of a reinforced plastic subreflector of the same

weight, diameter, and mounting configuration as the aluminum subreflector, was put
into effect. The plastic unit, which had a surface of revolution rather than a shaped sur-
face, was covered with a fine, expanded aluminum mesh. Although its performance
would not be as good as that of the full aluminum subreflector, it would suffice to allow
work to be completed to the 64-m performance level at DSS 43. The plastic subreflec-
tor was fabricated in California in two halves, shipped to the antenna site near Canberra
and installed without incident.

Until the aluminum subreflector for DSS 63 arrived, a concrete weight was installed in
its place so that servo tests and counterweight adjustments could proceed on schedule.
The subreflector eventually arrived in time for the RF testing to take place as scheduled.

Surface Panels

The surface of the 70-m antenna was over 4,185 square meters in area and was con-
structed of 1,272 individual aluminum panels of 17 different sizes. Each panel could be
adjusted individually to conform to the desired antenna surface shape which, when com-
plete, was accurate to 0.6 mm (root-mean-square) over the entire surface. The panels
were made by a special new facility in Italy, using a metal-bonding process to produce
the high-precision panels to an rms tolerance of 0.1 millimeter. Apart from some cur-
rency problems in 1987 caused by devaluation of the U.S. dollar, the work proceeded
smoothly, and all three sets of panels were delivered to the sites on time.

Madrid Site Implementation

With the necessary s d o l d i n g and a 500-ton crane in place, disassembly of the DSS
63 antenna began on I August 1986. Spanish contractors had already fabricated and
delivered the quadripod and all the new structural elements needed to replace the old
ribs and extend the diameter to 70 m. Reinforcing for the antenna backup structure
and elevation wheel was to be welded in place. Additional counterweights would be
needed to balance the added weight of the new subreflector, quahipod, and primary
reflector. Three months into the job, the contractor advised JPL that the downtime
would have to be extended by eight weeks due to various accumulated delays in the
work. Efforts to recover the lost time proved futile. Since the antenna conversions were
planned to be done in serial fahion, this delay would seriously afFect completion of the
Canberra and Goldstone antennas. It was decided to start the DSS 43 downtime on 1
February 1987, as originally planned, even though this would mean that two 64-m
antennas would be out of service simultaneously until DSS 63 was completed. At the
same time, the Canberra and Goldstone schedules were extended fiom six to eight months
based on the Madrid experience. This decision was soon vindicated when the Chllenger
accident in January 1987 resulted in a three-year delay to Galileo, leaving the Voyager
Neptune Encounter in August 1989 as the main driver for project completion.

The structure was complete and the 1,272 panels installed in mid-December 1986.
Alignment was delayed due to some of the worst winter weather in Europe in 100 years.
Using the concrete dummy weight for the subreflector, the counterweight was balanced
and the antenna servo systems tested and calibrated.

The aluminum subreflector finally arrived in March, and was installed on the quadri-
pod so that RF gain measurements could be made. The first set of results indicated a
serious deficiency in antenna performance. All 1,272 panels were readjusted with even
worse results. This was a grave situation which had to be understood before the align-
ment work on the Canberra antenna could begin. An intense effort, called in DSN
parlance a "Tiger Team," set about analyzing the problem. The "kicker braces" were
tightened and another panel alignment was carried out using a new measurement tech-
nique called holography. These actions were successful in assuring that the 70-m antenna
exceeded the performance of the 64-m by 2.1 dB.
On this basis, the antenna was returned to service on 27 July 1987, with a plan to com-
plete the final, more detailed measurements at a later date.

Canberra Site Implementation

Working under a somewhat different funding arrangement from the other two sites, it
was Australian rather than American contractors who ~rovidedthe new rib trusses, rein-
forcing steel, quadripod, counterweight, and services for erection of the Canberra antenna.
With all material and equipment in place, the downtime started on 1 February 1987,
as scheduled. Work ~roceededsmoothly despite some loss of time due to misalignment
of the truss modules and a problem with mounting some of the surface panels.

Because the reason for the anomalous performance on the Madrid antenna had not been
determined by the time the contractors were ready to start panel alignment on the
Canberra antenna, it was decided to set that antenna using holography for the final
alignment settings. Although this time consuming effort added three weeks to the com-
pletion date, DSS 43 was returned to service with 70-m performance on 21 October
1987. Although the measured antenna performance was somewhat less than that meas-
ured at Madrid (attributed to the use of the plastic subreflector), it met the 70-m
performance specifications and remained in operational service for over two years before
finally being replaced by the aluminum version in December 1989.

Goldstone Site Implementation

The implementation of the 64-m to 70-m antenna extension at DSS 14 benefitted

immeasurably from the DSN experience at the other two sites. The U.S. contractors
had the structural material on site, ready for installation at the start of the downtime
set for 1 October 1987. There were no problems in aligning the trusses at Goldstone,
and the installation of quadripod, rib modules, and panels was quickly accomplished as
planned. The aluminum subreflector from the Spanish contractor finally arrived on-site
in February 1988. Since the "Tiger Team" had not resolved the Madrid antenna per-
formance anomaly by that time, it was decided to set the panels as had been done at
Canberra, with a rough setting first, followed by a holograph-aided procedure.

As a result of the experience gained in Spain and Australia, the find panel settings and
the gain and efficiency measurements were conducted without problems, enabling the
DSS 14 antenna to be returned to service with 70-m performance on 29 May 1988,
two days ahead of schedule. Various aspects of the antenna work described above are
shown in Figure 4-16.
Figure 4-1 6. Work in progress on the 60-m-to-70-m antenna extension at DSS 14, Goldstone,
December 1987.

So ended the 64-m Antenna Rehabilitation and Performance Upgrade Project. When it
was finished, the DSN had a complete subnetwork of 70-m antennas whose perform-
ance met or exceeded the design values under all conditions and whose serviceability
had been vastly improved as a result of many years of continuous operational experi-
ence. It met the mission support requirements of the Vqpager 2 encounter of Neptune
in 1989 and would enhance the downlink and uplink for all flight projects thereafter.
Nevertheless, this antenna would be the final chapter in the DSN relationship with large
antennas. Studies had shown that larger antennas were associated with diminishing returns
in cost per/dB of performance gain.

The path to future enhancement of the downlink lay in the array of smaller, higher efi-
ciency antennas, and the pursuit of other avenues of telecommunications technology, as
future events would show.
NETWORKS CONSOLIDATION

In the past, the DSN managed by JPL provided tracking and data acquisition support
for NASA deep space planetary missions. Managed by Goddard Space Flight Center
(GSFC), a similar service was provided for NASA Earth-orbiting missions by a world-
wide network of tracking stations known as the Spaceflight Tracking and Data Network
(STDN). GSFC had for some time been developing a replacement for the STDN which
would consist of three geostationary satellites with a capability for tracking and relay-
ing, to a single ground station, data from Earth-orbiting satellites with a periapsis of less
than 15,000 km.This new facility, called the Tracking and Data Relay Satellite System
(TDRSS), but known in general terms as the Space Network, became operational in
1984. The few remaining satellites with Highly Elliptical Earth Orbits (HEEO), which
lay outside the TDRSS zone of coverage, were serviced by a portion of the STDN called
the ground network or GSTDN. Since the HEEO missions to be supported by GSTDN
would have operational parameters that were in many respects similar to deep space mis-
sions supported by the DSN, the question naturally arose as to whether there was any
combination of GSTDN and DSN that could result in a more economical way of sup-
porting those missions that could not be supported by the TDRSS.

To address this question, NASA Headquarters commissioned a study to be carried out

jointly by representatives from its Office of Space Tracking and Data Systems (OSTDS),
JPL, GSFC and the overseas Complexes. Known as the Networks Consolidation Working
Group, this body first met in May 1979 and completed its study in October 1979.46

In this study, a consolidated network comprised of various elements from the present
DSN and the GSTDN showed significant cost savings and greater capability than the
sum of the separate networks. This option was planned around the then-existing plan
for the next generation of DSN improvements calIed the Mark IVA Model, augment-
ed with GSTDN facilities and resources to support HEEO missions. A top-level overview
of the DSN and GSTDN Network Facilities that would eventually comprise the
Consolidated Network is shown in Figure 4-17.

In January 1980, a Networks Consolidation Program (NCP) office was established, in

the Tracking and Data Acquisition (TDA) Office at JPL, as a consequence of NASA
approval of the recommendations of the Networks Planning Working Group.
Simultaneously, program o&ces were estabIished at GSFC and in OSTDS at NASA
Headquarters. JPL and GSFC management and technical counterparts were identified,
and monthly working sessions were initiated. Based on the work of these groups, a man-
agement plan, transition plan, mission support plan, and a contingency plan were developed
Figure 4-17. Elements of the DSN and GSTDN, 1982.

and published in 1981.47 Budgetary considerations, changes in requirements, the poten-

tial mission set, antenna cost trade-offs, staffing, and implementation schedules became
matters of great concern as the requirements and constraints of the Network Consolidation
program were integrated with those of the Mark WA program.

The major elements of the GSTDN and DSN which were to be consolidated were the
tracking antennas and their associated support facilities. The current location of these ele-
ments at Goldstone, Canberra, and Madrid is shown in Figure 4-13. In the existing Mark
I11 DSN, and also in the GSTDN, each antenna was associated with its own support
facilities and control room. (At the overseas complexes, the two stations share the same
operations and control building.) In the Mark IV DSN, the fundamental elements of
each consolidated DSCC would consist of multiple antennas operated and controlled
from a single Signal Processing Center (SPC). As far as economically possible, all anten-
nas at each complex were to be colocated with the 64-m antenna to permit unattended
operations. Data output from each of the three complexes is routed from the SPC via
the GCF to the Central Communications Terminal (CCT) at JPL. The data are distrib-
uted from JPL to the Network Operations Control Center (NOCC), the Mission Control
and Computing Center (MCCC), and other Project Operations Control Centers (POCC).

The 1981 Mission Support Plan considered the mission set for the years during which
the Network Consolidation-MarkTVA ImplementationTask would take place (1982-1785)
in the following way:

Active Missions Active Missions

Viking Lander International Comet
Pioneer 10, 11, 12 Explorer (ICE)
Vyugeer 1, 2
Planned Missions Planned Missions
Galileo Planned Missions Active Magnetosphere Particle

This set of missions would form the basis for the schedule and sequence in which the
huge implementation task would be carried out.

To minimize the impact to ongoing missions and to protect the critical mission readi-
ness dates for future missions, a set of "ground rules" was established. Implementation
for both the consolidation effort and the Mark IVA effort would have to be planned
around these constraints. The "ground rules" included coverage for the Shuttle and Earth
satellites, and set the conditions under which a 64-m antenna or a 34-m antenna could
be taken out of service for modification. This meant that both projects had to be accom-
plished with no interruption of service to the flight projects. To satisfy these constraints,
three separate and parallel streams of activity were planned:

First, an interim configuration would be implemented around the 34-m subnetwork

between March 1983 and February 1984 to support the two HEEO missions, AMPTE
and ICE, beginning in March 1984. During this period, DSSs 12, 42, and 61 would
be configured in sequence.

In ~arallel,a consolidation configuration would be implemented between March 1783

and January 1985. In this period the actual consolidation aspects of the task would be
accomplished.
Finally, while this work was in progress throughout the Network, the stream of activity
directed to completing the final Mark TVA configuration by mid-1985, in time to sup-
port the Eyager 2 encounter of Uranus in the following January, would be moving ahead.

It began in June 1983 at Goldstone. All the existing Mark I11 equipment was installed
in the DSS 12 Control Room to create an interim configuration. Ongoing mission oper-
ations were conducted from there. Similar rearrangements of the equipment at DSS 42
and DSS 61 were made to create interim configurations at the overseas sites. This pro-
vided one fd subnetwork of antennas in the interim configuration to support ongoing
mission operations round the Network.

Under the Consolidation plan, the first antenna to be moved would be the 26-m anten-
na from the GSTDN station at Honeysuckle Creek near Canberra (DSS 44). Starting
in March 1983, it would be dismantled and re-erected near the 64-rn antenna on the
DSN Complex site at Tidbinbilla where it would be identified as DSS 46. GSFC would
provide the necessary electronics equipment to support a single uplink and downlink.
The data and communications circuits from DSS 46 would be routed through NASCOM
direct to the POCC at GSFC. The relocation was complete by January and System
Performance Tests verified the integrity of the station in July, and readiness to support
AMPTE in August 1984.

The 26-m antenna, DSS 16 at Goldstone, did not need to be moved. Microwave, and
later fiber-optic, links would be used to connect it to the new Signal Processing Center
on the Goidstone Complex. The existing equipment would be reduced to a single uplink
and downlink wirh communications direct with the POCC at GSFC.

In January 1984, the second relocation of a 26-m antenna would begin. It was to be
moved from the GSTDN site at Fresnadillas, Spain, to a site near the 64-m antenna on
the DSN Complex at Robledo and identified as DSS 66. GSFC provided the electron-
ics to support a single uplink and downlink, and communications circuits were routed
via NASCOM directly to the POCC at GSFC. The relocation was completed in October,
and System Performance Tests verified the integrity of the station in January 1985.

With this move completed successfdy, the tasks related to the Networks Consolidation
phase were now accomplished.48
MARK NA IMPLEMENTATION

To accommodate the new Signal Processing Center (SPC10) at Goldstone, the existing
control room at DSS 14 had to be enlarged and modified. This work began as soon as
the interim configuration at DSS 12 had begun supporting mission operations. While
SPC 10 was under construction, DSS 14 was taken offline for repair work on the hydro-
static bearing and repair work on the concrete pedestal. (Note: The existing control
rooms at the two conjoint stations were already large enough for their SPC and no sig-
nificant modifications were needed.) A few weeks later, construction work began on DSS
15, a new, 34-m high-efficiency (HEF) antenna, at a site close to both the DSS 14
antenna and the new SPC. Later that year work began on the construction of a simi-
lar antenna-DSS 45 at the Canberra Complex.

The old DSS 11 Control Room at the Goldstone Pioneer site was reactivated as a min-
imum hardware test bed and called the Nenvork Test Faciliry (NTF). It was used as a
staging area to assemble the equipment, get it turned on, and get the connections with
the Local Area Network (LAN) working. This was the first time the DSN had used a
LAN, and there was some concern about moving from the familiar (but now obsolete)
star-switch controller system and going into a new, unproven, LAN-based system.

A key factor in the Implementation Plan required each set of complex equipment to be
assembled and tested by the people who were going ro use it. Teams from the Madrid
and Canberra complexes arrived at Goldstone for this purpose. They were the ones who
were going to install it and they had to understand it and learn how to use it. They
were not reticent in expressing their opinion of the so-called "user-friendly" design. The
Australian Team thought "user-surly" would be more appropriate.

By October 1984, DSS 15 had been completed and equipment installed, the Mark N A
design had been validated, and SPC 10 was completed. System Performance tests with
DSSs 14, 15, and 16, and SPC 10 were started.

At the Canberra Complex, DSS 45 was completed, the Mark-NA equipment installed
in SPC 40, and System performance tests with SPC 40, DSS 43, and DSS 45 were
completed by January 1985. A photograph of the Canberra Deep Space Communications
Complex, taken in early 1985, is shown in Figure 4-18. The Signal Processing Center
(SPC lo), 64-m antenna (DSS 43), 34-m standard antenna (DSS 42), recently com-
pleted 34-m HEF antenna (DSS 4 9 , and the re-located 26-rn X-Y antenna (DSS 46)
are clearly seen.
Figure 4-1 8. Canberra Deep Space Communications Complex, January 1985.

A 34-m high-efficiency antenna at the Madrid Complex was not planned for this time,
and with the relocation of DSS 66 at the end of 1984, implementation of the Mark
IVA equipment into SPC 60 could begin immediately. System Performance Tests with
SPC 60 and DSS 63 were completed in May 1985.

All that remained to complete the Mark WA configuration throughout the Network was
to upgrade DSSs 12, 42, and 61, all still in the interim configuration, to the Mark IVA
status. With these three stations added to those already integrated with the new SPCs,
the Mark IVA upgrade task would be completed.

The final configuration of the Mark IV Model DSN, which includes the antennas and
capabilities provided by the consolidation program, is shown in Figure 4-19.
 COMPUTING CENTER

COMM - GROUND COMMUNICATIONS HE0 - HIGH EARTH ORBITER

DS - DEEP SPACE POCC - PROJECT OPERATIONS CONTROL CENTER
DSS - DEEP SPACE STATION SPC - SIGNAL PROCESSING CENTER

Figure 4-19. The DSM Mark IVA-85 configuration.

There were four other elements of the DSN, which aIthough not directly involved in
consolidation tasks were deeply involved in the Mark I11 to Mark IV transition. These
were the Ground Communications Facility (GCF), the Network Operations Control
Center (NOCC), and the two test facilities, CTA 21 at JPL and MIL 71 at KSC. The
transition plan for the GCF allowed the GCF to simultaneously support the existing
Mark I11 and the gradually evolving Mark IVA configurations. As the Mark IVA upgrades
were completed at the complexes the GCF communications channels were switched over
to the Mark IVA GCF, and the old Mark I11 equipment was deactivated.

A similar approach was used for the NOCC, due to the need to support a hybrid
arrangement of Mark III/Mark IVA, while the interim configuration was current in the
Network. Cut over to the Mark IVA NOCC was effected as each Complex came into
service in the Mark WA configuration.

Both CTA 21 and MIL 71 contained single string Mark I11 equipment. At CTA 21
this was adequate for spacecraft compatibility testing, and hardware or software devel-
opment associated with the interim configuration. In early 1985, when the validation
and testing of the Mark IVA equipment at Goldstone was complete, one of the new
Mark IVA strings from SPC 10 replaced the obsolete Mark 111 equipment. At MIL 71,
the Mark I11 equipment would be used to support the AMPTE launch. Later, it would
be replaced by Mark WA equipment.

Inevitably, there were delays to the original schedule and changes to the original plan.
Throughout its entire life, the Mark TVA project was adversely afFected by atu-ition of
the original budget.

A typical 1982 budget profile against which the progress of the Mark IVA and Networks
Consolidation programs would be measured is illustrated in the table below:

Typical Budget Profile (1982)

.,'/:.*
l <
/.d>
. .,. .,-
w ,, ,,- ."
. .-.. c r. ~ , ~ ~ . ~ ~ ~ . ~ ~r<-
~ ~ ~ . ~ ~ , C ~ ~ ~ A ~ ~ . ~ ~ ~ " - I . - I . - I . - I . - I . - I ~ . ~ . ~ . ~ , ~ . ~ . ~ . ~ ~ . ~ < . ~ C ~ . ~ . ~ ~ ~ ~ ,~.r.r~."c~C~CCci.i -/.
~ ~ ~ . ~:/xi~ ~ < ~ . ~ ~
, ~ ,/r.-ii. iF.>-.i+tttpP
~ ~ ~ ~ . ~ C ~P , ~ I ~ ~ , " I , ~ ~ . ~C~.rLii-ijll/-l
~. ITTTTCCiPCCcCiii. .~. i si

Fiscal Year 1981 1982 1983 1984 1985 Total

Networks
Consolidation ($M) 6.5 21.5 20.4 16.1 5.3 69.8
Mark IVA ($M) 10.0 19.5 22.1 21.9 14.9 88.4
Total ($M) 16.5 41.0 42.5 38.0 20.2 158.2

Under constant pressure from the 1980 Reagan Administration, the NASA budget was
cut several times during the course of the NCPIMark IVA program. These reductions
inevitably impacted the funding available for the work in progress, resulting in programmatic
changes, deferment of some tasks, and schedule delays. The DSN could only deal with
this continuing condition by deleting planned capabilities from the original design.

Despite these setbacks, the program was completed in time to support the Voyager
Uranus Encounter in January 1986. There were many liens outstanding against the sys-
tem and little time left for operations verification testing. It was mainly due to the heroic
efforts and dedication of the operations and engineering people at the three complexes
that the Mark IVA worked so well at the time it was needed.

It was the perception of those who were in the best position to know that the dtimate
cost of the Mark IVA project, including the Consolidation elements, was 10 to 20 per-
cent less than the overall cost estimated in 1982. This, however, resulted from mandated
budget cuts met by deleting capabilities, rather than from DSN management or engi-
neering processes.

Although there were some questionable features of the internal design, particularly in
the subsystem operation, the overall goal of centralized control was met and amply
demonstrated in the years that followed. The main purpose of the Mark IV Network
upgrade to centralized control was to reduce operations costs by reducing the number
of operators required to operate each antenna. That this goal was successfdly achieved
is shown graphically in Figure 4-20.

In retrospect, the conversion of the DSN Mark I11 Data System to the Mark IVA Model
was considered, at the time, to be the most complex implementation project undertak-
en in the DSN since its inception, surpassing even the MDS effort of five years earlier.
Initiated in 1980, the objective of the Mark IVA project was to reduce operations costs,
improve reliabiliry and maintainability, and to increase telemetry reception capability for
the Voyager Uranus encounter and all subsequent planetary missions.

That these immediate goals had been achieved soon became evident in the successfd
Uranus Encounter and the events that soon followed. The validation of the long-term
goals became apparent as the DSN moved through the remaining decade and a half of
the century. Additions to the Mark N A were made as the increasing mission set required,
but the original concepts on which this major undertaking were based gave no cause
for reconsideration. It could well be said the Mark NA-85 model was indeed the fore-
runner of the "modern" DSN.49
f \
Mark I V

-
Oper~~ions
:I;,": Reengtneering
9---D-
4000

3000

Operations
Cost
FY 97
Dollars
loo0

o e Unattended Operations
1980 1 9 ~ I988 1992 1996 2000 M04 8 Improved Reliability
e J u t In The Logistics
Fiscal Year
\ 1

Figure 4-20. DSN antenna operating costs, 1980-2000. The data shows that the Mark IVA
Network upgrade reduced antenna operations costs from over $3,000 per hour in 1980, when 14
operators per antenna were required, to $2,400 per hour in 1990, when only 4 operators were
required. Further economies in operations costs lay in the future.

Signal Processing Centers

The three Deep Space Communications Complexes (DSCCs) that composed the Mark
IVA DSN were located at Goldstone, California; Tidbinbilla near Canberra, Australia;
and Robledo, near Madrid, Spain. There was one Signal Processing Center (SPC) and
four antennas at each of the DSCCs. (The fourth antenna, a 34-m high-efficiency type
was added to the Madrid DSCC in 1987.) High-speed and wideband communications
circuits linked each complex to the Network Operations Control Center (NOCC) at
JPL. The necessary communications terminal and data transmission facilities at each
Complex and at JPL were included in the Mark N A DSN, shown in completed form
in Figure 4-19 on page 263.

Each antenna within a Complex had some locally mounted equipment-antenna drive
and control equipment, low-noise amplifiers, receiver front ends, and transmitters. The
antenna pointing equipment, microwave instrumentation, transmitter control, receiver,
and metric data were located at the SPC. The SPC aIso contained the equipment which
could be assigned to any of the designated antenna "links," such as the telemetry, com-
mand, radiometric, radio science, and interferometry processing equipment. Nonassignable
equipment such as that required for frequency and timing, test and training, and glob-
al communications was also Iocated in the SPC. Equipment associated with the 26-meter
 - MONlTOR PND CONTROL

Figure 4-21. Signal Processing Center, Mark IVA configuration.

Earth-Orbiter link was not controlled from the SPC until a few years later when it, too,
was integrated with the SPC.

In the Mark IVA Model, the Signal Processing Center at each DSCC was configured
to support the operation of each antenna individually, or to array any combination of
the antennas. A simplified bIock diagram of an SPC is shown in Figure 4-21.

Within each SPC, rhe equipment could be grouped into three "links" by operators work-
ing from the Link Monitor Console. Each "link" comprised the equipment necessary to
support one spacecraft with a downlink and uplink as required by rhe parricular mis-
sion sequence. This included the antenna and equipment associated with the tracking,
receiver, telemetry, command, radiometric, radio science, monitoring and communica-
tions hctions. Each "link" was then assigned to a particular spacecraftlmission by an
operator working from the Complex Monitor ConsoIe according to an established sched-
ule. Other subsystems provided test support, frequency and timing, maintenance, etc.
Interactions between the two monitor consoles and the hardware and s o h a r e com-
prising the "links" were carried by an extensive Local Area Network.

Compared to the former Mark I11 configuration, this arrangement resulted in a very
significant reduction in the number of personnel required to conduct continuous mis-
sion operations at each Complex. Improvements in operability, reliability, maintainability,
and the ability to reconfigure rapidly to work around anomalous situations, were soon
apparent. The objectives of the Mark IVA project had certainly been achieved.

The control and data processing equipment for the 26-m Earth Orbiter tr* station was
not integrated into the SPC. W1th the exception of the telemetry equipment which was updat-
ed, the equipment remained as it was when the stations were transferred &om the GSTDN
to the DSN in 1983. These stations could support only one uplidddownlink at a time, but
a cross-support connection with the SPC enabled a spacecrafi link through the 26-m anten-
na to be processed by one of the "links" at the SPC. Likewise, a spacecrafe link through the
34-m standard antenna at the SPC could be connected to the Earth Orbiter station for data
processing. Either of these operational arrangements was known as the 'ccross-support" mode.

MARKIVA NETWORK
OPERATIONS
CONTROL
CENTER

The Network Operations Control Center (NOCC) at JPL, was also upgraded as part of
the Mark IVA project. The NOCC Mark IVA configuration is shown in Figure 4-22.

Data fiom the SPCs was transported by the GCF to the Central Communications
Terminal. There, the data was switched to the JPL Mission Control Center (for JPL
missions) or to a remote Project Operations Control Center (for non-JPL Missions).
The data was also routed in parallel to the NOCC, where data monitoring and net-
work control functions were performed. The data monitoring functions were performed
in the real-time monitor processors which were retained from the Mark I11 model but
upgraded with Mark IVA compatible sohare.

Mission support data products such as DSS tracking schedules, mission sequence of events,
prediction data for antenna pointing, and setting uplink and downlink parameters, were
also generated in the NOCC and provided to the SPCs as required for mission support.

The Network was controlled from the NOCC by station controllers who were provid-
ed with suitable displays to enable them to carry out these functions. During the Mark
IVA period the physical arrangements in the NOCC were greatly improved by substi-
tuting convex shaped monitor consoles for the original circular consoles. These
 /

TO REMOTE
MISSION OPERATIONS

-
<=3
MONITOR AND CONTROL

DATA PATHS
REAL-TIME MONITOR PROCESSORS

* CCT = CENTRRL COMMUNICATIONSTERMINAL

NOCC = NETWORK OPERATIONS CONTROL CENTER

.' /

Figure 4-22. Network Operations Control Center, Mark IVA.

accommodated larger monitor screens (14-inch) arranged in two tiers of five, provided
better viewing conditions and increased the work capacity and physical comfort of the
individual controllers.

Mark IVA Ground Communications Facility

The communications links between the SPCs and the Central Communications Terminal
(CCT) at JPL were also upgraded as part of the Mark IVA project.50 The GCF Mark
IVA model permitted a more cost-effective utilization of available circuits without los-
ing the flexibility to protect critical data. At each SPC, new communications processors
were now capable of multiplexing data from all of the antennas on the Complex (data
links) on to a single 224 kbps circuit for transmission to JPL. In addition, critical data
could be transmitted on a single 56 kbps circuit with error detection and correction
from the Complex to JPL. Complete, error-free and gap-free records of all data trans-
mined fiom the SPC would be generated automatically by the Data Records Subsystem
residing in the Central Communications Terminal at JPL. This Intermediate Data Record
tape file would be the end product, deliverable to each flight project as a permanent
record of the data captured from each spacecraft during each tracking station pass. Future
plans called for a modular expansion of the transmission data rate capability to 1,544
kbps, enabling error correction on all data transmitted from the complexes.

At Goldstone, new microwave links were required to connect DSS 12 to the new SPC
10 now located 21 h away. Networks Consolidation added the former STDN stauon,
DSS 16, to the DSN and necessitating the construction of a new fiber-optic link, 8
miles in length, to bring its baseband data to the Goldstone GCF communications ter-
minal for transmission to JPL.

Consistent with the future introduction of packetized telemetry plans were made to
expand GCF capability to accept variable size data blocks which would be sized to match
the characteristics of the data stream being transmitted. This capability wodd further
contribute to Network efficiency by minimizing the amount of data conditioning need-
ed for gound transmission by the older Mark 111 system,

DSN MARK IVA SYSTEMUPGRADES

Mark TVA Telemetry

At the Deep Space Communication Complexes (DSCCs), the DSN Telemetry System
was changed in two major ways as a result of the Mark TVA implementation project.51
First, the addition of baseband combining equipment allowed for the arraying of up
to seven antennas at each Complex. Arraying the 64-rn antenna with two 34-m anten-
nas in this way would provide a gain of 1.84 dB relative to the gain of a single 64-m
antenna. The enhanced downlink performance derived from arraying was driven by the
Voyager project requirement for support during the Uranus Encounter where the DSN
antenna array at CDSCC was supplemented with the 64-m antenna at the Parkes Radio
Astronomy Observatory.

Second, the telemetry system was configured to support up to three missions simulta-
neously with one or rwo of them being highly elliptid Earth-orbiting (HEO) missions.

In addition to these major changes in capability, the Mark IVA telemetry system pro-
vided for demodulation of Manchester-coded data modulated directly on the carrier with
coded data rates up to 211.2 kilosymbols per second (ksps). This capability would be
used by many of the H E 0 missions. For planetary missions, maximum likelihood decod-
ing of short-constraint length convolutional. codes and sequential decoding of
long-constraint length codes was provided. Precise measurement of received signal level
and system noise temperature, was available to all DSN users.

At each Complex, new features oriented towards improved DSN operations included
centralized control and real-time status reporting by the Mark IVA Monitor and Control
System. Original data records ~roducedby each of the four telemetry groups could be
played back to the NOCC when necessary by local manual control or by automatic
response to GCF instructions from the Central Communications Terminal at JPL.

The principal features of the Mark IVA telemetry arrangement at a DSCC are shown
in Figure 4-23.

At each Complex there was one 64-m antenna and two 34-m antennas. The 64-m
antenna and 34-m standard antenna could receive an S-band and an X-band carrier
simultaneously. The 34-rn high-efficiency antenna could receive one X-band carrier only.
The high system noise temperature at S-band made it unsuitable for deep space recep-
tion at S-band. The telemetry data processing equipment was arranged in four groups,
each of which contained two data channels. Each channel could process one data stream
from either an H E 0 or deep space spacecraft depending on its capability. Hardware in
Groups 1 and 2 closely resembled the Mark I11 model. Hardware in Groups 3 and 4
contained the new Baseband Assembly (BBA), which was a new design incorporating
the baseband combiner, subcarrier demodulator, and symbol synchronizer functions. A
hnctional block diagram of the new BBA is shown in Figure 4-24.
 64-m
ANTENNA

RECEIVTWC~TER I
SUBSYSTEM
-
----.----..-----------
- - - - c -

Ag. 1. Telemetry System: DSCC block diagram

Figure 4-23. Principal features of the DSCC Mark IVA Telemetry Subsystem.
Figure 4-24. Functional block diagram for Mark IVA Baseband Assembly (BBA).

In addition to performing the fhctions required to format the decoded data for trans-
mission to JPL via the GCE the Telemetry Processor Assembly (TPA) carried out the
hnctions required by the centralized Monitor and Control system.

The key characteristics of the Mark IVA 1985 Telemetry System are tabulated in
Figure 4-25.

Mark WA Command

The existing Mark 111 store and forward command functions were carried f o m d into
the Mark IVA system. This design allowed mission operations controllers at JPL or other
Mission Operations Centers to prepare large fdes of spacecrafi commands in advance
and to forward them to the DSCC at the beginning of a spacecraft track for transmis-
sion at designated times. The system design provided elaborate and well-proven procedures
to protect command data file integrity during transfer to the DSN station as well as
bit-by-bit confirmation of proper radiation from the DSN transmitter.

In the Mark IVA model, the DSCC Command System utilized elements of the upgrad-
ed antenna, microwave, transmitter, receiver-exciter and monitor, and control subsystems.52
These capabilities were configured, controlled, and driven by the new Mark WA Command
Subsystem as shown in Figure 4-26.

Four identical strings of command processing equipment, each containing a com-

mand processor assembly (CPA) and command modulator assembly (CAM), could
be connected via a switch assembly to the exciters of any of the S-band transmitters
associated with the 64-m or 34-rn antennas as shown in Figure 4-26. At the proper
time for radiation, validated command data files previously stored on disk by the
CPA would transfer to the CMA for subcarrier modulation before being passed to
the appropriate exciter for generation of the command-modulated S-band carrier for
radiation by the selected transmitter. Suitable interfaces with the monitor and con-
trol system provided for real-time status and configuration monitoring. Although the
DSN command system was limited by the available transmitters to S-band uplink
transmissions only, the need to provide an uplink on X-band for future missions was
recognized, and development work to that end was already in progress in the engi-
neering sections.

Mark WA Tracking

In 1980 and 1981, the DSN Tracking System Mark 111-79 model had been extensively
modified to improve irs performance, operability, and adaptability to support anomalous
spacecraft conditions (notably, the Voyager 2 uplink tuning problem).53 The addition of
a new microprocessor-based exciter fi-equencycontroller or Digitally Controller Oscillator
(DCO) in place of the former programmed oscillator (POCA) to the receiver-exciter sub-
system, provided a large repertoire of precision, time-dependent frequency ramps,
controlled via the Metric Data Assembly or a computer terminal. Under the control of
the Metric Data Assembly, the DCO would generate the special uplink tuning necessary
to accomplish the special upfink tuning required to support the Voyager 2 spacecraft, as
well as standard tuning for all other spacecraft uplink operations.

The Meteorological Monitor Assembly transmitted non-real-time meteorological data to

the NOCC for use in dibrating radio science and VLBI observations.

The quality of ranging data generated by the Planetary Ranging Assembly, as the angu-
lar separation berween a spacecraft and the Sun decreases, was improved by the addition
of selectable higher frequency components to the range code and better filters to decrease
waveform distortion.

In the NOCC, older software used to generate frequency and pointing predicts for the
stations, based on a Probe Ephemeris Tape supplied by the flight projects. It was rede-
veloped to make better use of the capabilities of the Sigma V computers in the NOCC
Support Subsystem (NSS).
 UPLINK-DOWNLINK: OF THE DEEPSPACE
A HISTORY NETWORK

In the Mark IVA implementation, further improvements were made to most of the ele-
ments of the Tracking System, and all of them were integrated with the new DSCC
Monitor and Control Subsystem (DMC) to provide centralized control for all subsys-
tems and collection and distribution of the monitor data required by each subsystem.

At the NOCC, a new, Digital Equipment Corporation VAX 111780 processor was
added to perform two, key Tracking System hnctions. The first was an orbit deter-
mination function, similar to that carried out by the JPL flight projects, but intended
specifically for DSN support of the new cooperative and reimbursable missions. The
other was the predicts generation and distribution function previously carried out in
the Sigma V-based NSS. Using radiometric data generated by the tracking stations,
the NOCC Navigation Subsystem (NAV) could generate orbit data files and state vec-
tors for use by the NSS, or for dissemination to flight project navigation teams, or
on a reimbursable basis to external users such as the European Space Operations Center
located in Darmstadt, Germany; the German Space Operations Center at
Oberpfdenhofen, Germany; and the Japanese Tracking and Control Center Iocated
at Tsukuba, Japan.

More than any of the other DSN systems, the Tracking System reflected the impact of
changes to existing hardware and software and procedures introduced by the Mark IVA
implementation. The quality and quantity of the S-band and X-band radiometric data
being delivered to the flight projects, especially Voyager, in the immediate post Mark
WA period was not satisfactory. In March 1985, a special team was formed to moni-
tor performance and recommend changes to improve radiometric data delivery prior to
the Voyager Uranus Encounter in January 1986. Many changes to improve the oper-
ability of the Mark NA Tracking System were proposed and implemented over the
following months. By January 1986, the usable data return had improved to an accept-
able level to meet the lrqyager 2 encounter navigation requirements. The excellent quality
of the radiometric data provided by the DSN was considered to have been a strong con-
tributor to the remarkable targeting accuracy of the satellite Miranda, as I/oyger 2
transmitted the Uranian system.

Over the next few years, several major new capabilities would be added to the DSN
Tracking System. These would include an X-band uplink, Media Calibration Subsystem,
the Sequential Ranging Assembly, and a real-time interface between the radiometric and
navigation delivery systems.
AN ABRUPT TRANSITION

Unlike the gradual transition between the Viking Era and the Voyager Era, the transi-
tion from the Voyager Era to the Galileo Era was marked by an abrupt and tragic
event-the Space Shuttle CbuLhnger disaster in January 1986. At that time, both the
Ulysses and Galileo missions had been included in the DSN mission set for the Voyager
Era for several years (see Figure 4-1). Both spacecraft were to be launched by the Space
Shude using a high-energy, liquid-hydrogen fueled, Centaur G upper-stage combina-
tion. They would be the first spacecrafi to be launched from the Shuttle in this manner.

Since both spacecraft were bound for Jupiter, both launches were planned to take advan-
tage of the May 1786 Jupiter launch opportunity. The Ulysses launch window opened
on 15 May, and the Galileo launch window opened on May 20. Agreements between
the two flight projects had been made for the resolution of conflicts for launch priori-
ty (in case of launch pad delays) or overlap of view periods at the DSN tracking stations.

The DSN planned to support both missions with the recently completed Mark IVA
network configuration, 64-m antennas, 34-m standard antennas, 34-m high-efficiency
antennas and the newly consolidated 26-m antennas.

In January 1986, both spacecraft were well advanced in their prelaunch preparations at
Kennedy Space Center. At each of the DSN complexes, however, attention was focused
on the critical Voyager Uranus Encounter operations then in progress. At the same time,
a team from the DSN Operations OEce at JPL was en route to each of the complex-
es to discuss support for the Galileo and Ulysses launches Iater in the year. They had
reached Madrid when the news came. Two days later the Voyager spacecraft made its
closest approach to Uranus.

Although the weeks that followed were a period of great uncertainty and conEusion in
NASA and the Galileo and Ulysses projects, DSN was not immediately affected. Voyager
and other inflight missions continued to receive DSN attention. Then, on 12 February
1986, both flight projects were informed that the planetary missions would not be
launched in May 1986, and they were directed to work towards the next Jupiter launch
opportunity in June 1787. The change for the missions affected the DSN much more
significantly than was realized at the time, as subsequent events would show.

For various reasons, the launches did not take place. The reasons for the further defer-
ment of the launches, their effect on the DSN, and the eventual accomplishment of
Ulysses and Galileo missions are discussed in chapter 5, "The Galileo Era."
Endnotes

1. Bruce C. Murray, Phnetary Report XVII (Number 3, MayIJune 1997): 14.

2. Bruce C. Murray, Journey into S'ce (New York: W W. Norton, 1989).

3. David Morrison and Jane Sanz, Vyage toJupiter, NASA SP-439 (Washington, DC:
NASA, 1980).

4. E. K. Davis, "Deep Space Network Preparation PIan for Mariner Jupiter Saturn
1977," Document 618-706, Jet Propulsion Laboratory, Mariner Jupiter Saturn
Project Office (15 December 1976).

5. David Morrison and Jane Sanz, Voyge to J~piter,NASA SP-439 (Washington, DC:
NASA, 1980), pp. 58, 60.

6. Andrew J. Butrica, "From Engineering Science to Big Science," chapter 11, NASA
SP-42 19 (Washington, DC: NASA, 1998), pp. 25 1-56.

7. David Morrison and Jane Sanz, &age tolapiter, NASA SP-439 (Washington, DC:
NASA, 1980), p. 108.

8. J, E. Allen and H. E. Nance, "Voyager Mission Support," TDA Progress Report PR

42-53, July-August 1979 (15 October 1979), pp. 4-5.

9. N. A. Fanelli and H. Nance, "Voyager Mission Support," TDA Progress Report PR

42-54 September/October 1979 (15 December 1979), pp. 19-23.

10. M. R. Trader, "Deep Space Network Preparation Plan for Voyager (Saturn),"
Document 618-706, Jet Propulsion Laboratory Voyager Project Office; Rev. A (25
March 1980).

11. N. A. Fanelli and H. Nance, "Voyager Mission Support," TDA Progress Report PR
42-61 November/December 1980 (15 February 1980), pp. 22-26.

12. Jet Propulsion Laboratory Voyager Project Office, "Voyager Bulletin; Mission Status
Repo;t No. 52" (27 August 1980).
13. Jet Propulsion Laboratory, Voyager Project Oace, "Voyager Bulletin; Mission Status
Report No. 57" (7 November 1980).

14. N. A. Fanelli and H. E. Nance, "Voyager 1 Mission Support (I),Voyager 2 Mission

Support (TI)," TDA Progress Report PR 42-66, September/October 1981 (15
December 19811, pp. 2 4 2 9 .

15. See endnote 2.

16. C. T. Stelzried, "The Venus Balloon Project," TDA Progress Report PR42-85,
January-March 1986 (15 April 1986), pp. 191-98.

17. R. A. Preston and J. H. Wilcher, "The Venus Balloon Project," TDA Progress
Report PR 42-80, October-December 1984 (15 February 1985).

18. J. M. Urech, A. Chamarro, J. L. Morales, and M.A. Urech, "The Venus Balloon
Project Telemetry Processing," TDA Progress Report PR 42-85, January-March
1986 (15 April 1986), pp. 199-21 1.

19. K. M. Liewer, "Selection of Radio Sources for Venus Balloon-Pathfmder DDOR

Navigation at 1.7 GHz," TDA Progress Report PR 42-87, July-September 1986 (15
November 1986), pp. 279-84.

20. See endnote 16.

21. R. Stevens, "Implementation of Large Antennas for Deep Space Mission Support,"
TDA Progress Report PR 42-76, NovemberIDecember 1983 (15 February 1984),
pp. 161-69.

22. C. T. Stelzried and A. L. Berman, "Antenna Arraying Performance for Deep Space
Telecommunications Systems," TDA Progress Report PR 42-72, October-
December 1982 (15 February 1983), pp. 83-88.

23. J. W. Layland et al., "Interagency Arraying Study Report," TDA Progress Report PR
42-74, April-June 1983 (15 August 1983), pp. 117-48.

24. D. W. Brown et d.,"Parkes-CDSCC Telemetry Array; Equipment Design," TDA

Progress Report PR 42-85, January-March 1986 (15 May 1986), pp. 85-1 10.
25. R. Stevens, R. L. Riggs, and B. Wood, "Pointing Calibration of the Mark IVA DSN
Antennas (for) Voyager 2 Uranus Encounter Operations Support," TDA Progress
Report PR 42-87, July-September 1986 (15 November 1986), pp. 206-39.

26. Jet Propulsion Laboratory, Voyager Project Office, "Voyager Bulletin; Mission Status
Reports Nos. 72-8 1." See reports beginning 4 November 1985.

27. A. L. Berman and C. T. Stelzried, "Helios Mission Support," TDA Progress Report
PR 42-67 NovemberIDecember 1781 (15 February 1982), pp. 18-24.

28. W. E. Larkin, "Tracking and Data Acquisition Support for the Viking 1975 Mission
to Mars; June 1978 to April 1780," Jet PropuIsion Laboratory Publication 82-18
(15 April 1982). D. f . Mudgway, "Tracking and Data Acquisition Support for the
Viking 1975 Mission to Mars; May 1980 to March 1983," Jet Propulsion
Laboratory Publication 82- 107 (15 May 1983).

29. R. E. Nevarez, "Pioneer 10 and 11 Mission Support," TDA Progress Report PR 42-
6 1, NovemberIDecember 1980 (15 February 1981), pp. 17-21. D. Lozier, "Pioneer
Spacecrafe,"personal communication, 1 August 1997.

30. G. M. Rochell, "Pioneer 11 Saturn Encounter Support," TDA Progress Report PR

42-53, JulyIAugust 1979 (15 October 1779), pp. 10-20. R. B. Miller, "Pioneer 11
Saturn Encounter Mission Support," Tracking and Data Acquisition (TDA)
Progress Report PR 42-52, MayIJune 1979 (15 August 1979), pp. 4-7. R. B. Miller,
"Pioneer 11 Saturn Encounter Mission Support," TDA Progress Report PR 42-54,
SeptemberIOctober 1779 (15 December 1979), pp. 24-27.

31. J. W. Layland, "ICE Telemetry Performance," TDA Progress Report PR 42-84,

October-December 1985 (15 February 1986), pp. 23-213. N. A. Fanelli and D.
Morris, "ICE Encounter Operations," TDA Progress Report PR 42-84, October-
December 1985 (15 February 1986), pp. 17685.

32. J. I? Goodwin, "Usuda Deep Space Center Support for ICE," TDA Progress Report
PR 42-84, October-December 1985 (15 February 1986), pp. 186-96.

33. Jet Propulsion Laboratory, "Deep Space Network Mission Support Requirements,"
JPL Document D-787, Rev. 0 (July 1987).
34. National Aeronautics and Space Administration, Ofice of Space Operations,
"Reimbursable Policy for the Use of Ground Network Facilities,') Ground Networks
Division, Internal Document (19 April 1991).

35. National Aeronautics and Space Administration, Office of Space Communications,

"Obtaining Use of Office of Space Communications (OSC) Capabilities for Space,
Suborbital and Aeronautical Missions," NASA Management Instruction, NMI
8430.1C (31 December 1991).

36. R. J. Wallace, personal interview with D. J. Mudgway (13 August 1997).

37. I? L. Parsons, "Antenna Microwave Subsystem (Mark IVA)" TDA Progress Report
PR 42-79, July-September 1984 (15 November 1984), pp. 165-69.

38. M. F. Pompa, "The New 34-Meter Antenna," TDA Progress Report PR 42-85,
January-March 1986 (15 May 1986), pp. 127-38.

39. Jet Propulsion Laboratory, "Deep Space Network to FIight Project Interface Design
Handbook," DSN Document 8 10-5, Module TCI-30 (1997).

40. D. H. McClure and F. D. McLaughlan, "64 Meter to 70 Meter Antenna Extension,"

TDA Progress Report PR 42-79, July-September 1984 (I 5 November 1984), pp.
160-64.

41. R. Stevens, "Report on Maintenance, Rehabilitation, and Upgrade of the DSN 64-
meter Antennas," TDA Office Internal Memo (22 March 1982). B. Wood, "Report
on the Mechanical Maintenance of the 70-Meter Antennas," DSN Document 890-
257, Jet Propulsion Laboratory, p. 5; Document D-10442 (31 January 1993).

42. D. M. McCIure and F. D. McLaughlin, "64-m Antenna Rehabilitation and

Performance Upgrade Projecr Report," private communication (13 August 1997).

43. D. M. McClure, "Repair of DSS-14 Pedestal Concrete," TDA Progress Report PR

42-81, January-March 1985 (15 May 1985), pp. 1 3 6 4 8 . H. McGinnes,
"Rehabilitation of the 64-m Antenna Radial Bearing," TDA Progress Report
PR 42-65, July-August 1981 (15 October 198l), pp. 151-61.

44. See endnote 21.

45. See endnote 42.

46. M. L. Yeater and D. T. Herrman, "Networks Consolidation Program," TDA Progress

'
Report PR 42-59, July/August 1980 (15 December 1980), pp. 83-84. M. L. Yeater,
and D. Herrman, "Networks Consolidation Program," TDA Progress Report PR
42-65, July/August 1981 (15 October 1981), pp. 19-24.

47. M. L. Yeater et al., Jet Propulsion Laboratory, Deep Space Network, "Networks
Consolidation Program Plans and Mark IVA Implementation Plans," DSN
Document 803-008, Vol. 1, Management Plan (9 October 1980);Vol. 11, Transition
Plan (15 June 1981); Vol. 111, Mission Support Plan (15 December 1981); VoI. IV,
Mark IVA Implementation Plan (15 May 1984); Vol. V, NCP Contingency Study
(15 August 1981); Vol. VI, Mark IVA Contingency Plan (15 June 1983).

48. D. D. Gordon, "Mark IVA 26-Meter Subnet," TDA Progress Report PR 42-79,
July-September 1984 (15 November 1984), pp. 152-59.

49. R. J. Wdace and R W. Burt, "Deep Space Network Mark IVA Description 1986,"
TDA Progress Report PR 42-86, April-June 1986 (15 August 1986), pp. 255-69.

50. R. A. Crowe, "The GCF Mark rV Implementation and Beyond," TDA Progress
Report PR 42-82, April-June 1985 (15 August 1985). M. R. Traxler, Jet Propulsion
Laboratory, "Deep Space Network Preparation Plan for Voyager (Uranus)
Interstellar Mission," DSN Document 870- 19 (1 April 1984).

51. D. L. Ross, "Mark TVA DSCC Telemetry System Description," TDA Progress
Report PR 42-83, July-September 1985 (15 November 1985), pp. 92-100.

52. H. C. Thorman, "DSN Command System Mark IV-85," TDA Progress Report PR
42-84, October-December 1985 (I 5 February 1986), pp. 13542.

53. J. A. WacMey "The DSN Tracking System," TDA Progress Report PR 42-63,
March-April 1981 (15 June 1981), pp. 8-14. J. A. Wackley, "The Deep Space
Network Tracking System, Mark IVA, 1986," TDA Progress Report PR 42-85,
January-March 1986 (15 May 1986), pp. 139-46.
CHAPTER
THE GALILEO ERA

Unlike the transition fiom the Viking Era to the Voyager Era, the transition to the
Galileo Era was clearly defined by the aftermath of a tragic event of great significance
to NASA in general and to the Deep Space Network (DSN) in particular. O n 28 January
1986, 73 seconds after a perfect lift-off from the Cape Kennedy launch pad, the Space
Shuttle Challenger exploded in full view of a worldwide television audience. The inves-
tigation that followed would forever change the course of DSN support for the two
major deep space missions of that time, Ulysses and Galileo.

At the time of this disaster, both the Galilee and U . e s spacecrafi were at Cape Kennedy
being prepared for separate launches in May 1986, from the Challenger. After deploy-
ment in Earth orbit by the Shuttle, both spacecraft would be injected into their individual
mission orbits towards Jupiter by NASA's high-energy liquid hydrogen-fuelled Centaur
booster rockets.

Both Ulysses and Galiho had been carried by the DSN as future missions for several years,
and DSN plans to support the May launches were well advanced. After reaching Jupiter,
Ulysses was to swing up and out of the plane of the ecliptic to enter a solar orbit pass-
ing over the poles of the Sun. En route to Jupiter, GaIileo would release an atmospheric
Probe targeted to enter the Jupiter atmosphere at the same time as the spacecrafi arrived.
On arrival, Galileo would be inserted into an orbit around Jupiter from which it would
first return the telemetry data from the atmospheric Probe before making imaging and
other science observations from numerous close encounters with the Jovian satellites.

Bruce Murray1 describes what happened next: "A presidential commission to investigate
Challenger's demise was immediately formed. Chairman William Rogers soon conclud-
ed that 'the decision to launch was flawed.' By then, a sobered NASA had already
acknowledged that Galileo could not be launched in May 1986. Galileo and Ulysses
would have to slip at least thirteen months more, to June 1987, when Earth and Jupiter
next wodd be in favorable locations" for a Centaur-based planetary launch. Meanwhile,
based mainly on crew safety considerations, objections to the use of Centaur as a Shude-
borne upper stage were increasing. In June 1986, the NASA Administrator canceled the
Centaur as a Shude payload. Without the Centaur, the 1987 missions were not possi-
ble. The only option for an upper stage injection vehicle left to Gal& and Ulysses was
the much lower energy solid motor booster developed by the Air Force, d e d the Inertial
Upper Stage (IUS).
Ultimately, both Galilee and Ulysses were rescheduled for launching by the Shuttle in
October 1989, using the solid motor IUS as an upper stage injection vehicle. Except
for dates, the Ulysses mission was not altered by this change; however, the G d e o mis-
sion was significantly changed in ways that materially affected the DSN. It became a
new mission. It would be referred to as the Venus-Each-Earth-Gravity-Assist(VEEGA)
Galileo mission, and it was planned for launch in October 1789, with Jupiter arrival in
December 1995. in April 1991, just over a year into the mission, an intractable prob-
lem with the spacecrafi high-pin antenna led to a complete redesign of the mission
once again. The timely DSN response to this potenti* disastrous situation was a major
factor in the ultimate recovery of the Galileo mission objectives.

These momentous events would cause the Galileo project to follow a path that wodd
&ect DSN engineering, implementation, and operations in a major way for the next ten
years. For historical purposes, therefore, it is convenient to designate the period 19861396
as the Galileo Era. The DSN mission set for the Galileo Era is shown in Figure 5-1.

f 3
Galileo Era
Deep Space Mission Set

PI0 Vg2 Vgl Uls Gll Mgn MO MGS MP Cas tP HE0 ICE NR GR RA Gio
n Galileo Era Future Missions
1 Primary Missions Secondary Misions

Figure 5-1. Galiteo Era Deep Space mission set.

 UPLINK-DOWLINK: OF THE DEEP
A HISTORY SPACENETWORK

The following paragraph gives an overview of the Galileo Era mission set and missions
that it included.

Following successful flybys of Venus in 1985, the Vgu 1 and Vgu 2 spacecraft contin-
ued on to close encounters with Comet Halley in March 1986. Renamed Pathfinder,
they relied on DSN tracking support to refine the orbit of Comet Halley for use by
Giorto and several other comet-bound missions in 1986. DSN support for all the for-
eign missions to Comet Halley in 1986 was provided on the basis of cooperative or
reimbursable agreements.

Using radar rechniques to penetrate the dense Venusian atmosphere, the Magellan mission
to Venus in 1989 observed and mapped the surfice features of the entire planet. At the
end of its long extended mission, it gathered information on aerobraking techniques which
would be of great value in the design of the hture Mars Global Surveyor mission.

Mars Observer was launched in September 1992 and reached the vicinity of Mars in
August 1993. Three days before it was to be inserted into orbit for a Mars mapping
mission, the downlink was lost. Despite persistent efforts by the DSN, communication
with the spacecraft could not be restored. It was suspected that the spacecraft had been
destroyed by an explosion in the retro-propulsion system.

Reflecting a renewal of scientific interest and a substantial constituency for Mars, two
new Mars missions were launched in 1996. Both launches used expendable launch vehi-
cles. Mars Global Survyor, a similar mission to the ill-fated Mars Observer, was launched
in November 1996. On arrival at Mars in August 1997, it was to use aerobraking tech-
niques derived from the Magellan data to trim the initial orbit to the correct shape,
before beginning a full global surface mapping mission.

Mars Patbinder, launched in December 1996 was the first of the "faster, better, cheaper"
class of missions, which was to characterize the Cassini Era and the missions of the new
millennium. Using a unique inflated "bouncing ball" system, its landing on the Mars sur-
face in August 1997 was to be the first since the Viking Landers', twenty years earlier.

Towards the end of the Galileo Era, the first asteroid mission was launched. The pri-
mary objective of the Near Earth Asteroid Rendezvous (NEAR) mission was to orbit
the near-Earth asteroid Eros in a 50-krn orbit for the purpose of determining the gross
physical properties, mineralogical composition, and morphology of the asteroid surface.
Launched in February 1996, it would arrive at Eros in January 1999 afier an Earth
swingby in January 1998.

Throughout the Gdileo Era, the DSN continued to support Troygers 1 and 2,
International Cometary Ejcplorer (ICE),
and Pioneer 10 as extended missions.

By 1990, the number of Earth-orbiting missions for which the DSN was given respon-
sibility had increased to the point where they were managed in the DSN scheduling
system as a unique class, termed high-Earth orbiter (HEO) missions. Mostly, these were
inflight missions which were more suited to the ground-based network of the DSN than
to the space-based network of the Tracking and Data Relay Satellite System (TDRSS).
At about the same time, the number of non-NASA spacecraft for which the DSN pro-
vided tracking and data acquisition services on a reimbursable basis to NASA began to
increase significantly. These also were defined uniquely as "reimbursables" and were man-
aged within the DSN, independent of the deep space missions. From this point on,
DSN resources were assigned according to predetermined priorities to the deep space,
HEO, and reimbursable missions.

We turn now to examine in more detail the remarkable events that took place in the
Galileo Era and the DSN involvement in all of them. The DSN transition to the Galdeo
Era began with its continuing support for the foreign missions to Comet Halley.
NETWORK OPERATIONS

The Comet Halley Missions

In "Journey into Space," former Jet Propulsion Lab Director Bruce Murray describes his
intense efforts to interest NASA in a U.S. mission to Comet Halley in late 1985 and
1986. It would, he pointed out, be seventy-six years, the year 2061, before another
opportunity arose. Murray started lobbying early, in 1980, right after President Reagan
had taken office and media interest in JPL was still sparked by the stunning visual prod-
ucts of Voyager's encounter with Saturn.

"Where is the United States headed in space?" Murray was asked.

"Hdey," Murray replied. What Murray had in mind was a Halley Intercept Mission
(HIM), a NASA spacecrafi that would intercept the comet's path and photograph and
make scientific observations of the structure and composition of the Halley nucleus.
Murray pushed hard for the support of Reagan's new NASA administrator James Beggs,
presenting HIM as a "unique, affordable opportunity for NASA and the United States."
At NASA Headquarters, the arguments went back and forth for months. Central to the
debates were the costs and scientific merit in light of other foreign missions already
planned, and the rising demands of the Shuttle program for limited NASA funds. It
was all to no avail. In September 1981, NASA Headquarters officially directed JPL to
cease d further activity associated with the HIM. JPL would have to settle for the
United States mission then flying near Halley, ICE, or those observing from afar, Pioneers,
and for DSN support of the foreign missions.

In discussing the DSN role in the Comet Halley missions in 1986, Charles Stelzried,
Leonard Effron, and Jordan Ellis wrote, "In man's continual quest to explore and under-
srand the nature and origin of the solar system, 1985 through 1986 may well be noted
as the 'Year of the Comet.' The first in-situ measurements of a comet occurred on 1I
September 1985, with the passage of the International Cometary Explorer (ICE) through
the tail of Comet Giacobini-Zinner, 7800 kilometers down-stream of the nucleus. This
unprecedented encounter was followed six months later by the spectacular rendezvous
of a fleet of five spacecraft in the vicinity of Comet Halley during the four-day long
period March 9- 13, 1986. The Halley armada included Japan's Sakigake (MS-T5) and
Suisei (Planet-A), the Soviet Union's Vgu-l and Vgu-2, and ESA's Giotto. Twelve days
later, ICE also passed within Halley's sphere of influence. In addition to the aforemen-
tioned six spacecrh, two other interplanetary explorers turned their instruments in the
direction of Comet Halley in 1986. The long-lived Pioneer 7 and the Venus-orbiting
Pioneer 12 contributed to the library of space-based measurements made during the
Halley exploration period . . . . It was also an unprecedented period in the arena of
international cooperation between national space agencies and the worldwide scientific
community." 2

Together with two other national space agencies, the European Space Agency (ESA) and
the National Space Development Agency (NASDA) of Japan, NASA (through the DSN)
was a major contributor to the Comet Halley mission initiatives. The DSN provided
telemetry command, and navigation support for Gio~oin the near-Earth phase and for
the comet encounter. The DSN also provided telemetry, command, and navigation sup-
port for the near-Earth phase of the Japanese missions, as well as for the short period
around the actual comet encounter. Following the completion of their Venus Pathfinder
missions described in the previous chapter, the Soviet Vga-I and Vega-2 spacecraft, in
order to improve their encounter orbit determination, made use of the Very Long Baseline
Inteferometry (VLBI) techniques that the DSN had used for the Giotto mission. The
Comet Halley observations made by ICE, Pioneer 7, and Pioneer 12 were carried out
as part of the continuing DSN support for these missions.

Giotto

DSN support for the Giotto mission to Comet Halley in 1986 was provided on the
basis of a reimbursable agreement between the European Space Agency (ESA) and NASA.
The Giotto spacecrafe was a spin-stabilized vehicle that used S-band and X-band down-
links and an S-band uplink to communicatew i h the DSN antennas. The DSN committed
to providing telemetry, command, and navigation support for the near-Earth phase and
for the Comet Encounter. The uplinks and downlinks were fully compatible with DSN
telecommunication standards, and there were no special or additional implementation
requirements imposed on the DSN by the Giotto mission. It was recognized that the
late cruise and encounter periods in February and March 1986 would coincide with the
Uranus Encounter period of Vyageer 2 and that schedule conflicts for DSN antenna
time were to be expected.

Giotto was launched aboard an Ariane-1 launch vehicle from Kourou, French Guiana,
on 2 July 1985, and it was injected into a heliocentric orbit by the spacecrafi's onboard
propulsion system the following day. DSN support began with DSS 46 at Canberra,
Australia, as the initial acquisition station. For the next seven days, continuous support
was provided by the 26-m subnet and the telemetry and radio metric data was routed
to the European Space Operation Center (ESOC) at Darmstadt, Germany, via JPL and
the NASA Communications (NASCOM) switching center in Madrid, Spain. Afier that,
DSN support was reduced to one pass per day on a 34-m antenna, DSS 12 or DSS
42. Over a two-week period in mid-September 1985, the DSN participated with ESA
in the first of several Navigation Campaigns to validate the radio metric data collected
by the ESA tracking stations. Radio metric data (two-way Doppler and ranging) from
the station pairs DSS 12/42 and DSS 61/42 was provided in the form of Orbit Data
Files for comparison with the ESA data taken during the same passes.

Preparations for encounter with Halley began in October with an extensive series of
rehearsals using Giotto as a data source and the actual sequences that would be used
during encounter. As the test progressed, anomalies were identified and corrected, and
the procedures and techniques were refined. Some changes were made to the data flow
paths for delivering the Madrid and Canberra data to ESOC. All Giotto data fi-om these
sites would be routed to NASCOM at the Goddard Space Flight Center in Greenbelt,
Maryland, rather than to JPL. NASCOM would transmit the data directly to ESOC
via satellite link. Because this configuration reduced the transmission path by 6,000
miles, it was expected to make a significant reduction in the number of transmission
errors at the high encounter data rate of 46 kbps. The rehearsals and Navigation Campaign
passes continued right up to the start of the encounter.

The DSN responded to a Giotto spacecrafi emergency in late January by making DSS 12
available for support. However, the downlink from the distressed spacecrafi was too weak
for reliable telemetry data acquisition, and the project requested support from the Goldstone
64-rn antenna. DSS 14 was released &om Vcn/dger2 support of Uranus encounter by nego-
tiation between the Voyager and Giotto Project managers. Although this was its first support
of Giotto, the station was able to acquire the downlink signal and flow the telemeuy engi-
neering data to ESOC within 37 minutes for analysis. After it was determined that the
spacecraft antenna was rnispointed, the ESA station in Carnarvon, Australia, was used to
transmit a corrective command when the spacecrafi came into view at that longitude.

Three days of continuous support by the 64-m subnetwork to cover Halley's Comet
encounter events started on 11 March 1986 and was accomplished without incident.
Two seconds before the time of closest approach, the downlink signal was lost due to
mispointing of the spacecrafi high-gain antenna. This was attributed to nutation of the
spacecrafi spin axis induced by comet dust particles impacting the spacecrafi. Although
nutation dampers on the spacecrafi slowly compensated for this effect, 34 minutes of
close-encounter, real-time, telemetry data was lost. Some of this was later recovered fi-om
the digital recordings made at the tracking stations.
Nevertheless, the project reported that the overall encounter results were spectacular and
that pictures had been obtained to within about 1,500 kilometers of the nucleus before
the downlink was

Analysis by ESOC, using radio metric data fiom the DSN and the ESA Deep Space
Tracking System (DSTS), augmented with the Ega-I and Vga-2 Halley observations
(described in the previous chapter) indicated a miss distance of 610 kilometers with an
uncertainty of 40 kilometers.

An independent estimate of the spacecraft arrival time and position at Halley was derived
at JPL fiom DSN radio metric data obtained during the final Navigation Campaign
(just prior to encounter). A predicted comet miss distance of 6 10 kilometers was obtained
with an uncertainty of 138 kilometers. The times of closest approach agreed to within
two seconds.

It was suggested that differences in the comet ephemerides and spacecraft attitude infor-
mation used by DSN and ESA could easily account for the small disagreement in the
miss distance sol~tions.~

Phobos

After a hiatus of some ten years following the Mariner and Viking missions to Mars,
scientific interest in exploration of that planet resumed in 1988 with the Soviet Union's
Phobos 88 missions. The USSR Space Research Institute (abbreviated IKI in Russian)
in Moscow was the mission planning organization. More than a dozen countries were
involved in putting together the scientific payload. It would be the first international
deep space initiative of any scope since the successful Halley flyby missions, Vega and
Giotto, in March 1986.

The mission was to be devoted to studying the planet: Mars and its environment, par-
ticularly its natural satellite Phobos, which was thought ro be an asteroid captured by
Mars. The study of Phobos was the prime scientific objective of the mission and was
to be conducted by means of two Landers on the Phobos surface. One of them, a long-
term automated Lander (abbreviated DAS in Russian), would carry celestial mechanics
experiments, spectrometers, seismometers, and cameras. The other lander, called a "Hopping
Lander" or "Frog," was fitted with a jumping mechanism to allow it to move from place
to place, analyzing the chemical composition and physical and mechanical properties of
the surface at each location.
The Landers were to be deployed to the surface of Phobos from a "bus" spacecrafi
whose orbit had been carefully adjusted to be synchronous with that of Phobos. The
bus would remain in synchronous orbit for about a month, making topographical
studies of Phobos before initiating the approach and Lander descent sequences. The
bus was to descend to within 50 meters of the surface before releasing the Landers
to make individual, controlled descents to rhe surface. Once on the surface, the
Landers would begin their automated science missions, the DAS from a single fured
location, the Frog from various locations determined by its 20-meter jumps. The bus
would return to its circular orbit, from which it would observe Mars, the Sun, and
the interplanetary environment for a few months. The total duration of the mission
was to be fifieen months.

Communications with the spacecraft were planned for every five days during cruise and
every three days in the orbital phase. In the interval, data were to be recorded on board
the spacecraft. The link with the DAS was to be established three times per day once
during each orbit of Phobos, and last 20 to 40 minutes. The Frog would transmit its
data directly to the "bus" for recording and later retransmission to Earth.

The main receiving station was at Eupatoria in the Soviet Union. From there, data was
transmitted to IKI in Moscow for processing and distribution to the experimenters via
the French Space Agency, Centre National &Etudes Spatiales (CNES), in Toulouse, France.

DSN involvement in Phobos followed naturally from its involvement with the Vega and
Haley missions two years previously. Of particular interest to the DSN was the oppor-
tunity that the Phobos DAS Lander provided for refinement of the ephemerides of Mars
and Phobos, and for improving the accuracy of the existing radio-to-planetary frame-tie
mea~urements.~

Ranging measurements between Earth and the DAS Lander were to be used to deter-
mine very precisely the orbit of Phobos around Mars. They wodd also improve current
knowledge of the orbit of Mars, which had been most accurately determined prior to
this by the ranging measurements of the two Viking Landers from 1976 to 1978. Together,
these measurements should allow the Mars orbit to be determined to an accuracy of
about 5 nanoradians, relative to the orbit of the Earth around the Sun. This was much
better than the then-current level of knowledge of the "planetary reference frame."

In addition to the Mars ephemeris improvement, the Phobos mission would aIso pro-
vide a rare opportunity for delta VLBI measurements, which could be used to tie together
the "planetary" and "radio" reference frames used for interplanetary navigation.
Each Lander had a "VLBI broadcast mode," consisting of two coherent tones spaced
14.71425 MHz apart on the downlink frequency of 1.7 GHz (C-band). In this mode,
delta-VLBI measurements between a Lander and a nearby compact radio source in the
sky would locate the Lander, and therefore Phobos, in the "radioy' frame. The ranging
measurements described above would be used to tie the positions of Phobos and Mars
together, so that the delta-VLBI data effectively placed Mars in the "radio frame," there-
by completing the "planetary to radio" frame tie. It was on the strength of this argument,
and its importance to future JPL navigation technology, that the DSN went forward
with plans to support the Phobos missions.

Apart from the problem of finding suitable compact radio sources for the delta-VLBI
observations, no significant additional effort was required to meet the Phobos require-
ments for celestial mechanics. However, in two other areas, a very considerable effort
was required.

The Soviet-designed Phobos Landers were complex spacecrafi capable of receiving com-
mands, transmitting telemetry, providing two-way range and Doppler and supporting
delta VLBI measurements. Spacecraft-to-Earth communications were carried by a 1.7-
GHz (L-band) uplink and a 5.01-GHz (C-band) downlink.

During each Phobos-Earth view period, the Lander telemetry data was transmitted in
2,048-bit frames encoded in one of three different modes. One of these modes (con-
volutional code, K = 6, r = 112) could be decoded using the standard DSN MCD
capability. To accommodate the other two modes, special software was developed at
Madrid as a logical follow-on to the Venus-Balloon experience. This was then integrat-
ed into a Phobos telemetry data proce~sor.~

Since neither the uplink nor the downlink frequencies were compatible with the DSN
standard operating frequency bands, it was necessary to design and implement a special
dual-frequency feed to handle simultaneous two-way communications with the Landers.
The dual-frequency feed was composed of an L-band feedhorn enclosing a coaxially
mounted C-band surface-wave antenna.' This feed geometry provided greater than 38
dB of isolation between the C-band uplink and the L-band downlink, and it could safe-
ly handle 15 kW of uplink C-band transmitter power. The concept for this design was
derived from the L-band feedhorn that had been implemented at all 64-m antennas for
the earlier Vega and Halley mission^.^ The way in which the horn was positioned in the
antenna to correctly illuminate the subreflector is shown in Figure 5-2.
f

L-BAND HORN

INCOMING SIGNAL

- SUBREFLECTOR

INCOMING SIGNAL

\ J

Figure 5-2.Phobos L-band feedhorn mounted on the 64-m antenna.

Following the pattern of previous Soviet space projects, the Phobos mission comprised
two nearly identical spacecraft. Each spacecraft was launched on a Proton launch vehi-
cle in July 1988 from Baikonour in the Soviet Union. Pbobos-I was disabled by a gound
control error before reaching Mars orbit. Pbobos-2 completed the Earth-Mars transfer
phase and was successllly inserted into an elliptical Mars orbit in January 1989.

Using Soviet and French onboard imaging data supplemented with rwo-way Doppler
and VLBI data provided by the DSN, the complex process of changing the elliptical
spacecrafi orbit to be circular and "synchronous" with the orbit of Phobos began. On
21 March, the spacecraft entered a nearly circular orbit with a period of 7.66 hours,
which kept it close (about 275 krn) to Phobos. On 27 March, without prior warning,
all communications with the Pbobos-2 spacecrafi were lost. Extended efforts to reestab-
lish contact with the spacecraft were unsuccessfd.

Despite the premature termination of the mission, all was not lost. A usefd body of
Phobos frame-tie data had been obtained before the spacecrafi was lost. The data con-
sisted of two passes of VLBI observations of the Phabos-2 spacecrafi and angularly nearby
quasars. The data was recorded with the DSN wide-channel bandwidth VLBI systems
during an hour of observing passes on Goldstone-Madrid and Goldstone-Canberra base-
lines on 17 February and 25 March. Sparse though this data was, its subsequent reduction
and analysis yielded satisfying results for the "firame-tie" experiment. Prior to 1989, the
planetary and radio source reference fi-ameshad been aligned with an accuracy of approx-
imately 250 nanoradians. The analysis of the Phobos data reduced this uncertainry to
20 to 40 nanoradians, depending on direction?

Background for Magellan

By the time the Galileo Era began, Venus was one of the most visited planets in the
solar system. Since the early 1960s, no less than twenty spacecraft from the United States
and the Soviet Union had carried out missions to Venus. Despite all that attention, there
existed only a sketchy, general knowledge of the planet's surface because the dense, con-
stant cloud cover precluded surface observations by conventional optical means. In the
late 1970s arid early 1980s, the Pioneer Venus (U.S.) and Venera (USSR) spacecraft had
used radar to image its surface. Earth-based radar, too, had been used to penetrate the
clouds and provided valuable surface data, but this coverage was limited by the fact that
Venus always presented the same hemisphere to Earth view. Although these endeavors
had provided answers to many intriguing scientific questions and disclosed most of Venus's
large-scale geography, they raised many more questions that remained unanswered.

To pursue these questions further, NASA planned to launch a new radar imaging space-
crafi called Venus Orbiting Imaging Radar (VOIR) from the Space Shuttle in 1984.
The program was subsequently modified to defer the launch to 1988, and the mission
name became Venus Radar Mapper (VRM). The Challenger accident in 1986 forced a
further deferment of the launch to April 1989. The new Mission was called Magellan.lo

The plan was for Magellan to arrive at Venus in August 1990, spend eight months in
orbit about the planet, and map most of surface at a resolution nearly ten times better
than the resolution of existing surface maps. On arrival, the spacecraft's rocket propul-
sion system would be fired to inject the spacecraft into an elliptical orbit around Venus.

After establishing the correct orbital conditions and checkout of the radar systems, the
science acquisition phase of the mission would begin. This phase would last 243 Earth
days or one Venus day, the time required for Venus to make one complete rotation
under the elliptical orbit of the spacecraft. During the part of each orbit when the space-
craft was near the planet surface, the spacecraft would be oriented so that the high-gain
antenna pointed slightly to one side of the ground track, and the radar was turned on.
The surface return data would be recorded on the spacecrd tape recorders until the
radar was turned off at the end of the mapping sequence about 30 minutes later. Altimetry
and radiometer data were also to be taken during this period.
f 3
SEOUENCE START TURN (5.0 MINI

(37 2 MIN; 806.4 KBPS)

NORTH SWATH

ANGLES ARE TRUE ANOMAUES

ORBV PERIOD = 3.15 HRS
PERIAPSISAT = 250 KM
PIAVeACK (64.8 MIN; 268.8 KBPS)
i J

Figure 5-3.Magellan Venus orbit mapping profile.

As the spacecraft moved further away from the surface toward apoapsis, it would be reori-
ented to point the HGA toward Earth and the stored radar and spacecrafi engineering
data would be transmitted to the DSN. Playback, at a downlink data rate of 268.8 kilo-
bits per second, would take place in two periods of about one hour each, with an interruption
of 14 minutes between them for a gyro star calibration. Data-taking on the succeeding
revolution would start about 10 minutes after completion of playback.

These data-taking and playback cycles were to be repeated on each orbit, at regular inter-
vals of three hours and nine minutes. Figure 5-3 shows these extremely busy orbital
activities and highlights the small margin of time between them.

On each successive orbit, the radar would scan a new swath of surface as the planet
rotated under the spacecraft orbit. Based on a very precise knowledge of the position of
the spacecraft determined from two-way and three-way Doppler data, the Synthetic
Aperture Radar (SAR), altimetry, and radiometry data would be processed into images
and maps for scientific study.
f 3

HGA (FROM VOYAGER)

FORWARD EQUIPMENT
ALTIMETER HORN
PROPELLANT TANK
(FROM SHUTTLE)

EQUIPMENT BUS
(FROM VOYAGER)
SOLAR ARRAY

STAR-48 SOLID R
MOTOR (PAM-D)

\ I
Figure 5-4.Diagram of Magellan spacecraft in cruise configuration.

Magellan Mission to Venus

The Magellan spacecrafi was powered by two Sun-tracking solar panels and stabilized in
three axes using gyros and a star sensor for attitude reference. The spacecrafi carried a
solid rocket motor for Venus orbit insertion, as we11 as a small hydrazine system for tra-
jectory correction maneuvers and certain attitude control functions. Low-gain and
medium-gain antennas and a 3.7-m high-gain antenna (HGA) rigidly attached to the
spacecraft provided S-band and X-band uplinks and downlinks with the DSN. The
HGA also functioned as the SAR mapping and radiometer antenna during orbital oper-
ations. A separate horn antenna, for the radar altimeter, was fixed to the side of the
main body structure. A diagram of the Magellan spacecraft is shown in Figure 5-4.

The capabilities of the DSN Mark IVA systems in 1989 were adequate to meet the
tracking and data acquisition requirements of the Magellan mission. However, the require-
ments for 70-m antenna time could not be fully met due to conflicting requirements
from other high-priority missions such as Galilee, UGysses, and %yager 2. In such cases
where the 70-m antenna conflict could not be resolved through schedding, Magellan
was downloaded to an arrayed configuration of two 34-m antennas at the same loca-
tion. Further economy in antenna usage was effected by making a small modification
to one of the two receivers at all 70-m and 34-m standard antennas, which allowed
both of the Magellan subcarriers to be received through a single X-band receiver chan-
nel. Since Pioneer 12 was also in orbit about Venus, the second receiver channel at the
same antenna could then be used to pass the S-band Pioneer 12 signals.

The downlink coverage of important events associated with the launch and initial
acquisition sequence were the subject of an immense amount of planning and coor-
dination by the DSN in the year or two preceding launch.I0 DSN responsibility, in
this so-called "near-Earth" phase of the mission, was profoundly changed by the NASA
decision to use the Space Shuttle rather than the former expendable launch vehicles
for all future planetary launches. The original "near-Earth" planning for Shuttle launch-
es was related to GuliZeo, since at that time, Galileo was to be the first of the planetary
spacecraft to use the Space Shuttle as a launch vehicle." As it turned out, Mugelhn
became the first planetary spacecraft to use a Shuttle launch, and the successful track-
ing coverage, data flow, and generation of predicted data products during the
"near-Earth" phase of the first Shuttle-based launch would validate the thoroughness
of DSN planning. The interagency agreements, interfaces, and communication con-
figurations developed for MagelLzn were used on all subsequent planetary launches,
beginning a few months later with Gulileo.

The Magellan mission to Venus began on 4 May 1989, eleven years after the launch of
Pioneer Enw, the last lanet tar^ spacecrd? to use an expendable launch vehicle. Magehn
thus became the first planetary spacecraft to be launched from a reusable space vehicle,
and it was for JPL the first planetary mission since Vqager I in September 1977.

After deployment from the Space Shuttle Atlantis cargo bay, the IUSISpacecraft combi-
nation coasted to the correct point for injection. At that point, the two-stage IUS injected
the spacecraft into an Earth-Venus transfer orbit with an arrival date of 10 August 1990.
The spacecraft then separated from the spent IUS, and initial acquisition of the Mugelhn
downlink was accomplished without incident by the first viewing, using DSN 26-rn and
34-m antennas at Canberra.

Mapping began in September 1989 and continued for three cycles of 243 days each.
During this period, MageLhn mapped over 98 percent of the Venus surface. The star-
tling detail produced by the Magelhn SAR data is obvious in the image of an impact
crater in the Atalanta Region of Venus, reproduced in Figure 5-5.

In January 1992, partway into Magellan's third cycle, one of two telemetry modulators on
the X-band downlink fiom spacecrafr transmitter A Med, leaving the spacecraft unable to
transmit the high-rate mapping data. TXle S-band downlink and telemetry performance were
Figure 5-5. Magellan SAR image of an impact crater in the Atalanta Region of Venus.

not affected. In an effort to continue the mapping mission, the project switched back to
transmitter B. However, transmitter B, which had been used earlier in the mission, had a
history of generating spurious signds and noise which severely degraded its normal use for
transmitting the mapping dara. Various schemes to continue the transmission of the rnap-
ping data at lower data rates were tried with varying degrees of success. Finally mapping was
ended on 14 September 1992, and the radar was turned off on the fourth cycle. Transmitter
B continued to provide low-rate engineering data on the X-band downlink, and the S- and
X-band Doppler data were used to obtain a high-resolution global gravity field map.

Between cycles 4 and 5, the existing elliptical orbit necessary for radar mapping was modi-
fied to near circular for gravity studies by aerobraking and propulsive maneuvers. This was
the first time an interplanetary spacecraft had used a planet's atmosphere to signkicantly
change a spacecraft orbit. The then-new rechnique was used very effectively until almost ten
years later, when a new generation of lightweight spacecraft began to arrive once more at
Mars. Repeated aerobraking maneuvers rapidy reduced the orbital ~eriodfiom over 3 hours
to 94 minutes, and its apoapsis altitude from 8,469 kilometers to 542 kilometers. With the
near-circular orbit, Magelhn could collect high-resolution gravity data over the poles, an area
for which there was no previous data. A gravity field map covering 95 percent of the globe
resulted &om the fiftA and sixth cycles. In addition to the gravity field measurements, dur-
ing several occdtadons that occurred during this period, Mugehn perfbrmed bistatic radar
experiments to measure the specular reflection of the radio signal from the surface, and radio
science experiments to determine the characteristics of the atmosphere, ionosphere, and inter-
planetary medium. The DSN was an integral part of all of these activities.

Due to funding limitations that precluded continuing the mission further, MageZhn was
intentionally destroyed by an aerobraking maneuver that caused the spacecraft to enter
the atmosphere on orbit number 15,032, near the end of the sixth cycle. All subsys-
tems remained functional to the end, albeit with some loss of redundancy. The last
confirmed downlink contact with Magellan was made by DSS 65, Madrid, at 10:04:35
UTC on 12 October 1994. This was almost five and a half years after the first down-
link contact with the DSN following its launch in May 1989.

In retrospect, the Magellan mission was an enormous success, returning more data than
allother planetary missions combined. Despite difficulties with the X-band downlinks
toward the end of the mission, the mapping data return from the mission far exceeded
the primary mission objectives. Over 98 percent of the planet was imaged at resolutions
as small as 75 meters, and a high-resolution gravity map covering more than 95 per-
cent of the planet surface was generated.12

Voyager at Neptune

The year 1989 was a most notable year for space exploration in general and the DSN
in particular. The notable events of 1989 began in January with the arrival of the Soviet
spacecraft Phobos-2 at Mars. In February, the third and final satellite in the new Tracking
and Data Relay Satellite System (TDRSS) was placed in geosynchronous Earth orbit. The
Magellan launch in late April marked the first new planetary spacecraft launch in the 12-
year period since the 1 launch in 1977. It was also the first planetary spacecraft to be
launched on the Space Shuttle. In June, the Voyager Neptune Encounter period began.
The s~acecrafimade its closest approach to the Neptune surface in August, and the prime
mission continued through September. The second Space Shuttle launch of a planetary
spacecraft took place in October with the GuZileo spacecraft bound for Jupiter. The year
ended with the long-awaited launch of the Hubble Space Telescope in December. The
DSN was involved to a greater or lesser extent in all of these events. Here, we discuss
the Voyager Neptune Encounter, in which the DSN was involved in a major way.
f \

VOYAGER 2 AT

NEPTUNE
EARTH AND SUN

Voyager2s closest approach to Neptune on August 25 (GMT) will be about 3,000 miles
above the planet's cloudtops. Five hours later, the spacecraft will fly within about 23,900
miles of the moon Triton.
L 1

Figure 5-6.Encounter geometry for Voyager 2 at Neptune.

The Neptune Encounter would be Vyager 2's closest encounter with any of the plane-
tary bodies in io twelve-year journey through the outer solar system. The close flyby
had been designed to maximize science observations of the planet's magnetic field, charged
particles, and deep atmosphere, as well as to bend the flight path close to Triton, while
limiting the risk of radiation damage to the spacecraft. Figure 5-6 shows the Neptune
Encounter geometry.

While the previous Uranus Encounter took place at a spacecraft-to-Earth range of about
20 AU, the distance to Earth would be about 30 AU for the Neptune Encounter.
Although the spacecraft would be 1.6 billion kilometers (1 billion miles) further from
the antennas of the DSN, it was necessary to keep the downlink telemetry rates about
the same as were used at Uranus in order to return the vast amount of science data that
was expected from this very unique event. How was this to be accomplished?
Two techniques were to be employed--data compression on the spacecraft, and anten-
na arraying in the DSN.

In the spacecraft, the telemetry data was modified by editing and compression tech-
niques to reduce the number of data bits needed to transmit a full-size image without
loss of resoiution. This improvement in the eaciency of information content of the data
stream was carried out in the Flight Data System before the telemetry signal was trans-
mitted from the spacecraft.

On Earth, the signal-gathering sensitivity of the DSN had been sigdlcantly increased by
the enlargement of the 64-m antennas to 70-m antennas in 1987. Further improvement
could be obtained by arraying 34-m antennas with the 70-m antenna at each Complex.
The Uranus encounter had demonstrated the efficacy of adding the 64-m Parkes anten-
na to the CDSCC array in order to enhance the DSN capability even further at that
longitude. The Neptune Encounter required all of that and more. Plans were made to
supplement the 70-ml34-m Goldstone array with the 27 (25-m) antennas of the National
Radio Astronomy Observatory's (NRAO) Very Large Array (VLA) facility in Socorro,
New Mexico. In addition, the 70-m antenna of the Institute for Space and Astronautical
Science (ISAS) at the Usuda Observatory in japan was to collect radio science data in
sync with the DSN srations and the Parkes Radio Astronomy antenna in Australia. This
would happen during the periods when the spacecrafi passed behind the rings, Neptune,
and Triton as viewed horn Earth. The telemeuy modulation was to be rurned off dur-
ing these complex occdtation sequences to provide maximum power in the S-band and
X-band downlinks to enhance the quality of the radio science data.

By August 1988, the Goldstone-VLA link had been installed and progressively checked
at the maximum general science and engineering data rates of 14.4 kbps and 21.6 kbps
that would be required during Encounter operations. System tests with the
new in-flight s o h a r e verified the design parameters, and gave confidence that the cor-
rect dowdink signal-to-noise ratios, error rates, and signal margins were being achieved.
Much remained to be done before the DSN was fully ready for encounter, but a review
of the Ground Data System in February of that year found that progress to that point
was satisfactory.

At the beginning of 1989, byager 2 was 4.36 billion km (2.71 billion miles) from
Earth-299 million km (186 million miles) fi-om Neptune. The Radio Science Team
was comp!eting an eight-week period of solar occultation observations, and both the
DSN and Voyager Operations Teams were engaged in intensive testing and training exer-
cises to ensure maximum readiness for the encounter events. Bright cloud features had
begun to appear on the Neptune images and, with the exception of the well-understood
receiver problems, the spacecraft was in good health.

In April, a large dark spot rotating round the planet in 17 to 18 hours was detected in
the growing detail of the Neptune images by the Imaging Team.

The readiness of all elements of the DSN to support the encounter was reviewed again,
in depth, on 1 June. The review included the Canberra, Madrid, and Goldstone Complexes,
the NOCC, GCFINASCOM, Very Large Array, Parkes Radio Telescope, and the Usuda
Deep Space Station. It was determined that with minor exceptions, all the new capa-
bilities were in place and had been demonstrated, including the new interagency capabilities
and interfaces. The review also found that appropriate arrangements were in place for
dealing with personnel or equipment contingencies and protecting the sites against radio
frequency interference, and that the operations crews bad demonstrated an acceptable
level of operational competence. The DSN was declared ready.

As defined by the Mission Plan, the Neptune Encounter period began on 5 June 1989,
81 days before closest approach. The first 62 days were called the "observatory phase"
and consisted of continuous observations of the Neptune system combined with numer-
ous pre-encounter checkouts and calibrations of the science instruments. By the time
the "far encountery'phase began on 6 August, the spacecrafi had reached the poinr where
the image of the planet and the rings required at least two narrow-angle camera frames
to cover them. Detailed ring and satellite imaging began, and infrared observations of
Neptune were made. As byager 2 approached the planet, imaging scientists were able
to track the features in the clouds to determine wind speeds.

The "Near Encounter" period, fiom 24 to 29 August, contained the highest value Neptune
science data. It included a close swing over Neptune's north polar region, a flyby of the
satellite Triton, a view of distant Nereid, numerous magnetic field measurements, and
searches for other possible rings and satellites. Voyager passed 4,900 kilometers (3,000
miles) above the cloudtops of Neptune and, five hours later, passed within 40,000 kilo-
meters (25,000 miles) of Neptune's largest moon, Triton.

The five-day "near encounter" period was covered by all three of the DSN complexes
at various times. The complicated arrayed antenna configurations at VLA-Goldstone
(VGTA) and Parkes-Canberra (PCTA) both operated perfectly to return the high-vol-
ume engineering and science data at the desired downlink data rates of 14.4 and 21.6
kilobits per second.
Figure 5-7. Voyager 2 views three most prominent features of Neptune. Although Neptune is
the smallest of our solar system's four gaseous planets, its volume is nearly 58 times that of Earth.
Three of the most prominent features in Neptune's atmosphere are captured in this photograph
reconstructed from two images taken by Voyager 2. At the north (top) is the Great Dark Spot. To
the south of the Great Dark Spot is the bright feature "Scooter," so-named because it rotates
around the globe more rapidly than the other features. Still further south is the feature called "Dark
Spot 2," which has a bright core.

The remarkable images of Neptune and Triton reproduced in Figures 5-7, 5-8, and 5-9
are typical of the period. They were made possible by (among other things) the preci-
sion of the Voyager navigation and the high quality of the telemeq downlink. Both o f
these depended on the DSN fox their realization.

Vquger 2 discovered six new moons and a number of rings at Neptune. It found an amaz-
ing assortment of icy terrains and a form of ice volcanism on the s h c e of Triton. Ra&o
science ocdration data fiom the DSN, Parka, and Usuda antennas were used to charac-
terize the amospheres and ionospheres of Neptune and Ziton and to investigate the structure
o f the rings system. Celestial mechanics determined the gravity field of the planet and the
Figure 5-8. Voyager 2 views south polar cap of Triton. Triton, Neptune's largest moon, is about
two-thirds the size of Earth's Moon and orbits Neptune at about the same distance as the Moon
orbits Earth. Nearly two dozen individual images were combined to produce this comprehensive
view of the Neptune-facing hemisphere of Triton. The large south polar cap at the bottom of the
image is highly reflective and slightly pinkish in color.

masses of its satellites. National Geographicdescribed the Neptune Encounter events as "Voyager's
Last Picture Showyy13 and the Voyager mission as the "Voyage of the Century.'"*

Once past Neptune, Vqagger 2 continued to observe the Neptunian system on a continu-
ous basis for five more weeks until early October 1989. Observations of the dark side of
Neptune and calibrations were carried out during the post-encounter period. Although this
was Vqagger 2 s last planetary encounter, the spacecraft continued to return data on mag-
netic fields and particles and to make other science observations in space as it joined Vqager
1 and Pioneers 10 and 11, moving toward the edge of the solar system and beyond, into
interstellar space. It would now be known as the Voyager Interstellar Mission and would
remain a presence in the DSN "inventory" of deep space missions for many years to come.
Figure 5-9. Voyager 2 views Neptune and Triton. This dramatic view of the crescents of Neptune
and its largest moon Triion was acquired by Voyager 2 about three days after its closest approach
to Neptune. The spacecraft was then 4.86 million kilometers (3.01 million miles) from Neptune, and
5.22 million kilometers (3.24 million miles) from Triton. The smallest details discernible are approx-
imately 90 kilometers (60 miles) in size.

The Neptune Encounter was over, and the DSN was changing its focus once again to
the beginning of a new planetary mission to follow where Vqageer 2 had gone ten years
earlier. It was the long-delayed Space Shuttle launch of the Galileo mission to Jupiter.

Background for Galileo

Carried forward by the dramatic successes of the Viking Mars Landers and the Pioneer
and Voyager flybys of Jupiter in the mid-1970s, JPL proposed an ambitious mission to
Jupiter for the 1980s. Known as Jupiter Orbiter Probe (JOP), the spacecraft combined
a powerful orbiter with an atmospheric enuy probe and would be launched directly to
Jupiter by a Titan-Centaur expendable launch vehicle. However, late in 1975, NASA
decreed that all new planetary missions would use the Space Shuttle, and at the time
of its approval by Congress in 1977, JOP was to be rhe first mission to do so. In January
1978, JOP became oEcially known as the Galileo Mission to Jupiter and was sched-
uled for launch in January 1982 on the thirtieth Shuttle flight. For a direct flight to
Jupiter, the Galileo spacecrafiwould need the high-energy, liquid hydrogen-fuelled Centaur
as an upper stage. The spacecrafi, with attached Centaur, would be deployed into Earth
orbit by the Shuttle. At the proper time, the Centaur would be fired to inject Galileo
into a direct trajectory towards Jupiter. That was the plan, but it never materialized.

Over the next six years, the Galileo project became reluctantly involved in a series of
NASA changes to the Shuttle launch schedule and performance, as well as changes to
the availability of a suitable upper stage. Each of the changes, none of them due to
the Galiho spacecrafi itself, necessitated a change to the mission design, with conse-
quent impact on the supporting services required, including the DSN. Despite the
de-moralizing effects of so many changes on the technical design teams at JPL, the
tenacity of Galileo Project Manager John Casani held the critical elements of the proj-
ect together. In that period alone, the Galileo mission was redesigned three times before
reaching the final version destined for a May 1986 launch. This version would use the
Centaur upper stage to achieve a direct ascent trajectory to Jupiter, with arrival two
years later in August 1988.15

That plan, too, never materialized. The Galt'ho spacecrafi had been shipped from JPL
to Cape Kennedy in December 1985, while preparations for its planned Shuttle-Centaur
launch the following July were in progress. In January 1986, the Space Shuttle Challenger
exploded shortly after launch and, in the afiermath of the ensuing enquiry, Centaur
was cancelled as a Shuttle payload. For the fourth time, the Galileo mission had to be
redesigned.

What emerged from this effort was to be called the Galileo Venus-Earth-Earth-Gravity-
Assist (VEEGA) mission. VEEGA would be launched in October 1989 with a lower-risk,
but also lower-energy upper stage called an Inertial Upper Stage (IUS). Since the ener-
gy available from the IUS was insufficient to rake Galileo directly to Jupiter from Earth
orbit, VEEGA would take advantage of the gravitational fields of Venus and Earth to
make up the deficit. Afrer leaving Earth orbit, Galileo would fall inward towards the
Sun to intercept Venus in February 1990. The Venus flyby was designed to accelerate
Galileo back toward a close Earth flyby in December 1990. A similar maneuver would
use Earth's gravitational field to accelerate Galileo into a wide orbit round the Sun for
a second and final gravitational assist from Earth two years later, in December 1992.
Then, Galiho would finally be moved in the right direction and with s&cient ener-
g to reach Jupiter in December 1995, six years h e r leaving Earth. The convoluted
VEEGA trajectory is shown in Figure 5-10.
 JUPITER 100

DEC 7,1995
PROBE
RELEASE
LAUNCH
JUPITER

10118/B$ 2H0190 12/08/90 10/2!3191 12IOEl92 8/28/93 12/07/95 11/07/87

i 1

Figure 5-1 0. The Galileo Venus-Earth-Earth-Gravity-Assist (VEEGA) Trajectory.

The Galdeo spacecrafi was to consist of an Orbiter and an atmospheric Probe arranged
as a composite unit as shown in Figure 5-1 1.

The Probe would enter the Jupiter atmosphere for the purpose of determining the com-
position and structure of the atmosphere as it descended through the various cloud
layers. The Orbiter was to be inserted into Jupiter orbit for the purpose of making obser-
vations of the Jupiter system, including its satellites, during the 22-month prime mission.

Though there were grand plans for the original Galileo mission to Jupiter, the acrual
mission, due to unforeseen circumstances, was carried out in quite a different way, as
we shall see. From the time of its launch in October 1989 until the planned HGA
deployment in April 1991, the Galilea spacecraft relied on S-band (2.3 GHz) uplinks
and downlinks for its telecommunications support by the DSN. These links were used
for telemetry, command, navigation, and radio science. After the HGA deployment, the
mission was ro have used the X-band (8.4 GHz) capabilities of both the spacecraft and
the DSN t3 provide the additional telecommunicationslink performance margin required
to meet all of the mission objectives at maximum Earth to Jupiter range.
 STAR SCANNER

PLASMA SCIENCE
DUST DETECTOR
ROPROPULSION MODULE

RADIOISOTOPE
.........................................
ABOVE: SPUN SECTION
BELOW: DESPUN SECTION
THERMOELECTRIC

PROBERELAYANTENNA SCAN PLATFORM,CONTAINING:

PHOTOPOLARIMETER RADIOMETER
NEAR-INFRARED MAPPING SPECTROMETER
SOLID-STATE IMAGING CAMERA
ULTRAVIOLET SPECTROMETER

Figure 5-11. The Gdileo spacecraft cruise configuration. In the composite spacecraft diagram,
the high-gain antenna (HGA) is shown in its unfurled position. At launch, the antenna is stowed in a
"furled" position, rather like a closed umbrella. At a suitable point in the mission, when gravitational
forces and solar conditions are appropriate, a latching mechanism is released to allow the HGA to
slowly unfurl into the parabolic shape required for it to perform as an antenna. This antenna is a key
element in the spacecraft's X-band downlink communication channel to the DSN 70-m antennas on
Earth. Without it, an X-band downlink is not possible. Without an X-band downlink, the High Rate
Telemetry Data Channel, on which the entire mission depended for its prime science data return, is
not possible.

However, when the antenna was unlatched in April 1991, the unthinkable happened.
The antenna failed to unfurl properly and, despite intensive efforts over the following
two years to free what appeared to be a stuck element of the unlatching mechanism,
the HGA remained unusable. The X-band downlink was useless.

In February 1993, a decision was made to redesign the mission using the available low-
gain antennas (LGAs) to support the existing S-band uplinks and downlinks. DSN and
spacecraft engineers hastened to initiate design and implementation plans to provide the
new capability that: would be needed to conduct orbital operations in 1996 and 1997
on the S-band LGAs.

Five months before arriving at Jupiter, the Galileo Probe was released from the space-
craft to continue its trajectory directly to the point of entry into the Jupiter atmosphere.
A few days h e r releasing the Probe, the Orbiter carried out a deflection maneuver
 PROBE TWUECTORY

Figure 5-12. The Galileo Orbiter and Probe arrival at Jupiter, 7 December 1995.

to place it on a trajectory that would overfly the Probe during its entry and descent.
When the Probe reached that point, it entered the Jovian atmosphere. After surviv-
ing the deceleration forces and deploying a parachute, it descended through the
atmosphere, making scientific observations as it passed through the cloud layers. This
data was transmitted on an L-band radio link to the Orbiter, which was in view of
the Probe as it descended through the cloud layers. The Orbiter relayed a small sam-
ple of this data directly to Earth as it was received from the Probe during its descent.
To ensure that the critical Probe data was preserved, the Orbiter also recorded the
full set of Probe data for later retransmission to Earth. After the Probe relay sequence
had been completed, the Orbiter used its retropropulsion engine to insert itself into
permanent Jupiter orbit. A diagram of the complex arrival sequence is shown in
Figure 5-12.

With the successful recovery of the Ml set of Probe data, the first set of mission objec-
tives had been accomplished. The remainder of the Galileo mission was directed toward
the accomplishment of the second set of mission objectives, namely, the study of the
Jupiter system including its satellites.
Jupiter orbital operations began in December 1995, immediately following Jupiter Orbit
Injection 001). For approximately the next two years, the Orbiter wodd make scien-
tific observations of the Jupiter system, including imaging from eleven close encounters
with the Jupiter satellites Ganymede, Callisto, and Europa. The modifications and improve-
ments that had been made to the S-band downlink as a consequence of the failed X-band
antenna were M y validated. The new science data returned by the spacecraft exceeded
the most optimistic expectations of the scientists.

The Galileo prime mission officially ended in December 1997, with the spacecrafi
in fully operable condition. By that time, a proposal to extend the mission for an
additional two years to further study Europa had been approved. DSN capabilities
required to support the extended mission were somewhat less than those required for
the prime mission, and on this basis, the Galileo Europa mission went forward into
the Cassini Era.

Gdileo Mission to Jupiter

This was to be the second deep space mission to use the Space Shuttle as the launch
vehicle (Magellan was the first). The Shuttle-Centaur launch in May 1986 would have
placed Galileo on a direct ascent trajectory to Jupiter, with arrival two and a half years
later in December 1988. The change from the former expendable type of Iaunch vehi-
cle to the Space ShurtIe necessitated major changes to the procedures and data flow
configurations used by the DSN to provide the project with telemetry and command
links with the spacecraft during the launch and Earth-orbit period. This was largely due
to the events associated with the Shuttle onorbit phase and the direct launch involve-
ment of additional NASA Centers, including Johnson Space Center in Texas and Lewis
Research Center in 0hio.l"

The DSN planned to support the 19% Galileo mission with the Mark IVA config-
uration described in the previous chapter. The Galileo downlinks would be on S-band
and X-band, and all complexes could support the downlink telemetry rates from 40
bits per second to 134 kilobits per second. The normal uplink would be on S-band,
with 20-kilowatt transmitters at the 34-m stations and 125-kilowatt transmitters at
the 64-m stations. However, the DSN planned to demonstrate the then-new X-band
uplink technology for command, Doppler, and ranging purposes by means of 20-kilo-
watt X-band transmitters located at the high-efficiency (HEF) antennas only. The
command data rare for S- and X-band would be 32 bits per second. The Network
planned to support the launch and cruise phases of the Galileo VEEGA mission with
the configuration shown in Figure 5-13.
'f \

COMM = GROUND COMMUNICATIONS TS = fRANSMlT S-BAND

MIL 71 = MERRm ISLAND LAUNCH AREA RS = RECEIVE S-BAND
(FLDRIDA) POCC = PROJECT OPERATIONS CONTROL CENTER RX = RECEIVE K-BAND
NOCC = NETWORK OPERATIONS CONTROL CENTER TX =TRANSMIT X-BAND
GSFC = GODDAAD SPACE FUGKl CENTER NASCOM = NASA COMMUNICATiONS
DTF = DEVELOPMENTAND TEST FACILITY
\ J

Figure 5-1 3. Network configuration for Galileo launch and cruise, 1989.

Network loading studies had identified many conflicts for Network support in 1988 and
1989. These conflicts were due to the combined requirements of Galileo at Jupiter,
Voyager at Neptune, and Ulysses in solar orbit, together with requests for extended 64-
m support from ICE and Pioneers 10 and 11. It was anticipated that these would be
resolved as the mission progressed. In addition to long-term conflicts, the ESA space-
craft, Ubsses, was also at the pad preparing for a Shuttle-Centaur launch. Resolution of
the Ubsses and Galilee competing requirements for launch priority had been documented.

In January 1986, the Network stations were completing tests to verify the operational
configurations of hardware and s o b a r e that would be required for the forthcoming
Galileo mission. The new and complicated inter-Center interfaces needed to deliver
spacecraft telemetry to the project at JPL through the Shuttle phase were being vali-
dated. The spacecraft radio system had been checked for full compatibility with the
DSN frequencies, receivers, and transmitters. The DSN was ready to proceed to the
final prelaunch phase.
Following the ChuZZenger accident, the DSN suspended further effort on Galileo-related
tasks except for technical support to the mission design team. The project desperately
searched for an alternate way of delivering the heavy Galilee spacecrah to Jupiter with-
out the high-energy Centaur launch vehicle. The search ended in August 1986 with the
discovery of the 1989 VEEGA trajectory, which was adopted as the new mission base-
line design.

While the new trajectory satisfied the limited injection capabilities of the IUS upper
stage, it carried with it several drawbacks of great significance to the DSN. One was
the flight time of six years; another was that the spacecraft would arrive at Jupiter at
maximum Earth-Jupiter range with a southerIy declination of -23 degrees rather than
minimum range with a northerly declination of +18 degrees, as would have been the
case for the 1986 launch. The maximum range translated into minimum downlink sig-
nal strength. In addition, the southerly declination meant that the most favorable location
for tracking Galileo at Jupiter would be the single site at Canberra, Australia, in the
southern hemisphere, rather than from two sites, Goldstone and Madrid, in Earth's
northern hemisphere.

With a 1989 launch, an arrival in 1995, and a two-year orbital objective, it would be
1997 when the mission ended. The DSN was being asked to commit to a capability
almost ten years in advance. This was an unprecedented degree of commitment for
NASA Headquarters to approve, considering that it had not been a simple matter to
get approval for even much shorter-term agreements.

Quite apart from that, by 1995, the spacecraft RTG power system would no longer be
adequate to run the spacecraft X-band transmitters at fdl power. They would have to
run at low power. This fact, combined with the adverse effect of increased range, meant
that the downlink signal power available to the DSN receivers would be about one-fifth
of the original design value. This was nowhere near enough to capture the huge amount
of high-rate science data called for in the original mission design.

Having come this far with the mission, the designers were not inclined to reduce their
requirements, and there was no way to find a solution before launch, now less than two
years away. It would be up to the DSN to find a solution to the problem, commit to
implementing it, and have it in full operational readiness when GuIileo arrived at Jupiter
on 7 December 1995.

The DSN plan to address this situation is illustrated in Figure 5-14 which shows the effect
of Galileo arrival at a southerIy deciination on the available downlink telemetry data rates.
Figure 5-14. Galileo 1989 VEEGA mission: predicted X-band downlink performance Jupiter
arrival, 7 December 1995. The horizontal scale in the figure shows the UTC time on 7 December
1995, (this is also called Earth Received Time or ERT) at which the downlink telemetry data rates
shown in the upper vertical scale would be available at each of the Deep Space Communication
Complexes. The lower vertical scale depicts the available downlink power margin relative to the
maximum data rate of 134 kilobits per second.

As the figure shows, the Probe data relay of paramount scientific importance would take
place in view of the southern hemisphere stations, where the view periods would be longer.

The M y arrayed 70-m and 34-m antennas ar Goldstone and Canberra could support
the high-rate telemetry for all of this period. However, the less vital but still important
10 encounter events would occur in the downlink telemetry "gap" between the setting
Madrid stations and the rising GoIdstone stations. It was imperative that this gap be
filled or significantly reduced in order to capture the 10 Encounter data.

In June 1987, a combined GalileoIDSN team of telecommunications engineers presented

John Casani, then Galileo Project Manager, with a proposal for improving downlink
telemetry performance by 1 dB. The plan was to add a new convoiurional decoder of
type (15, 114) (see chapter 7 for an explanation of convolutional coding) to the exist-
ing (7, 112) convolutional coder on the spacecraft Telemetry Modulation Unit (TMU)."
A suitable switch would allow the TMU to be commanded to select either one. An
improvement of 1 dB in downlink telemetry performance would narrow the data gap
during the Io Encounter as shown in Figure 5-4. The increase in science data rerurn
would be of great value to the Galileo project. However, because the decoders in the
DSN Mark TVA telemetry system were designed to handle the existing 7, 112 code, an
entirely new, very large decoder would be needed far the 15, 114 code. The DSN under-
took to develop such a machine and have it operational in the Network by 1995. The
costs were estimated at about $1M for GaliIeo and about $ 3 4 M for the DSN.ls

Furthermore, the DSN proposed that the downlink "gap" could be closed even hrther
by combining the Galileo high-rate telemetry obtained by the DSN with a similar stream
of recorded data from the Very Large Array (VIA) antennas of the National Radio
Astronomy Observatory (NRAO) in Soccoro, New Mexico. Also, the VLA would pro-
vide a valuable backup in the event that weather or a technical failure disabled the
Goldstone 70-m antenna at the critical time of 10 Encounter. The possible addition of
the VLA to the Galileo Ground Data System had been under study for some time by
a special team led by James W. Layland, a telecommunications specialist in the TDA
Planning Oflice. The team had identified a number of technical problems for which
solutions were not available prior to launch.19

Encouraged by the successful demonstration of VLA support for the Voyager Neptune
Encounter, which would take place in August 1989, the DSN was confident that a solu-
tion to the ~roblemsrelated to VLA support for Galileo data would be found before
Galileo reached Jupiter, and planning proceeded on that basis.

Convinced by the arguments of the telecommunications engineers, Casani agreed to add

the new convolutional coder to the spacecraft telemetry system. This was completed
shortly before launch, just in time for the DSN engineers to verify the decoding algo-
rithms on which the design of the new DSN decoder would be based. This complex
device was to be developed by JPCs Communications Systems Research Section, where
the engineer responsible for the task,Joseph I. Statman, dubbed it the Big Viterbi Decoder
(BVD). The goal was to deliver a demonstration model at Goldstone in May 1991, and
fully operational versions at Goldstone, Madrid, and Canberra by May 1995.

By November 1988, a DSN support plan for the 1989 mission had been approved by
NASA. Galileo would be supported by the 34-m standard antennas, the 34-m high-
efficiency antennas and the 70-m antennas which had been upgraded &om 64-m diame-
ter in 1987-88.

In order to capture the high-rate (134 kdobits per second) telemetry data at maximum
Jupiter range, the DSN planned to use its full arraying capability of 70-m and 34-m
antennas supplemented for critical 10 Encounter period by the VWI. The VLA data was
to be recorded and delivered to JPL for non-real-time combining with the recordings
of the DSN arrayed data.

The S-band uplinks would be provided by 400-kW transmitters at the 70-m antennas and
by 20-kW transmitters at the 34-m antennas. In addition, 20-kW X-band transmitters
were installed at each of the three 34-m HEF antennas and would be used to demon-
strate the viability of X-band uplink technology in a critical operational environment.

For spacecraft navigation purposes, the DSN would generate the standard Mark TVA radio
metric data rypes, together with the necessary media calibration data. In addition, to meet
the Galileo navigation requirement for angular measurement accuracy of 50 nanoradians
(I sigma), the DSN Very Long Baseline Interferometry (VLBI) system would process the
special tones provided on the spacecraft S-band and X-band downlink radio carriers.

Radio science data generated by the Mark IVA Radio Science System during solar con-
junctions and Jupiter occultations was to be supplemented with data for rwo other
unique experiments.'' The Faraday Rotation Experiment required the DSN to record
the right hand and left hand polarization signals at the output of the S-band orthomode
feed on the 70-m antennas. The data would later be used by the Radio Science team
to determine the rotation of the plane of poIarization (Faraday rotation) of the linear
polarized S-band downlink as it passed through the magnetic field in Jupiter's iono-
sphere. Investigators would use this data to study the magnetic field of Jupiter at lower
altitudes than was previously possible.

The objective of the other experiment was to detect the presence of g-avitational waves
in the interplanetary medium. The success of this experiment was critically dependent
on the sensitivity and stability of the two-way Doppler &equency measuring system.
Two-way X-band Doppler was now available at the 34-m HEF antennas and the nec-
essary frequency and phase stability had been designed into the microwave and
receiver-exciter subsystems for these antennas. Although the fractional frequency stabil-
iry of the X-band RF carrier signals remained to be verified after launch, expectations
were high that the desired level, at least 5 parts in 1015 (5 x 10-15),would be achieved,
and that the hitherto undetected gravitational waves would be observed during the solar
oppositions in April 1994 and June 1995.

NASCOM was to provide wideband digital communications circuits with a maximum

capaciry of 224 kbps between all Comptexes and the Mission Control and Computing
Center (MCCC) at JPL. This would be sufficient to transport the 134 kbps telemetry
plus ancillary data in real time from the sites to JPL. To enable the German Space
Operations Center (GSOC) to support Galileo cruise operations, provisions were also
made for low-rate data, voice, and teletype services from the GSOC at Wefieim, Germany,
to the MCCC at JPL.

In the weeks preceding launch, the scene at Cape Kennedy was colored by the activi-
ties of a group of environmentalists protesting the use of radioisotope thermoelectric
generators (RTGs) as a prime power source for Galileo, despite the assurances of NASA
that the risk of endangering the environment or the population was minimal. NASA
asserted its confidence in the detailed risk analysis studies that had been made for this
and several earlier planetary spacecrafi carrying similar RTG power sources, and the
launch went ahead as planned.

Because this was a combined STSIIUS launch, a high degree of coordination and tech-
nical interfacing with other NASA Centers was involved. In the last few weeks before
launch, the DSN participated in integrated simulation tests with the Kennedy Space
Center (KSC), Johnson Space Center USC), and the Air Force Consolidated Space and
Test Center (CSTC). In these tests, state vectors were received from CSTC and processed
by the DSN Navigation Team to produce predictions for eventual use by the initial
acquisition station in making first contact with the spacecraft. Data was correctly trans-
ferred across the various interfaces and delivered in a timely manner to the correct
destinations. Following a final "Operationd Readiness Test," the DSN was declared ready
for launch on 17 September.

STS-34, carrying the Galilee spacecraft bound for Jupiter, was finally launched from
Cape Kennedy on 18 October 1989. The launch was perfect and the subsequent deploy-
ment of the IUS and Galileo spacecraft from the Shuttle cargo bay on rhe f i f i Earth
revolution was completely normal. Following deployment, all the critical events leading
to initial acquisition of the spacecraft downlink, first by the 26-m antenna at Goldstone
and shorrly after by the 34-m antenna at DSS 12, played out as exactly as planned. The
success of the initial acquisition sequence testified to the excellence of the inter-Center
coordination described above. Galileo was on its way to Jupiter via a close encounter
with Venus in February 1990 and Earth flybys in December 1990 and December 1992.
Over the next few weeks, as the Earth-to-spacecraft range continued to increase, the
downlink signal level reached the point where the 70-m antennas were needed to pro-
vide the telemeuy data required for spacecraft management and control. This created
severe conflicts for other DSN users whose needs were, of necessity, of lesser priority
than those of Galileo. The situation was exacerbated in mid-December when the 70-m
antenna at Madrid became immobilized due to a failed elevation bearing. The DSN
responded immediately, and a team of JPL engineers arrived at DSS 63 two days later
to start repair work. Under intense pressure because of the rapidly approaching Venus
encounter, and working under extremely difficult winter conditions, the team was able
to replace the massive bearing and return the antenna to operations on 25 January, fif-
teen days before Venus Encounter.

The GaIileo closest approach to Venus occurred on 9 February 1990, during the mutu-
al view of DSS 43 and DSS 63. Playback of the first images of Venus began over DSS
14 on 13 February. Three hours into the Galilee pass, the Goldstone antenna had to be
shut down because of exceptionally high winds at Goldstone. Despite the resulting loss
of data, reports from the Galileo science team commented favorably on the high quali-
ty of the downlink data delivered by the DSN. The Galileo Navigation Team used the
Doppler and ranging data generated by the DSN to confirm that the gravity assist fiom
the flyby resulted in the expected equivalent velocity increase of 2.2 k m per second.

Within a few days of Venus Encounter, the sustainable downlink telemetry data rate on
even the 70-rn antennas had fallen to 40 bps, and by March it had fallen still further
to 10 bps. This was due to an unfavorable pointing angle of the spacecrafi low-gain
antennas used for transmitting the downlink signal to the DSN. During the first year
of its orbit, the spacecraft was oriented with the spin axis pointed towards the Sun. In
this attitude, the Sun shields would protect the spacecraft and the furled X-band high-
gain antenna from damaging solar thermal radiation, and the spacecraft would depend
on its low-gain antennas for uplink and downlink. When the spacecraft orbit passed
beyond the Earth orbit (I AU), solar heating would no longer be a problem. The space-
craft would be oriented toward Eatth, the big high-gain antenna would be deployed and
the X-band uplink and downlinks could be used. Until then (April 1991), the S-band
low-gain antennas would have to suffice.

Following repair of the DSS 63 elevation bearing, it was decided to inspect and carry
out a bearing maintenance program on all three 70-m anrennas to assure their integri-
ty for the rest of the Galileo mission. This work, which required three weeks of downtime
at each antenna, was started in May 1990 and completed s u c c e s s ~ yat DSS 43 in
January 1991. Fortunately during most of this time, the spacecraft was approaching the
first Earth flyby and the need for 70-m support was considerably reduced.

As the spacecrafi began to approach Earth in August, the DSN intensified its effort to
deliver high-quality Delta Differential One-way Ranging (DDOR) for use by the space-
craf? navigation team in refining the spacecraft targeting parameters for the first Earth
Encounter.

The 26-m and 34-m stations at Canberra, Madrid, and Goldstone supported the
operationally intense four-hour period of Earth-1 Encounter activity of 8 December
1990 without incident. During this period, very high downlink signal levels were
able to sustain telemetry data rates of 134 kilobits per second to provide the proj-
ect with 100 percent of the desired science data. Closesr approach to Earth occurred
at an altitude of 960 kilometers over the Caribbean Sea and out of view of the DSN,
between Madrid set and Goldstone rise. The effort expended by the DSN on the
DDOR campaign prior to Encounter was well justified when the GaliIeo Navigation
Team estimated that the improved targeting errors resulting from the use of the
DDOR data translated into a spacecraf? fuel savings of approximately 6 kilograms,
an important factor for consideration in the eventual design of the Jupiter satellite
tour, still five years in the future.

The geometry of the first Earth flyby is shown in Figure 5-15. A dramatic picture of
Antarctica taken by the GaIileo spacecraft camera several hours afier the Earth flyby at
a distance of 200,000 km (124,000 miles), is seen in Figure 5-16. The downlink data
from which this picture was constructed was received at the Goldstone 34-m antenna
DSS 12 at 134 kbps a few hours after closest approach on 8 December 1990.

In the immediate post-Encounter period, rfie project took advantage of the prevailing
downlink performance to return all of the Earth Encounter science data at high data
rates. However, as the Earth spacecrafi range increased, the downlink performance mar-
gins rapidly decreased, and by the end of January 1991, the telemetry data rate was
back to about 40 bps. By then, all three 70-m antennas were available due to cornple-
tion of elevation bearing work.

One of the final events in rhe Venus to Earth-1 portion of the Galileo mission was
to be the unfurling of the high-gain antenna (HGA) in April 1991. At that time,
the danger of solar heating was past and the spacecraft could be safely oriented towards
Earth to point the big HGA directly at the DSN antennas. The X-band uplink and
downlink would become the prime telecommunication channels. The command to
 >

CLOSEST APPROACH
TIME = December 8, 1990, 1235 P.M. PST
ALTITUDE = 960 kilometers (597 miles)
LATITUDE = 25 deg N
LONGITUDE = 63 deg W

i J

Figure 5-15.Geometry of the Galileo first Earth flyby.

unfurl the HGA had already been transmitted to the spacecraft and verified for com-
pleteness. It lay resident in the spacecraft computers waiting for the correct time to
execute, and that time had finalIy arrived. A new phase of the Galileo mission was
about to begin.

However, when the command to unfurl the HGA was activated, the redundanr deploy-
ment motors ran for eight minutes before automatically switching off. A normal deployment
should have taken three minutes. Several hundred trials had been successfully complet-
ed beforehand, but the spacecrafi spin changed by less than was expected with a full
deployment, and the Sun sensors indicated some parrial obscurement. Something was
drastically wrong with the HGA.

The rest of the spacecraft appeared to be completely normal, and the mission contin-
ued on S-band while the HGA "anomaly" was investigated. The conclusion was that
three of the eighteen ribs in the HGA structure had remained stuck in the stowed or
furled position, and the "anomaly" now became the "problem" of how to recover the
HGA. It took the HGA Recovery Team two years to solve this problem as they sub-
jected the antenna structure to repeated heating and cooling cycles and "hammering
shocks from the antenna drive motors.
Figure 5-1 6. Galileo image of Earth: Antarctica.

Galileo requirements for DSN operations coverage of the lengthy heating and cooling
cycles were greatly increased, and conflict resolution for DSN antenna time became a
major effort. The immediate impact of the HGA problem on DSN implementation
plans was loss of the Big Viterbi Decoder demonstration, planned for May and June
when the X-band downlink should have been available. The X-band uplink demon-
stration at DSS 15 during the solar opposition in March as part of the Gravitational
Wave Experiment was also a casualty.

Meanwhile, the spacecraft continued outward bound towards the apoapsis of its two-
year orbit around the Sun (see Figure 5-10). On the fringe of the asteroid belt in
October 1991, it would make a close pass by the Asteroid Gaspra before returning
to Earth for its final gravity-assist flyby in December 1992. By mid-1991, Optical
Navigation (OPNAV data, because of its potential for extreme anguIar precision,
had become a key data type for the Galileo Navigation Team. In this scheme, the
target image, the asteroid Gaspra, is photographed against a known star background
from which the precise angular position of the target can be determined. Using the
DDOR data to determine the spacecraft position enables the spacecraft to be tar-
geted very precisely to the desired position for a close flyby. Because the downlink
performance was now in the 10 bps to 40 bps range, it was not possible to recover
the images in real time. The strategy that was employed involved recording the images
on the spacecrafi tape recorder and then transmitting the playback data to the DSN
at whatever data rate the downlink could sustain. Recovery of the necessary images
was, therefore, very time-consuming for 70-m antennas. The DSN expended nearly
400 hours of 70-m antenna time in recovering the OPNAV data for the Gaspra
Encounter alone.

Closest approach to Gaspra occurred on 29 October 1991 as the spacecraft passed

within 1,600 km of the asteroid, a very precise maneuver indeed, and an impres-
sive demonstration of the power of OPNAV, Playback of all the Gaspra science at
10-40 bps occupied many months following the actual encounter and was not com-
pleted until November of the following year, 1992, just prior to the second Earth
Encounter.

The second Earth Encounter took place on 8 December 1992, with near-perfect tar-
geting results. The altitude at closest approach was only 303 km on this occasion and
occurred over the south Atlantic Ocean, again out of view of any DSN tracking sta-
tion. Because the gap in DSN coverage would be about two hours in duration, an
elaborate arrangement of smaller antennas in Santiago, Chile; Perth, Australia; and Okinawa,
Japan, was set up to cover the gap. They would accomplish this by recording the down-
link Doppler signal and returning it to JPL via the NASA Tracking and Data Relay
Satellite System (TDRSS). The purpose of this effort was to search for the occurrence,
at closest approach, of an orbit anomaly, ~ o s s i b associated
l~ with fundamental gravita-
tional or general relativity theory Although high-quality data was provided from all of
the sites, no orbit anomaly was found.

The targeting precision for the second Earth flyby was near-perfect. GaliLeo passed with-
in one kilometer of its intended path, one-tenth of a second ahead of its predicted time.
This, the last of the three gravity assists, added 3.7 km per second to the spacecrafi's
speed in its solar orbit. Galileo was finally on an elliptical path to intersect the orbit of
Jupiter, 780 million km from the Sun, on 7 December 1995. As the spacecraft depart-
ed the environs of Earth on 16 December 1992, it captured the unique picture of Earth
and the Moon shown in Figure 5-17.

Except for the HGA problem, and that was a very big exception, the success of the
Galileo mission to this point had been extraordinary. But the primary mission could
not begin until the spacecraft reached Jupiter, and that lay three years ahead. With
the downlink getting weaker every day, and no sign of progress in recovering the
Figure 5-17. Earth and Moon from Galileo, December 1992.

HGA, the question of what could be done to save the mission was receiving a great
deal of attention. Because of the long lead time involved in implementing new capa-
bility in the DSN, it was now necessary to make a decision as to whether to accept
the loss of the HGA as inevitable and redesign the mission for S-band only, or to
continue with the original X-band design in the hope that the HGA would eventu-
ally be recovered.

In announcing his decision in February 1993, Galileo Project Manager William J. O'Neill
declared:
 We are now proceeding to implement the Galileo mission using the LGA
(S-band). We are absolutely confident of achieving at least 70 percent of
our primary objectives including 100 percent of the Probe mission and sev-
eral thousand of the highest resolution images of Jupiter ever planned. All
the instruments will make their most important observations, and we will
monitor the Jupiter magnetosphere virtually continuously, albeit at low data
rate. The success of Galileo without the HGA at Jupiter will be a techno-
logical triumph. Developing the extensive, new spacecraft flight software and
ground software and hardware, including state[-]ofi-]the [-I art enhancements
in the Deep Space Network, is a big challenge to complete in just three
years. The project and the DSN are MIy up to this challenge. Galileo will
indeed fulfill its promise and be a magnificent mission at JupiterS2'

The new course was set, and the DSN moved forward upon it.

While attempts to free the antenna were in progress, Galileo and DSN engineers explored
the possibility of continuing the mission using only the existing S-band capability and
the low-gain antennas (LGAs) on the spacecraft. Earlier studies had shown that improve-
ments in the coding and efficiency of the telemetry downiink, combined with a substantial
enhancement in DSN receiving capability at S-band and suitable data processing at JPL,
could indeed produce a viable S-band It was estimated that as much as 70
percent of the original mission objectives could be achieved.

The DSN began design and development work almost immediately, and by February
1993, when the decision to commit the Galileo mission to S-band was made, the DSN
had already made substantial progress toward meeting the new requirements. The DSN
continued to support the ongoing mission at low data rates consistent with the existing
S-band capability, while moving toward the new configuration that would be required
for orbital operations in December 1995, as the new hardware and software for the new
DSCC Galileo Telemetry (DGT) subsystem became a~ailable.~' The development of the
DGT is discussed later in this chapter.

Two key elements in the design of the S-band mission were related to new spacecraft
flight software. These would be transmitted to the spacecraft to reprogram its comput-
ers in two phases. The Phase 1 set of flight software would compensate for the fact that,
without the HGA, the &-important Probe entry data could not be returned to Earth
in real time. It would have to be recorded on the spacecraft tape recorder for later play-
back at the low data rates now determined by the LGA mission design. The spacecraft's
redundant command and data system memory would be used to provide data storage
backup for the tape recorder in case the critical Probe data got "lost" during its deiiv-
ery to the project.

Once that data was safely received by the DSN, the Phase 2 flight sofrware could be
sent to the spacecraft for reprogramming its onboard computers. The Phase 2 software
would provide programs to condense the huge amount of science data collected by the
Orbiter during its two-year orbital mission, without losing any of the scientifically impor-
tant information. That data could then be transmitted to Earth at a data rate consistent
with the capabilities of the S-band downlink.

Without these enhancements, the data transmission rate via the LGA at Jupiter would
be limited to 8-16 bps. When coupled with the DGT and the M y arrayed configu-
ration of the DSCCs, including the Parkes Radio Astronomy antenna at Canberra, the
data rate codd be increased by as much as a factor of 10, up to 160 bps.

Although the original X-band HGA mission design had been based on a downlink data
rate of 134,000 bps, most of the science investigations had been adapted to the lower
data rates, and, with the exception of the Jupiter a+mospheric circulation observations,
few of the science experiments planned for the original mission had to be abandoned.

Through most of 1993, the Galileo S-band downlink was able to sustain a telemetry
data rate of 40 bps on the 70-m antennas, but as the Earth-to-spacecrafi range contin-
ued to increase, the data rate fell to 10 bps by the end of the year. The most notable
event in 1993 was rhe close flyby of another asteroid named Ida. Because the optical
images being used by the Navigation Team to target the asteroid required 30 hours of
70-m antenna time to retrieve at the low data rate, few images were provided. Nevertheless,
they were sfiicient to satisfy the targeting specifications, and good imaging of the aster-
oid was obtained during the August 1993 encounter sequence. Only five of the Ida
images could be played back in 1993. The remainder were played back over a period
of four months in the following year when the data rate returned to 40 bps. One of
the images contained the interesting view of Ida with its own tiny satellite, shown in
Figure 5-18. The image was taken 14 minutes before closest approach on 28 August
1993, at a range of 10,870 km.

In general, the Galileo mission moved through 1994 in the cruise mode with prevail-
ing d o d i n k conditions similar to those of the previous year. In Jdy the Gdziho spacecraft
was in a unique position to observe the Shoemaker-Levy Comet impacts on Jupiter. Five
of the eleven science instruments onboard the spacecraft were able to make observations,
Figum 5-18. Asteroid Ida and its moon.

which were stored on the spacecraft tape recorder for later playback at 10 bps as the
necessary antenna time became available during the rest of the year.

Toward the end of 1994, the first prototype models of several of the components of the
DGT became available for engineering demonstration tests with live spacecraft data.24
The new Block V Receiver (BVR) was used in downlink acquisition and tracking tests
with a live signal from the spacecraft. This was a special sohare-based receiver designed
to track the suppressed carrier downlink the spacecrafi would transmit when it activat-
ed the Phase 2 flight software, in the orbital operations phase, in March 1996. Performance
tests of the Full Spectrum Recorder (FSR) showed excellent agreement with theoretical
predictions for signal-to-noise-ratio, harmonic contribution, and tracking loop capabili-
ty. Galileo data that had been acquired simultaneously at DSS 14 and DSS 15 and
recorded earlier in the year were successfully combined using a prototype of the Full
Spectrum Combiner (FSC). The mean combining gain achieved with the experimental
data was within 0.02 dB of the gain expected from perfect combining. Later demon-
strations of intercontinental combining, using Galileo data fiom DSS 43 and DSS 14,
gave equatly impressive results.
Verification of the operational performance of the DGT continued in 1995 as the instal-
lations progressed at each of the DSCCs. By September 1995, the BV3R had been fully
implemented and demonstrated at all 70-m stations, and the Galileo spacecraftwas switched
to the suppressed carrier downlink mode of operation that, together with the BVR at
the 70-m stations, would be standard for the rest of the mission.

The Galileo S-band downlink was able to sustain a data rate of 8 to 16 bps through-
out 1995, as the most critical events of the entire mission approached. As a precaution
against the occurrence of problems with the 70-m antennas during these periods, each
of the antennas was taken out of service for a few weeks for hydrostatic bearing and
subreflector maintenance. DSN and Galileo mission operations personnel rehearsed the
time-critical sequences and verified the data flow paths and conf-igurations. In February,
the new Phase 1 fight software had been uplinked to the spacecraft. The long com-
mand sequences were received correctly by the spacecraft, were Loaded into the computers,
and began operating in March. All was ready as the time for Probe release approached.

The Probe release sequence began on 5 July 1995. After a system status check, the Probe
transferred to its own internal power supply and the "umbilical" cable connecting Probe
to Orbiter was disconnected. The spacecraft turned to the correct attitude for pointing
the Probe towards Jupiter, and, in response to a "GO" command, the Probe was released
at 10:07 P.M. PDT, 12 July, to travel the rest of the way on its own. Nothing more
would be heard fiom the Probe until it established its uplink relay to the Orbiter as it
descended through the Jupiter atmosphere. That would be the last 60 minutes of its life.

Three weeks later, the Orbiter executed a deflection maneuver to change its trajectory
slightly so that it would not enter the Jupiter atmosphere on arrival, but would arrive
at the correct point for Jupiter Orbit Insertion (JOT).

Both the Orbiter and the Probe arrived at Jupiter during the overlapping view periods
of DSS 43 and DSS 14 on 7 December 1995. In the final hours of its approach to
Jupiter, the Orbiter was able to capture some images of distant 10 as it crossed the orbit
of the satellite. Limited by the capabilities of the S-band mission, the 10 images were a
far cry from the finely detailed pictures that the original mission design would have pro-
duced. Nevertheless, they were adequate to show the changes that had taken place since
the Voyager flyby 17 years before.

The Orbiter made its closest approach to Jupiter at 2:46 P.M. PST, and the Probe
started transmitting its science data to the Orbiter thirteen minutes later. Although
the Probe data was not being sent to Earth in real time, telemetry from the Orbiter
 UPLINK-DOWNLINK:
A HISTORY NETWORK
OF THE DEEPSPACE

verified that the L-band signal from the Probe was being received and recorded by
the Orbiter, much to the relief of everybody concerned. By 4:14 P.M., recording of
the Probe data was complete and the spacecraft was prepared for the next major event,
insertion into Jupiter orbit. This was accomplished with great precision by the 400-
Newton retropropulsion engine, built by Messerschmicht-Bolkow-Blohm (MBB) and
provided by Germany as a partner in the Galileo project. The first orbit of Jupiter
began. Preliminary indications were that 57 minutes of Probe data had been received
and recorded by the Orbiter prior to loss of Probe signal-a most remarkable feat of
mission design and spacecrafi rechnology. Actual tape playback of the Probe science
data would begin at the end of January 1996 and was expected to take about two
months to complete at 10 to 16 bps. The Phase 1 s o h a r e had proved its worth.

From the DSN perspective, support for all the events of the critical Probe entry and JOI
sequence had been very nominal. The BVR receivers had continued to track the down-
link signal throughout the retroengine burn, and alI desired telemetry and radio metric
data products had been delivered. There were no anomalies or problems to be reported.

In each of the three years after Gulz'leo entered its final Earth to Jupiter cruise phase in
1992, the DSN had supported a major scientific event, two asteroid encounters and the
Comet Shoemaker-Levy observations. A solar conjunction and a solar opposition had
provided good data for radio science observations. Gravitational wave observations, in
conjunction with Ulysses, had also been supported. An emergency control center for
Galileo had been established at Goldstone and could have supported all operations had
the Mission Control Center at JPL been unavailable for any reason.

Many resource scheduling problems that appeared intractable early in 1993 were even-
tually resolved by DSN scheduling to fully meet Gglileo requirements for 70-m tracking
time in each of the three years. The 34-m Network was used only for test purposes dur-
ing this time. The BVR had reached maturity and was now a standard operational
component of the almost-completed DGT. Over the period of January 1993 to December
1995, the DSN maintained an average downlink telemetry data capture rate of 98.1
percent. In the same time, it correctly transmitted over 220,000 uplink commands to
the GuZileo spacecraft.

With the completion of the Probe and JOI events just described, the mission entered the
final ~hase:the orbital operations phase. During the next two years, the spacecraft would
make swera! close encounters with three of the Jupiter satellites, using the gravity assist
technique to move from one orbit to another in sequence. The first encounter would be
with Ganymede in June 1996, following an adjustment of the spacecraft orbit to avoid
repeated exposure to the Jupiter radiation environment at each periapsis passage.

By the beginning of 1996, all the components of a single-antenna version of the DGT
were onsite and being te~ted.~5After a successful Ground Data System operations demon-
stration in May, using simulated Phase 2 flight software, this version was put online for
live Galilee support. A short time later, inflight loading of the real Phase 2 s o b a r e
began. It took over sixty-five continuous hours to transmit the program to the space-
craft, the longest command sequence ever transmitted by the DSN Command System.
There were no transmission errors. AEter verifying that both computers on the space-
craft had been correctly loaded and the data properly processed by the ground data
system, the spacecraft was commanded into the new mode of operation. It irnmediate-
ly began transmitting the new downlink packetized telemetry at 32 bits per second for
processing by the new DGT at the DSN stations.

The DGT single-antenna version was used to support the first Gan~rnedeEncounter in
June 1996. At this phase of the mission, the single-antenna DGT, in conjunction with
the new downlink telemetry format, could support 40 bits per second over DSS 14 and
DSS 63, and 80 bits per second over DSS 43, already a substantial improvement over
the former performance of the downlink. The science data could be played back in days
rather than weeks, enabling the first Ganymede data to be presented at a press confer-
ence on 10 July, just rwo weeks after closest approach.

A second Ganymede Encounter in September was supported by the single-antenna

DGT configuration in a similar way, with equally satisfactory results. In September and
October, the DSN had been engaged in end-to-end ground data system testing of the
DGT in its full operational arraying mode. At SPC 40 in Canberra, the Galileo down-
link was processed through the multiple-antenna version of the DGT, with signals from
DSS 14 in Goldstone being combined, in real time, with signals from DSS 43, DSS
42 at Canberra, and DSS 49 at Parkes. After processing at SPC 40, the resulting
enhanced telemetry stream was sent to JPL via the GCF for delivery to the project.
During the tests, the spacecraft was run at all data rates up to 160 bps, the highest bit
rate available. These rates required 111 arraying capability to recover the data but were
demonstrated successfdly. The demonstrations revealed a number of configuration and
procedural problems that were rapidly corrected. There were no anomalies related to
the design of the DGT.

Now confident in its capability to support encounter operations in the fully arrayed
mode, the DSN used the DGT for the Callisto Encounter in November with very sat-
f
1
ULTRACONE

1 1 I 1
300 MHz

BLOCK V BLOCK V BLOCK V

RECEIVER RECEtVER RECEIVER

~ D G T
FSR
+I
BVR
oGT
FSR
+I
BVR
DOT
FSR
CHANNEL
+
1
BVR
CHANNEL
CHANNEL CHANNEL CHANNEL CHANNEL
t - C
1 I
GOLDSTONE CANBERRA MADRID

, GROUND COMMUNICATIONS

L 1

Figure 5-19. The DSN in full array configuration for Galileo Jupiter orbital operations.

isfactory results. Although the introduction of arrayed operations into the Galileo down-
link sequences resulted in increased operational complexity, it provided almost tenfold
increase in the downlink data rate that would otherwise have been possible.

Based on the operational performance of the DGT in supporting the Catlisto Encounter
and on confidence in the DSN ability to deal with the increased operational complex-
ity of l l l y arrayed operations, the Project Manager elected to schedule every Galilee
pass from then to the end of the mission in that mode. A functional block diagram of
the DSN in the 111 array configuration is shown in Figure 5-19.

The Europa Encounter in December 1996 and the six remaining satellite encounters of
the Galifeo prime mission in 1997 were all conducted in the array mode. As it worked
through these events, DSN experience in handling the complex operational processes
involved in red-time arraying, rapidly improved to the point where intercontinental oper-
ations becane routine. The DSN had reached a new level of competence in a very
complex operational environment. A typical data rate profile for the Galilee downlink
in 1997 routine array operations is shown in Figure 5-20.
f -l

2 160 - * -
ui 140'- -
2 120 - -I
4
$ 100'- -
80 -
a
kY :::
20 - 'NOT AVAILABLE ON THE SPACECRAFT
-
-
-

DSS 14 DSS 14
DSS 34
DSS 49
DSS 42
DSS 43
DSS 63
1 I I I I I I I 1 I I I
0800 1000 1200 1400 1600 1800 2000 2200 2400 0200 0400 0600
DSN VIEW PERIODS. UTC

\ J

Figure 5-20. Full DSN array operations for Galileo: typical downlink data rate profile.

As it moved towards the conclusion of the Prime Mission in 1997, GaZdeo provided
its sponsors with a cornucopia of imaging and other scientific riches. Typical of the
imaging results is the striking photo of the surface of Europa shown in Figure 5-21.
This image was recorded by the spacecraft during the Europa Encounter on 20 February
1997. It was downlinked as part of the total Europa science data playback sequence
at 40 to 80 bps during subsequent DSN array passes over the following five weeks.

The quality of this image and the many others like it that followed is a tribute to the
innovative ideas of the DSN and to the Galileo engineers who were involved in the
effort to recover a viable dowdink for the mission after the loss of one of its key ele-
ments, the HGA.

Background for Ulysses

In 1977, NASA and ESA agreed upon a joint mission to study the surface of the Sun
and other solar phenomena. It would be called the International Solar Polar Mission
(ISPM) and would comprise two spacecraft travelling on separate trajectories passing over
Figure 5-21. Galileo image of icebergs on the surface of Europa.

opposite poles of the Sun. One spacecraft wodd be b d t by ESA, the other by NASA,
who in addition would provide the launch vehicle. Both spacecraft would carry a com-
plement of experiments drawn from science communities in the United States and Europe.

ISPM was a natural successor to the Helios mission of a decade earlier. Helios had
explored the solar environment from an equatorial orbit around the Sun. ISPM was to
make similar observations from a polar orbit over the Sun. To create an orbit passing
over the poles of the Sun, however, was much more difficult than creating an orbit lying
entirely in the plane of the ecliptic. To accomplish this, mission designers planned to
make use of a high-velocity gravity assist from Jupiter to change the plane and direc-
tion of the spacecrafi trajectory to the extent necessary to achieve the desired north and
south polar passages.

The solar polar trajectory that was eventually initiated by the single spacecraft launch
of 1990 is shown in Figure 5-22. In the original plan, there were to have been two
spacecraft orbiting the Sun in opposite directions.
 \

North Polar Pass

....-

South Polar Pass

June - November 1

Figure 5-22. Ulysses solar polar trajectory.

-
100 days
2

In keeping with the NASA policy at the time of the original agreement, ISPM would
use the Space Shuttle as the primary launch vehicle. A single launch was planned for
1983. The rwo spacecraft would be mounted in tandem on a single injection vehicle.
After burnout and separation from the upper stage, each spacecraft would be injected
into its individual trajectory towards Jupiter.

In 1980, because of NASA difficulties with development of the Space Shuttle, the launch
was delayed to 1985. A year later, due to NASA problems with funding, the NASA
spacecrafi was canceled, and the launch of the remaining ESA spacecrafi was hrther
delayed to 1986. At the time, the circumstances surrounding the launch delays and can-
cellation of the NASA spacecrafi raised a storm of protest at ESA, the aftermath of
which persisted for many years in the European science community. Eventually, the ESA
spacecrafi was completed and placed in storage, and the ISPM mission was appropri-
ately renamed Ulysses, h e r the legendary Greek hero who, after his return from the
Trojan wars, set out to explore "the uninhabited world beyond the Sun."
f- \

HIDDEN IN M I S CONFIGURATIONVIEW:
ENERGETICPARTICLE COMP
AND IMERSTEUA GAS COS

HIGH GAIN ANTENNA

IONSAND GAMMA BURST

REACTION CONTRO
EQUIPMENT (RCE) TANK
RCE THRUSTER CLUSTER

AXIAL BOOM
3 DRIVE
\
COSMIC RAYSISOLAR
PARTICLES

WIREBOOM

\ 1

Figure 5-23.The Ulysses spacecraft inflight configuration.

In 1987, in the aftermath of ChZZenger, NASA and ESA jointly agreed that Ulysses would
be launched in October 1990, following the Magelhn and GaZileo launches in 1989.26

The Ulysses spacecraft had a mass of 370 kg and was radiation-resistant and spin-stabi-
lized. The flight configuration is shown in Figure 5-23. Prominent features included the
1.65-m high-gain antenna (HGA), the Radio Isotope Thermoelectric Generator (RTG),
and the 5.5-m radial boom, which provided an electromagnetically clean environment
for certain experiments. A 72-m dipole wire boom and a 7.5-m axial boom served as
antennas for the Unified Radio and Plasma-wave experiment.

The main platform of aluminum honeycomb provided a mounting surface for all the
electronic units and for the Reaction Control Equipment and its h~drazinefuel tank.
Science instruments other rhan those mounted on the radial boom were also mounted
in the body of the spacecraft.

The spacecraft carried redundant transponders, each of which included S-band receivers
and 5-watt transmitters, and redundant 20-watt X-band transmitters. Located on the
spin axis of the spacecraft, the HGA permanently pointed toward Earth to provide S-
band uplink and downlink and X-band downlink communications with the DSN. Two
low-gain antennas (LGAs) provided hemispherical coverage in the forward- and rear-
ward-looking directions for S-band only, and omnidirectional uplink and downlink
communications when the spacecraft was in the early phases of the mission.

Two fully redundant tape recorders supplemented the telemetry system to ensure con-
tinuous collection of scientific data throughout the mission. When contact with the DSN
was not available, data was recorded on the spacecraft for later playback when DSN cov-
erage became available. A variety of selectable downlink data rates between 128 bits per
second and 1,024 bits per second were provided for recorded or real-time data.2' 27

The UZysses spacecraft carried nine scientific experiments, all related to the investigation
of a wide range of fields and particles phenomena of the Sun and its environment. In
addition, two radio science experiments making use of the spacecraft and DSN uplinks
and downlinks were to be carried out. The solar corona experiment would study the
densiry, velocity, and turbulence spectrum of the coronal plasma near the Sun. The
Gravitational Wave Experiment (GWE) would search for low-frequency gravitational
waves. The GWE experiments would be correlated with similar observations made by
GaZdeo in order to confirm the detection, and determine the directional characteristics,
of any gravitational wave that might be detected.

Ulysses's requirements on the DSN for tracking and data acquisition support were com-
paratively modest in comparison to those of Galileo and other JPL flight projects.28The
Mark N A 70-m antenna upgrades and the two 34-m subnets had been completed by
the time of the Ulysses launch, so the DSN was able to meet most of Ulysses's require-
ments with one 34-m subnet while Galileo and MageZlllzn (in extended mission status)
were accommodated on the 70-m and remaining 34-m antennas. For a few weeks dur-
ing solar conjunctions and oppositions, radio science experiments, including the Gravity
Wave Experiment, were conducted on the 70-m antennas at the rate of one or two pass-
es a day.

The DSN configuration for Ulysses support was similar to that shown for Galileo in
Figure 5-13. Within the SPCs, the Mark IVA telemetry, tracking, command, radio sci-
ence, and monitor and control systems were M y capable of handling all the Ulysses
uplink and downlink requirements. Telemetry data and S-and X-band radio metric data
were delivered to the project at JPL via the Mark IVA GCF. Navigation requirements
for VLBI data were handled by the NOCC Navigation Subsystem. There were no require-
ments for arraying.
Ulysses Solar Mission

Placed into a 300-kilometer Earth orbit by a perfect Space Shuttle launch on 6 October
1990, Ulysses was powered into a trajectory towards Jupiter by a combined Interim Upper
Stage (1US)IPayload Assist Module (PAM) injection vehicle. About 30 minutes after the
PAM burnout, the spacecraft S-band downlink was acquired by the 26-m and 34-m
antennas at Canberra, and the Earth-Jupiter phase of the Ulysses mission began. The
first of several large trajectory correction maneuvers (TCMs) to target the spacecraft to
the Jupiter aim point was carried our over a four-day period in mid-October.

In early November, shortly after deployment of the Axial Boom, the spacecraft X-band
antenna beam unexpectedly developed an oscillatory motion called nutation. Although
the resulting variations in downlink signal threatened future use of the X-band down-
link, the nutation effect gradually disappeared as the Axial Boom became shadowed from
solar thermal effects by the spacecraft, whose attitude slowly changed to keep the HGA
pointed toward Earth. Radio science observations made during the first solar opposition
in December 1990 showed no deleterious effects from HGA nutation. A small TCM
carried out in July 1991 brought the trajectory within 250 kilometers of the desired
Jupiter aim point.

Moving very rapidly along the corrected trajectory, and with uplink and downlink oper-
ating normdy, Ulyrses reached the Jupiter aim point on 8 February 1992. DSN radio
metric data confirmed that the closest approach to Jupiter was 6.31 Jupiter radii (Rj),
as expected. Note: The GaZiZeo spacecrafi made its closest approach to Jupiter at 4.0 Rj
in 1995 with special protection to withstand the Jupiter radiation environment.

Ulysses began moving in a new orbital plane, and, for the first time, a planetary space-
craft was travelling out of the plane of the ecliptic. The trajectory took it on a big loop
back toward the Sun, passing under the south polar region of the Sun, back across the
equator, and over the north polar region. It was these two regions, around 80 degrees
south and north latitude, that would be of greatest scientific interest.

Throughout 1992 and 1993, all spacecraft, science, and routine antenna pointing oper-
ations were pedormed without incident. DSN support on the 34-m antennas was routine,
with 70-m passes being scheduled for special events.

On 26 June 1994, at 70 degrees south latitude, the spacecrafr entered what scientists
consider to be the south polar region of the Sun. For the next four months, the space-
craft collected science data on the complex forces at work in that region. Where
possible, DSN schedules were adjusted to accommodate the demand for increased
tracking time.

In August, the spacecraft moved into the Sun-Earth region, where the spacecraft's Axial
Boom was again illuminated by the Sun. It was known from earlier experiences during
the mission that this illumination would cause uneven heating of the Axial Boom, which,
in turn, caused a slight wobble in the attitude of the spacecraft. The wobble was con-
trolled by an onboard control system that maneuvered the spacecraft to keep the HGA
pointed to Earth. Ground controllers in constant contact with the spacecraft carried out
this technique and dowed the onboard system to detect and correct unwanted motion.
The European Space Agency's tracking facility at Kourou, French Guiana, was modified
and brought online to help the DSN antennas in Canberra provide 24-hour coverage
of the spacecraft for this purpose.

The spacecraft reached its most extreme latitude of 80.2 degrees south in mid-September,
and Ulysses mission scientists began reporting important discoveries in the structure and
composition of the Sun's solar wind and magnetic field at a science conference in Noordwijk,
Netherlands.

In February 1995, as the spacecraft approached a position in back of the Sun as viewed
from Earth (opposition), the spacecraft S-band transmitter was turned on to comple-
ment the X-band transmitter in conducting the solar corona radio science experiment.
During the "opposition period," the DSN Radio Science System would record changes
to the S-band and X-band spectra induced by passage through the solar corona. The
Ulysses Radio Science Team would use these data to measure the electron content of
the Sun's corona.

Ulysses recrossed the Sun's equator (ecliptic) at a distance of 1.3 AU on 12 March 1995.
With all systems in good operating condition, it headed toward the north polar region.
When it reached 70 degrees north latitude on 19 June 1995, the second and final phase
of the primary mission began. Over the Sun's north polar region, Ulysses continued to
observe such solar phenomena as the speed of the solar wind, the rate of loss of mate-
rial from the Sun, the outward pressure of the solar wind, its impact on the shape of
the heliosphere and coronal structure, and the entrance or repulsion of high-energy cos-
mic rays at these high latitudes. This period of high science activity was supported by
the Canberra, Madrid, and Goldstone stations at a rate of 16 hours per day. The obser-
vations continued through September, until the spacecraft left the high-latitude region
of the north solar pole.
 North Polar Pass
-
September 3 December 12,2001
Maximum Solar Latitude Ulysses Orbit Begins
80.1 degrees North

Aphelion
April 1998
(5.4 astronomical
units)

South Polar Pass

September 8,2000 -January 16,2001
Maximum Solar Latitude 80.1 degrees South

I
Figure 5-24. UIysses second polar orbit.

On 29 September 1995, five years &er it began, the end of the Ulysses primary mis-
sion was reached. UCysses then began the long journey back out to the orbit of Jupiter,
making fields and particles observations as it progressed. O n reaching the giant planet's
distance of 5.4 AU on 17 April 1998, the spacecraft would once again head back on a
second high-latitude trajectory to reach the vicinity of the Sun in September 2000. A
diagram of the second Ubsses polar orbit of the Sun is shown in Figure 5-24.

The DSN continued to provide 34-m antenna support for the "routine phase" of the
Ulysses extended mission, at a level of about 10 hours per day through 1996 and 1997.
With the spacecraft in good condition and sustaining requirements minimal, the DSN
expected to support Ulysses through a second solar passage in the years 2000 and 2001.

Background for Mars Observer

Mars observe^ was to be the next step in NASA's long involvement with exploration of
the Red Planet, which had begun thirty years before with the Mariner 4 spacecraft and
the first crude pictures of the planet's cratered surface. The Iast spacecraft to visit Mars
had been the twin Viking Orbiters and Landers in 1976, on a mission to search for
signs of life fiom the distant past. The primary objectives of the Mars Observer mission
were to idenufy and map the chemical and mineral composition of the surface, meas-
ure the topography of the surface landforms, defme the gravitational field, and search
for a planetary magnetic field. It was also intended to study the carbon dioxide, water,
and dust in the Martian atmosphere. The mission was to last approximately one Mars
year, equivalent to about two Earth years. The spacecraft carried seven science instru-
ments designed to accomplish these objectives.

Mars Observer also carried a radio-receiver package provided by the French space agency,
Centre National &Etudes Spatiales (CNES). It was intended that the data from several
penetrators and experiment packages, which were to be placed on the surface by the
Russian Mars 94 spacecraft toward the end of the mission, would be relayed to Earth
by this equipment.

The mission was lost three days before it was due to enter Mars orbit to begin its pri-
mary mission. When the downlink fell silent at approximately 6 P.M. PDT on 21 August
1993, the spacecraft had traveled almost eleven months into the cruise phase of the mis-
sion with no problems and had taken its first and only photograph of Mars.

For months &er the loss of the downlink, ground controllers tried to reestablish com-
munication with the spacecraft but were unsuccessfd. NASA quickly established an
independent review board to determine the most likely cause for the failure and began
to lay plans for a new mission to return to Mars.

Mars Observer Mission

The Mars Observer spacecraft was the first JPL ''all X-band spacecraft." It represented the
final transition of X-band technology to full operational status in the DSN. Ironically, it
also carried with it an experiment to demonstrate the next generation of telecornmunica-
tions technology, the use of Ka-band for space communications. Equipped with an X-band
uplink and two X-band downtinla, the spacecraft used a high-gain antenna and a low-
gain antenna for transmitting and two low-gain antennas for receiving the X-band uplink.

The new DSN 34-132 HEF antennas with X-band uplink and downlink capability were
selected to provide most of the daily c o m m ~ c a t i o n swith the spacecraft. While Mars
Observer benefited from the general upgrades to the Mark NA, several specific improve-
ments were made to support the mission. These improvements included the Standard
Format Data Unit (SFDU) headers for all data types except command; the Reed-Solomon
decoders at each Complex; and the X-band acquisition aid antenna, mounted on the
26-m antenna at Canberra, to ensure rhe initial acquisition of the X-band downlink at
first view afier launch.

By agreement with the Multimission Operations Systems Office (MOSO) and the Mars
Observer project, all DSN data, with the exception of radio metric data, was to be rout-
ed to the project data base via the new Advanced Multimission Operations System
(AMMOS). Radio metric data would be routed to the DSN Multirnission Navigation
Team for preprocessing before delivery to the project.29

The Mars Observer launch marked a change in NASA launch policy back to expend-
able launch vehicles for interplanetary launches, after ten years of Space Shuttle-based
launch policy. For the DSN, this resulted in some simplification of prelaunch interfaces
and data flow paths compared to those required by Shuttle-based launches, but it also
meant that new interfaces and agreements had to be negotiated, and data flow paths
verified, before launch.

Mars Obseruer was placed into Earth orbit on 25 September 1992, by a Titan XI1 first-
stage vehicle launched from the Cape Canaveral Air Force Station, Florida. A Transfer
Orbit Stage (TOS) booster later injected the spacecraft onto a direct ascent trajectory
to Mars. Initial acquisition of the X-band downlink via the spacecraft low-gain anten-
na was rapidly accomplished by the Canberra 26-m and 34-m antennas, and the spacecraft
settled into routine cruise activity, supported by the DSN 34-m subnet. After the first
30 days of continuous tracking, the DSN planned to support the cruise portion of the
mission at a rate of one or two passes per day until the end of July, when full 34-m
support on the HEF antennas would again be required for orbital operations. Science
instrument calibration and checkout was completed, and the instruments began return-
ing data.

In January 1993, the spacecraft switched to the high-gain antenna with no problems,
and a downlink telecommunications experiment using a low-power Ka-band transmit-
ter on the spacecraft was carried out. The Ka-band Link Experiment (KABLE) was
designed by the DSN to test the capabilities of Ka-band (9-millimeter wavelength) radi-
ation for use in future downlink telecommunications. The DSN research and development
station, DSS 13 at Goldstone, was provided with the necessary antenna and receiving
equipment to perform the testse30Excellent results were obtained, and the plan was to
repeat the experiment periodically throughout the cruise period.
During rhe cruise period, over a span of three weeks, Mars Observer participated in a
gravitational wave search with Ulysses and GaIiZeo. The X-band Doppler data was deliv-
ered to the Radio Science Team for Iater analysis.

Reports from the Flight Team indicated that all spacecraft subsystems and instrument
payload were working well. The first picture of Mars was taken on 26 July, one
month before the spacecraft was due to arrive at the point of Mars Orbit Insertion
(MOf). As the flight teams began preparing for MOI and the start of orbital oper-
ations, the DSN increased its tracking support to a full three passes per day on the
34-HEF subnet.

MOI was programmed to start on 24 August with a 29-minute burn of the space-
craft propulsion system. O n Saturday, 21 August, as part of the pre-programmed orbit
insertion sequence, the spacecraft downlink was turned off to protect the transmitter
tubes from possible shock effects as the propellant fuel tanks were pressurized. The
spacecraft was programmed to turn the downlink back on afier pressurization had
been completed.

The expected downlink never appeared. A declared spacecraft emergency condition gave
Mars Observer first priority in use of DSN antennas for downlink recovery efforts. The
spacecraft failed to respond to repeated uplink commands to turn the transmitter on,
to switch to the low-gain wide-beam antenna, switch to a backup timer, and various
other actions related to possible fdure scenarios.

All three DSN complexes responded to the emergency with continuous commanding
sequences and closely-monitored receiver and signal detection searches. As the days passed,
each possible filure scenario was explored, but to no avail. At the end of September,
an attempt was made to activate the radio beacon on the CNES radio package in the
hope that its weak (1-watt) downlink might be detected to gave some indication of the
location and condition of the spacecraft. The Jodrell Bank radio telescope in England
was enlisted to aid dne DSN 70-m antennas in the search. Again, no signal was detect-
ed after several days of searching.

Soon afier the loss of downlink occurred, an independent review board was established
by NASA to determine the most likely cause of the failure. This anomaly investigation
was augmented by a special technical review board set up by JPL. The findings of both
boards were generally consistent and pointed to a rupture in the pressurization system.
If this were the case, Mars Observer would have tumbled out of control until the pres-
surant and fuel were completely exhausted. Then, electrically damaged beyond recovery,
the spacecrafi would have overshot Mars to become a mass of space debris in a per-
petual elliptical orbit around the Sun.

The Pioneer Missions

The Pioneer program was officially terminated by NASA Headquarters on 31 March

1997. For thirty-two years, since the launch of Pioneer 6 in 1965, the family of Pioneer
spacecraft had maintained a presence as prime or extended missions on the DSN track-
ing schedules.

Pioneers 6 through 9 were launched from 1965 through 1968 into heliocentric orbits
to make particles and fields measurements inward and outward from these orbits. In the
later years, they were also used to observe the ultimate longevity and failure modes of
aging flight components such as solar panels, Sun sensors, data handling systems, and
radio transmitters and receivers. Pioneer 7 was last tracked by the DSN in May 1983.
The DSN continued to maintain downlinks with Pioneers 6, 7,and 8 until the Pioneer
program was terminated in 1997. Although Pioneers 7 and 8 had developed problems
by then, Pioneer 6 was still operating normally. An unofficial attempt to reestablish con-
tact with Pioneer 6 was made from DSS 63 in July 1997. Responding to a sequence of
commands to switch to its backup transmitter, the spacecraft activated a coaxial switch
and turned on the backup Traveling Wave Tube (TWT) amplifier for the first time in
30 years. Although the signal received by the DSN was weak (-165), and the signal to
noise ratio was poor (3 dB), a small amount of Plasma Analyzer data was received. The
DSN continued to monitor the spacecraft from time to time for the engineering inter-
est it provided on long-duration deep space missions.

Pioneers 10 and 11 were launched in 1972 and 1973 to explore Jupiter and Saturn. The
planetary encounters were used to place them on trajectories that would take them toward
the edge of the heliosphere and eventually out of the solar system. Pioneer 11 reached a
distance of 44.1 AU before its downlink was lost on 30 September 1995. It was sur-
mised that Earth had moved out of the spacecrati high-gain antenna beam after the
spacecraft could no longer be oriented to keep it pointed in the right direction. The
downlink from Pioneer 10 was still active when the program was terminated in 1997. As
with Pioneer 6, DSN interest in the longevity and performance of the downlink prompt-
ed an attempt to contact the spacecraft fiom DSS 14 on 31 July 1997. The spacecrafi
downlink was detected at a signal level of -178 dbm, and good two-way S-band Doppler
was obtained. At that time the spacecraft was estimated to be at a distance of 68.65 AU.
For most of their long lives, the Pioneer 10 and I 1 spacecraft were able to benefit kom
the periodic enhancements to DSN downlink capability which were driven not by the
Pioneers but by the requirements of other spacecrafi. In that sense, they never became
obsoiete and never reached the predicted limit of the DSN capability to maintain an
uplink and a downlink with them.

Pioneers I 2 and 13 were Venus Orbiter and Venus Multiprobe missions, respectively.
Launched about three months apart, they both reached Venus in December 1978. Pioneer
13 was destroyed on atmospheric entry, but Pioneer 12 remained active in Venus orbit
for 14 years. Over the years, the orbit gradually decayed to the point where it could no
longer be raised to prevent aerodynamic damage to the spacecraft. The mission ended
when no downlink was detected by the DSN after periapsis passage on 9 October 1992.

These pioneering spacecrafi paid enormous scientific dividends over the more than thir-
ty years of their diverse explorations of the solar system. Relatively inexpensive by planetary
spacecrafi standards, the Pioneer spacecraft established a basis for h t h e r exploration by
more complex, more costly ~pacecraft.~' Furthermore, they amply demonstrated the value
of simple, highly reliable design principles in producing long-lived spacecraft downlinks
for deep space communications.

International Cometary Evlorer (ICE)

Following its successful passage through the nucleus of the Comet Giacobini-Zinner in
September 1985, the ICE spacecraft continued to move along its highly elliptical orbit
around the Sun as the DSN moved fiom the Voyager Era into the Galileo Era. Its scien-
tific objective was to gab knowledge of how the Sun controls Earth's near-space environment.

Throughout the Galileo Era, ICE remained on the DSN tracking schedules as an extend-
ed deep space mission, supponed on the 34-m subnet for a few passes per week as
tracking time became available. Although the spacecrafi transmitted no telemetry, the
downlink carrier was recorded by the DSP during each short duration pass, and the
data was delivered to the Radio Science Investigator for analysis.

Voyager Interstellar Mission

The Voyager Interstellar Mission (VIM) began in October 1989, right after the Vqyager
2 Encounter of Neptune, and included both Voyager spacecraft. The objectives of this
prolonged mission were to sample the interplanetary medium, conduct ultraviolet stel-
lar astronomy, and search for the helio~ause,the region where the Sun's magnetic influence
wanes and intersteuar space begins. The spacecraft were expected to operate for anoth-
er twenty-five to thirty years, after which the supply of electrical power generated by
the RTGs would drop too low to operate the science insuurnenrs.

Vyageer I was exiting the solar system at the rate of 525 million k m per year on a head-
ing 35 degrees above the ecliptic plane. By late 1992, it was more than 50 AU from
Earth. At the same time, Vyageer 2, diving below the plane of the ecliptic at a 48 degree
angle, had reached a position more than 38 AU from Earth. All seven science instru-
ments needed for the VIM were operating well; others designed for the earlier planetary
mission had been turned off in 1990 to conserve power.

After the Neptune Encounter, the level of DSN support for Voyager was reduced to
periodic contact on the 34-m antennas consistent with the new requirements of the
VIM and the availability of antenna tracking time.

With the exception of the Plasma instrument on Vyageer I, all instruments on both
spacecraft continued to collect science data and return it to the DSN. The condition of
the radio subsystems on both spacecraft remained unchanged after the initial failures in
1978, and the workarounds developed to cope with them had since been improved and
become normal operating procedure.

In 1997, the Voyager project developed a plan to extend the Voyager Interstellar Mission
(VIM) for the next twenty-five to thirty years.32It was estimated that by about the year
2020, the spacecraft would no longer be able to generate sufficient electrical power for
continued operation of the science instruments. Until then, both spacecrd had the
potential to maintain their present capability for acquiring fields, particles, and wave sci-
ence data in the interplanetary and possibly h e interstellar regions.

A completely new sequencing strategy was required for spacecraft operations to provide
for acquiring the desired science data, dealing with spacecraft anomalies, and maintain-
ing each spacecrafi's high-gain antenna continuously pointed toward Earth.

Not only were the VIM sequence designers constrained by the limited available onboard
computer memory (1,700 words total), but the duration of the VIM introduced many
considerations of special interest to the DSN.

The baseline sequence was stored in the spacecraft computer memory; it contained, in
addition to a continuously executing set of spacecraft operating instructions, the HGA
pointing information required to maintain the uplink and downlink with the DSN
through the year 2020. All of the spacecraft events that required 70-m coverage for
best downlink telemetry data return were planned to occur over the Goldstone Complex
for Voyager I and over the Canberra Complex for Voyager 2. The timing of spacecrafi
events had to be adjusted by the onboard sequence to keep them synchronized with
station view periods. It was also necessary to avoid designated station quiet periods-
Thanksgiving, Christmas, and New Year's Day-and to allow for the steady increase
in one-way light time of approximately one half-hour per year of flight. For a short
mission, these were trivial matters, but in terms of a potential thirty-year mission, they
became very significant.

In September 1997, both spacecrafi had been operating in deep space for twenty years.
Vyager 1 was 68 AU from the Sun; Voyger 2 was 53 AU from the Sun. Traveling at
the speed of light, the radio signal from the spacecraft took nine hours to reach the
DSN antennas on Earth. The DSN receiving capability had improved over these years
to the point where the new 34-m antennas could support a downlink data rate of 160
bps at these great ranges. Return of the recorded plasma-wave data was possible at 7.5
kbps from Kyager 2,and 1.5 kbps from Vqyager 1 on the 70-rn antennas using the high
power X-band transmitters on the spacecrafi. Most of the Voyager support was there-
fore carried on the 34-m subnet. In a typical week in 1997, the DSN scheduled 60 to
80 hours of antenna time for each Voyager spacecraft, only 5 percent of which was pro-
vided by a 70-m antenna.

Earlier in their missions, Vyageers 1 and 2 had revealed the scientific beauty of Jupiter,
Saturn, Uranus, and Neptune. In doing so, they challenged the DSN to match the
spacecraft's ability to capture the data with a capability to receive it. As both spacecraft
slowly stretched the existing communications links toward the boundary of the solar sys-
tem, the DSN faced the same challenge. The Voyager Interstellar Mission affbrded the
DSN its first opportunity to demonstrate uplink and downlink communications "to the
edge"-and beyond.

General

In 1986, the DSN carried nine Earth Orbiter missions on its tracking schedules. Two
of these missions were of the reimbursable type, one French and the other German in
origin. All the Ear& Orbiters were supported on the single 26-m subnet, which had
become a DSN responsibility under the Networks Consolidation program of 1983-85.
By 1997, the number of Earth Orbiter missions on the DSN schedules had increased
to twenty-three, seven of which were the reimbursable type of Japanese, German, or
French origin. AU. missions were divided among the 7-my 26-m, and 34-m networks
as appropriate.

The growth of DSN involvement with Earth orbiting missions is shown in the table
below. The number of deep space missions being tracked by the DSN in the same time
period is included for comparison.

Growth of Earth Orbiter Missions Supported by rhe DSN, 1986-97

Missions 1986 1988 1990 1992 1994 1995 1996 1997

NASA
Earth
Orbiter 7 5 4 4 7 19 16 23
Non-NASA
Earth
Orbiter 2 5 7 3 1 5 3 4
Deep Space 12 11 13 13 12 10 7 10

Under the terms of a reimbursable agreement, the space agency owning the spacecrafi
was charged an hourly fee by NASA for the use of the DSN antennas. In the early
agreements, the costs for antenna time, supporting services, technical management, and
travel were detailed separately. Later, these costs were combined into a standard fee based
on the antenna type.

The reimbursable rate for the 26-m or 34-m antennas was established in 1793 at $2,825
(U.S.) per hour. By 1995, the rate had been reduced to $2,080 and hrther reduced to
$1968 by 1997, due to economies in manpower required for station operations func-
tions. The reduction in human resources required for DSN operations reflected the
greatly expanded capabilities of improvements in the Mark IVA DSN Monitor and
Control System. Corresponding hourly cost (in U.S. dollars) for the 70-m antennas was
$5,650 in 1773, reduced to $4,160 by 1795 and to $3,936 by 1997.

The NASA strategic plan of 1995 called for the "transfer of operational activities, as fea-
sible, to other federal agencies or commercial operators." In response, the Telecommunications
and Mission Operations Directorate (TMOD) conducted a study to estimate the eco-
nomic viability of using a commercial service as an alternative ro NASA-provided tracking
and data acquisition services for l o w - E d orbiter (LEO) and high-Earth orbiter (HEO)
missions. The results of the study were presented to the NASA Space Operations Office
(Code 0 ) in October 1995 and included a recommendation that the DSN 26-m subnet
be closed by the end of fiscal year 1997. The antennas were to be decommissioned, and
the equipment not needed by JPL was to be declared "surplus property" and disposed of
appropriately Procurement action for commercial services equivalent to those provided by
the 26-m subnet was initiated with a target transfer date of September 1997.

Before these recommendations were implemented, however, a new proposal for a more cost-
effective solution to the disposition of the 2G-m network was evaluated by TMOD. This
resulted in a final recommendation that the 26-m subnet be automated and utilized to sup-
port low- and high-Earrh orbiting missions through the year 2001. By that time, support
for low-Earth orbiters was to be provided by commercial services and support for the high-
Earth orbiters would continue to be provided by the M y automated 26-m subnet.

W1th the authority of the new Space Operations Management Organization (SOMO), direc-
tives to this effect were issued to the Mission Services Managers at JPL and other NASA
Field Centers in March 1997 by the Space Operations Data Services Manager. W~thinJPL
this directive provided TMOD with the formal instructions it needed to proceed with automa-
tion of the DSN 26-m subnet and the eventual transfer of low-Earth orbiting missions to
a commercial service. Planning toward these objectives began to move forward.

Emergency Mission Support

The introduction of the Tracking and Data Relay Satellite System (TDRSS) from 1983 to
1985 as an alternative to ground-based stations for tracking satellites in low-Earth orbits
required that provision be made for emergency support in the event of failure of the TDRSS
satellites, or failure of the ground station at White Sands, New Mexico. In 1983, a DSN
study had shown that for a nominai cost, the 26-m subnetwork of antennas being trans-
ferred to the DSN under the Networks Consolidation program could be utilized to provide
the necessary support for a limited set of emergency sit~ations.~~
After approval of the con-
cept by the NASA Office of Tracking and Data Acquisition (OSTDS),the necessary intehces
and operational procedures were developed to allow timely notification of an emergency sit-
uation by the Flight Projects Directorate at GSFC, and a prompt response by the DSN
Operations organization at JPL. The procedures provided for the exchange of information
needed by the DSN stations to point the antennas, acquire the uplinks and downlinks,
transmit commands, and record and deliver the desired telemetry and ranging data.

The original list of eight Earth-orbiting missions approved for emergency DSN support
under this agreement included the Hubble Space Telescope, the Space Shuttle, and the
TDRSS satellites. Through the next decade, Hubble, Shuttle, and TDRSS remained on
the DSN schedules as "emergency support only" missions, while changes and additions
to the original list brought the total to twelve by 1997.

Except for the coverage provided by the Goldstone 9-m and 26-m antennas for emer-
gency Shuttle landings at Edwards Air Force Base at Mohave, California, the need for
emergency support under the terms of this agreement occurred only once or twice per
year in the twelve years after its inception. Shuttle support was phased out of the 26-
m schedules in 1997.

Beginning in 1995, international agreements for ~ r o v i d i nemergency

~ support for low-
Earth orbiting spacecraft were negotiated among NASA (U.S.), NASDA (Japan), CNES
(France), and DLR (Germany). Under these agreements, the various space agencies agreed
to provide emergency support for foreign spacecraft on the antennas of each other's
tracking networks, as appropriate to the particular situation.34Emergency support was
defined to be a period of unscheduled tracking, telemetry, and/or command support
necessary for spacecraft survival in the event of a spacecraft emergency. The support
would be provided by existing resources on a best-efforts basis, would be of short dura-
tion, and would involve no exchange of funds or liability Only S-band uplinks and
downlinks would be used for spacecraft communications, and the support would be lim-
ited to satellites launched for peaceful purposes only. A key element of the agreements
was a listing of the mission sets for which emergency support might be requested. The
DSN antennas designated for this type of service included DSS 16 and DSS 27 at
Goldstone, DSS 46 at Canberra, and DSS 66 at Madrid.

Lower level agreements in the form of Interface Conu-01 Documents (ICDs) were devel-
oped to manage the technical interfaces, communication configurations, protocols, operational
procedures, and delivery of data products. The reciprocal exchange of telecornmtlnications
services between the agencies was viewed by the parties to the agreement as mutually ben-
eficial. At the time of this writing, there have been no calls for support of this kind.
NETWORK ENGINEERING AND IMPLEMENTATION

Reporting on the maintenance, rehabilitation, and upgrade of the DSN 64-meter antennas
in March 1782,35the DSN chief engineer, Robertson Stevens, focused atrention on the his-
tory and status of the hydrostatic azimuth bearings and azimuth radial bearings, particularly
those at DSS 14 and DSS 43. He also recommended a plan for continuing repair or replace-
ment of components and the continuing upgrade of maintenance and operating procedures.
The report also argued convincingly that it was indeed pracrical to edarge the antennas to
70-meter diameter and to extend the useful operating &equency for these antennas.

Over the next several years, while the Network continued to support all flight mis-
sions described in the previous chapter, these recommendations were implemented.
This work included the three 70-m upgrades, the last of which was completed at DSS
14 in May 1788.

By 1989, the Goldstone antenna had been in continuous operarion for 23 years, and the
overseas antennas had been in operation for 16 years. The specified design life for parts
subject to wear was 10 years. The antennas had been specified and designed more than 27
years earlier, They had performed remarkably well, but with a resurgence in the number of
planetary missions planned for the G d e o Era, even more was to be demanded of them.

To this end, a DSN Antenna Rehabilitation Team36was established in March 1989 to

identify the resources needed to maintain the new 70-m antennas past the year 2000.
The team recommended five high-priority tasks it believed would, if promptly imple-
mented, extend the useful operating life of the antennas for the next twenty-five years.

The tasks were to develop a trend analysis program to provide early detection of incip-
ient failures; to provide a means of lifting the antenna in the event that a pad or ball
joint needed to be replaced, as had occurred in Spain and Australia in 1976; to repair
or replace aging structural and mechanical elements of the antenna before further dete-
rioration took place; to rehabilitate the elevation and azimuth gearboxes, since the gear
drive assemblies ranked next to the hydrostatic bearing in the number of problems expe-
rienced over the years; and to replace the subreflector positioner and controls because
these had been a longstanding source of trouble.

The hydrostatic bearings at Spain and Australia were believed to be in good condition
and were not expected to require any special attention in the foreseeable future. However,
there was less confidence in the hture of the hydrostatic bearing at Goldstone. It was
hoped that existing procedures to inhibit rust formation would be successful and that
major rework would not be necessary. Only time would tell.

It was the opinion of the team that rhe basic design was sound and that, under nor-
mal operating conditions, the structure was not subjected to high stress levels. At the
time, cyclic stresses were considered to be infrequent and of a low level, so fatigue or
catastrophic failure was thought to be unlikely.

Elevation Bearing Failure

About the time that the 70-m Rehabilitation Team was publishing these findings (November
1989), the GalZeo spacecraft was r e a h the point where it became dependent on contin-
uous support from the 70-m antennas to sustain the downlink telemetry data rate of 1,200
bps needed for its rapidly approacfing Encounter with Venus on 9 February 1990. Closest
approach would occur during the DSS 43 overlap with DSS 63, and full 70-m coverage
was essential to the success of this first gravitational-assist event in the Galilee mission to
Jupiter. On 13 December 1989, in the midst of this intense 70-m activity, the right-side
inboard elevation bearing on the DSS 63 antenna failed, and the antenna was immobilized.

The failure occurred about 9:00 A.M. (local time) while the antenna was being used to
investigate some anomalous Doppler problems at low elevation angles. It appeared that
one or more of the rollers in the roller bearing race that supported the right inboard
elevation bearing had cracked. The bearing sustained further damage when the antenna
was driven back to the zenith position for safety. The extent of the damage can be appre-
ciated by the photograph of the damaged bearing shown in Figure 5-25.

Obviously, a very serious antenna bearing now existed. The situation was exac-
erbated by pressure born GaZiZeo, the approaching holidays, and arduous working conditions
due to inclement winter weather in Spain.

The DSN immediately dispatched several antenna engineers to Madrid to investigate the
problem. Dale Wells, one of the engineers involved in the mechanical maintenance of all
three large antennas since they were first built, remained on site to direct the repair work,
while other technical staff at JPL coordinated the shipment of spare bearings and other parts.

The unusual nature of the repair work required special tooling, much of which was fab-
ricated on site. The right side bearing, carrying two million pounds of antenna weight,
was lifted with hydraulic jacks, and removal of the damaged bearings began. This proved
time. It was decided to continue with regular analysis of bearing for detection of metal
particles. This work, requiring only two days of antenna downtime at each site, was car-
ried out during Aprii and May of 1990.

The team also directed that the loads on each of the inboard and outboard elevation
bearings on all three antennas be equalized to correct the additional, unequal loading that
resulted from the 70-m upgrade. This work was sufficiently important to require 21 days
of antenna downtime, beginning immediately at DSS 63. It was fortuitous that the Galilee
spacecraft was approaching its first Earth fly-by around this time, and consequently its
need for 70-m support was minimal. The work was completed at DSS 14 in January
1991. In some cases, before the load equalization adjustment, the inequality of load shar-
ing between the inner and outer bearings was as much as 80 percent to 20 percent. Wlth
the work completed, this was reduced to, typically, 55 percent to 45 percent or better.

Finally, the Team recommended that the original roller bearings, which were manufactured
with 314-inch axial holes to facilitate the heat treatment process, should be replaced with
solid roller bearings. Orders for new bearings were placed, and the work of replacement
began in May 1991 at DSS 63. Usiig their experience and tooling kom the previous bear-
ing replacement effort, the engineers completed the work in just 17 days. To economize on
antenna downtime, replacement of the elevation bearings at DSS 43 and DSS 14 was car-
ried out in 1993 and 1994, in conjunction with the upgrade of the elevation gearboxes.

Gearbox Rehabilitation

The primary motive power to drive the huge 70-m antennas in the azimuth and ele-
vation directions is provided by hydraulic motors activated by computer-controlled servo
systems. The high-speed output shafts of the hydraulic motors are connected to the var-
ious low-speed gears (pinions) and quadrants. These physically rotate the antenna by
using reducer gearboxes-four for the elevation drive and four for the azimuth drive.

Ever since the antennas were first built, persistent problems with the gearboxes had been
constant cause for concern and led to expensive and antenna-time consuming repair efforts.
Until about 1984, most of the gearbox problems were related to lubrication matters, but
in 1985, a significant increase in the wear patterns on all antenna drive components, par-
ticularly the gearboxes, became evident. This was attributed to the introduction of the
1985 Mark-NA servo modifications described in the previous chapter. The new servos
drove the antennas at maximum velocity between the "stow" and "track" positions, and
vice versa, causing additional wear and even inducing oscillations in the overall system.
In addition, the increasing DSN tracking schedule in that period called for more rapid
changes between multiple spacecraft; also, VLBI, clock synchronization, and Star Catalogue
passes required frequent and rapid slewing of the antennas, with fast starts and stops.

By 1986, the situation had become serious enough on the DSS 14 antenna to warrant an
examination and evaluation of wear on the elevation b d gear and pinion by a profes-
sional consultant from the University of California at Berkeley. On the basis of tt.lls evaluation,
the elevation pinion gears at DSS 14 and DSS 43 were replaced with specially hardened
gears in 1987 as part of the 70-m upgrade. Gear replacement was not required at that
time at DSS 63. In April 1988, an expert consultant specializing in gear train systems was
sought to examine and report on the condition of the gearboxes themselves. The two spe-
cific recommendations called for replacing the drive train gearbox components with
case-hardened and precision-ground gears, and adding magnetic particle detectors to the
lubrication systems to remove wear debris. DSN engineers responded with a contract to
the PhdadeIphia Gear Company in 1990 for a completely new set of components to
upgrade or replace the existing 24 gearboxes on all three 70-m antennas. The new design
incorporated the experts' opinion and the DSN engineers' experience with the original
design, which had provided approximately thirty years of continuous operational service.

Installation of the new gearboxes began with DSS 43 in March 1993 and proceeded
around the Network as downtime could be scheduled, to DSS 14 in September 1993
and DSS 63 in June 1994.

Subreflector Drive Problems

The considerations that led the 70-m Antenna Rehabilitation Team in 1989 to recom-
mend complete replacement of the subreflector suspension and drive systems resulted
from the inadequate design of the 70-m components. Constrained in many ways by the
design of the former 64-m components, they simply were not robust enough to handle
the much larger and heavier subreflector used on the 70-m antennas. Frequent failures
and problems occurred, and several modifications and temporary work-arounds were
necessary to keep the antennas in operation. Because the estimated cost of the work was
3.6 million dollars, budget constraints at the time prevented the recommendations fiom
being implemented immediately, and the matter was deferred.

However, before anything more could be done in this regard, nature itself intervened.
On Sunday 28 June 1992, just as the DSS 14 antenna was being rotated to the hori-
zon in preparation for a Pioneer 10 spacecraft track, the Goldstone area was hit by two
severe earthquakes within a few hours of each other.
In the shaking that followed, the 24,000-pound subreflector broke loose from the three-
axis drive shafi and impacted the quadripod suspension smcture. The cast aluminum
subreflector was severely damaged, and its precision-contoured surface was cracked and
holed. AIl antennas at Goldstone were halted while damage was assessed. As it turned out,
only DSS 14 had sustained permanent damage, and repair started immediately. At first it
was t h o 4 t that the subreflector would have to be replaced, and action was started to
ship the fiberglass subreflector that had been used several years earlier at DSS 43. However,
it was eventually decided that the subreflector and its drive system should be repaired in
place, over 71 meters above the ground, with the antenna at zenith. Over the next three
weeks, all the damaged parts of the subreflector and its three-axis drive system were either
repaired or replaced, and the antenna was returned to full operation on 22 July 1992.
After necessary mechanical and optical realignment adjustments were made, it was esti-
mated that the loss of RF gain attributable to damage inflicted by the earthquake (and
subsequently repaired) was no greater than 0.06 dB at X-band. This small performance
loss was wirhin acceptable limits, and the antenna was approved for full operational use.

On 1 July 1992, right after the earthquake occurred, an EngineeringAnalysis and Corrective
Actions Team was formed to analyze the damage, assist with the return to service actions,
and recommend corrective action to improve the protection of the 70-m antennas hom
future earthquake damage. In 1995, as a result of the team's findings, new earthquake-
resistant subreflector positioners were designed, built, and installed on all 70-m antennas.
The 70-m antenna mechanical specificationswere changed to better match an earthquake-
prone environment, and hture antenna designs reflected the need for seismic protection.
Seismic monitoring procedures were established for the Goldstone antenna using accelerom-
eter sensors located at strategic points in the base and apex areas of the antenna.

There have been no Grther earthquake incidents in the DSN as of this writing.

Toward the end of the Galileo Era, the funds available for antenna maintenance were
significantly affected by overall reductions in the DSN budget. Out of this situation,
yet another review board emerged to report on the physical condition of the 70-m and
34-m antennas. Based on these findings, the board was to recommend cost-effective
measures that would assure their continued reliable operation in the forthcoming peri-
od of diminished resources.

The board was chaired by Richard I? Mathison, then chief engineer, and included sev-
eral experts from the National Radio Astronomy Observatory (NRAO) with experience
in large antennas, and some with knowledge of telescope (Keck) mounting structures.
In late 1996, the board inspected dne antennas at each of the three sites and closely
examined specific onsire maintenance problems with the engineers involved.

The board concluded that the condition of the antennas and the level of maintenance
at all sites was uniformly good. It expressed concern for incipient failures in several of
the steel and concrete elements of the 70-m structures and recommended a continuing
program of monitoring and inspection to establish trends in performance deterioration.
To reduce antenna maintenance costs, the board suggested that the DSN should clear-
ly establish an absolute minimum level of maintenance required for reliable operation
for all 70-m antennas. This would enable proper allocation of funding between the
requirements for minimum maintenance, preventative and correcuve maintenance, and
the replacement and procurement of spare parts.

In 1997, it was generally believed that the 70-m antennas were in sound mechanical
condition, that their ~otentialstructural and mechanical weaknesses were well under-
stood and under control, and that: with the recommendations of the Mathison Review
Board in effect, they would continue to provide reliable uplink and downlink service
well into the new century.

Parkes-Canberra Telemetry Array

As early as 1981, advanced systems planners in the DSN had begun to evaluate the pos-
sibility of enhancing the downlink performance of the DSN 64-m antennas for special,
short-duration mission events of high science value, such as the hture Voyager Encounters
with Uranus and possibly Neptune, by arraying other large, non-DSN antennas with
those of the DSN. These ideas followed naturally from the DSWs long history of array-
ing with its own 64-m and 34-m antennas, summarized in Figure 5-26.

Because of its past associations with the DSN and reasonably close proximity to Canberra,
the 64-m antenna of the Radio Astronomy Observatory in Parkes, New South Wales,
Australia, was the first non-DSN antenna to be considered for this purpose. This facil-
ity, owned and operated by the Australian Government's Commonwealth Scientific and
Industrial Research Organisation (CSIRO), formed the basis for much of the early design
work on the DSN's own 64-m antennas twenty years earlier.

After the necessary high-level agreements between CSIRO and NASA were completed,
a technical design for a Parkes-CDSCC Telemetry Array (PCTA) was initiated.37At that
 UPLINK-DOWNLINK: OF THE DEEPSPACENETWORK
A HISTORY

f 3

Location
Mssion Date" of Antenna Combiners"
antennas types

Pioneer 8 1970 Spain 26-rn + 26-rn Passive BB

Mariner Venus-Mercury
(Mercury) Sepr. '74 U.S. 64-rn + 2 (26-m) R&D BB
Voyager 1
(Jupiter) Mar. '79 U.S. 64-m + 34-rn RTC
Voyager 2
(Jupiter) Ju!. '79 U.S. 64-m + 34-rn RTC
Pioneer 11
(Saturn) Sep. '79 U.S. 64-rn + 34-rn RTC
Voyager 1
(Saturn) Aug. '80 ALL DSCCs 64-rn + 34-rn RTC
Voyager 2
(Saturn) Aug. '81 ALL DSCCs 64-m + 34-m RTC
International Cornerary
Explorer (Giacobini-Zinner) Sep. '85 ALL DSCCs 64-m + 34-m Passive BB
Spain1U.S. 64-rn + 64-m SSRC R&D
Auscralia/Usuda 64-rn + 64-rn SSRC R&D
Voyager 2
(Uranus) Jan. '86 Spain 64-rn + 34-rn RTCIBBA
U.S. 64-rn + 2 ( 3 4 - 4 RTCIBBA
Australia 6Crn + 2 ( 3 4 4 RTCIBBA
+Parka 64-m + LBC
+ SSRC R&D
Voyager 2
(Neptune) Aug. '89 Spain 70-rn + 2 ( 3 4 4 RTCIBBA
U.S. 70-m + 2 ( 3 4 4 RTCIBBA
+ VLA 27 (25-m) + VLBC
+ SSRC
Australia 70-m + 2 (34-m) RTCIBBA
+Parkes 64-rn + LBC
+ SSRC

" Listing of month and year indicates encounter period.

The Spain and U.S. 64-rn antennas also combined dual channels wirh passive BB.
' BB = Baseband (symbol-modulated subcarrier).
LBC = Long Baseline Combiner (baseband at -300 km).
RTC = Red-Time Combiner, first version (baseband at -30 km).
RTCIBBA = Operational R T C p a r t of BBA (Baseband Assembly).
SSRC = Symbol-Stream Recording and Combining (non-real time).
-
VLBC = Very Long Baseline Combiner (baseband at 1,000 km).

L J

Figure 5-26. Twenty years of telemetry arraying in the DSN.

time, the PCTA was designed specifically to meet the downlink requirements of the
January 1986 Voyager Encounter with Uranus. It was estimated that the addition of the
64-m Parkes antenna to the CDSCC ~ o ~ g u r a t i oofn one 64-m antenna arrayed with
two 34-m antennas could increase the downlink capability for the VoyagerlUranus
Encounter by as much as 50 percent.
Prior to NASA's becoming interested in seeking the support of CSIRO for the use of
the Pxkes antenna for Voyager, the European Space Agency (ESA) had already com-
pleted similar agreements for Parkes support of the Giotto mission to Comet H d e y
The antenna had been in service for twenty years, and a major upgrade to its facilities,
computers, and servo and pointing capability had just been completed. Under this agree-
ment, CSIRO would further upgrade the antenna for operation at X-band. Fortunately
for NASA, the ESA X-band configuration for Giotto was compatible with the config-
uration required for Voyager, and arrangements were made with CSIRO for the two
agencies to share antenna equipment and tracking time from late 1985 through the
encounters with Uranus and Comer HalIey in January and March 1986.

The low-noise, X-band receiving equipment that was to be shared consisted of a feed-
horn provided by CSIRO, a microwave assembly provided by JPL, and two masers built
in the U.S. to the standard JPL design with funding provided by ESA. The front-end
equipment package was integrated with French designed down converters by ESA at
Darmstadt, West Germany A video-grade microwave link, of Japanese manufacture, was
supplied by Telecom Australia, to transfer the Parkes data stream to the Canberra Complex,
some 350 km distant, for combining with the DSN data stream.

At first, DSN engineers, with their long history of successful in-house experience, viewed
the international flavor of the PCTA front-end package with some reservation. However,
its performance under actual operational conditions lefi no room for skepticism, and it
performed well for many years afterward, with a minimum of problems.

The functional block diagram of the Parkes-CDSCC Telemetry Array shown in Figure
5-27 identifies the various areas of responsibility among the DSN, ESAIParkes, Telecom
Australia, CSIROIParkes, and the PCTA at Parkes and CDSCC.

As might be expected, the transmission time delay on the 350-km microwave link from
Parkes to Canberra introduced additional complexity to the combining process. The
delay was found to be quite stable and equal to 1.2 milliseconds. The effect of this and
other more subtle dynamic delays was provided for by special circuits in the design of
the Long Baseline Combiner.

Most of 1983 was occupied with negotiating the unique NASA-ESA-CSIRO agreements
and developing interface design control specifications to ensure that all the elements of
the PCTA would work together when assembled end-to-end. In 1984, fabrication of the
individual elements, including implementation of the intersite microwave link, was in
progress. Testing and verification for compatibility with each other and with the Voyager
f >
64 rn ---------------------

MARK NA DSN
W PPRKES

-------
CSlRO PARKS

FACIIillES FREOUWCYAND TtMING

CSRO P M W - PCTAw - ~ T E I E C O M L P C T A

Fig. 2 Parkes=CDSCCTelemebyhf, funcbonalblock diagram

L J

Figure 5-27. Parkes-CDSCC telemetry array. In this diagram, the amplified X-band downlink
signals from the antennas at Parkes and CDSCC were first changed down to the 300-MHz
band to feed the receivers which delivered baseband (subcarrier plus data) signals at their out-
puts. The baseband outputs were treated differently at the two locations. At Parkes, the baseband
signals were digitally recorded at the Mark Ill Data Acquisition Terminals (DATs) while simulta-
neously being transferred to CDSCC via the Telcom Australia microwave link. At CDSCC, the
baseband signals from each of the three DSN antennas were first combined in the Baseband
Combiner. The combined composite signal then followed two parallel paths. In one path, sub-
carrier demodulation and frame synchronization functions were performed by a Demodulation
and Synchronization Assembly (DSA). After decoding and formatting, the data stream was
returned in real time to JPL. On the other path, the baseband signal fed both a DAT (similar
to the Parkes set-up) and a Long Baseline Baseband Combiner. It was in this latter combin-
er that the Parkes and composite CDSCC signals were finally brought together for baseband
combining to provide the enhancement expected from the overall PCTA system. This opera-
tion was followed by data handling processes identical to those just described. A study of the
diagram will show how the DATs allowed these same operations to be performed in non-real
time whenever the need arose.
and Giotto spacecrafi began in 1985. The Parkes antenna began supporting Voyager and
Giotto passes in September 1985, following a final set of system performance tests to
validate the end-to-end system and a few more weeks of operational verification testing
to refine operations procedures. By the time the Voyager spacecraf? began its observa-
tory phase in November 1985, the PCTA was running in full operational status and
providing the predicted downlink enhancement that was essential for the successful Voyager
Encounter of Uranus on 4 January 1986.

At the range of Jupiter, 5 AU from the Sun, the maximum downlink data rate from
the Voyager spacecraft had been 115.2 kbps. By the time the spacecraft reached Saturn,
at a range of 10 AU, the maximum data rate had fden to the range of 44.8 kbps to
14.4 kbps. When the distance doubled again to 20 AU at Uranus, the data rate would
have been about one-fourth of that value, without the PCTA. The enhancement of the
downlink provided by the PCTA allowed the mission controllers to recover telemetry
virtually error-free from the spacecraft at data rates of 21.2 kbps to 14.4 kbps.

The demonstrated operational success of the PCTA during the Voyager Uranus Encounter
confirmed the DSN's confidence in interagency arraying as a viable adjunct to the DSN
capability for planetary missions. The future would provide two important occasions on
which the DSN would have good reason to call on these techniques. The first would
be to enhance an existing mission, Voyager at Neptune, the other to save a disabled
mission, Galileo at Jupiter.

About the time that the PCTA technical studies were being compieted in 1982, a broad-
er based study of interagency arraying as a potential benefit to DSN capability for
downlink support of planetary missions was initiated. Led by James W. Layland,38 this
study was directed toward determining "which other facilities might be feasibly and ben-
eficially employed for the support of Voyager at Uranus, and examining the
VoyagerINeptune Encounter and such other future events and options as might appear."
In light of subsequent events connected wirh the Galileo mission to Jupiter ten years
later, this proved to be a very prophetic vision.

At its conclusion in early 1983, the study recommended that the existing plans for sup-
port of Voyager at Uranus should be completed. This culminated in the PCTA described
above. The study also recommended that the arraying configuration for the Voyager
Neptune Encounter should consist of the full array of DSN antennas, plus Parkes, plus
the Very Large Array (VLA) at Socorro, New Mexico. There were some qualifications
to these recommendations related to budget matrers, but they did not affect the course
Figure 5-28. Enhanced downlink performance at Uranus.

of history. The comparative downlink performance data on which these recommenda-

tions were based are shown in Figures 5-28 and 5-29.

The recommendations from this study set the stage for the continuing DSN inter-
est in interagency arraying and contributed materially to the success of the Voyager
Encounter with Neptune in 1989, and eventually to the Galileo Encounter with
Jupiter in 1996.

VIA-Goldstone Telemetry Array

Sponsored by the National Science Foundation (NSF) and operated by the National
Radio Astronomy Observatory (NRAO), the Very Large Array (VLA) is a premier radio
astronomy facility located at Soccoro, New Mexico. It consists of 27 antennas, each 25
meters in diameter, arranged in the form of a "Y' with a 20-km radius. This was the
faciIity that, arrayed with the DSN antennas in Goldstone, Cdifornia, would be known
as the VLA-GDSCC Telemetry Array (VGTA).
 49 . I I i I i
3

48 - -
-- - - - --------------- 29.9
47 - -

DATA RATE
DB Hz KILDBITS/ SEC

00 04 08 12 16 20 24
GMT HOURS
\ I
Figure 5-29. Enhanced downlink performance at Neptune. In these figures, each curve repre-
sents one of the array options from the study. The left-hand scale represents the performance of
the downlink in terms of the ratio of data signal power (P,) to noise power (No) expressed in dB.
The right-hand scale shows the Voyager downlink telemetry data rates that correspond to the
downlink performance values. These were the downlink data rates which theory predicted would
be available 90 percent of the time when weather and other natural effects were taken into account.
The reduction in downlink data rate that results from a 6-dB loss in downlink performance, which
in turn corresponds to a doubling of Earth-to-spacecraft range, is very evident. The beneficial result
of adding a non-DSN antenna to the existing DSN capability is represented by the performance
curves. The bottom time scale shows the GMT time at which the Voyager spacecraft would be
visible at each longitude during the respective encounters. These figures convey, in a very dramatic
way, the essential reason for the strong DSN interest in the technique of antenna arraying.

The VGTA and its Australian counterpart, the PCTA, would be the DSN key to the
Voyager Encounter with Neptune in August 1989.

By that time, the range to the spacecraft would have doubled again from 20 AU at
Uranus to 40 AU at Neptune. Before that point was reached, however, the DSN 64-m
antennas would have been upgraded to 70-m diameter and this, together with the two
34-m antennas and the VGTA, would more than double the downlink capabjliry at
Goldstone and offset the loss due to increased range. The PCTA would produce a sim-
ilar result at the Canberra longitude. It was expected that these enhancements would
support data rates at Saturn similar to those that had been used at Uranus, despite the
increased range penalty.

By early 1985, a Memorandum of Agreement and Management Plan for a joint JPL-
NRAO VGTA Project had been signed by NSF and NASA. The JPL TDA Engineering
Ofice would have responsibility for overall planning and management, supported by
implementation and preparation managers at JPL and NRAO. The JPL effort was led
by Donald W. Brown, and the NRAO effort by William D. B r ~ n d a g e . ~ ~

Under the terms of these agreements, NASA agreed to bear &e costs of installing low-
noise, X-band arnplifiers on all 27 VLA antennas, as well as the direct costs of preparing
and operating the VLA for VoyagerlNeptune support. This would include at least forty
monthly tests with Goldstone between late 1984 and early 1989, in addition to forty
spacecraft tracks during the Neptune Encounter period in late 1989.

The onsite engineering, installation, test, and operation of the PCTA were managed, as
they had been for the Uranus Encounter, by the staff at the CDSCC in cooperation
with the staff at Parks.

A high-level block diagram of the VGTA is shown in Figure 5-30.

In the process of reaching the final operational configuration described above, a num-
ber of technical problems that were unique to the VGTA design demanded considerable
attention before they were finally resolved.

The problems began with the low-noise front-end amplifiers for the VLA antennas. The
baseline plan for the X-band low-noise receivers on the VLA antennas included cooled
Field Effect Transistor (FET) low-noise amplifiers. By the time the first three receivers
had been installed in 1985, new technology offered the promise of significant improve-
ment in system noise temperature by the use of the High Electron Mobility Transistor
(HEMT) low-noise amplifiers. This translated into an increase of 1.5 dB in the expect-
ed overall sensitivity of the VLA. As the HEMT amplifiers became available beginning
in 1986, tests co&rmed the expected improvements, and ultimately all of the VLA
antennas were equipped with HEMT amplifiers. It was calculated that when equipped
with HEMT receivers, the VIA would equal nearly three 64-m DSN antennas in terms
of downlink performance, a valuable resource indeed.
Figure 5-30.VLA-GDSCC telemetry array. In concept and operation, the VGTA closely resem-
bles the PCTA described above. The essential features of the VGTA as eventually implemented
were 1) X-band reception at the VIA and at GDSCC; 2) full-spectrum combining of 27 separate
signals at VLA; 3) carrier demodulation to baseband at both sites; 4) transmission of the baseband
signal from V I A to GDSCC via an Earth-satellite link; 5) standard baseband combining of one 70-
meter and two 34-meter antennas at Goldstone; 6) baseband combining of the VLA and Goldstone
signals in the Very Long Baseline Combiner at GDSCC; 7) convolutional decoding, signal pro-
cessing, and data transmission to GDSCC to JPL; and 8) symbol stream recording at both sites
to back up the real-time system and to allow for symbol stream playback and non-real-time com-
bining at Goldstone.

For the combining (summing) process to work correctly, the signals from the twenty-
seven separate VLA antennas must all be in phase with one another. By 1987, with the
aid of new, more powerful computers at the VLA, phase measurements on all antenna
baseline combinations could be made simultaneously. This allowed adjustments for phase
variations along the signal paths to each antenna, caused by fluctuations in the tropo-
sphere, to be made in near-real time. The resulting ability to track out the effects of the
troposphere, wen in the severe summer thunderstorms characteristic of the VLA site,
was instrumental in keeping the signal combining procedure working efficiently even in
the worst summer weather experienced during the Neptune Encounter period.

An item of concern in the initial planning for use of the VLA for Voyager telemetry
was the existence of a data gap in the signal output from each of the VLA antennas.
The data gap was 1.6 milliseconds in length and occurred approximately 20 rimes per
second. During the gap period, vital monitoring and "housekeeping" information passed
between the VLA antennas and their remote control center.

Early studies (1982) of the effect of the on Voyager's coding scheme indicated
that the error correcting capability of the outer Reed-Solomon code would bridge the
gap and yield error-Gee performance, comparable to standard performance with perhaps
0.5 dB loss in threshold level for the combined VGTA output. These estimates were
verified with simulated Voyager data and VGTA hardware and s o h a r e during tests at
JPL in 1985 and 1986.

The procurement and installation of a fully redundant Earth satellite link with dedicat-
ed transponders and Earth stations at each site were carried out by a contractor. Following
the resolution of initial start-up problems and frequency response adjustments at both
ends of the link, the performance of the satellite link was found to be similar to that
of the microwave link associated with the PCTA.

During 1985, it became clear that the commercial power supplied to the VLA was too
unstable to meet the DSN standards for primary power. Frequent voltage transients and
outages occurred during inclement weather conditions. The operational deficiencies of
the existing commercial power system were overcome by installing two diesel generators,
each of 1,400-kW capacity, to provide primary power for the facility This was followed
by replacement of the deteriorating underground wiring that supplied power to the
antennas out to the limit of the 20-km "Y' arrangement.

While an availability requirement of 80 percent was imposed on the PCTA Uranus

Encounter configuration, the favorable experience led to an implicit design goal on the
order of 90 percent for the entire VGTA for the Neptune Encounter. This required set-
ting much higher reliability standards for the individual elements of the K.4 system,
including the twenty-seven individual antennas. In the beginning, it was not at all clear
that thirty antennas (three DSN plus twenty-seven VLA) could be sustained and oper-
ated repeatedly with confidence in their reliability, given the diverse geographical and
organizational aspects of the array. Because of these concerns, system reliability was given
special attention throughout the implementation phase, and critical elements, such as
an online computer and rubidium fiequency standard, were provided by the DSN for
use at the VLA.

The majority of the equipment supplied by JPL was installed at the VLA and Goldstone
in late 1988. As each major assembly was completed, it was tested at the system level
before being shipped to VLA or Goldstone for integration with the onsite systems.
Monthly tests with the VLA and operational training at both sites continued through
the year. These included demonstration tracks with all antennas on the Voyager space-
craft, where the array data was processed through the end-to-end system and delivered
to the Voyager project for evaluation of its quality.

By May 1989, the VGTA was ready to provide full operational support to the Voyager
mission as the spacecraft began its observatory phase of the mission. It was anticipated
that, fiom this point on, the resolution in the Voyager images would exceed that of the
best E d - b a s e d observations. These expectations were completely fulfded by the per-
formances of the VGTA and the PCTA as the mission progressed. Some examples of
the high-quaIity images that were downlinked at 21.6 kbps fiom Neptune were shown
in Figures 5-7, 5-8, 5-9 of the Voyager section.

AIX of the carell, pre-Encounter preparations, enhanced by the proficiency of the NRAO
and DSN operations personnel at the VLA and Goldstone, paid off handsomely dur-
ing the Neptune Encounter. Statistics for the period 26 April through 28 September
1989 showed that during the forty Voyager Encounter passes supported in that period,
the VLA signal to the DSN at Goldstone was available 99.959 percent of the time. As
for the full VGTA and PCTA configurations which included the DSN antennas, data
for the same period showed that more than 99 percent of the data transmitted by the
spacecraft, from a distance of 40 AU, was captured by the Earth-based antennas of the
interagency arrays.

The successful implementation and operation of interagency arrays in Australia and the
United States provided a downlink with the capability of supporting telemetry from
Neptune at 21.6 kbps for the full view period when the Voyager spacecrafi was over
Goldstone and Canberra (at the standard DSN 90-percent weather confidence level).
Together with the expansion of the 64-meter antennas to 70 meters, interagency array-
ing had enabled a DSN downlink that effectively doubled the science data return from
the Voyager Neptune Encounter.
The arguments that drove the DSN to move from S-band to X-band for operation of
their downlinks were not so compelling for the uplink. While the immediate irnprove-
ment in downlink performance that followed directly from the frequency ratio of S-to
X-band was of critical importance to the extension of DSN downlink capability to the
outer planets, there were other means by which the S-band uplink capability could be
extended. Prior to the early 1 9 8 0 ~improvement
~ in uplink performance was achieved
by increasing the power of the DSN S-band transmitters, first from 10 kW to 20 kW,
then to 100 kW, and finally to 400 k W At the same time, the effective radiated power
in the uplinks was enhanced by the increased gain that resulted from the addition of
new, larger antennas to the Network. Nevertheless, within the DSN Advanced Systems
Development Section, a small (but very effective) X-band uansmitter development pro-
gram had been in progress for many years. This program was directed specifically toward
supporting the DSN Planetary Radar X-band transmitters had also been used
to transmit timing signals from Goldstone to Canberra and Madrid by bouncing the
time synchronization signal off the These transmitters were designed for pulsed
operation where, although the peak power level was hundreds of kilowatts, the average
power was on the order of tens of watts. As distinct from planetary radar, planetary
communications required continuous wave (CW) operation at the leveI of tens of kilo-
watts. The generation of tens of kilowatts of stable, CW, X-band power, and the power
dissipation or cooling problems that were associated with it, was a challenging problem.
The DSN needed a powerhI rationale to embark upon such a task. It first appeared in
1978 with the %atheringinterest in the search for gravitational waves.

A proposal was made to detect gravitational waves by means of the Doppler signarure
imprinted on a CW, two-way radio link between Earth arid a spacecraft crossing the
interplanetary medium,42The key to success for such an experiment lay in the use of
1) a high fi-equency for the two-way radio path and 2) extremely high stability for the
frequency of transmission. Together, this amounted to a requirement for a 20-kW, W,
X-band transmitter with a frequency stability of one part in ten raised to the power of
fifteen (1 x 10-15).

Apart from the potential for collaborating in a major scientific discovery, the DSN was
highly motivated to gain the other benefits that were associated with a highly stable X-
band uplink. These benefits related principdy to improving the quality of the DSN
radio metric data used for spacecrafi navigation by effectively removing the transmission
media effects. Improved accuracy of the DSN data would reduce the level of uncertainty
in the orbit determination calculations, on which the spacecraft navigators depended for
knowing the precise location of their spacecraft in interplanetary space.

The DSN undertook a phased development program in 1979 to add an X-band uplink
capability to the Network. In che first phase of this program, an experimental model of
an X-band transmitter and its associated equipment would be built and tested at DSS
13, the DSNYsresearch and development station at Goldstone. In the second phase, a
coherent X-band transponder would be added to an existing spacecraft prior to launch,
to be used as an inflight technology demonstration of the end-to-end X-band system.
The add-on would not intexfere with operation of the spacecraft's normal S-band uplink.

By early 1981, a prototype 20-kW X-band transmitter and a highly stable exciter to drive
it had been developed and were under test in the DSN labs at JPL. Also, arrangements
were in place to add the X-band uplink to the NASA spacecraft, one of two spacecraft
planned for the international Solar Polar mission to be launched in 1985.43The other
spacecraft was to be provided by ESA and would eventually be renamed UZysses.

Despite the cancellation of the NASA spacecraft in 1981, development of the experi-
mental X-band uplink continued. At DSS 13, a full-scale ground system, including the
Doppler extractors, X-band to S-band down-converter, and a Block I11 receiver, was set
up and intensively tested. It was then planned to carry out the inflight X-band tech-
nology demonstration with the Guileo and Knus l&&r Mapper (later called Magellan)
spacecraft in 1984, using DSS 13 as the single ground station for the e~periment.~~

In a 3 June 1982 memo to the Assistant Laboratory Director for Telecommunications

and Data Acquisition, the DSN Chief Engineer, Robertson Stevens, drew attention to
the progress that had been made in the development of a stable X-band uplink at DSS
13. He also identified the advantages that would accrue from a move to X-band for
operational use in the DSN. Relative to S-band uplinks, Stevens cited, as examples,
improvements in range and Doppler measurements, immunity to signal distortion by
charged particles along the transmission path, improved telemetry threshold in the two-
way mode of Earth-to-spacecraft communications, and a tenfold improvement in command
capability. Benefits not directly related to the uplink performance were to be found in
lower cost spacecraft implementation and testing, as well as relief from future radio fre-
quency interference problems.

He concluded by recommending that the operational Network should support the Galileo
technology demonstration as a preamble to the initiation of a fully operational X-band
uplink capability in the DSN.45
As a result, the DSN system engineering offices began writing technical requirements
for the antenna, microwave, receiverlexciter, and transmitter subsystems that would be
necessary to support the addition of an operational X-band uplink capability to the new
34-m high-efficiency (HEF) antennas, then being planned for implementation in 1984.

Shortly thereafter, Galileo was postponed to 1986 and Magellan to 1988, and the priv-
ilege of carrying out the first X-band uplink inflight demonstration fell to Galileo.

Frank I? Easterbrook and Joseph I? Brenkle made persuasive arguments for a joint
Gravitational Wave Experiment (GWE) between Galileo (X-band) and Ulysses (S-band
only), which provided the incentive needed to establish the GWE as a formal scientif-
ic as well as a technology demonstration as envisaged by the DSN.

In a c a r e l y worded agreement that reflected the uncertainty of the times with regard
to both budget and X-band operational performance, the DSN committed to imple-
menting a 20-kW X-band uplink transmitter and three 34-112HEF antennas to support
a technology demonstration which would include Doppler, ranging, two-way radio loss
measurements, and commanding technique^.^^ It was believed that particular attention
to phase stability would achieve a fractional phase stabiIiry approaching 5 x 10-15to per-
mit the detection of gravitational waves. The capability would be available in Canberra
(DSS 45) in January 1987, at Goldstone (DSS 15) in July 1987, and in Madrid (DSS
65) in January 1988.

This plan was in process of being implemented when, as a result of the Chdllenger disas-
ter in January 1986, both Galileo and Magellan were postponed to 1989, and Ulysses to
1990. In the aftermath of this affair, the HEF antennas were completed as planned, but
the installation of the X-band uplink was deferred for about three years to meet a new set
of requirements for the 1989 Galileo and Magellan missions. The new dates for the avail-
ability of an X-band uplink in the DSN were to be DSS 15 in March 1991, DSS 65 in
September 1992, and DSS 15 in January 1993. In the meantime, a method was found
for controlling the frequency and phase stability of the X-band transmitters to meet the
striigent requirements imposed by the GWE.48Implementation progressed rapidly,

In fact, the X-band uplink at DSS 15 was completed and declared operational on 22
January 1990. By the end of the year, M y operational X-band uplinks were complete
at DSS 45 and DSS 65. Tests were run to prepare them for the start of Galileo X-
band operations when the high-gain antenna was unfurled in April 1991 for immediate
support of Magellan. Ironically, Galileo never did get to use the X-band uplink because
of the HGA failure as discussed earlier in this chapter, However, the spectacular suc-
cess of the Magellan mission to Venus firmly established the X-band uplink as a pow-
erful new DSN capability for Doppler, ranging, commanding, navigation, and radio
science applications. The stringent requirements for X-band frequency stability were
used to great advantage for very precise orbit determination and detection of gravity
anomalies on Venus.

The early planning in 1994 for the new 34-meter beam waveguide antennas had made
provision for 20-kW X-band transmitters. Budget considerations required these trans-
mitters to include major components that were surplus equipment from tracking stations
operated by the Goddard Space Flight Center. Subsequently, it was found that the effec-
tive radiated power needed to meet the technical requirements of future flight missions
could be achieved with 4-kW? vis-A-vis 20-kW, transmitters on the new BWG antennas.
Furthermore, the overall cost of installing new 4-kW transmitters was estimated to be
about the same as retrofitting the surplus 20-kW transmitters. On these grounds, it was
decided that the new BWG antennas would be implemented with 4-kW X-band trans-
mitters and the work was re-directed accordingly.

By mid-1997, the X-band transmitters were compiete at all three 34-m BWG anten-
nas and, after a short period of operational testing, were put into service a few months
later. The DSN could now operate two subnets of 34-m antennas with X-band uplink
capability.

As the Galileo Era drew to a close in 1996, the X-band transmitters of the DSN were
providing uplinks for Mars Global Surveyor, Mars Pathfinder, and Near-Earth Asteroid
Rendezvous missions. In 1997, the giant C k n i spacecraft depended on X-band uplinks
for its mission to Saturn. In 1998, Deep Space 1, the first of the New Millennium mis-
sions, and the Mars Surveyor Orbiter and Mars Surveyor Lander missions would also use
X-band uplinks. Further into the future, long-term plans within the DSN called for 20-
kW X-band transmitters on the 70-m antennas in the first years of the new millennium.

In 1997, 20 years had passed since the DSN, motivated by an interest in the search for
gravitational waves, had moved toward X-band for its uplinks. X-band uplinks had become
a reality the improvements cited by Stevens in 1982 had all been fully realized, but the
search for !gravitational waves using this ~owerfirlmedium for their detection remained
a challenge for the fircure.
The Block V Receiver (BVR) was an all-digital receiver that had been under develop-
ment as the Advanced Receiver and was characterized by high phase stability and extremely
narrow tracking loop bandwidths, both highly desirable features for the Galileo and other
weak downlink application^.^^

The BVR first appeared for initial testing with the Galileo downlink at Goldstone in
February 1994. As a result of these tests, some design modifications were made. In a
further series of tests at the end of the year, the BVR successfully demonstrated its abil-
ity to track the weak Galileo downlink at all modulation settings, including the fully
suppressed carrier mode. A telemetry data stream was flowed to JPL, and Doppler data
were extracted and delivered to the Metric Data assembly. Repeated tests with the Galileo
downlink in early 1995 confirmed the BVR under M l operational condi-
tions and showed excellent agreement with the performance predicted by theory.

Compared to the existing DSN Block IV receiver, the BVR showed greatly reduced
high-frequency noise characteristics; it was virtually free of cycle slips even when the
antennas were pointed close to the Sun, and noise in the Doppler data was reduced by
at least a factor of three.

The role of the BVR in the development of the DSN Galileo Telemetry Subsystem (DGT)
is described later. However, the new capabilities brought to the DSN downlinks by the
BVR were not limited to its Galileo application. Its ability to track downlinks in the sup-
pressed carrier mode; the extremely narrow bandwidth (about 0.1 Hz) of its carrier tracking
loop; rapid, automatic signal acquisition; and great stability and versatility made it an
obvious replacement for the older Block 111 and Block IV receivers throughout the Network.
The time was propitious for the DSN to make such changes. The number of antennas
in the Network had more than doubled in recent years, as had the number of simulta-
neous downlinks now handled by the DSN at each Complex. A separate receiver was
necessary for each link, and the existing analog-based Block IV receivers were near the
end of their lifespan. New receivers were needed, and the high-performance, multiple-
function capability of the all-digital BVRs made them a logicd choice for the new era.
The 70-rn antennas were equipped with BVRs in May 1995, followed by the 34-m HEF
and 34-m BWG antennas a year later. A few of the old Block I11 and Block IV receivers
were retained for use with the remaining SDAISSA units on the 34-m STDN antennas
and the 26-m Low-Earth Orbiter antennas.
The BVR not only performed residual carrier tracking, which was the only function
performed by the Block III and Block IV receivers, but it also performed suppressed
carrier tracking arid carried out subcarrier and symbol demodulation functions. It there-
fore replaced not only the former Block I11 and IV receivers, but the former SDAISSA
and DSA units as well. In addition, it carried out the functions of signal estimation,
Doppler extraction and IF signal distribution, thereby replacing the signal precision power
monitors, Doppler extractors and IF distribution equipment that had been an integral
part of the Mark EVA DSN.

The BVR accepted an S-band or X-band signal from a low-noise amplifier on the anten-
na to which it was assigned by the station controllers and, after performing the functions
described above, distributed the following outputs for subsequent data processing:

1. two symbol streams to the assigned telemetry channel assemblies (TCAs) for decod-
ing, fi-arne synchronization, and telemetry data delivery;

2. two baseband outputs to rhe baseband assemblies (BBA) for combining with sim-
ilar signals from other antennas;

3. two open-loop IF signals to other subsystems, such as radio science and VLBI, for
special signal spectrum processing; and

4. two range-modulated signals to the Sequential Ranging Assembly (SRA) for rang-
ing measurements.

Local Area Networks provided for communication and control between the BVR and
other subsystems, and between the BVR and station controllers at the Link (LMC) or
Complex (CMC) Monitor and Control level.

THEDSN GALILEOTELEMETRY
(DGT) SUBSYSTEM

As it gradually became apparent that the Galileo HGA problem might not be correct-
ed in time to save the mission as originally designed, attention began to turn towards
viable options for continuing the mission in a modified form, using the spacecraft low-
gain antenna (LGA).

In October 1991, the Telecommunications and Data Acquisition (TDA) Office at JPL
chartered a thirty-day study to idenufy a set of options for improving the telemetry per-
formance of the Galileo downlink at Jupiter, using the LGA, in the event that the HGA
could not be recovered. The study team was led by Leslie J. Deutsch, Manager of
Technology Development in the TDA Office.

The jobs of tuba player for a Dixieland jazz band and organist for a choir, church, and
synagogue might seem unlikely roles for the mathematician most responsible for lead-
ing the DSN effort to recover the Galiko mission to Jupiter. However, all three roles
were at one time played by Dr. Leslie J. Deutsch. Born and raised in Los Angeles,
California, "Les" Deutsch spent much of his early professional life developing electron-
ic music technology for his father's Deutsch Research Laboratories. After earning a doctorate
in mathematics horn the California Institute of Technology in 1980, Deutsch joined
JPL in the Communications Systems Research Section, the section originally led by
Walter K. Victor twenty years earlier. Deutsch became manager in 1986.

In 1989, he took charge of the Technology Development program in the Tracking and
Data Acquisition Ofice, and it was while there that he brought the considerable resources
of that group to focus on the Galileo problem. The full, scope of the research carried
out under that program is discussed separately in chapter 6.

A pleasant man of medium build, Deutsch always appeared to be busy. He was as much
at ease discussing an abstruse mathematical representation of some physical problem in
the DSN as he was executing a complex m u s i d phrase on the pipe organ or tearing
off a tuba riff in a jazz band.

In an interesting sidelight to his involvement with Galileo, Deutsch was invited to accom-
pany a small party from NASA and JPL to Padua, Italy, in 1997 to participate in a
"Three Galileos" conference. Sponsored jointly by the University of Padua, the German
and Italian space agencies, and NASA, the conference honored "Galileo, the Man,"
"Galileo, the Telescope," and "Galileo, the Spacecraft." Deutsch contributed a technical
paper50 on the Galileo telemetry recovery effort. He was also honored with an invita-
tion to present an organ recital in the Basilica of St. Anthony at Padua that featured
music of Galilee's time. Dr. Deutsch continued to manage the program of technology
development for the DSN until 1998, when he transferred to another area of JPL to
study new technologies for planetary spacecraft.

Deutsch began by establishing a set of assumptions on which the team would base its
considerations. Because of the HGA problem, all telemetry was being transmitted fiom
the spacecraft on an S-band carrier through the LGA. The design of the spacecrah pre-
cluded the transmission of X-band telemetry through the LGA. If nothing was done to
improve the S-band downlink performance by the time the spacecraft arrived at Jupiter
in December 1995, the telemetry data rate would be only 10 bits per second, compared
to the rate of 134,000 bits per second that had been designed into the original X-band
mission.

The question facing the team was, "What could be done by making changes to the
DSN or the spacecrafi, or both, to improve this situation in the time available, and how
much would it cost?"

Because the period of the study coincided with the GaliIeo Encounter with asteroid
Gaspra, technical participation kom the spacecraft and mission design areas was limit-
ed. Deutsch presented the team's report on 5 November 1991.

Out of the eight options that were identified, the following four were recommended for
further evaluation:

Arraying Antennas

Increase the effective aperture area of the DSN antennas by building new ones or
arraying existing DSN antennas with large, non-DSN antennas such as were then
in operation in Australia, Japan, Germany, and Russia. In addition, the S-band per-
formance of the existing 70-m antennas could be improved by installing special
low-noise feeds called "ultra-cones."

Compression

Use data compression techniques on the spacecrafi to increase the effective telemeuy
rate by reducing the number of data bits that needed to be transmitted for each image.

Coding

Improve downlink capability at low data rates by adding more advanced coding to
the existing coded telemetry data stream. This wodd require inflight reprogram-
ming of the spacecrafi computers.

Modulation

Increase the effective power received by the ground antennas by changing the mod-
ulation index at the spacecraft transmitter to fully suppress the S-band carrier. This
 would have the effect of putting all the transmitted power into the subcarrier side-
bands that carried the information content of the downlink.

Estimates of the performance gain, cost, and uncertainty associated with each option
were made and evaluated. The study concluded by emphasizing the urgent need for the
TDA and the Flight Project Ofice (FPO) to jointly develop a workable plan to imple-
ment the necessary changes in the spacecraft and in the DSN in the limited time available.

The positive results of the largely TDA-driven, thirty-day Galileo Options study, along
with its emphasis on the need for timely action, led to the chartering of a firther study
which would involve both TDA and the Flight Projects Offlce (FPO). It was to be
called the Galileo S-band Mission Study and would run from 9 December 1991 through
2 March 1992. The co-leaders of the study group would be Leslie J. Deutsch, repre-
senting TDA, and John C. Marr, Manager of JPfi Flight Command and Data Management
Systems Section, representing FPO.

The objectives of that study were to

assimilate and verify the information from the previous TDA Galileo Options Study;

solicit additional ideas and assess their performance benefit and feasibility;

create a conceptual design for the end-to-end telemetry system for the Galileo S-band
mission;

generate a rough cost estimate for this design for submission to the program oEces
at NASA Headquarters which were responsible for hnding TDA (Code 0 ) and FPO
(Code S) in time for the Fiscal Year 1994 budget cycle; and

make specific recommendations for the implementation of an end-to-end data system

and hand these over to the organizations that would perform the work.

Whereas the TDA study had addressed only the high-rate science telemetry downlink,
this study included real-time, low-rate science and spacecraft engineering data, uplink
command and radio metric data, and navigation data. It also stipulated that no improve-
ments to the downlink would be made until the important Probe science data had been
successfdly returned to E d using the existing downlink capability.
The basic conclusion of the study was that a viable Galileo S-band mission was indeed
feasible, based on a design that could meet a somewhat reduced, but still very accepr-
able, set of science objectives. Such a design could be implemented in time for the start
of the Galiieo orbital mission, and the total cost to NASA would be about 75 million
dollars spread over seven years.

The results of the study were presented at NASA Headquarters on 20 March 1992 by
Deutsch and Marr, supported by the Galileo Project Manager, William J. O'Neil, and
the Galileo Project Scientist, Torrence V. John~on.~'

O'Neil came straight to the point. There was no way of knowing whether the HGA
would eventually be deployed, he said, and NASA should therefore prepare to complete
the Galileo mission with the spacecraft LGA. He said that a key design feature of the
new LGA mission would limit the data return from the tape recorder to one full tape
recorder load for each of the ten targeted satellite encounters. That feature, coupled with
the proposed enhancement of the downlink by additions to the spacecraft and DSN,
gave reason for great optimism that a worthwhile scientific endeavor could still be suc-
cessfully accomplished.

In addressing the mission science objectives, Johnson demonstrated that the proposed
LGA mission would, in effect, yield 70 percent of the science return expected from
the original HGA mission. That new science, he said, would represent a major advance
beyond the current level of Jupiter knowledge based on the results of the Voyager
mission.

Marr described the reprogramming changes that would have to be made in the spacecraft
computers to add the new data compression and encoding algorithms; he also described the
costs, risks, and implication of these changes in data format on Galileo mission operations.

The four options recommended for enhancement of the downlink by the DSN were
essentially a refined version of those that had originated in Deutsch's original study.
Suppressed carrier modulation would yield 3.3 dB improvement; advanced coding, 1.7
dB; ultracones for 70-rn antennas, 1.6 dB; and antenna arraying, 0-4 dB depending on
which antennas were used. Together with the spacecrafi modifications, these enhance-
ments would provide a capability in the DSN to return one full load of tape-recorded
data after each satellite encounter. They also satisfied the project requirements for receiv-
ing continuous engineering data and low-rate science data. The options for arraying
antennas were not limited in this presentation to just the DSN. They included the cost
of adding the large antennas at Parkes (Australia), Bonn (Germany), and Usuda (Japan)
to the DSN arrays.

Following the presentations, action items to study h e e alternative approaches were assigned,
and the JPL team returned to Pasadena to prepare for a final decision meeting set for
10 April 1991.

The three alternative approaches related to when and how the various arraying options
would be implemented in the DSN. This decision largely affected h d i n g profile issues
that were of paramount importance in the NASA Headquarters view of the LGA mission.

The final meeting was attended by the newly appointed JPL Director, Edward C.
Stone, in addition to the NASA Associate Administrators for Space Communications
and Space Science.

While the previous meeting emphasized the technical aspects of the LGA mission and
addressed the costs in a rather general way, this meeting concentrated specifically on
costs and the time rate of expenditure that would be incurred by the three alternative
approaches previously identified. Since support of the LGA mission would require so
much 70-m antenna time at the expense of other ongoing and future missions, the issue
of DSN loading was also included in the DSN presentation.

The major cost drivers for the DSN were known to be the decoder and decompressor,
the ultracones for the 70-m antennas, the signal combiners, communication links between
antenna sites, and the cost of renting time on whichever of the non-DSN antennas
would be chosen for arraying. Except for the latter, each of the costs was well under-
stood, easily identifiable, and bounded. The arraying costs were not.

At a breakfast meeting on the morning of the presentations, the new JPL director expressed
concern over the uncertainty associated with arraying costs and the rather overwhelm-
ing number of options available to choose from. His concern prompted Deutsch to
propose a last-minute option that he had budgeted out during an impromptu lunchtime
session in a vacant office at NASA Headquarters. The "Deutsch" proposal limited all
arraying operations to the Canberra site. The Australian antenna at Parkes would be the
only non-DSN antenna used for arraying, and DSS 43 would be the only antenna fit-
ted with an ultracone. This approach offered a more manageable proposal on which to
base a decision and was included as part of the total DSN presentation. The totd DSN-
related cost was estimated to be 38.3 million dollars spread over the six fiscal years 1992
through 1997, with the major cost impact in 1994. The spacecrafi and mission opera-
/ -I

OTHER
APERTURES

Figure 5-31. The concept of DSN telemetry for the Galileo LGA mission.

tions costs were 31.8 million dollars over the same period. This strategy was successful
and was accepted as the way to recover the science data from Jupiter. Taken together,
the spacecraft and ground system proposals would provide 70 percent of the original
science data value at 1 percent of the original data rate for an overall cost increase of
less than 5 percent.

The Galileo LGA mission as eventually approved by NASA satisfied most of the proj-
ect requirements for real-time data and playback data return. It also provided for the
new spacecraft capability but deIayed the DSN enhancements until late 1996. It includ-
ed DSN arraying centered on Canberra, as proposed by Deutsch. Software "hooks" were
to be inserted in the design for the possible later inclusion of arraying with Parkes or
Usuda, if funds could be identified.

The conceptual arrangement of telemetry proposed for DSN support of the Galileo LGA
Mission is depicted in Figure 5-31.

Three types of receive antennas were to be used. These included the DSN 70-m anten-
nas and the DSN 34-m antennas, with the non-DSN antenna at Parkes, Australia, as
an option. Each antenna was to be equipped with a low-noise amplifier (LNA). The
DSN 70-m antennas and the Parkes antenna would be equipped with ultracones and
low-noise maser amplifiers, the others with cooled High Electron Mobility Transistor
(HEMPT) amplifiers.
The LNAs were to be followed by receivers that would translate the incoming S-band
signals to a lower frequency, filter out the uninteresting parts of the bandwidth, and
digitize the resulting signal. The digital signals would be recorded on tape and sent to
a central site for combining with the signals from the other antennas. There was also
an option to send the digitized signal to the central site in real time.

At the central site, the signals from each antenna would be aligned with one another
and combined (summed). The alignment would be accomplished by looking for the sig-
nature of the packet synchronization markers in the signals. The summed signal would
then be demodulated. During much of the mission, there would not be enough signal
strength to perform any demodulation before this summation. In some cases, data could
be lost in the aligning process. When this happened, post-processing of the tape-record-
ed signals would be used to recover the data.

FoIlowing demodulation, the composite signal would be decoded in two stages. First,
the effective (14, 114) convolutional code would be decoded. Then the Reed-Solomon
code would be decoded, Finally, the effects of the two spacecraft compression algorithms
would be undone in two decompressors. Subsequent processing of the science data would
be performed by the Galileo Mission operations organization and the Science Teams.

From a technical point of view, the Galileo LGA Mission represented a remarkable
episode of conceptual design in the areas of spacecraft data systems, planetary mission
design, and DSN downlink engineering. However, before these concepts could become
a reality, they had to be h d e d and implemented. While the NASA Headquarters peo-
ple went off to wrestle with the h d i n g problem on the short time scale required, the
JPL people returned to Pasadena to initiate the implementation tasks.

Although we are only concerned here with the DSN effort, it needs to be said for the
record that this evolution was an intensively interactive process. A satisfactory end result
could only have been achieved by keeping all three elements-spacecraft, DSN, and
mission operations-focused on the end-to-end performance of the downlink, while the
implementation tasks proceeded as separate entities.

The task of implementing the DSN portion of the GaliIeo LGA mission was assigned to
Joseph I. Statrnan, whose previous experience with developing the Big Viterbi Decoder for
the original Galileo mission in 1989 was reported earlier. h Ask Implementation Manager,
Staunan became responsible for transforming the conceptual ideas into hardware and sofi-
ware and installing and testing this equipment at the three DSN sites around the globe.52
This task was to be accomplished within the cost estimates that had been presented to (and
accepted by) NASA, and completed in time to support multiple Galileo Encounters with
the satellites of Jupiter in less than forty-eight months. Although most of the technology
was available, none of it had found practical application in this way or on this scale, before.
Some arraying technology had been used in the earlier Voyager encounters with Uranus
and Neptune, and better techniques had been the subject of continuous study since then.53
It had, however, been centered at Goldstone and had not involved intercontinental array-
ing between Goldstone and Canberra. The application of coding to enhance the performance
of telemetry downlinks was well established in the DSN, and an advanced version of con-
catenated, convolutional coding had been added to the Galileo spacecrafi downlink prior
to launch.54This was intended for use at very high data rates (134 kbps) during the 10
Encounter on the original X-band mission. Because it could not operate at the low data
rates (10-100 bps) which would be typical of the LGA mission, it was of no M e r use.
Nevertheless, the studies that had gone into developing that coding system were to be
extended and applied in a new way to the LGA mission.55

Under the guidance of Statman, a comprehensive set of system and subsystem require-
ments and a cost breakdown for the DSN Galileo Telemetry (DGT) Subsystem, as it
came to be called, was prepared and presented to a DSN Review Board for approval on
3 December 1992. Five days later, GaZileo made its second and final close flyby of Earth
and was on a direct path to Jupiter, just three years away. There was no time to lose.

In the course of transition fiom conceptual form to physical form, the DGT had been
somewhat modified. Decompression of the received data function had become a Galileo
mission operations function, and the arraying h c t i o n had been limited to the one 70-
m and three 34-m DSN antennas at Canberra. The combiner at CDSCC, however, was
to make provision for combining the signals fiom up to seven antennas. This would per-
mit a digitized signal fiom the DSN 70-m antenna at Goldstone to be added to the
Canberra array to create real-time intercontinental arraying during Goldstone-Canberra
overlapping view periods. The use of non-DSN antennas was not considered in this review.

The form in which the DGT was implemented at CDSCC is shown in Figure 5-32.

Although it was not considered part of the DGT, the Block V Receiver (BVR), shown
in Figure 5-32, was an integral part of the DSN configuration for Galileo. Its primary
purpose was to track the M y suppressed carrier downlink from a single 70-m antenna
to extract telemetry and the two-way Doppler data essential for radiometric-based nav-
igation of the GaZileo spacecraft. When multiple-antenna arraying was used, the BVR
provided Doppler data only and the DGT provided telemetry as described above. Its
implementation proceeded in parallel with, but a year ahead of, the DGT.
Figure 5-32. Functional diagram of the DSN Galileo Telemetry Subsystem for the Canberra Deep Space CommunicationsComplex.
The S-band signals from each of the antennas were first stepped down by 300 MHz to feed the Full Spectrum Recorders (FSR). After
recording, the separate digitized signals were summed in a Full Spectrum Combiner (FSC) and delivered to a software receiver, where
the detection and demodulation functions were carried out in a Buffered Telemetry Demodulator (BTD).= 7he complex decoding opera-
tion on the inner and outer codes was performed by a special Feedback Concatenated Decoder (FCD)- that delivered a decoded data
stream to the ground communications equipment for formatting and transmission to JPL via the GCF. A system controller performed all
the configuration and data management tasks via several local area networks. The arrangement at Goldstone was similar, except that
there was no ultracone and no Full Spectrum Recorder. At Madrid, the arrangement was simpler stifl, consisting only of the decoder
and system controller.
Repeated tests with the Galileo downlink in early 1995 confirmed the BVR perform-
ance under fi.8 operational conditions and it was accepted for future GaliIeo S-band
mission support.

By May 1995, BVRs had been installed at dl Complexes and soon demonstrated their
abiliry to track the Galileo downlink in the suppressed carrier mode, deliver telemetry,
and extract Doppler data.

In September 1995, based on these impressive results, the Galileo downlink was final-
ly switched to the suppressed carrier mode that, supported with the BVR, would become
the standard mode of operation for the rest of the mission. Less than three months later,
Galileo arrived at Jupiter, relying on the BVR to track its Iow-powered S-band down-
link and recover the most important data of the entire mission, the Probe entry science
data. Over the next several months, the BVR lived up to everyone's expectations, and
in recovering all of the Probe playback data, it allowed the Galileo project to accom-
plish its primary mission objective.

The tests conducted at Goldstone in early 1994 to demonstrate the BVR had also been
used to demonstrate the proposed full-spectrum recording and fdl spectrum combining
techniques58using prototypes of each of these elements of the DGT The Gdileo down-
link signals fiom DSS 14 and DSS 15 were recorded simultaneously by the prototype
FSR and were later combined with the prototype FSCe5'The results were well within
the predicted performance limits. Later experiments would verify similar performance in
combining the downlinks from the intercontinental array of DSS 43 with DSS 14.

Over the course of the following two years, the operational versions of these units were
built, installed, and tested at the Complexes. Since arraying was to be used only at Canberra,
that was the only site where a fd-spectrum combiner was installed. Also during this peri-
od, an ultra-low-noise, receive-only feed system (ultra-cone) was added to the Canberra
70-m antenna to reduce the S-band system noise temperature hom its normal value of
15.6 kelvin (K) to 12.5 K By reducing still fiuther the noise on the downlink, this addi-
tion enhanced the performance of the downlink and increased the rate at which it could
return data from the spacecraft. The DGT was rapidly taking real form and substance.

Although the DSN had received approval (and h d i n g ) fiom NASA Headquarters for
the use of multiple-antenna arraying techniques to support the Galileo LGA mission, this
support was limited to DSN antennas only. As design studies matured, however, it became
evident that, even with the aIready-approved downlink enhancements in the DSN, the
Galileo science data return would be especialty low during the six-month period when the
spa~ecraftwas firthest from Earth, November 1996 through April 1997. This was the peri-
od when four of the ten satellite encounters would occur. Galileo and DSN concern with
this situation in March 1994 led the TDA Executive Committee to recommend that the
DSN array configuration planned for Canberra be augmented with the 64-m antenna of
the Australian radio telescope at Parkes. The cost of making the necessary modifications
to the DSN, making additions to the Parkes antenna, and renting time on this antenna
was estimated at 4.3 million dollars. This would be met by funds transferred &om the
Galileo project. The estimated enhancement in data return was about 10 percent.

NASA approval was q d y followed by an Australian agreement to participate in the Galjleo

LGA mission, and the DSN began work on the changes needed to add Parkes to the Canberra
array. Working arrangements with Parkes were quickly reestablished based on the agreements
that had been made for Parkes support of the Voyager mission a few years earlier.

The DGT at Canberra supported the 70-m antenna, plus three 34-m antennas at CDSCC,
plus the 64-m antenna at Parkes, in addition to the 70-m antenna at Goldstone when-
ever overlapping view periods permitted.

While DSN engineers had been focused on implementing the BVR and the DGT in
the Network, the Galileo project focused on the changes that were required in the space-
craft: to make the end-to-end downlink work. At the same time, the project had to carry
the mission forward to the point at which the new flight software could be transmitted
ro the spacecraft. This point was reached on 13 May 1996, when the DSN began radi-
ating a continuous series of commands to load the spacecraft computers with the new
Phase II flight sohare. The continuous sequence was completed without incident nine
days later on 22 May, and the spacecrafi was enabled to begin operating in the new
mode the following morning.

By that time the DGT, including the FCD decoder, was in place at all three Complexes
and ready to handle the new concatenated convolutional (14, 114) coded downlink at
all bit rates in the suppressed carrier mode.

The Ganymede Encounter on 28 June 1996 was the first critical test of the new flight soft-
ware and the DGT ability to process it. Since it was not yet ready for 11I arraying, the DSN
used single 70-m antennas to return the Encounter data. As the tape recorder playback data
began to arrive over the new downlink for the first time, the Galileo project manager declared
the images returned from the Ganymede Encounter to be "absolutely simmhg" and a trib-
ute to everyone involved in meeting the challenges of the LGA mission.@The end-to-end
design of the new downlink had been proven in critical flight operations conditions, and the
initial optimism for a successful conclusion to the Galileo Mission was M y vindicated.

The Callisto Encounter in November 1996 was the first occasion on which multiple anten-
na intercontinental arraying at CDSCC was used in critical real-time operations. The ~LIU
DSN array capability, augmented with the Parkes antenna, allowed the GaIileo spacecraft
data rate to be raised fiom 40 bps to 120 bps during overIap periods. This Encounter,
too, was extremely successful and demonstrated the fLlll operational capacity of the DGT
Based on the performance of the DGT in supporting the Callisto Encounter, the ensu-
ing recorded data playback, and confidence in the DSN ability to deal with the increased
operational complexity of 111 array operations, the DGT and associated array mode was
designated as the standard configuration for all subsequent Galileo operations in the DSN.

The operational complexities and remarkable downlink perfbrmance resulting &om this arrange-
ment throughout the rest of the Galileo mission were discussed earlier in this chapter.

The weekly edition of "Significant Events in the DSN" for 29 August 1997 reported,
"After successllly completing an extended series of mission verification tests on the X-
band uplink and downlink capability, the 34-m beam waveguide antenna in Canberra
(DSS 34), began operational tracking support this week. This antenna will now be avail-
able to provide support [to] such missions as NEAR, Mars Global Surveyor, Mars Pathfinder,
and Cassini, in addition to continuing support for the Galileo array." It further report-
ed, "System performance testing began on schedule this week on the new 34-m beam
waveguide antenna in Madrid (DSS 54). This will be followed by mission verification
testing prior to the antenna being scheduled for operational tracking support on 1 October
1997." And so they were.

Together with DSS 24 at Goldstone, which had been completed in February of 1995,
the completion of DSS 34 and DSS 54 gave the DSN a full subnet of three 34-m beam
waveguide (BWG) antennas, which represented a most significant increase in uplink and
downlink capability.

Based on the successful design of the existing 34-m high-efficiency (HEF) antennas, the
new BWG antennas now extended the existing DSN uplink and downlink capabilities
on S-band and X-land to include Ka-band (32 GHz).
The design of the 34-m BWG antennas embodied the most advanced principles of anten-
na and microwave design. It also represented the culmination of nearly ten years of DSN
engineering, research, and development, which began with a JPL report on Ka-band down-
link capability for Deep Space Communications by Joel G. Smith in December 1986.61

In advocating the extension of DSN downlink, and ultimately uplink, capability to Ka-
band, Smith argued that, just as the move from L-band (0.96 GHz) to S-band (2.3
GHz) had offered a possible increase in performance by a factor of 5.74 (7.6 dB), and
the move from S-band to X band (8.4 GHz) a further increase of 13.5 (1 1.3 dB), the
move fiom X-band to Ka-band (32 GHz) offered a potential increase in performance
of 11.6 dB. These improvements were based on the theory that the capacity of a radio
link between two well-aimed antennas is roughly proportional to the square of the oper-
ating frequency. In practice, the actual realizable gain is reduced by effects in the propagating
media and losses in the various physical components of the link.

However, these latter effects could be controlled more easily at higher frequencies where
microwave components, particularly the surface area of the antenna itself, could be small-
er, stiffer, and mechanically more precise. Furthermore, measurements had shown that
degradation of downlink reception at Ka-band, due to weather effects, was not as severe
as had previously been thought.

Based on these considerations, Smith proposed a course of action which would culmi-
nate in an operational Ka-band downlink capability in the DSN by 1995. It began with
the development of a new research and development antenna at Goldstone to "verify
the various approaches to be used in upgrading the existing 34-m and 64-mI7O-m anten-
nas to good efficiency at Ka-band."

As Smith pointed out, there were ocher reasons, independent of the Ka-band decision,
for building a new research and development antenna at Goldstone, but there was a
strong synergism between the two.

In a companion paper,G2Smith laid out all of the arguments for a new antenna, includ-
ing those related to Ka-band, and proposed a specific design based on the existing HEF
antennas but modified to incorporate new beam waveguide technology for the transport
of microwave energy at low level (receiving) or high level (transmitting) between the
antenna and the receiving or transmitting devices.

At that time, 1986, the beam waveguide concept had been around for several years, but
it had not been employed in the design of any DSN antennas. It had, however, been
used recently in the design of several major non-DSN antennas, most notably the 64-
m antenna at Usuda, Japan, and the 45-m radio astronomy antenna at Nobeyama, Japan.
R C. Clauss had made an evaluation of these installations and became a strong advo-
cate for the new technology.63

The supporting arguments included significant simplification in the design of high-power

water-cooled transmitters and low-noise cryogenic amplifiers, improved accessibility for
maintenance and adjustment, and avoidance of performance degradation associated with
the accumulation of rain and moisture on the feedhorn cover. Beam waveguide tech-
nology allowed these systems to be located in a fied, nonrotating area beneath the
antenna azimuth bearing. As a result, the long bundles of power and signaI cables and
cryogenic gas lines that were required on the existing antennas could be replaced with
short, f ~ e d ,nonflexing connections. Better frequency stability, higher reliability, and
improved performance could be expected.

A beam waveguide system is shown in conceptual form in Figure 5-33.

The flat reflectors were used to redirect the beam; the curved reflectors were used to refo-
cus the beam. As the antenna rotated in elevation, or azimuth, the transmitting or receiving
beam of microwave energy was constrained to the beam waveguide path by the action of
the reflecting and refocusing surfaces. The various reflectors were enclosed by a large pro-
tective tube or shroud along the beam waveguide path. Although this tube was a highly
visible element of a BWG antenna, it played no sigdicant part in the beam waveguide
transmission process.

Over the next two years, 1989 and 1990, a design for a new BWG research and devel-
opment antenna at DSS 13 was completed; funding was obtained; and construction of
the new antenna ~ o m m e n c e d It
. ~ would be located about 300 feet south of the exist-
ing 26-m antenna at the Venus site at Goldstone, and it was scheduled for completion
in mid-1990. It was intended to serve as the prototype for a whole new generation of
DSN antennas using the new BWG-type of antenna feed system.

The mechanical and microwave optical components in the BWG path are visible in the
diagram of the completed DSS 13 BWG antenna shown in Figure 5-34.

From July 1990 through January 1991, the new BWG antenna was tested as part of its
post-construction performance evaluation. Antenna efficiency (gain) and pointing per-
formance measurements at X-band and Ka-band were carried out using unique portable
test packages installed ar either of the two focal points, F1 and F3. Celestial radio sources
 \

PMNE REFLECTOR 4

MAIN REFLECTOR

REFLECTOR 1
- REFLECTOR 2
(CURVED OR FIAT)

FEED NETWORK

\ J

Figure 5-33. Beam waveguide antenna in conceptual form. In this diagram, Reflectors 4 and
3, mounted along the elevation axis of rotation, brought the beam from the subreflector on the
main antenna down to the alidade structure. There, Reflectors 2 and 1, along the vertical axis,
brought the beam to a stationary equipment room below the alidade that housed the transmitting
and receiving equipment.

of known flux density were used in the calibration process. The values of peak efficiency
of 72.38 and 44.89 percent, at X-band and Ka-band respectively, measured at the beam
waveguide focus, met the functional requirements for antenna performance and agreed
well with the predicted design values.65Further testing followed, as frequency stability,
noise temperature, and the G/T ratio were evaluated and optimized.

Because system noise temperature is a critical parameter in deep space communication

systems, the system noise temperature of the new antenna was of special concern. In
the early tests, the system noise temperature was found to be higher than expected due
to the spill-over losses of the BWG mirrors having a greater effect than previously
 SHAPED-SURFACE
SUBREFLECTOR --\

TRIPOD 3 4 4 SHAPED-
SURFACE
REFLECTOR

REAR VlEW

Figure 5-34. DSS 13 beam waveguide antenna.

New feedhorns were designed and tested and new methods for determining noise tem-
perature in beam waveguide systems were de~eloped.~'

While the construction and testing of the prototype antenna at DSS 13 was in progress,
submissions for funding for a f l y operational version of a BWG ar Goldstone (DSS
24) were being made. At this time, the Galileo mission to Jupiter was progressing nor-
mally and the high-gain antenna had not been deployed. The planned close flyby of 10
in December 1995 over Goldstone demanded the full X-band support of the 70-m and
HEF antennas at Goldstone, leaving little or no resources for other flight projects. Using
this as justification, further funding for two more BWG antennas was requested from
NASA Headquarters and, at the expense of some institutional compromises, was award-
ed. Now there was funding for a duster of three BWG antennas at Goldstone, DSS 24,
DSS 25, and DSS 26. These were placed on contract with an American construction
 SIDE ELEVATION
J
Figure 5-35. Design for operational 34-meter beam waveguide antenna. In the operational ver-
sion of the 34-m beam waveguide antenna, the alternative bypass microwave system was eliminated
and the alidade structure was strengthened to add stiffness and to accommodate the additional
drive and ancillary services.

company, TW Systems, h c . All three antennas would be located in the Apollo Valley
at Goidstone, and work began immediately.
By the time construction began, the original design for the DSS 24 antenna had been
modified to incorporate many of the lessons learned from the DSS 13 experience and the
intensive program of pedormance measurements that had been conducted on that anten-
na. In particular, the antenna structure had been modified to improve its "gravity pedorrnance"
by controlling the antenna distortion for best fit to the ideal parabolic shape, as the anten-
na moved through the full range of elevation angles. Other improvements related to the
azimuth cable wrap-up and to the substitution of nomechanical "RF Chokey' bearings
for the mechanical tube bearings used at the rotating junctions of the waveguide shroud.
Also, the four luge (30-inch diameter) wheels which supported the alidade structure on
the azimuth track had been "surface hardened" to improve wearability, based on opera-
tional experience with the existing HEF antennas. After a short period of use, these wheels
cracked and had to be replaced with "through hardened wheels to stand up to the loads
and wear induced by continuous use under operational conditions. The design features of
the operational version of the DSN 34-m BWG antenna are illustrated in Figure 5-35.

Construction of DSS 24 began in February 1992 with the blasting of a large hole, 30
feet deep and 60 feet in diameter, into solid bedrock to accommodate the pedestal room
with its 24-fi- high ceiling. Because this was the first antenna to be built, it revealed
the presence of several unsuspected problems in design and planning and, acting some-
what as a pathfinder for the others, was not completed until May 1994.

Microwave performance testing of the new antenna was carried out during the summer
of 1994 using the same techniques &at had been employed at DSS 13. Antenna effi-
ciency, pointing calibrations, and system noise temperature were all included at S-band,
X-band, and Ka-band.G8 In addition, microwave holography was used to adjust the reflec-
tor panels on the main antenna for optimum eff1.ciency.

The values for gain and efficiency obtained from these tests are tabulated below:

DSS 24 Peak Gains and Efficiencies

Band, Frequency (GHz)

S, 2.295 56.79 71.50
X, 8.45 68.09 71.10
Ka, 32.00 78.70 57.02

The results of these measurements confirmed the soundness of the basic design and the
improvements that had accrued from the modifications mentioned above.
Experience pined from the DSS 24 task enabled the remaining two antennas to be
built in a parallel fishion to save time. They were completed in July and August 1996,
respectively, about 18 months after construction started in 1994.

Unlike DSS 13, which was built for deep space communications research and develop-
ment purposes, the other BWG antennas were intended for operational use in the Network.
Consequently, after the construction phase was completed, they were subjected to an
intensive program of testing under operational conditions before they were declared oper-
ational and accepted for tracking flight spacecraft as an element in an array or in a
stand-alone mode. The "operational" dates for all the BWG antennas are given in the
table below, and they generally followed the completion of the antennas by several months.

DSN plans for BWG antenna construction in Spain and Australia were significantly
a c t e d by the dramatic events, described earlier, that occurred on the GuZiZeo spacecrafi
in April 1991. To save the mission, DSN and Galileo engineers had proposed an alter-
native mission using the spacecraft's S-band, Low-Gain Antennas and an array of S-band
antennas in Australia. This situation provided the DSN with the high-profile justifica-
tion needed for adding a fourth BWG antenna at Canberra (DSS 34). With design and
specification details already available from the Goldstone task, hnding for construction
of this antenna was quickly approved.

International bidding was opened for the erection of a fourth BWG antenna in Australia
with an option for two more when funding became avaiIable. (It was recognized that
significant savings could be realized by building the antennas in groups of three). The
contract was won by the Spanish firm Schwartz-Hautmont, and construction began at
the Canberra site in July 1994.

Initially, problems with water percolation in the large excavation for the pedestal room,
along with difficulties with the contract and Australian unions, caused significant delay
to the construction schedule. Eventually these problems were dealt with, work resumed,
and the antenna was completed in November 1996, very close to the original sched-
ule. M e r a short period of operational testing, the antenna was accepted for operational
support and immediately began supporting Galileo on S-band as part of rhe Canberra
array. Later, the X-band uplink was added, and after a period of mission verification
testing, it began operational tracking support with X-band uplink and downlink in
August 1997, as mentioned above. A photograph of the completed antenna is shown
in Figure 5-36.
Figure 5-36. DSS 34 BWG antenna at the Canberra Deep Space Communications Complex,
Australia, 1996.

Two years after the Galileo high-gain antenna problem changed the course of BWG his-
tory in the DSN, another spacecraft event occurred which once again changed the course
of the program.

In August 1993, eleven months afier launch and three days before insertion into Mars
orbit, communication with the Mars Obsmer spacecraft was lost and never recovered. In
the aftermath of this loss, NASA hasrily proposed and fimded a new mission to Mars to
be called Mars Glabal Surveyor (MGS). Its arrival at Mars toward the end of 1997 would
be just in time to create a conflict for X-band uplink support over Madrid with the
Casini spacecraft scheduled for launch in October of that year. Once again there was a
strong programmatic need for an additional BWG antenna in the Network, this time at
the Madrid site. But this time, there was no funding immediately available. Fortunately,
the DSN was able to reprogram sufficienr money saved from the contracts for the first
and second groups of three antennas to fund the option for the fifth BWG antenna at
Madrid. The contractor, Schwartz-Hautmont, began construction in June 1995. Despite
a water problem similar to the one that had occurred during construction in Australia,
the antenna was completed by August 1997. System and mission verification tests fol-
lowed, and the antenna was accepted for operations support in October 1997.

To a large degree, the disposition of S-band and X-band capability among the different
BWG antennas was driven by the hnding available for procurement of the necessary
electronics packages. Although all of the antennas were capable of operation at Ka-band,
the use of Ka-band for spacecraft telecommunications links, except for Cassini, had not
matured beyond the need for a demonstration capability at one BWG antenna by the
end of 1997. The need for a full subnet capability was expected to develop in the years
ahead. A &-band downlink capability existed only at DSS 25.

h late 1994, two 34-m BWG antennas that had formerly belonged to the U.S. Army
and were situated near the Venus site at Goldstone were transferred as surplus property to
JPL. On transfer to the DSN, they became identified as DSS 27 and DSS 28 and were
designated as High Speed Beam Waveguide (HSB) type. These antennas were azimuthlele-
vation mounted and, before they could be put into service, had to be stiffened and
strengthened to meet DSN standards for reliability and stability under conditions of con-
tinuous use. DSS 27 was equipped to provide backup for the one remaining 26-m at
Goldstone, DSS 16, and to supplement DSN capability for high-Earth orbiter missions
like Infrared Space Observatory (ISO) and Solar Heliospheric Observatory (SOHO). DSS
27 thus became a 34-m BWG that had the performance of a 26-m S-band antenna with
one multihction receiver and a low-power, 200-W transmitter. Like DSS 16, it was oper-
ated remotely from the SPC located some ten miles away from the antenna site. Activation
of DSS 28 was postponed for later, when &ding for suitable electronics would become
available. The photograph in Figure 5-37 shows a cluster of three 34-m BWG antennas
at Goldstone in 1995.

The times at which the various BWG antennas came into operational service in the
Network and their uplink and downlink operating bands as of December 1997 are sum-
marized in the table below.
Figure 5-37. Cluster of three 34-m BWG antennas at Goldstone, 1995.

DSN 34-Meter Beam Waveguide Antennas

DSS In Service uplinks Downlinks

13* July 1990 s, XY Ka S, X, Ka
24 Feb. 1995 S s, x
25 Aug. 1996 X x, Ka
26 Aug. 1996 X X
27 July 1995 S s
28 Oct. 2000 NIA NIA
34 Nov. 1996 s, x s, x
54 Oct. 1997 s, x s, x
* DSS 13 for research and development use only.
NtA: Electronics not available as of January 1998.
With the completion of DSS 54 in 1997, the DSN finally had a complete subnetwork
of three 34-m BWG antennas, all of which were capable of operation at S, X, and Ka-
band. By then, the trend for & m e missions was towards shorter passes and higher data
rates, a natural application for Ka-band uplinks and downlinks, as Joel G. Smith point-
ed out in the 1988 papers described earlier. In this context, the DSN was well positioned
to cope with mission requirements in the immediate hture. The possibility of a justifi-
cation for more antennas seemed unlikely, and in terms of antennas, the DSN had
reached a plateau that appeared to extend well into the foreseeable future.
SIGNAL PROCESSING CENTER UPGRADE TASK

The progressive enhancement of uplink and downlink capabilities through the Mark
11, Mark 111, and Mark IVA models of the DSN has been traced in previous chap-
ters. By 1993, the major gods of the Mark IVA configuration had been achieved.
New antennas had been added to the DSN, Networks had been consolidated, and
the Signal Processing Centers (SPCs) had been integrated. Centralized control was
emerging as the established process for conducting mission operations at the SPCs.
Spacecraft data received at the Deep Space Communication Complexes (CDSCCs)
could now be returned to JPL at higher data rates and with fewer errors, thanks to
complementary improvements in the NOCC and GCF. Fewer errors meant less need
for lengthy data recalls and faster delivery of the final data records to the flight proj-
ects. Despite a few remaining functional and operational problems associated with
the introduction of these changes, the Nenvork was running smoothly and at max-
imum capacity.

In 1993, and for the remainder of the Galileo Era, implementation of new uplink and
downlink capability in the DSN proceeded more on the basis of individual tasks high-
ly focused on specific objectives, rather than as part of a generd overall enhancement
of the entire Network, as had been the case in the past. Three such tasks, the Galileo
telemetry, Beam Waveguide Antennas, and X-band uplinks, are examples of this change
in approach and have already been discussed.

While these tasks were in progress, however, a fourth major task involving changes and
additions at the subsystem level in most of the major systems of the DSN was being
carried out at the SPCs. It was called the SPC Data Systems Upgrade and covered the
period 1990 through 1993.

Signal Processing Centers Upgrade

The drivers for the SPC upgrade were the future Mars Observer and International Solar
Terrestrial Physics (ISTP) missions. Between them, these two hture missions placed
heavy demands for 26-m support, for X-band uplinks, and for additional telemetry
downlinks, which exceeded the capability of the existing DSN.
 - MONITOR

Figure 5-38. Signal Processing Center configuration, 1993.

There were four main objectives for this task:

1.Upgrade the telemetry system to support data rates up to 2.2 Mbps and provide Reed-
Solomon decoding capability at the SPCs.

2.Add the 26-m Earth orbiter antennas to the existing SPC capabiliry for centralized
monitor and contr~lof all Complex subsystems.

3. Increase the number of telemetry and command links to five per complex and expand
the capability to control them by adding two more Link Monitor and Control (LMC)
computers and consoles.

4. Replace the obsolete Modcomp If computers with new Modcomp 7845 machines,
which had significant growth potential.

The scope oithe SPC upgrade task is identified in the SPC configuration diagram shown
in Figure 5-38.
In the planning and execution of this task, great care was taken to avoid impacting the
SPC ability to continue full-scale mission operations throughout the entire period. This
constraint led to a phased implementation plan that began in 1990 and ended in 1993.
The SPC implementation plan was carefully integrated with similar and interlocking
plans for the GCF and the NOCC. However, other events soon overtook the original
plans, so by 1993, the original objectives had been extended. It was not until about
1995 that the task of upgrading the SPCs was completed.

SPC Telemetry

As part of the SPC data systems upgrade, the DSN Telemetry System, which began the
Galileo Era with four groups (strings) of data processing equipment, was expanded to
five strings, each capable of handling two data streams from a single spacecraft at select-
ed data ratesg9Each group was configured, monitored, and controlled by a Group Controller
that set the spacecraft data channels, monitored their performance, logged the data, and
formatted it for transmission to JPL via the GCF.

Groups 1 and 2 were composed of the original Mark WA assemblies, subcarrier demod-
ulator (SDA), Symbol Synchronizer (SSA), Maximum Likelihood Convolutional Decoder
(MCD), and Telemetry Processor (TPA). Their convolutional decoding capability was
limited to constraint lengths of 24 or 32 at rate 112 for sequential decoding at data
rates of 3 bps to 5 kbps, and a constraint length of 7 at rate 112 or 113 for data rates
10 bps to 134 kbps, for maximum likelihood decoding. Uncoded data in the range 2
bps to 15 kbps was processed by the TPA.

These capabilities were expanded in Groups 3 and 4 by the addition of new functions
and more advanced assemblies. For combining signals from up to eight receivers simul-
taneously, a baseband combiner assembly (BBA) was added. In addition to the combining
function, the BBA also performed subcarrier demodulation and symbol synchronization
and various combinations of dl three functions. A more advanced version of the MCD,
MCD I1 extended the data rate for maximum likelihood decoding to 2.2 Mbps, while
frame synchronization at the frame transfer level was performed by a new frame syn-
chronizer assembly (FSA). For telemetry data streams employing concatenated coding,
a Reed-Solomon decoder for data rates of 2 bps to 2.2 Mbps was provided. Control
and monitoring functions for each Group were performed by a new TPA, which also
extended the upper limit for uncoded data horn 15 kbps to 2.2 Mbps and carried out
sequential decoding functions on either channel in the range 3 bps to 5 kbps.
With the exception of baseband combining, the capabilities of Groups 3 and 4 were
replicated in Group 5.

Although the older Mark IVA assemblies in Groups 1 and 2 were eventually replaced
with equipment which was equivalent to that in Group 5, the former data rates and
formats were retained in these rwo Groups.

Over the next several years, improvements were made as the need arose. Responding to
requests from the DSCCs, a self-test capability was added to the Telemetry Group Controller
(TGC) and Telemetry Channel Assembly (TCA) s o b a r e in 1995 to reduce the time
required by the station operators to carry out pre-track readiness checks.

Following a long period of complex development and an exhaustive series of tests at

Goldstone, new Block 111 Maximum Likelihood Decoders (B3MCD) were added to the
telemetry subsystem at each Complex in 1997. The B3MCD could be selected as an
alternative to the standard MCD in one channel of each of telemetry Groups 3, 4, and
5. The B3MCD responded to the requirements of Mars Pathjnder and Cassini for max-
imum likelihood decoding of convolutional codes of constraint length 15 and rate 116,
in addition to the standard (K = 7, R = 112) decoding provided by the DSN. The addi-
tion of the B3MCD provided 1 dB to 2 dB improvement in bit signal-to-noise ratio
compared to the existing (7, 112) MCD.

The B3MCD represented not only a major increase in downlink capability, but also a
major accomplishment in new VLSI technology. It grew out of a 1988 concept by J. I.
Statman for a powerful new decoding device for the long constraint length (15), high-
rate (114) convolutional code to be employed on the Galileo 1989 X-band mission for
returning 10 Encounter science at a high data rate (134 kbps) from J~piter.~'The new
Big Viterbi Decoder (BVD) grew out of two contemporary developments in the DSN's
Advanced Systems program-the successful search for a Iong constraint length code with
a "2 dB additional coding gain,"71 and the burgeoning expertise in Very Large Scale
Integrated (VLSI) technology.72

Early progress was rapid, and by mid-1991 a prototype had been completed and run
successfully. Using experience gained from the prototype and improved VLSI chips, a
new "single board" decoder was being constructed for actual implementation in the
DSN. The new BVD would be programmed to decode convolutional codes having con-
straint length of 2 to 15 at rates of 112 to 116, and it would handle the highest Galileo
telemetry data rate of 134 kbp~.'~
 Baseband Inputs SPC bcal A r r Newark
fmm all (12)
Block V Receivers

Station

Pmcessor

I DSS Telemetry Subsystem Group, (1 of 8)

+
Data Return
to JPL

High Rate
Local Area Nehvork

Figure 5-39. DSN Telemetry Subsystem, 1997.

But the grear expectations that were held for the BVD would not be fulfilled. In the
wake of the Galileo HGA problem in 1991, further work on the BVD was suspended,
and Staunan was redirected to manage the DGT task in an effort to recover downlink
data from the Galileo S-band mission at data rates of 10 bps to 100 bps.

Later, the need for an improved decoder for the Telemetry System prompted the revival
of the BVD concept. However, it was an entirely new development task using even more
advanced VLSI technology that resulted in the design for Block 3 Decoders that were
eventually produced for implementation at all three complexes in 1997.

This brought the DSS Telemetry Subsystem to the generic configuration shown in func-
tional form in Figure 5-39.

The number of telemetry groups at each complex varied according to the number of
antennas to be supported, eight being the maximum, at Goldstone. A Baseband Patch
panel provided flexibility for switching or combining multiple inputs from up to twelve
Block V receivers. Each group contained the assemblies necessary for performing the
telemetry-related hnctions of baseband or symbol stream combining, demodulation, bit
and frame synchronization, and convolutional, Reed-Solomon, or sequential decoding.
Output data was delivered to a high-rate Local Area Network for transport to JPL via
the GCE To form a link for a specific spacecraft pass, the elements of any group could
be configured from a centralized control location via the SPC Locd Area Network.

By the time the DSN made the transition horn the Galileo Era to the Cassini Era in
1997, the DSN Telemetry System had reached the functional form in which it would
support the new missions of that era, Mars Global Surveyor, Mars Pathfinder, and Cassini.

SPC Command

The DSN Command System also was improved to meet increased demands for more
simultaneous uplinks with higher performance, as part of the SPC upgrade period of
the mid-nine tie^.'^ Not only were the number of independent command channels increased
from four to five, but switching arrangements were added to d o w station controllers
greater flexibility in switching any command channel to any exciter-transmitter combi-
nation. Command data rates and subcarrier frequencies were increased, and provision
was made for the command system to be operated directly from the NOCC or a remote
Mission Operations Center (MOC).

At each complex, each of the five groups in the command subsystem comprised a Command
Processor Assembly (CPA), Command Modulation Assembly (CMA), and Command
Switching Assembly (CSA).

The CPA contained s o h a r e that provided electrical interfaces with all other related sub-
systems in the Complex and with the NOCC and MOC. It accepted the serial,
pulse-code-modulated (PCM) command data stream from the MOC and delivered it
to the CMA for modulation on to a subcarrier.

The CAM generated the desired subcarrier in the frequency range of 100 Hz to 16 kHz
and modulated it with the command data received from the CPA. Phase Shifi Keyed
(PSIS) or Frequency Shifi Keyed (FSK) modulation was available at a command data
rate ranging from I bps to 2,000 bps.

The CSA connected the command-modulated output of the CMA to a designated exciter
for generating the S-band or X-band carrier, which was finally radiated by the trans-
mitter and assigned antenna. These functions are shown in Figure 5-40, which illustrates
the DSCC Command System in 1996.
r
b- DEEP SPACE STATIONS+
ANTENNA MECHANICAL
s SIGNAL PROCESSING CENTER --+ COMMU~dUnOMS
GROUND
4
3

AND ANTENNA
MICROWAVE SUBSYSTEMS

34-METER
HIGH EFF[CIENCY

RCIIWSWILL BE YIA GCF

26-METER INTERSEANALOG COM-
MUNICATIONSSUBSYSTEM

< 1

Figure 5-40. DSCC Command System, 1996.

In 1997, an additional 34-m antenna, the BWG, with a 4-kW X-band transmitter, was
added to each complex and integrated with the command system by means of the
Command Switch Matrix.

In the upgraded SPC, assignment of command groups to specific exciters, transmit-

ters, and antennas was carried out by a single cccornplex"operator. The resulting
"uplink" was then configured by a "link" operator to support a specific spacecraft in
the next scheduled tracking and commanding period. Afier the correct configuration
and operation of the link had been validated, it was turned over to the MOC for
use in transmitting the desired commands to the spacecraft. Because the DSN Command
System had the potential to cause unintended, undesirable, or irreversible actions in
a spacecraft, particular care was taken to ensure that the command data radiated by
the transmitter was identical to the command data input to the CPA. This assurance
was provided by verification and confirmation functions as an integral part of the
command system.

Afier the final installation of CPAICMA software in September 1995, each complex had
the capability to support five active command uplinks simultaneously, two of which
could be on X-band. Any uplink could be assigned to any of the active antennas on
the complex and could (after proper validation by the DSN) be operated for the pur-
pose of sending commands from a remote MOC (such as a NASA Center other than
JPL, or an agency other than NASA).

SPC Tracking

In addition to the telemetry and command subsystems discussed above, DSN plans for
upgrading the SPCs included rhe tracking subsystem. As a result of the Network consol-
idation program some years earlier, the 9-m and 26-m antennas, formerIy operated by
GSFC as p m of the Spaceflight Tracking and Data Network (STDN), had become a
DSN responsibility. DSN plans in 1990 called for them to be added to the DSCC Tracking
Subsystem and integrated into the SPCs at all complexes by May or June of 1995.75

In a fimctional sense, the Tracking Subsystem consisted of four basic groups of equip-
ment (hardware and software), one for each type of antenna-namely, 70-m, 34-m STD,
34-m HEF, and 9-ml26-m-where the 9-111126-m antennas were represented by a sin-
gle type. The principal application of the 70-ml34-m group was in tracking deep space
or high-Ear& orbiting spacecraft. The principal application of the 9-ml26-in group was
in initial acquisition of all spacecraft for which the DSN was to provide tracking and
data acquisition services, and in tracking low-Earth orbiting spacecraft.

AU 70-m134-m groups were identical and contained a Metric Data Assembly (MDA) and
a Sequential Ranging Assembly (SRA). Their primary functions were to generate radio met-
ric data consisting of Doppler, range, merenced range versus integrated Doppler (DRVID)
and uplink: tuning data. Performance monitoring and data validation were secondary h c -
tions. The primary functions were applied to S-band and X-band uplinks and downlinks
as determined by the capability of the antennas to which they were assigned. Each group
was permanently associated with a specif-ictransmitter-exciter-receiver combination and, like
all other subsystems in the SPC, could be configured and controlled fiom the centrally
located DSCC Monitor and Control (DMC) station via a Local Area Network

With the exception of the MDA, which was replaced by a Metric and Pointing Assembly
(MPA), rhe composition of the 9-ml26-m groups was similar to the 70-ml34-m groups.
Their primary functions, however, were very different. The main purpose of the MPA
was to point the 9-ml26-m antenna, a function which was carried out by a special
Antenna Pointing Subsystem on the other antennas.
Figure 5-41. DSCC Tracking Subsystem, 1995.

Using trajectory data unique to its specific site and spacecraft of interest supplied by an
external entity (generally GSFC), the MPA generated spacecraft signal acquisition pararn-
eters in the form of topocentric angles, Doppler frequency and range delay. These data
were used to point the antenna in various modes of tracking and to validate the Doppler
and ranging performance of special ranging equipment used on these antennas for sup-
port of older Earth-orbiting missions. The MPA also used angle and acquisition data to
generate predictions for any colocated antenna. This capability was frequently used to
point the 34-m antennas during the initial downlink signal acquisition sequence fol-
lowing a spacecraft launch, when the spacecraft trajectory was not known with sufficient
precision to point the narrow-beam 34-rn antennas. The MPA controlled and moni-
tored the SRA in a manner identical to the MDA.

The functions just described for all four types of antennas are shown in block diagram
form in Figure 5-4 1.
In this figure, each group of the DSCC Tracking Subsystem is represented by the designa-
tor, DTK. A fif& group was added in 1997 to accommodate h e new 34-m BWG antennas.

Implementation of the new MDA at the SPCs in Madrid, Canberra, and Goldstone
was completed in early 2993 without significant problems. This was followed a few
months later by the MPA, so that by the end of the year, the DSCC Tracking Subsystem
as originally planned was in place and supporting mission operations.

By that time, however, several major events had occurred that would materially afkct
the just-completed DSCC Tracking Subsystems. These events were the failure of the
Galileo HGA and subsequent emergence of the DGT and BVR to support a Galileo
S-band mission, encouraging results from development tests of an X-band exciter and
approval of plans for another set of 34-m antennas (BWG). The BVR would replace
the existing Block IV receivers, the new exciter would bring X-band uplinks to the DSN,
and the BWG antennas would increase the number of tracking system groups to five.
In all three instances, new MDA s o h a r e and hardware was required.

Although an early version of the BVR appeared at Goldstone at the beginning of

1993 for testing with the Galileo downlink, it was not until May of 1995 that it
was implemented in Eull operational form at all complexes and ready to begin accept-
ance testing with the new MDA s o h a r e . These tests revealed problems with missing
Doppler data between the BVR and the MDA and required more changes to the
MDA software. The DSN was to switch to the BVR for Galileo downlink mission
operations on 18 September 1995. Also at that time, the Galileo downlink would
be switched permanently to the suppressed carrier mode. Galileo was approaching
Jupiter, and good Doppler data was vital to the success of the Galileo Jupiter Orbit
Insertion sequence. Under pressure of these impending changes, the new MDA sofi-
ware was developed; checked in the DSN test facility at JPL, DTF 21; tested at
Goldstone (DSS 14); shipped to the other complexes; and installed in eight weeks.
It was soon followed by a new sofnvare package for the SRA, which, together with
a new hardware board, allowed it to generate ranging data with the BVR. It was this
sofnvare that was used to support the highly successful GaliIeo mission operations at
Jupiter in December 1995.

Subsequently, more changes were made to the MDA s o b a r e to add operability enhance-
ments requested by the users at the Complexes, to improve the Monitor interfaces, and
to provide an interface to the Block V X-band exciter. Although this latter capability
was not driven by Galileo, it would be required by other spacecraft: Mars Glabal Surveyor,
Mars Pdtbjnhr, and Cmsini, all to be launched within the next year or two.
THEGALILEO
Em: 1986-1996

By 1977, when the new beam waveguide antennas came into operation, the DSCC
tracking subsystem had been greatly expanded at each Complex to accommodate the
additional X-band uplink and downlink.

In addition to the DSCC Tracking Subsystem, there was another important compo-
nent of the overall DSN tracking system located at each complex. This was called
the DSCC Media Calibration Subsystem (DMD). The two principal functions of the
DMD were to make continuous measurements of ground weather parameters and to
make continuous measurements of ionospheric conditions at the site. These data were
formatted and transmitted to the NOCC for use by the Navigation Team in cor-
recting radio metric data for media effects before it could be used for orbit
determination purposes.

The local ground weather observations were made by a meteorological monitoring

assembly and had gradually been improved over the years. However, the ionospheric
data had for many years been supplied to the DSN by other agencies and were based
on ad hoc observations of an aging Earth satellite (ETS-6) from several remote-observ-
ing sites. This unsatisfactory method of determining total electron content (TEC) was
replaced in 1772 by a system that measured TEC continuously at the local site. Based
on Global Positioning Satellite (GPS) technology, the new GPS Receiver/Processor
Assembly (GRA) could automatically acquire and track at least four GPS space vehi-
cles when they were above the local horizon without requiring operator intervention.
The GRA computed the differential group delay and differential phase delay between
the L1 and L2 carriers from the space vehicles and estimated the slant TEC. The
DMD made provisions for monitoring, validating, and recording the TEC data prior
to its transmission to JPL.

OPERATIONS
NETWORK CENTER
CONTROL (NOCC)

Located at JPL, the NOCC furnishes coordination, control, and monitoring services to
the Network and serves as the single data and communications interface between the
Complexes and the external users of the DSN. Although these functions were not changed,
they were significantly affected by the SPC upgrade, and necessitated an upgrade to
existing NOCC capability.

Concurrent and interlocked with the upgrade of the SPC, the task of upgrading the
NOCC was directed toward complementing the new capabilities of the SPC and improv-
ing the reliability and operating efficiency (operability) of the Center. The primary functions
of the NOCC-providing a control interface between the Complexes and external users
such as the Flight Projects, and providing a central facility for the control and moni-
toring of the DSN-were not changed.

The upgraded NOCC gave its operators better information with which to monitor the
performance and status of the tracking stations, resulting in a reduced need for verbal
interaction with the station operators. The upgraded SPC gave the station operators
more information with which to identify and correct problems independently of the
NOCC operators. The new ability of the SPCs to perform frame synchronization and
decommutation of telemetry data and to calculate the deviation of actual performance
from predicted values was a key factor in this area.

Other areas of activity in the NOCC saw similar benefits. A manual procedure for
reporting frequency and timing offsets was replaced by an automatic procedure, and the
need for a member of the radio science team to be present at a station during radio sci-
ence activity was eliminated. Faster identification of operational errors and the reduction
of command, tracking, and monitor communications traffic between the NOCC and
the complexes led to significant improvement in the quality of the VLBI navigation data
generated by the Network.

The NOCC upgrade was carried out as planned and, when completed in 1993, repre-
sented a significant advance in the efficiency of centralized control of the Network.
Except for a change in its hnction as an interface between external users and the DSN,
the NOCC continued to function as an integrd parr of the DSN without further mod-
ification through the end of the Galileo Era.

In order to make the changes that were planned for the SPC, it was necessary to upgrade
the existing communication facilities at the three Signal Processing Centers (SPCs), as
well as those at the Central Communication Terminal (CCT) at JPL. This became the
Ground Communication Facility (GCF) Upgrade Task, and it ran concurrently with,
and with the same constraints as, the other upgrade tasks.

Prior to the GCF upgrade, the interchange of data between the subsystems at each
SPC was carried on a single Ungerman-Bass Local Area Network (LAN). To handle
the additional telemetry, command, and tracking equipment thar came with the SPC
upgrade, the GCF task provided two new LANs, each connected to a new Station
Communications Processor (SCP). One LAN was linked to the old UBI LAN via a
"gateway" and was designated as the SPC LAN. The other was designated as the Hi-
rate LAN. Interactive traffic between the ComplexILink Controllers on the UBI LAN
and the telemetry, command, and tracking subsystems on the SPC LAN flowed through
the "gateway." Traffic between these subsystems and the NOCC moved through the
SCP The Hi-rate LAN handled only telemetry, transporting the output from all five
telemetry channels to the SCl? The SCP provided digital recording and formatting
functions for the data on both LANs before transmitting it to the Central
Communications Terminal (CCT) at JPL via wideband satellite and gound comrnu-
nication circuits provided by NASCOM.

The CCT at JPL was the terminal for all voice and data circuits between the NOCC
and each of the complexes. Located at the CCT, the Central Communications Processor
(CCP) could be viewed as a mirror image of the Station Communications Processor
with the GCF transmission subsystem, and it included the far-reaching communica-
tion channels of NASCOM, forming the Iink between them. A LAN connected the
CCP with a digital recording subsystem, a monitor and control subsystem, and three
"gateways." It was through these "gateways" that the GCF digital communications sub-
system connected the Complexes and the NOCC to the outside world. The NOCC
gateway provided for all NOCC traffic. An "External User" gateway conducted traf-
fic with remote Mission Operations Control centers such as GSFC or space agencies
in foreign countries, while the third "SFOC" gateway handled the data flow berween
the SFOC* and the DSN.

A simplified diagram of the subsystems of the GCF as it was configured in 1992 is

given in Figure 5-42.

The Big Pipe and Little Pipe elements of the GCF transmission system were descrip-
tors for the NASCOM communication channels. They delivered data to the CCT at
672 kbps and 64 kbps, respectively, from Madrid and Canberra, and 1.2 Mbps and 224
kbps from Goldstone.

Implementation of these changes was to be phased in with incremental changes in

SPC capability and was in progress when hrther changes were imposed by the require-
ments of the Galileo mission redesign in 1993. To meet the arraying requirements of

A t various times, the faclig at JPL j o m which ]PL inflight missions were controlled was known by dzfferent
names that r$ected the changing scope of its activities. Some of these names inc1zrde the Space Flight Operations
Facility (SFOg, the Misszon Control and Computing Center (MCCC), the Space Flight Operations Center (SFOC),
a d the Multimission Ground Data System (MGDS). At the time of the GCF upgrade, i.t was called the SFOC
f \

DSCC SUBSYSTEMS

MGDS
(SFOC)

NOCC
0 TRANSMISSION INSS, RTI
(NAV)
(VLBf)

ODR

GDC - GCF DIGITAL COMM. SUBSYSTEM REMARK: IDR

0 -
GMC GCF MONITORAND CONTROL SUBSYSTEM OTHER GCF SUBSYSTEMS
GDR - GCF DATA RECORDS SUBSYSTEM * GVC - GCF VOICE SUBSYSTEM
GTM - GCFTRANSMISSION SUBSYSTEM GTY - GCR TELETYPE SUBSYSTEM

-1 MIL 71
1 --I JWCCT1-
\ J

Figure 5-42. Ground Communications Facility, 1992.

the DGT, a special channel having a capacity of 400 kbps was established between
Goldstone and Canberra. This channel, which consisted of two standard intersite cir-
cuits, was routed through the CCT for monitor and control purposes and was specified
to require an error rate better than 2 x 105 with a 1-way, 2-hop delay that would not
exceed 0.75 second.

As part of the original SPC upgrade, new capability had been added to all DSN sub-
systems to allow them to transfer data files among each other. This was extended in
1993 to include the SFOC by providing suitable hardware and s o h a r e in the GCF
digital communications subsystem for routing and monitoring non-real-time data file
transfers. This service eventually came into regular use for automatic post-pass trans-
fer (PPT) of data files between the complexes and the flight project mission control
centers, and it proved to be of great value in streamlining the conduct of routine
DSN operations.
A SUCCESSFUL CONCLUSION

The successful conclusion of the Galileo Prime Mission in December 1997 brought the
Galileo Era to a close. The effort devoted to conducting mission operations, for both
the spacecraft and the DSN, was scaled back but continued, at a much reduced level,
as the Europa Continuation Mission. By that time, the Galileo project had been on the
DSN "books" in various forms for nearly rwenty years. Also, NASA had recently announced
a new initiative for "faster, better, cheaper" spacecraft. Casini, the last of the planetary
"mega-spacecraft," had already been launched and was on its way to Saturn.

The Cassini Era was at hand.

Endnotes

1. Bruce Murray, Journey Into S p c e (New York: W W. Norton, 1989), pp. 227, 228,
321.

2. C. Stelzried, L. Effron, and J. Ellis, "Halley Comet Missions," TDA Progress Report
PR 42-87, July-September 1986 (15 November 1986), pp. 240-42.

3. C. T. Stelzried and T. Howe, "Giotto Mission Support," TDA Progress Report PR

42-87, July-September 1986 (15 November 1986), pp. 243-48.

4. N. A. Mottinger and R. I. Premkurnar, "Giotto Navigation Support," TDA Progress

Report PR 42-87, July-September 1986 (15 November 1986), p. 249-62.

5. R. Linfield and J. Ulvestad, "Source and Event Selection for Radio-Planetary Frame-
Tie Measurements Using the Phobos Landers," TDA Progress Report PR 42-92,
OctoberIDecember 1987 (15 February 1988), pp. 1-1 2.

6. K. M. Cheung and F. Pollara, "Phobos Lander Coding System: Software and

Analysis," TDA Progress Report PR 42-94, AprilIJune 1988 (15 August 1988), pp.
274-86.

7. l? H. Stanton and H. F. Reilly, Jr., "The L-/C-band Feed Design for the DSS 14 70-
Meter Antenna (Phobos Mission)," TDA Progress Report PR 42-107, July1
September 1991 (15 August 199I), pp. 364-83.

8. J. Withington, "DSN 64-Meter Antenna L-band (1688-MHz) Microwave System

Performance Overview," TDA Progress Report PR 42-94 AprilIJune 1988 (15
August 1988), pp. 294-300.

9. C. E. Hildebrand, B. A. Iijima, and W. M. Folkner, "Radio-Planetary Frame Tie

from Phobos-2 VLBI Data," TDA Progress Report PR 42- 119, July-September
1994 (15 November 1994).

10. A. L. Berman, "Deep Space Network Preparation Plan for MagelIan Project," DSN
Document 870-73, JPL D-4852 (Pasadena, California: JPL, 15 March 1989).
11. A. L. Berman, D. J. Mudgway and J. C. McKinney, "The 1986 Launch of the
Galileo Spacecraft via the Space Transportation System," TDA Progress Report PR
42-72, October-December 1982 (15 February 1983), pp. 186-99.

12. Jet Propulsion Laboratory, "Magellan Spacecraft Final Report," Internal Report
MGN-MA-011 (Pasadena, California: Martin Marietta Technologies, Inc., January
1995), pp. 50-51.

13. Rick Gore, "Neptune; Voyager's Last Picture Show," National Geogrdphic Maguzine
(August 1990): 35-64.

14. Bradford A. Smith, "Voyage of the Century," National GeographicMagazine (August

1990): 48-65.

15. W. J. O'Neill et al., "Project Galileo; Mission and Spacecraft Design," JPL D-05 18,
Jet Propulsion Laboratory, January 1983 (compilation of papers presented at ATAA
2 1st Aerospace Sciences Meeting, Reno, Nevada, 10-1 3 January 1983).

16. A. L. Berman, D. J. Mudgway and J. C. McKinney, "The 1986 Launch of the

Galileo Spacecraft via the Space Transportation System," TDA Progress Report PR
42-72, October-December 1982 (15 February 1983), pp. 186-99.

17. S. Dolinar, "A New Code for Galileo," TDA Progress Report PR 42-93,
January/March 1988 (15 May 1988), pp. 83-96.

18. D. J. Mudgway, Jet Propulsion Laboratory, Telecommunications and Data

Acquisition Oace, interofice memorandum to R. J. Amorose, from "Galileo
Convolutional Coding Proposal," 6 July 1987.

19. J. W. Layland ec al., "Galileo Array Study Team Report," TDA Progress Report PR
42-1 03, JulylSeptember 1990 (15 November 1990)' pp. 161-69.

'
20. T. K. Peng, J. W. Armstrong, J. C. Breidenthal, F. Donivan, and N. C. Ham,
"Deep Space Network Enhancement for the Galileo Mission To Jupiter," IAF-86-
304 (Innsbruck, Austria: International Astronautical Federation, October 1986).

21. W. J. O'Neil, "The Galileo Messenger," Issue 31 (Pasadena, California: Galileo

Project Oflice, JPL, February 1993), p. 9.
22. L. Deutsch and J. Marr, "Low Gain Antenna S-band Contingency Mission," Galileo
Project Document 1625-501 (Pasadena, California: JPL, 10 April 1992).

23. J. I. Statman, "Optimizing the Galileo Space Communication Link," TDA Progress
Report PR 42- 118 (15 February 1994), pp. 11 6 2 0 .

24. E E. Beyer, D. J. Mudgway, and M. M. Andrews, "The Galileo Mission to Jupiter:

Interplanetary Cruise; Post-Earth Encounter through Jupiter Orbit Insertion," TDA
Progress Report PR 42-125, MarchIApril 1996 (15 May 1996), pp. 1-16.

25. P. E. Beyer, B. G. Yetter, R. G. Torres, and D. J. Mudgwa~,"Deep Space Network

Support for the Galileo Mission to Jupiter: Jupiter Orbital Operations From Post-
Jupiter Orbit Insertion Through the End of the Prime Mission," TDA Progress
Report PR 42-133, MarchIApril 1998 (15 May 1998), pp. 1-23.

26. Willis G. Meeks and Ed. B. Massey, "Ulysses: An Investigation of the Polar Regions
of the Heliosphere," personal communication, July 1997.

27. European Space Agency, "The Ulysses Data Book," European Space Agency
Document, ESA-BR 65 (Noordwijk, Netherlands: ESTEC, June 1990).

28. D. M. Enari, "DSN Preparation Plan: Ulysses Project," JPL Document D-1508
(Pasadena, California: JPL, 1 August 1985).

29. M. R. Traxler, "DSN Support of Mars Observer," TDA Progress Report PR 42-1 13,
January-March 1993 (15 May 1993), pp. 118-22.

30. T. A. Rebold, A. Kwok, G. E. Wood, and S. Butman, "The Mars Observer Ka-band
Link Experiment," TDA Progress Report PR 42-1 17, January-March 1994 (15 May
1994), p. 250.

31. Earl J. Montoya and Richard 0. Fimmel, "Space Pioneers and Where They Are
Now," NASA EP-264 (Washington, District of Columbia: NASA Educational
Affairs Division, 1987).

32. Richard, I? Rudd, "Voyager Approach to Maintaining Science Data Acquisition for
a 30-Year Extended Mission," Voyager Project Web site, http://jplnasa.gov/
voyager/30plan.html (Pasadena, California: JPL, November 1997).
33. N. A. Fanelli, "JPL Emergency Support ofTDRSS and Compatible Satellites,"TDA
Progress Report PR 42-82, April-June 1985 (15 August 1985), pp. 120-24.

34. N. A. Fanelli to D. J. Mudgway, "Emergency Cross Support Agreements," private

communication, 11 November 1997.

35. B. Wood, "Report on rhe Mechanical Maintenance of the 70-Meter Antennas,"

DSN 890-257, JPL D-10442 (Pasadena, California: JPL, 31 January 1993),
pp. 5, 207.

36. Donald H. McClure, "Rehabilitation of the DSN 70-Meter Antennas: The DSN
Antenna Rehabilitation Team Final Report," DSN Document 890-226 (Pasadena,
California: JPL, 1 November 1989).

37. D. W. Brown, H. W. Cooper, J. W. Armstrong, and S. S. Kent, "Parkes-CDSCC

Telemetry Array: Equipment Design," TDA Progress R e p ~ r tPR 42-8 5, January-
March 1986 (15 May 1986), pp. 85-1 10.

38. J. W. Layland et al., "Interagency Array Study Report," TDA Progress Report PR
42-74, April-June 1983 (15 August 1983), pp. 117-48.

39. D. W. Brown, W. D. Brundage, J. S. Ulvestad, S. S. Kent, and K. I? Bartos,

"Interagency Arraying for Voyager-Neptune Encounter," TDA Progress Report PR
42-102, April-June 1990 (15 August 1990), pp. 91-1 18.

40. Andrew J. Butrica, Ta See the Unseen;A History of Phnetary R d h r h o n o m y , NASA

SP-4218 (Washington, District of Columbia: NASA, NASA History Office, 1996).

41. J. R. Paluka, "100-kW, X-band Transmitter for FTS," Technical Report TR 32-
1526, Vol. VIII, January/February 1972 (15 April 1972), pp. 94-98.

42. A. L. Berman, "The Gravitational Wave Experiment: Description and Anticipated

Requirements," TDA Progress Report PR 42-46, MayIJune 1978 (15 August 1978),
pp. 100-08.

43. T. A. Komarek, J. G. Meeker, and R. B. Miller, "ISPM X-band Technology

Demonstration, Part 1. Overview," TDA Progress Report PR 42-62,
JanuaryIFebruary 1981 (15 April 1981), pp. 50-62.
44. J. G. Meeker and C. Timpe, "X-band Uplink Technology Demonstration at DSS
13,"TDA Progress Report PR 42-77, January-March 1984 (15 May 1984),pp. 24-32.

45. R. Stevens, Jet Propulsion Laboratory, Telecommunications and Dara Acquisition

Ofice, Interoffice Memorandum to I? T. Lyman, "X-band Uplink Experiment for
Galileo" (3 June 1982).

46. Frank B. Easterbrook, "Gravitational Wave Searches with Ground Tracking

Networks," LW-85-386 (paper presented at the 36th Congress of the Internationd
Astronautical Federation, Stockholm, Sweden, October 1985).

47. D. J. Mudgway, " 1986 Galileo Project NASA Support Plan," JPL Document D-5 85
(Pasadena, California: JPL, 15 November 1983).

48. R. M. Perez, "Improvernents in X-band Transmitter Phase Stability Through

Klystron Body Temperature Regulation," TDA Progress Report PR 42-109,
January-March 1992 (15 May 1992), pp. 114-20.

49. S. Hinedi, "A Functional Description of the Advanced Receiver," TDA Progress
Report PR 42-100, October-December 1989, (15 February 1990), pp. 131-49.

50. Universiry of Padua, "The Galileo Spacecraft: A Legacy for Future Space Flight"
(proceedings of the Three Galileos Conference, Padua, Italy, 1997).

5 1. Jet Propulsion Laboratory, Telecommunications and Mission Operations

Directorate, "Galileo Low Gain Antenna S-band Contingency Mission," presenta-
tion to NASA Headquarters, JPL Document 1625-497 (20 March 1992).

52. J. I. Statman, "Optimizing the Galileo Space Communication Link," TDA Progress
Report PR 42-1 16, October-December 1993 (15 February 1994), pp. 1 14-20.

53. J. W. Layland et al., "Galileo Array Study Team Report," TDA Progress Report PR
42- 103, JuIy/September 1990 (15 November 1990), pp. 161-69.

54. A. Mileant and S. Hinedi, "Overview of Arraying Techniques in the Deep Space
Network," TDA Progress Report PR 42-104, October-December 1990 (15
February 199!), pp. 109-15.
55. S. Dolinar, "A New Code for Galileo," TDA Progress Report PR 42-93, January-
March 1988 (15 May 1988), pp. 83-96.

56. H. Tsou, B. Shah, R. Lee, and S. Hinedi, '% Functional Description of the Buffered
Telemetry Demodulator (BTD)," TDA Progress Report PR 42-112, October-
December 1992 (15 February 1993), pp. 50-73.

57. S. Dolinar and M. Belongie, "Enhanced Decoding for the Galileo S-band Mission,"
TDA Progress Report PR 42-1 14, April-June 1993 (15 August 1993), pp. 96-1 11.

58. S. Million, B. Shah, and S. Hinedi, 'RComparison of Full Spectrum and Complex-
Symbol Combining Techniques for the Galileo S-band Mission," TDA Progress
Report PR 42- 116, October-December 1993 (15 February 1994), pp. 128-43.

59. T. T. Pharn, S. Shambayati, D. E. Hardi, and S. G. Finley, "Tracking the Galileo

Spacecraft with the DSCC Galileo Telemetry Prototype," TDA Progress Report PR
42-1 19, SeptemberlOctober 1994 (15 November 1994), pp. 221-35.

60. Jet Propulsion Laboratory "The Galileo Messenger," Issue 39, Gaiileo Project
Offlce, JPL (July 1996).

61. J. G. Smith, "Ka-band (32 GHz) Downlink Capability for Deep Space
Communications," TDA Progress Report PR 42-88, OctoberIDecember 1986 (15
February 1987), pp. 96-103.

62. J. G. Smith, "Proposed Upgrade of the Deep Space Network Research and
Development Station," TDA Progress Report PR 42-88, OctoberIDecember 1986
(15 February 1987), pp. 158-63.

63. R. C. Clauss and J. G. Smith, "Beam Waveguides in the Deep Space Network,"
TDA Progress Report PR 42-88, October-December 1986 (15 February 1987),
pp. 174-82.

64. T. Veruttipong, W. Imbriale, and D. Bathker, "Design and Performance Analysis of

the DSS-13 Beam Waveguide Antenna," TDA Progress Report PR 42-101,
MarchlApril 1990 (15 May 1990), pp. 99-1 13.
65. S. D. Slobin, T. Y. Otoshi, M. J. Britcliffe, L. S. Alvarez, S. R. Stewart, and M. M.
Franco, "Efficiency Calibration of the DSS 13 34-m Diameter Beam Waveguide
Antenna at 8.45 and 32 GHz," TDA Progress Report PR 42- 106, April-June 1991
(15 August 1991), pp. 283-89.

66. M. S. Esquival, "Optimizing the G/T ratio of the DSS 13 34 m Beam Waveguide
Antenna," TDA Progress Report PR 42-109 January-March 1992 (15 May 1992),
pp. 152-61.

67. W. Imbriale, W. Veruttipong, T. Otoshi, and M. M. Franco, "Determining Noise

Temperature in Beam Waveguide Systems," TDA Progress Report PR 42-1 16,
October-December 1993 (15 February 1994), pp. 42-52.

68. L. S. Alvarez, M. J. BritclifKe, M. M. Franco, S. R. Stewart, and H. J. Jackson, "The

Efficiency Calibration of the DSS 24 34-Meter Diameter Beam Waveguide
Antenna," TDA Progress Report PR 42-120 (15 February 1995), p. 174.

69. D. L. Ross, "DSCC Subsystem Functional Requirements; Telemetry Subsystem

1992-1995," JPL Document D-6111, 824-35 (15 May 1989).

70. J. Statman, G. Zimmerman, F. Pollara, and 0 . Collins, "A Long Constraint Length
VLSI Viterbi Decoder for the DSN," TDA Progress Report PR 42-95, July-
September 1988 (15 November 1988), pp. 13442.

71. C. R. Lahmeyer and K. M. Cheung, "Long Decoding Runs for Galileo

Convolutional Codes," TDA Progress Report PR 42-95, July-September 1988 (15
November 1988), pp. 143-52.

72. J. Statman, J. Rabkin, and B. Siev, "Big Viterbi Decoder Results for (7,112)
Convolutional Code," TDA Progress Report PR 42-99, July-September 1989 (15
November 1989), pp. 122-23.

73. I. M. Onyszchuk, "Testing Interconnected VLSI Circuits in the Big Viterbi

Decoder," TDA Progress Report PR 42-106, April-June 1991 (15 August 199l), pp.
175-82.

74. B. Falin, "DSCC Subsystem Functional Requirements: Command Subsystem

1992-1 996," JPL Document D-6395, 824-36 (1 August 1989).
75. Jet Propulsion Laboratory, "DSCC Subsystem Functional Requirements: DSCC
Tracking Subsystem 1992-1 996," JPL Document D-7335, 824-38 (1 July 1990).
THE CASSINI ERA

Throughout its life, the form of the DSN at any single point in time was shaped by
many forces acting on it simultaneously and ofren in different directions. These forces
included budget constraints from the NASA Office of Tracking and Data Acquisition
(Code O), the requirements for support of new missions from the NASA Offlce of
Space Science (Code S), the drive of new technology, the pressure of real-time opera-
tions, and the concerns of foreign agencies and competing requirements for limited
antenna time from radio astronomy and other space science constituencies. Dealing with
these often conflicting currents had become a way of life at all levels within the DSN
structure, and the remarkable progress recorded in the forthcoming pages attests to the
high degree of success that was achieved.

By 1995, however, the winds of change were blowing more strongIy than ever before
throughout NASA, and by 1997 their effect was being keenly felt in the DSN. The
changes wodd affect the established purpose, scope, and functions of the DSN, and

Cassini Era Mission Set

Deep Space Missions (Launch) Earth-Orbiter Missions
Galileo (1989) ISTP-Soho
Voyager (1977) ISTP-Polar
U/ysses (1990) ISTP-Wind
Pioneer 10 (1972) ISTP-Geotail
N-E Asteroid Rndv. (2117/96) IR Space Observatory
Mars Global Sun/eyor (11/7/97 ASTRO-D
Mars Pathfinder (12/04/96) TOMS-EP
Cassini (10/15/97) YOHKOH Solar-D
Space Tech. Res. Vehicle A
Radar Satellite
Roentgen Satellite
SURF Satellite-1
SSTI-Lewis
Hotbird-2,-3
HALCA
L J

Figure 6-1. Cassini Era mission set, 1997.

they first appeared in the Telecommunications and Mission Operations Directorate
(TMOD) Implementation Plan for Fiscal Year 1998.' This plan is discussed fully in a
later chapter.

This chapter will review the DSN and the missions it was supporting as the Galileo Era
came to an end and the Cassini Era opened to an environment of change.

A snapshot of the DSN operational tracking schedule in 1997 would have revealed the
portfolio of deep space and Earth-orbit missions shown in Figure 6-1. This represented
the DSN mission set for the first year or two of the Cassini Era.

Except for Ulysses, which was an ESA mission, all of the deep space missions were of
NASA origin. Half of the missions, Galileo, Voyager, Ulysses, and Pioneer 10, were older
missions that had been in the DSN for many years and have been discussed in previ-
ous chapters. The Near-Earth Asteroid Rendezvous, Mars Global Surveyor, Mars
Pathfinder, and Cassini missions were new in 1996 and 1997, and brought with them
new challenges and changes for the DSN.
DEEP SPACE MISSIONS

Through the end of the Galileo Era and the beginning of the Cassini Era, the ongoing
deep space missions, Galileo, Ulysses, Pioneer 10, and Myugers 1 and 2, presented no
new challenges for the DSN. They continued to return science and engineering data,
each according to its own remaining capability, and were provided with minimal track-
ing station support, sdcient only to maintain their viability as scientific missions. The
changes and upgrades that had been incorporated into the DSN since these missions
began had always maintained a capability to handle their diminishing requirements for
telecommunications and data acquisition support.

Amongst the aging missions, Galileo was the most active. As described in the previous
chapter, Galiho had completed its primary mission in December 1997 and continued
ar a much reduced pace to carry out further observations of Europa under rhe name
Galileo Europa Mission. It remained entirely dependent upon arraying of the DSN 70-
ml34-m antenna for its downlinks, since the Parkes antenna was no longer availabIe to
support the arrayed configurations.

As the Ulysses spacecrafi embarked on its second 6.2-year orbit over the poles of the
Sun, its science instruments continued to perform normally. The north and south polar
passes were planned for 2000 and 2001, respectively On this orbit, the properties of
solar winds in high solar latitudes during the maximum of the solar activity cycle were
to be investigated. Through 1997, Earth-pointing maneuvers and instrument calibra-
tions were made regularly, in preparation for observations during the fifth solar opposition
in March 1998. Daily passes on the 34-m antennas at S-band or X-band, and teleme-
try running at 512 bits per second to 2,048 bits per second, were typical of this period.

The continuation of the Vyuger I and 2 missions beyond the outer planets was calied
the Voyager interstellar Mission (VIM) and would continue through the year 2019.
Through 1996 and 1997, Myuger 1 and Vyugeer 2 maintained a presence on the DSN
tracking schedules with several passes per week for each spacecraft on the 3 4 m anten-
nas, with occasional support on the 70-m antennas.

At approximately 2:10 P.M. Pacific time on 17 February 1998, according to VIM proj-
ect manager Ed B. Massey, Vaydgeer 1 reached an Earth-to-spacecraft range of 10.4 billion
krn (6.5 billion miles), exceeding the record held by Pioneer 10 for 25 years, and became
the most distant human-made object in space. Almost 70 times farther from the Sun
than Earth (70 AU), with che radio signal taking 9 hours and 36 minutes to travel from
the spacecraft to Earth, Vyager 1 was then at the very edge of the solar system. The
downlink data rate was 160 bits per second or 600 bits per second on the 34-m anten-
nas and could be increased to 1.4 kilobits per second on the 70-rn antennas when
required for tape recorder playback. At a smaller distance of 52 AU, Vyager 2 was being
supported by the southern hemisphere stations in Canberra and could receive downlink
data at 7.2 kilobits per second.

JPL marked the event with a special press release? "Vyager 1 Now Most Distant Human-
Made Object In Space." The article explained the scientific si&cance of this remarkable
achievement:

Having completed their primary explorations, Vyager 1 and its twin Vqager 2
were studying the environment of space in the outer solar system. Although they
were beyond the orbit of all the planets, the spacecraft remained well within the
boundary of the Sun's magnetic field, called the heliosphere. Science instruments
on both spacecraft sensed signals that scientists believed were coming horn the
outermost edge of the heliosphere, known as the heliopause. The heliosphere results
from the Sun's emission of a steady flow of electrically charged particles d e d the
solar wind. As the solar wind expands supersonically into space in all directions,
it creates a magnetized bubble-the heliosphere-around the Sun. EventuaUy, the
solar wind encounters the electrically charged particles and magnetic field in the
interstellar gas. In this zone, the solar wind abruptly slows down from superson-
ic to subsonic speed, creating a termination shock. Before the spacecraft travel
beyond the heliopause into interstellar space, they will pass through this termina-
tion shock. Dr. Edward C. Stone, Voyager project scientist and director of JPL,
said, "The data coming back &om Vyager now suggest that we may pass through
the termination shock in the next three to five years. If that's the case, then one
would expect that within ten years or so, we would actually be very close to pen-
etrating the heliopause itself and entering into interstellar space for the first time."
 Reaching the termination shock and heliopause would be major milestones
for the mission because no spacecraft had been there before, and the Voyagers
would gather the first direct evidence of their structure. Encountering the ter-
mination shock and heliopause had been a long-sought goal for many space
scientists. Exactly where these two boundaries were located and what they were
like remained a mystery.

Although these were significant long-range telecommunications records for the DSN,
uplink and downlink communications with both spacecrafi remained routine matters.
With plenty of performance margin still remaining, the DSN expected to be able to
support the VIM for the next twenty years.

Pioneer 10

After an active life of 25 years, the Pioneer 10 mission came to an end on 31 March 1997,
with the last track at DSS 63. Launched on 2 March 1972 and tracked by the DSN ever
since, Pioneer 10 support had almost become a permanent feature of routine DSN activi-
ty. Although the downlink was still running at 8 bps, the signal was very weak, and it had
been determined that the science value of the data being returned was no longer sufficient
to jusufy further continuation of the mission. Pioneer operations had been conducted from
the mission control center at the Ames Research Center in Sunnyvale, California, for many
years, and the mission had many notable scientific achievements to its credit. By 1983, it
had become the first spacecraft to travel beyond the orbits of Neptune and Pluto. Following
a solar system escape trajectory away from the Sun, it continued to move M e r outward
in search of the heliopause, the region of the solar system where the influence of the Sun
itself finally ends and true interstdar space begins. Now, following a trajectory in the oppo-
site direction to that of Pioneer 10, the pursuit of that goal would be taken over by Voyager.
In the immediate future, the Pioneer 10 spacecd would be used by the DSN operations
group at JPL to train new station controllers for future missions.

The Near-Earth Asteroid Rendezvous (NEAR) mission was the first in NASA's Discovery
program of lower cost, highly focused planetary science missions. It was developed and
managed by the Johns Hopkins University Applied Physics Laboratory, Laurel, Mar~land.~

The primary objective of the NEAR mission was to inject a spacecraft into a 50-krn
orbit around the near-Earth asteroid Eros for the purpose of making scientific observa-
tions to determine its size, shape, mass, density, and spin rate. It could also determine
the morphology and mineralogical composition of the surface.

Launched in February 1996, the NEAR mission used an Earth gravity-assist flyby in
early 1998 to deliver the spacecraft to the asteroid Eros in January 1999. Orbital oper-
ations were to begin immediately following arrival and were planned to be about one
year in duration. In the early part of its mission, June 1997, NEAR flew by the aster-
oid Mathilde. It found Mathilde to be composed of extremely dark material with numerous
large impact craters, including one nearly six miles deep.

DSN support for the NEAR mission in 1996 and 1997 was conducted primarily on
the 34-m HEF or BWG antennas, and the X-band uplinks and downlinks carrying
telemetry, command, and ranging data were well within existing capabilities. Initial acqui-
sition and a trajectory correction maneuver in March 1997 were accomplished by the
DSN without incident.

During the critical phases of the mission and mission operations at Eros, NEAR required
continuous tracking support (three 8-hour passes per day) from the 34-m stations, with
occasional use of the 70-rn antennas. At other periods during the cruise phase, the track-
ing support was reduced to one or two passes per day.

In preparation for an Earth swing-by maneuver in early 1998, tracking support for
NEAR was increased to the level of three passes per day in December 1997.

Mars Global Surveyor (MGS) was the first mission in a new program of Mars explo-
ration called the Mars Surveyor Program. It was also the first of NASA's new "faster,
better, cheaper" family of missions. MGS was designed to deliver a single spacecraft to
Mars for an extended orbital study of the surface, atmosphere, and gravitational and
magnetic fields of the planet. A search for gravitational waves and a demonstration of
Ka-band downlink communications technology was to be conducted during the flight
to Mars. The spacecrah was to be launched during the November 1996 Mars oppor-
tunity, using a Delta launch vehicle with a PAM D upper stage. The transit time to
Mars would be about ten months. MGS was to be the first U.S. spacecrafi to orbit
Mars since the Viking Orbiters twenty years before and the first to use aerobraking
rather than propulsive maneuvers to adjust its orbit upon arrival. The use of aerobrak-
ing techniques had been demonstrated in mid-1993 during the find stages of rhe Magellan
mission to Venus and was considered to be a promising technique for reducing the
f \

Drag
Drag Flap
Flap

Nad~rDeck

L 2

Figure 6-2. Mars Global Surveyor spacecraft in mapping configuration. The spacecraft was
built and operated by Lockheed Martin Astronautics, JPL's industrial partner on the mission, in
Denver, Colorado.

spacecraft he1 expended in circularizing the spacecraft orbit around a planet. Repetitive
observations of the Mars surface and atmosphere were to be carried out from a nearly
circular, low-altitude (378 km) orbit over a ~ e r i o dof one Martian year (687 Earth days).

A n overview of the major components of the spacecraft is shown in Figure 6-2.

Like the earlier Mars Observer spacecrafi, MGS uplinks and downlinks were both on X-
band, and it also carried a Ka-band downlink beacon as a new downlink technology
demonstration. The downlink telemetry would employ Reed-Solomon and convolutional
(K = 7, R = 1/2) encoding and would range from 10 bits per second to 32 kilobits per
second for engineering data, and 4 kilosymbols per second to 85 kilosymbols per sec-
ond for Reed-Solomon encoded science data. In coding terminoIogy, a symbol is essentially
a Reed-SoIomon encoded data bit. The MGS spacecrafi transmitted 250 encoded bits
to represent every 218 bits of its raw science data, a ratio of 1.147 to 1.O.
The command uplink wodd operate at a nominal rate of 125 bits per second, although
several other rates were available. Radio metric data consisting of two-way Doppler and
sequential ranging data would be generated from the coherent X-band uplink and down-
link carriers, and a noncoherent signal from the spacecraft's ultra-stable oscillator (USO)
would be used to generate project data during the frequent occultation experiments.

The MGS mission required the tracking stations to be able to acquire the X-band down-
link and achieve telemetry "in-lock" within five minutes of spacecrafi view at the station,
for data rates of 18.6 ksps or greater. Reacquisition of the telemetry signal following each
Earth occultation during rhe mapping phase was to be accomplished within one minute
of spacecraft view. These requirements, which posed some problems for the DSN, could
not be solved before launch but were satisfied by the time the spacecraft reached Mars.

The DSN was required to deIiver at least 95 percent of the science data transmitted
from the spacecraft during the mapping phase to the Advanced Multimission Operations
System (AMMOS) at JPL. Because the Mars Orbiter camera instrument made exten-
sive use of data compression, the imaging science data could not tolerate significant gaps
in the "packets" of telemetry data delivered by the DSN. This produced a further require-
ment on the DSN to deliver science data "packets" containing no more than one packet
gap or error within 10,000 packets. This was equivalent to one error per 100 million
bits (1 x lo-') and would require some additional backup recording equipment in the
GCF to meet the stringent requirement.

Following initial acquisition of the downlink signal after launch by the 26-m antenna
at Canberra, h e mission was to be supported by the 34-m HEF antennas, DSS f 5, 45,
and 65. The 34-m BWG antennas would also provide support as they became available
in 1997. The 70-m antennas were to support the critical Mars Orbit Insertion sequence
and to be available to support any other situations which were declared to be "critical"
by the MGS flight operations manager.

DSN readiness to support the launch and cruise phase of the MGS mission was pre-
sented to a Telecommunications and Mission Operations Directorate (TMOD) Review
Board in October 1996. For the first time, it was part of a broad-based review which
included DSN and AMMOS as an integrated service of the TMOD. It was this inte-
grated service (which included its multimission operators) that would be used by the
MGS flight operations manager to conduct the mission afier launch.

The DSN presentation satisfied the Review Board that it had verified telecornrnunica-
tions compatibility with the spacecrafi, had adequately trained its operations personnel,
had developed an Initial Acquisition Plan covering all available launch windows, and
that all of the facilities necessary for launch and early cruise were in place. The uplink
and downlink requirements for MGS were well within the capabilities of the DSN sys-
tems that had been implemented earlier under the SPC upgrade program. Because the
issues of downlink acquisition time and telemetry data accountability were related to
operations at Mars, they were not included in this review. The DSN, together with
AMMOS, was ready to support the MGS launch.

At noon on 7 November 1996, Mars Global Surueyor lifted off launch pad 17A at Cape
Canaveral Air Station, Florida, on a three-stage Delta I1 launch vehicle bound for Mars. All
launch vehicle sequences were completed as planned and ended with separation horn the
spacecraft about 50 minutes after launch. An anomaly that occurred during deployment of
the spacecraft solar panels was not a cause for concern at the t h e , since it was believed it
could be corrected the following day. About 70 minutes &er launch, just after the space-
craft came in view of Canberra, the X-band downlink was very quickly acquired, fist by
the 26-m antenna (DSS 46), and then by the 34-m HEF (DSS 45), using the of&ets from
DSS 46. The DSN settled into routine operations as MGS cruise support began.

Over the next several weeks, spacecraft engineers continued to evaluate various solutions
to the problem creared by the solar array anomaly, although it posed no immediate
threat to the mission.

In January 1997, the spacecraft was turned to point its high-gain antenna toward Earth,
and with a stronger downlink available, the spacecraft immediately began transmitting
telemetry at the higher data rates. During a period of low activity in May, a program
of observations directed to the search for gravitational waves was carried out. This was
the long-awaited opportunity that had eluded the project investigators when the GaIileo
X-band mission failed. The high-stability X-band exciter and X-band uplink and down-
link were key factors in reaching the sensitivity needed to detect the presence of gravity
waves by their characteristic three-perturbation signature in the two-way Doppler data.
There was no immediate answer to whether the observations revealed the presence of
gravitational waves, since analyses of these data would take many months to complete.

Ka-band technology demonstration passes with DSS 15 and DSS 13 were conducted
over a two-week period beginning on 21 July. The Ka-band downlink was received at
DSS 13 in a coherent mode with the X-band uplink fiom DSS 15. Afrer a period of
good ranging data had been recorded, DSS 15 transferred its uplink to DSS 34 to gen-
erate compamtive data over the Goldstone/Australia path.
In July, with the spacecraft rapidly approaching Mars, mission controllers and DSN oper-
ators engaged in several simulation exercises to prepare for orbital operations. During
the simulations, the spacecraft radio transmitter was turned on and off over the course
of a six-hour period to simulate three actual orbits. Because capturing the project data
and acquiring the telemetry data under these rapidly changing conditions would be an
operationally demanding task for the tracking station operators and project team, these
simulations were used to familiarize them with the necessary operational procedures.

About three weeks before the actual MOI, the DSN team reviewed its state of readi-
ness to support the rapidly approaching orbit insertion and orbital operations phase of
the MGS mission. As in the Launch Readiness Review, the review included the status
of the AMMOS as well as that of the DSN.

The DSN team presentation focused on the steps that had been taken to meet the
telemetry delivery (data gaps) and acquisition (lock-up time) requirements that had been
of concern to the DSN prior to launch. It was shown that the implementation of reli-
able network servers in the GCF and the correction of a known antenna-pointing anomaly
at the BWG stations had accounted for most of the difference between actual measured
performance of the DSN and the performance required by the MGS specifications. The
remaining errors would be attributable to data gaps induced by downlink data rate
changes on the spacecrafe. Although lab measurements at JPL and past performance at
the stations confirmed that telemetry lock-up time would meet the requirement, some
uncertainty remained until the actual dowdink became available and performance was
verified under fdly operational conditions. An operations support plan for the MOI
sequence; availability of the necessary antennas, communications, and facilities; and a
demonstrated level of crew training completed the DSN presentation. There were no
outstanding issues or concerns. The DSN's Telecommunications and Mission Services
(TMS) manager, John C. McKinney, assessed the status as Network-ready to support
the Mars Orbit Insertion and Aerobraking events.

The DSN configuration used to support Mars Global Surveyor orbit insertion and orbital
operations is shown in Figure 6-3,

The spacecraft executed the orbit insertion sequence perfectly, beginning with a 22-
minute retro-engine burn on 11 September 1997, during the mutual view period of
Canberra and Madrid. Twelve minutes after the start of the engine burn, the first Mars
occultation occurred, providing the traclung station operators with their first opportu-
nity to demonstrate the benefits of the simulated acquisition training. When the spacecraft
emerged from behind the planet four minutes after the retro-engine burn had been com-
 I

Figure 6-3. DSN configuration for Mars Global Surveyor.This was the standard DSN configuration for the Cassini Era and has been
discussed earlier as part of the SPC upgrade task. It was used in various forms for all the missions in 1996 and 1997.
pleted, the downlink was acquired within seconds. Doppler data indicated that the space-
craft had entered a highly elliptical orbit within one minute of the intended 45-hour
orbit. The spacecraft was about 250 km above the Martian surface (periapsis) and about
56,000 krn above the surface at the farthest point (apoapsis).

A few days after the spacecraft began orbiting the planet, the spacecrafi magnetometer
detected the existence of a planet-wide magnetic field. This discovery was hailed by Vice
President Gore as "another example of how NASA's commitment to faster, better, cheap-
er Mars exploration is going to answer many fundamental questions about the history
and environment of our neighboring planet."

The first aerobraking maneuvers began with retro-engine burns at apoapsis on 17,
20, 22, and 24 September. As each periapsis passage began to take the spacecraft into
the upper reaches of the Mars atmosphere, atmospheric drag provided additional aer-
obraking effect. While the orbit slowly decreased, data taking progressed to the point
where the MGS Science Team was able to hold a press conference at JPL on 2
October to report their findings from dl six science experiments. "The spacecraft and
science instruments are operating magnificently," reported Dr. Arden Albee, the MGS
project scientist at the California Institute of Technology, Pasadena, California. "The
initial science data we've obtained from the walk-in phase of aerobraking are remark-
able in their clarity, and the combined measurements from all of the instruments
over the next two years are going to provide us with a fascinating new global view
of the planet."

Suddenly, during the fifteenth periapsis passage on 6 October, significant movement was
observed in the damaged solar array panel. O n 12 October the orbit was temporarily
raised to reduce the stress on the solar panel at each periapsis passage while the opera-
tions teams at JPL and Lockheed Martin Astronautics investigated this potentially alarming
situation. Two weeks later, with the spacecraft in a 35-hour orbit and closest approach
to the surface of 172 km, aerobraking was resumed at a much slower rate than before
to avoid putting undue stress on the unlatched solar panel. This decision would extend
the aerobraking phase by eight to twelve months and would change the final mapping
orbit. However, it would not significantly &ect the ability of MGS to accomplish its
primary mission's science objectives.

Apart from the faulty solar panel, the spacecraft was operating perfectly, and with DSN
support available from the BWG antennas, science data collection continued unabated.
An image of the giant volcano Olympus Mons taken during this period is shown in
Figure 6-4.
Figure 6-4. Mars Global Surveyor image of Olympus Mons. Mars Global Surveyor obtained
this spectacular wide-angle view of OIympus Mons from an altitude of 900 km (500 mi) above
the surface on its 263rd orbit around the planet on 25 April 1998. More than three times the
height of Mt. Everest, this giant volcano is almost flat, with its flanks having a gentle slope of
three to five degrees.

MGS completed its 100th orbit around Mars in January 1998. Assisted by the calm
state of the Martian atmosphere, aerobraking operations were proceeding satisfactorily.
The spacecraft had reached a 19-hour orbit around the planet, with a high point of
about 28,000 krn and a low point of about 120 km.

The target date for the start of mapping operations was March 1999, and it was hoped
that the duration of the mapping phase would still be one Martian year. The DSN was
providing support with the 34-rn HEF antennas on a routine basis with typical down-
link data rates of 2,000 bits per second for engineering and 21,333 symbols per second
for playback o f science data.4
Mars Pathfi~der

About one month after Mars C;labal Surveyor lifted off the launch pad at Canaveral Air
Force Station bound for Mars, the second spacecraft in the Mars Discovery Program
was launched from the same pad with the same destination, but with a different objec-
tive. Mars Pathinder (MPF) was primarily an engineering demonstration of key
technologies and concepts for eventual use in future missions to Mars employing sci-
entific Landers. MPF also delivered a scientific instrument pac!sge to the surface of
Mars to investigate the elemental composition of Martian rocks and soil and the struc-
ture of the Mars atmosphere, surface meteorology, and geology. In addition, MPF carried
a free-ranging surface Rover, which was deployed from the Lander to conduct technol-
ogy experiments and to serve as an instrument deployment mechanism. At launch, the
mass of MPF was 890 kilograms, compared to 1,060 kilograms for MGS. MPF would
arrive at Mars on 4 July 1997, seven months after launch, compared to 11 September
1997, ten months after launch, for MGS.

The MPF spacecraft comprised three major elements: the cruise vehicle, the decelera-
tion systems, and the Lander vehicle that contained the Rovera5

The cruise vehicle provided the major spacecraft functions of power generation, propul-
sion, and attitude control prior to entry into the Martian atmosphere. The deceleration
subsystem comprised an aeroshell, parachute, tether, retrorockets and inflatable airbags.
The tetrahedral Lander structure enclosed the science instruments, the Rover, and the
engineering subsystems necessary for cruise and surface operations. The radio transpon-
der and transmitter, along with attitude control and data handling electronics and sobare,
were also located on the Lander and were connected to the cruise stage via a detach-
able cable harness. The Lander structure was self-righting, with three side petals that
opened to establish an upright configuration on the surface from which the Rover could
be deployed. Power was provided by solar arrays mounted on the surface panels.

The MPF Lander with the Rover vehicle in a deployed configuration on the Mars sur-
face is depicted in Figure 6-5.

The duration of the primary mission was defined to be 30 days beginning at the time
of landing. Pathfinder requiremenrs for tracking and data acquisition lay within current
(1997) DSN capabilities, and the Telecommunicatons and Mission Services (TMS) man-
ager Dennis M. Enari concluded that no special implementation was needed. Both uplink
and downlink for MPF were on X-band and would be used for command, telemetry,
and generation of radio metric Doppler and range data. There were no requirements for
 Atmospheric Structure lnstnrment

Instrument Electron
Solar Panel
X-ray Spectrometer
Mars Pafhfinderlander

Figure 6-5. Pathfinder Lander deployed on the surface of Mars.

project observations, although the DSCC project equipment would be used to record
open-loop spectral data from the X-band downlink during the rapid and critical entry,
descent, and landing sequence. Descent Doppler profiles and key engineering telemetry
data would be reconstructed from this data after the event. Telemetry data rates would
range from 40 bits per second in the cruise phase to 22,120 bits per second during sur-
face operations. AU data would be convolutionally coded at either (R = 7, K = 112) or
(R = 15, K = 116). There would be no uncoded dara. Command data rates ranging
from 7 to 500 bits per second would be used on the X-band uplink.

During the entry, descent, and landing phase, both the 70-m and the 34-m HEF anten-
nas at Madrid would be tracking the MPF spacecraft simultaneously. At other times,
the tracking support would be provided by the 34-m HEF and 34-m BWG antennas,
with occasional use of the 70-m antennas for the downlink only. The 26-m antennas
at DSS 46 and DSS I 6 would provide support for the initial acquisition events only.
Mars Pathfinder blasted into space on 4 December 1996, the third day of its launch
window. The launch, ascent, orbital injection, and spacecraft separation events were all
normal. About 70 minutes after lift-off, the spacecraft made its first appearance over the
Goldstone radio horizon. Stations DSS 16 and DSS 15 quickly locked up their receivers
on the downlink signal and, less than two minutes later, began flowing telemetry data
to the mission controllers at JPL. On the evidence of these data, the spacecraft team
reported that all critical spacecraft systems, such as power, temperature, and attitude con-
trol, were performing well. "Everything looks really good and we're very happy," said
Tony Spear, Pathfinder project manager at NASKs Jet Propulsion Laboratory.

An event of some significance to the DSN occurred on 18 January 1997 when the new
Block 3 Maximum Likelihood Decoder (MCD3) was used for the first time to process
live inflight data from DSS 15 Goldstone.

The MCD3 grew out of the technology that had been used by Statman in developing
the Big Viterbi Decoder for the original Galileo X-band mission in 1989. Although that
effort was dropped when the spacecraft high-gain antenna problem forced a redesign of
the Galileo mission in 1991, the enhancement in downlink performance available from
newer, more powerfd convolutional codes remained viable. Although the encoding process
in the spacecraft was a relatively simple process, the new codes required much bigger
and extremely complex decoders in the DSN to take advantage of the improvement in
performance. In the intervening years, DSN engineers had pursued the development of
such a machine and, by the time of the MPF mission, were ready to demonstrate the
first machine of its kind in a real, inflight situation.

In this demonstration, the MPF telemetry downlink consisted of a convolutionally coded

(R = 15, K = 116) data stream, running at 4,070 bits per second. In light of the impor-
tance of this machine to later MGS, MPF, and Cassini mission support, its first successful
demonstration under operational conditions was a matter of considerable satisfaction to
the DSN.

The cruise phase passed quietly without the occurrence of any significant incidents.
Spacecraft and science instrument status checks were made routinely, and several trajec-
tory correction maneuvers were successfully executed. The DSN operations team carried
out several rehearsals to improve proficiency in the procedures that would be used in
July for the entry, descent, and landing sequence in July.

On 30 June, with all spacecraft systems in excellent operating condition and com-
manded by a sequence from its onboard computer, the MPF spacecraft began the
entry, descent, and landing phase of its mission. O n the morning of 4 July, an update
of the MPF orbit based on the latest DSN Doppler and range data indicated that
MPF was heading straight for the center of its redetermined 60-mile by 120-mile
landing ellipse and would enter the upper atmosphere at an entry angle just 0.75
degrees less than its original design value of 14.2 degrees. A few hours later, at 10:07:25
A.M. Pacific time, 4 July 1997, the MPF spacecraft landed successfully on the surface
of Mars, marking NASKs return to the surface of the Red Planet after more than
twenty years. A low-power transmission from an independent antenna on one of the
petals confirmed rhe landing and indicated that the craft had landed on Its base petal
in an upright position.

The first low-gain antenna transmission was received on schedule about four hours later.
It contained preliminary information about the health of the Lander and Rover; the ori-
entation of the spacecrafi on the surface; the entry, descent, and landing; and the
temperature and density of the Martian atmosphere. This was soon folowed by a high-
gain antenna transmission which contained the first images from the Lander. The next
day, 5 July, after clearing a problem with the petal that the Rover needed as a ramp to
reach the surface and rectifying a Rover-to-Lander communication link problem, the
mission control team moved the Rover vehicle (named Sojourner) down the ramp under
its own power and onto the Martian soil. Images of the Rover from the Lander and
vice versa, plus engineering data from both, confirmed that both vehicles were in their
proper positions and in good operating condition. With the downlink telemetry dara
rate set to 6,300 bps, science activities from the surface began in earnest.

During the first two weeks of surface operations, uplink and downlink communications
were interrupted several times due to Lander computer resets that unexpectedly switched
the low-gain and high-gain antennas, and to operational errors related to the timing of
the limited duration downlink sessions. Once these initial problems were cleared up, the
data rate was raised to 8,300 bps and the return of high-quality data from the MPF
Lander became a matter of routine operation for the DSN. Downlink sessions with the
Lander were of short duration, generally about ninety minutes, during which about GO
megabits of science and engineering data were returned. By the time MPF had com-
pleted its primary 30-day mission on 3 August, it had returned 1.2 gigabits of data,
including 9,669 images of the Martian landscape.

The designers of the spacecraft and the mission had every reason to be proud of their
efforts. "This mission demonstrated a reliable and low-cost system for placing science
payloads on the surface of Mars," said Brian Muirhead, Mars Pathflnder project man-
ager at JPL. "We've validated NASA's commitment to low-cost planetary exploration,
Figure 6-6. The MPF Lander camera views the Sojourner Rover in operation near Yogi Rock
on the surface of Mars, July 1997.

shown the usefdness of sending microrovers to explore Mars, and obtained signiiicant
science data to help understand the structure and meteorology of the Martian aunos-
phere, and to understand the composition of the Martian rocks and soil."

The downlink telemerry data from Mars, delivered by the DSN stations to JPL in real
time were subsequently processed into images by the MPF scientists and made available
to the media and on the World Wide Web. The images returned from both Lander and
Rover were remarkable indeed, but it was the robust, semi-autonomous Rover, Sojourner,
that captured the imagination of the public. To accommodate the swell of public inter-
est in following the mission via the World Wide Web, JPL engineers, in cooperation
with several educational and commercial institutions, constructed rwenty Pathfinder mir-
ror sites. Together, these MPF sires recorded 565,902,373 hits worldwide during the
period 1 J u l y 4 August. The highest number occurred on 8 July, when a record 47 mil-
lion hits were logged, more than twice the number received by the official Web site for
the 1996 Olympic Games in Atlanta, Georgia.
A typical high-quality MPF image transmitted from Mars to Earth over the DSN teleme-
uy downlink is shown in Figure 6-6. The inherent capability of the DSN downlink
contributed to the remarkable detail observed in this and other MPF images.

Downlink communications with the Lander continued with no sign of trouble until 27
September, when DSS 15 at Goldstone was unable to detect the presence of a down-
link at the scheduled transmission time. Declaration of a spacecrafi emergency by the
Pathfinder mission director authorized Enari to negotiate the release of the 70-m anten-
nas from Galileo and MGS support to assist with DSN attempts to recover the MPF
downlink. At the time, it was surmised that the downlink problems were most likely
related to depletion of the spacecrafi battery and uncertainties in the status of the onboard
clock. While the spacecrafi team investigated various scenarios to explain what might
have happened to the spacecrafi, the DSN went into emergency mode on a daily basis
for all MPF passes.

While these downlink recovery efforts were going on, however, there was a great deal
of excitement in the science community as the scientific results of the mission became
availabIe. A press release on 9 October 1997 reporred that "Mars was appearing more
like a planet that was very Earth-like in its infancy, with weathering processes and flow-
ing water that created a variety of rock types, and a warmer atmosphere that generated
clouds, winds, and seasonal cycles."

At an 8 October 1997 press briefing at JPL, Mars Pathfinder project scientist Dr. Matthew
Golombek observed, "What the data are telling us is that the planet appears to have
water-worn rock conglomerates, sand, and surface features that were created by liquid
water. If," he added, "with more study, these rocks turn out to be made of composite
materials, that would have required liquid water flowing on the surface to round the
edges in pebbles we see on the surface or explain how they were embedded in larger
rocks. That would be a very important finding."

Golombek also stressed the amount of differentiation--or heating, cooling, and recy-
cling of crustal materials-that appeared to have taken place on Mars. "We're seeing a
much greater degree of differentiation-the process by which heavier elements sink to
the center of the planet while lighter elements rise to the surface-than we previously
thought, and very clear evidence that liquid water was stable at one time in Mars'[s]
past. Water, of course, is the very ingredient that is necessary to support life," he added,
"and that le2ds to the $64,000 question: Are we alone in the universe? Did life ever
develop on Mars? If so, what happened to it and, if not, why not?"
Despite the most intense efforts by the DSN and spacecraft controllers to detect the
presence of a downlink, however weak or off-frequency it may have been, no downlink
could be found. There was conjecture that, without the heat generated by the battery-
powered transmissions, the spacecrafr temperature would fall below its operating limits
and the spacecraft computer that controlled the spacecraft transmission times would no
longer operate correctly. Whatever the cause, the downlink was never recovered, and h r -
ther efforts were discontinued in mid-October.

At the time of the last downlink, the Lander had operated for nearly 3 times its design
lifetime of 30 days, and the Sojourner Rover had operated for 12 times its design life-
time of 7 days. After the 4 July landing, it returned 2.6 billion bits of data, which
included more than 16,000 images from the Lander and 550 images from the Rover,
as well as more than 15 chemical analyses of rocks and extensive data on winds and
other weather factors. All MPF requirements of the DSN for telecommunications and
data acquisition support had been fulfilled.

The cost of designing and building the Pathfinder had been $171 million; that of the
Sojourner Rover, $25 million. Together, their combined costs would have been a mere
round-off error in the $3 billion (1997 dollars) cost of NASA's previous mission to Mars,
the Viking Landers in 1976. In that sense, the MPF mission had accomplished one of
its primary objectives. As Brian Muirhead, Mars Pathfinder project manager at JPL,
declared at the end of the mission, "This mission has demonstrated a reliable and low-
cost system for placing science payloads on the surface of Mars. We've validated NASA's
commitment to low-cost planetary exploration."

NASA seemed well satisfied. A few weeks later, NASA Administrator Daniel S. Goldin
recognized the efforts of all involved with the mission in a formal press release: "I want
to thank the many talented men and women at NASA for making the mission such a
phenomenal success. It embodies the spirit of NASA and serves as a model for hture
missions that are faster, better, and cheaper. Today, NASA's Pathfinder team should take
a bow, because America is giving them a standing ovation for a stellar performance."

At the same time that it was dealing with the highly visible Mars Pathfmder events
described above, the DSN was also handing the arrival and aerobraking maneuvers of
the Mars Global Surveyor and arrayed operations for the Galileo encounter with the
Jupiter satellite Callisto. While these dramatic events were unfolding at JPL and around
the Network, another deep space mission was being readied for launch at Cape Canaveral.
The Cassini Saturn Orbiter with the Huygens Titan Probe had been successfully mated
with the Titan IVICentaur at the Cape Canaveral Air Force Station, and launch was
planned for 13 October.

Cassini

The Cassini mission to Saturn and Titan was a cooperative endeavor by NASA, the
European Space Agency (ESA), and the Italian Space Agency (MI). Its genesis can be
traced to the early 1980s, when the international science community agreed upon a
combined Saturn OrbiterITitan Probe mission as the next logical step in the exploration
of the solar system, following the Galileo mission to Jupiter. It was seen as a follow-on
mission to the brief reconnaissance of Saturn, which had been carried out by the Pioneer
I 1 spacecraft in 1979 and the Vyuger 1 and 2 encounters of 1980 and 1981. It would
be an enterprise that would span nearly thirty years from its initial vision to its com-
pletion in the year 2008.

The Cassini mission was designed to make a detailed scientific study of Saturn; its icy
moons, magnetosphere, and rings; and the satellite Titan. Observations relative to the
first four science objectives were to be made by twelve science instruments carried on
an orbiting spacecraft (Cdssinz].The Titan observations would be made by six science
instruments onboard a probe (Huygens), which would be carried to the vicinity of Saturn
by Cassini before being released to descend to the surface of the satellite. In addition to
these experiments, the DSN Project System would be used to conduct at least three
gravitational wave searches en route to Saturn and to carry out numerous occultation
experiments with the planet, moons, and rings during the orbital phase of the mission.

Scheduled for launch in October 1997 using a Titan IVICentaur launch vehicle, the
spacecraft would require a unique set of four gravity-assist maneuvers to arrive at Saturn
almost seven years later, in July 2004. The gravity-assists would be provided by close
flybys of Venus in April 1998 and June 1999. It would then fly by Earth in August
1999 and Jupiter in December 2000. The Cassini interplanetary trajectory with the four
gravity-assist maneuvers is shown in Figure 6-7.

To reduce costs, the development of many mission operations capabilities was deferred
until after launch. Except for the three gravitational wave experiments, there were no
plans to acquire science data during the cruise phase or gravity-assist maneuvers. Spacecraft
operations during the interplanetary cruise would be centralized at JPL. During the
orbital phase of the mission, a system of distributed science operations would allow sci-
entists to operate their instruments from their home institutions with the minimum
interaction necessary to collect their data.
f \

SaturnArrival
1 July 2004
.
.":_.
, ..'.a
,. \
.
\
\

-----------------
Interplanetary T@ectory
L J

Figure 6-7. Cassini interplanetary trajectory.

Casini was a three-axis-stabilized spacecrafi, t!ne main body of which was formed by a
stack consisting of a lower equipment module, a propulsion module, an upper equip-
ment module, and the HGA. The various science instruments were mounted on pallets
and, rogether with the Huygens Probe and RTGs, were attached to the stack at appro-
priate points. The spacecraft contained twelve engineering subsystems in addition to irs
cornpIemenr of twelve science instruments.

At launch, the combined weight of Casini, the Huygens Probe, fuel, and the launch
vehicle adaptor was 5,712 kilograms, making it the heaviest interplanetary spacecraft
ever Launched by NASA. More than half of the launch weight was contributed by the
propellant needed for the 74-minute main engine burn that would inject Casini into
Saturn orbit. The spacecraft stood 6.8 meters high, with a maximum diameter of 4
meters due to the HGA. The major components of the spacecraft are identified in the
diagram shown in Figure 6-8,

The design of the Ckssini spacecraft was the end result of extensive uadeoff studies that
considered cost, mass, reliability, the availability of hardware, and past experience. Moving
parts were eliminated from the spacecraft wherever possible. Science instruments and
the high-gain antenna, which replaced the deployable type of antenna used on Galileo,
were permanently attached to the spacecraft, their pointing functions performed by
rotation of the entire spacecraft. Tape recorders were replaced with solid-state recorders
Figure 6-8. Cassini spacecraft with Huygens Probe.

and mechanical gyroscopes were replaced with hemispherical resonator gyros. The high-
gain antenna would be used for the short-duration (2.5 hours), S-band radio-relay link
between Cassini and the Huygens Probe, as well as for the permanent radio link between
Cmsinni and Earth.

Uplink and downlink communications on X-band were provided by the Radio Frequency
Subsystem. A Radio Frequency Instrument Subsystem provided unmodulated RF carri-
ers on the Ka-band and S-band for project experiments only during the cruise and orbital
phases. The X-band uplink carried command data in the range 7.8 bps to 500 bps and
sequential ranging modulation. The Ka-band uplink was unmodulated. Borh Radio
Frequency Subsystems included redundant Deep Space Transponders (DSTs). Originally
developed as a joint DSNIFlight project task at JPL in 1990 for use on X-band only,
the DST was first flown on the NEAR and Mars Pathfinder spacecraft in 1996.' It was
later upgraded to include Ka-band and was flown for the first time in that configura-
tion on Cassini.'

The X-band downlink carried telemetry data with a wide range of data rates, fiom 5
bps to 249 kilobits per second. The telemetry data was encoded with Reed-Solomon
(255, 223) coding concatenated with Viterbi (7, 112) or (15, 116) coding prior to trans-
mission. It could also be transmitted in uncoded form. In addition to telemetry the
downlink carried ranging modulation.

The versatile design of the DST provided X-band, Ka-band and S-band downlinks that
could be either coherent or noncoherent with the X-band uplink. The Ka-band down-
link could be coherent with, or noncoherent with, the Ka-band uplink. This allowed
the downlinks to be optimized for the various project investigations under a wide range
of conditions.

The Antenna Subsystem included the 4-meter-diameter, narrow-beam-width HGA that

was designed for transmitting and receiving on all four communication bands, and two
wide-beam-width, low-gain antennas that could receive and transmit on X-band only.

Working with Cassini mission designers, the Telecommunications and Mission Services
(TMS) manager, RonaId L. Gillette, negotiated tracking and data acquisition services for
the mission that generally fell within the existing (1997) or planned (2002) capability of
the Network. The 70-m, 34-m BWG, and 34-m HEF subnets would be required singly
or arrayed at various times during the mission. The initial acquisition and early cruise
requirements, when the spacecraft would be using its low-gain antennas, would be met
without the need for any new implementation. By 2001, however, the 20-kW X-band
uplinks on the 34-m HEF antennas would not provide adequate uplink margin for com-
manding and X-band transmitters, and exciters would be needed on the three 70-m
antennas. The DSN agreed to provide this new capability. The DSN also agreed to pro-
vide a Ka-band uplink and downlink at DSS 25 for the project experiments, particularly
the first gravitational wave experiment of 40 days' duration in 2001. Where the research
and development installation at DSS 13 had been equipped with an 80-watt uansmit-
ter, this first fully operational Ka-band capability in the Network would have a 4-kW
transmitter and provide simultaneous X-band and Ka-band downlink capability.

A new and innovative approach to orbital operations planning was expected to effect
economies in the scde of DSN coverage required to support this critical phase of the
mission. For the purposes of operational simplicity, DSN coverage was classified into
two categories: "high activity," when science opportunities would be most intensive, such
as targeted satellite flybys and Saturn periapses; and "low activity," when fewer science
opportunities would permit lower data return volume. The science community had agreed
that a data volume return of 4.0 gigabits per day for high activity periods and 1.0 giga-
bit per day for low activity periods would be adequate to accomplish their science objectives.
Because Saturn would be above the plane of the ecliptic during the period of orbital
operations (2004-08)) the daily view periods for the DSN antennas in the southern
hemisphere wodd be of shorter duration, and the received downlink signal-to-noise ratio
about 1.5 dB less, than they would be for the DSN antennas in the Northern Hemisphere.
For this reason, the DSN agreed to provide coverage, most of the time, for antennas in
the Northern Hemisphere, Goldstone, and Madrid. High-activity data return would be
accommodated by the 70-rn antennas and 70-ml34-m arrayed antennas in the Northern
Hemisphere. Low-activity periods would be covered by the 34-m Northern Hemisphere
antennas or the 70-m and 34-m Southern Hemisphere antennas, if necessary.

The sustainable data rates on the X-band downlink during Saturn orbital operations also
depended on the position of Earth in its orbit around the Sun and on the change in
elevation of the spacecraft as it made its daily pass over the tracking station. When all
these variables were considered, it was estimated that the downlink data rates would vary
over the range 14 to 166 hlobics per second.

In a joint effort to reduce the complexity and cost of daily orbital operations, the fac-
tors just described were used by the DSN and Cassini mission planners to develop a
data return strategy based on an average pass length of about 9 hours. This strategy
optimized the downlink capability with the defined science activity period while limit-
ing the data rate changes on the telemetry downlink to two or three per station pass.
Once developed, each strategy would remain in effect for a fixed period, such as 90
days, during which time data rates and pass lengths would not be changed. This arrange-
ment promised significant economies in mission planning and operations effort for both
DSN and the Cassini projects.

Afeer the Whire House gave approval to ~roceed,Cdssini was launched without incident
from Cape Canaveral, Florida, on 15 October 1997. White House approval to launch
was required by presidential directive due to the use of RTG units to power the Cdssini
spacecraft. This matter had been the subject of a number of environmental protest demon-
strations at Cape Canaveral in the weeks immediately preceding launch.

The %tan NIB and Centaur Launch Vehicle fhctioned perfectly "Right on the money,"
observed Cassini program manager Richard J. Spehalski. The sequence of post-launch events,
including initial acquisition of the downlink by Canberra stations DSS 46 and DSS 45, was
executed precisely as planned. First reports from the Casini mission controllers indicated that
the angular deviation of the trajectory from its design value was insignrficant at better than
0.004 degrees. Mission plans had called for an expected adjustment in the post-launch tra-
jectory of 26 meters per second, but night data derived fiom the DSN radio metric data in
the first fa^ hours of two-way tracking showed that a correction of only one or two meters
per second would be necessary to correct for the small injection error. Early DSN telemetry
from CDSCC also showed that all subsystems on the spacecrafi, including the solid-state
recorder, were operating normally "I can't recall a launch as perfect as this one," said Cassini
mission director Chris Jones; "everythmg we see is within predictions, with no fdures."

the spacecrafi high-gain antenna pointed toward the Sun and the X-band uplink and
downlink established through the low-gain antennas, the DSN began the first 30 days of
continuous tracking on the 34-rn antenna subnet at the rate of 21 passes per week. During
this time, mission controllers took advantage of the high data rate (14.2 kbps) available on
the downlink, due to the short spacecraft-to-Earth range and the relatively quiescent (no sci-
ence) state of the spacecraft, to carry out a series of engineering and science instrument
maintenance activities. These activities included a checkout of the Huygens Probe about one
week after launch. Because of the near-perfect post-launch injection conditions, the first tra-
jectory correction maneuver on 9 November required only a 2.7-meters-per-second adjustment
in spacecraft velocity to fine-tune its night path. Real-time telemetry at 948 bps fiom DSS
54 allowed the spacecraft controllers to observe the 34.4-second main-engine burn in progress
and provided reassuring evidence that this critical subsystem was hctioning correctly.

By early January 1998, the C'sini flight team had completed all spacecraft: activities planned
for the early phase of the mission. Onboard software, stored in the command and data
subsystem, had been directing spacecraft activities as planned, and the spacecrafi thrusters
were maintaining the spacecraft in its correct attitude in space. For the next 14 months,
Cassini would fly with the HGA facing the Sun to shield the spacecraft from the intense
solar radiation characteristic of the inner solar system. Throughout this period, the uplinks
and downlinks with the DSN would be maintained through either of the two low-gain
antennas, depending on the relative geometry of Earth, the Sun, and the spacecraft. Downlink
data rates and weekly DSN tracking schedules were adjusted to meet the predefined, returned-
data volume requirements. With all its subsystems working perfectly, and periodic
instrument maintenance and spacecraft housekeeping activities dominating its routine sched-
ule, Cassini moved steadily around its orbit toward an appointment with Venus on 26
April 1998, the first milestone on its seven-year voyage to Saturn8
EARTH-ORBITING MISSIONS

By 1997, almost two-thirds of the missions that composed the total DSN mission set
were missions of the Earth-orbital type. These are listed in the mission set for h e Cassini
Era, shown in Figure 6-1. Together with the Deep Space missions, these accounted for
a Network operations load of 30 to 35 mission events being handled by the NOCC in
a typical 24-hour period in 1997. In addition to the real-time mission events, a typical
operations day for the NOCC would include two to five data playback sessions with
the complexes.

Some of these spacecraft, like the Solar Heliospheric Observatory (Soho), Polar, Wind,
and Geotail, were part of the continuing International Solar and Terrestrial Physics
(ISTP) program and had been supported by the DSN for several years. Others, like
Hotbird, were supported by the DSN only for the launch and early orbit period, a
few days at most. Some were technology satellites; others were communications satel-
lites; and some were solely for scientific purposes. Except for the cooperative Space
VLBI mission HALCA and the reimbursable ESA mission Hotbird, a11 were NASA
missions. However, the cooperative and reimbursable options for DSN support of
future, non-NASA missions remained open. There were a number of other NASA
missions, not listed in this set, for which the DSN assumed responsibility for pro-
viding tracking support in the event of a spacecraft emergency. Appropriate authorities
and procedures to invoke support of this kind were permanently in place in the
NOCC.

Spacecraft in low-Earth orbit (apogee less than 12,000 Ism) were supported on the
26-m or 34-m antennas, depending on their requirements for downlink data recep-
tion and handling. Those in high-Earth orbit usually required 34-m HEF or 34-m
BWG antennas.

The HMCA spacecraft was a special case that warrants further discussion because of
the new technology associated with its entry into the DSN. It was developed for the
Japanese VLBI Space Observatory Program (VSOP) and was known by that name prior
to its actual launch. Following a successful launch in February 1997, it became HALCA
and participated in an international cooperative Space VLBI program for which the DSN
provided some of the Earth-based antenna support.
 VLBI OBSERVATORY
SPACE PROGRAM

The Space Very Long Baseline Interferometry (SVLBI) program was a two-mission, coop-
erative venture among NASA, Japan's Institute of Space and Astronautical Sciences (ISAS),
and the Russian Astro Space Center (ASC). NASA participated in both missions with
funding for the DSN; the National Radio Astronomy Observatory (NRAO) Very Long
Baseline Array at Socorro, New Mexico; and the NRAO tracking station at Green Bank,
West Virginia.

As early as 1983, Gerry S. Levy and his JPL associates had shown that the resolution
of an array of Earth-based VLBI antennas could be greatly increased by means of a spe-
cial spacecraft which would act as an extension of the Earth-based "radio ba~eline."~
The
SVLBI program applied and extended these basic ideas to improve the resolution of
radio-astronomy images of celestial radio sources by extending the use of VLBI tech-
nology into space. The SVLBI missions therefore had a space component and a ground
component, both of which heavily involved the DSN.

DSN involvement in the space component of the SVLBI project included the imple-
mentation of a new subnetwork of 11-meter antennas with Ku-band uplink and downlink
capability. This effort is described later in this chapter. DSN participation in the ground
component of SVLBI as a co-observer with the DSN 70-meter antennas is discussed in
the Radio Astronomy section of this book.

The space component of the SVLBI program was to consist of two spacecraft: the VSOP
(VLBI Space Obsematory Proram), provided by ISAS and launched in 1997, and the
RddioAstron, provided by ASC and due to be launched several years later. Each space-
craft would carry a radio-telescope antenna and suitable receiving equipment for making
radio-astronomy observations from a high-Earth orbit. The DSN's 11-meter antennas
would provide continuous uacking support to supplement the limited tracking resources
of the other participants in the program.

The VSOP antenna was capable of receiving signals in the three standard radio astron-
omy bands, L-band (1.6 GHz), C-band (5.0 GHz), and Ku-band (22 GHz). The digitized
radio astronomy data, together with time synchronization signals, was to be transmit-
ted to Earth over a Ku-band downlink at 128 Mbps. A Ku-band uplink carried time
and frequency reference signals derived from a Hydrogen-maser at the Earth-based receiv-
ing stations to the spacecraft for use as a frequency reference. Spacecrafi command and
control functions and generation of radio metric tracking data used an S-band uplink
and downlink that was completely independent of the Ku-band link. These functions
were exercised from the Thgoshima Control Center and were not part of the DSN
responsibility.

The VSOP spacecraft was successfully launched on 12 Februaq 1997, on the then-new
ISAS Mark-V launch vehicle, from the Kagoshima Space Center in Japan.lo The satel-
lite was renamed HALCA after launch. Following initial acquisition of the S-band downlink
by the 26-m station at Goldstone, the Ku-band uplinks and downlinks were activated
and communications were established with the 11-meter subnet consisting of DSS 23
(Goldstone), DSS 33 (Canberra), and DSS 53 (Madrid). Three perigee-raise maneuvers,
done from 15 to 21 February, were used to place HALCA in a high elliptical orbit with
an apogee of 21,400 krn, perigee of 560 krn, and orbit period of 6.3 hours. This high
elliptical orbit had in inclination of 31 degrees (the latitude of the Kagoshima Space
Center) and would enable VLBI observations on baselines up to three times longer than
those achievable with Earth-based antennas. First interference fringes between HALCA
and ground telescopes were found at the Mitaka, Japan, correlator in May, and at the
Penticton, Canada, and Socorro, New Mexico, correlators in June 1997.

Tracking station passes varied in length from 10 minutes to as much as 5 hours, depend-
ing on the position of the spacecraft in its orbit at the time of the station pass. The
short-period orbit provided two or three passes per day at each compIex. The DSN was
not required to perform any processing of the downlink data. It was simply recorded
on special wide-band VLBI-compatible tapes and shipped to designated locations in the
United States, Japan, and Canada. Special wide-band correlators at these centers processed
the spacecraft observations together with corresponding data from the ground observa-
tories, including the DSN 70-meter antennas, to produce the desired images of celestial
radio sources.
THE NETWORK

COMPLEXES
AND ANTENNAS

By 1997, when the Cassini Era became a reality in the DSN, the total number of anten-
nas operated by the DSN around the globe had increased to rwenry-four, twelve at
Goldstone and six at each of the Madrid and Canberra Complexes. The DSN also
engaged in spacecraft mission operations with many other institutions around the world,
each of which had one or more anrennas of its own.

The sites of the DSN antennas and Signal Processing Centers were identified according
to their geographical location, as shown in Figure 6-9. The sites of related antennas such
as Weilheim, Germany; Usuda, Japan; and Parkes, Australia, were identified in h e same
manner.

It was convenient to describe the Network in terms of several subnets accordmg to the met-
ric diamerer of the anrennas composing it. There were a 70-meter subnet, two 34-m subnets,
a 26-m subnet, an 11-m subnet, and several unique antennas such as DSS 13 and DSS
17 at Goldstone. An inventory of antennas in the DSN in 1997 is given in Figures 6-10,
6- 11, and 6- 12 for the Goldstone, Canberra, and Madrid complexes, respectively.

These tables show the various uplinks and downlinks that were available in 1997 to
support space operations on each of the antennas. Comprehensive though this listing
was, substantial additions were planned for it in the immediate future.

f \

Location Identifier for DSN and Related Sites

Number Range Geographic Description and Position*

00-29 JPUGoldstone and the VIA, Socorro, NM
30-49 Pacific Basin: Between L = 180 and L = 270
50-69 European Basin: Between L = 345 and L = 90
70-89 Atlantic Basin: Between L = 270 and L = 345
90-99 Miscellaneous
100-999 GPS and Future Use

*Note: L = Longitude in degrees, measured east from zero longitude.

i 1

Figure 6-9. Location identifiers for DSN and related sites.

F \

Goldstone Deep Space Communications Complex, 1997

DSS Diam., Type Uplinks Downlinks Operational

12 34 m, STD none none Decom'd. 2/96
13 34 m, BWG, RBD S, X, Ka S, X, Ka, Ku 6/90
14 70 m S S, x 64 m/70 rn, 5/88
15 34 m, HEF X s, x 8/84
16 26 m S S 1/67
23 I 1 m, OVLBl X X, Ku 2/96
24 34 m, BWG x, s s, x 1 2/95
25 34 m, BWG s, x S, X, Ka 1 8/96
26 34 m, BWG X X 8/96
27 34 m, BWG, HS S S 7/95
28 34 rn,BWG, HS S S 1O/OO

i /

Figure 6-10. Goldstone Deep Space Communications Complex, 1997.

r \

Canberra Deep Space Communications Complex: 1999

DSS Diam., Type Uplinks Downlinks Operational

33 11 m, OVLBl X, Ku X, Ku 6/96
34 34 m, BWG X s, x 11/96
42 34 m, STD S s, x Decom'd. 12/99
43 70 m S s, x 9/87
45 34 m, HEF X S, x 12/84
46 26 m S S 12/83

i 1

Figure 6-1 1. Canberra Deep Space Communications Complex, 1997.

c >

Madrid Deep Space Communications Complex, 1997

DSS Diam., Type Uplinks Downlinks Operational

53 11 m, OVLBI X, Ku X, Ku 6/96
54 34 m, BWG X s, x 10/97
61 34 m, STD S s, x Decom'd. 12/99
63 70 m S S/X 64 m/70 m, 7/87
65 34 m, HEF X S/X 4/87
66 26 m S S I 2/84

\ 1

Figure 6-1 2. Madrid Deep Space Communications Complex, 1997.

The 70-m subnet would be provided with X-band uplinks for first use with Cassini,
beginning with DSS 14 in 2000. Already, plans were being made to install an X-/Ka-
band dichroic plate on the DSS 25 BWG antenna. This would be the first srep in
providing an operational Ka-band downlink for use on the Deep Space-1 mission and
for project use with Cassini in 2000. The DSS 26 BWG antenna was being modified
to demonstrate the more stringent antenna pointing capability required for Ka-band
operation.

Driven by the need to reduce maintenance and operations costs, the closure of the
26-m subnet by the end of Fiscal Year 2001 had been under consideration since 1996.
Several alternatives, including the automation of routine telemetry, command, and
monitoring functions, the transfer of low-Earth-orbiting spacecrafi to NASA's Wallops
Island facility, and eventually to a commercial ground network, were being actively
pursued in 1997.

Wgys of making more effective use of DSN antenna time were demonstrated in sever-
al areas in 1996 and 1997. The "Multiple Spacecraft Per Antenna" concept was
demonstrated effectively at Goldstone in September 1997 with the successfd tracking
of Mars Ghbal Surveyor and Mars Pathfinder from a single antenna. Both spacecraft were
at Mars and both lay within the beamwidth of a single 34-m BWG antenna (0.063
degrees at X-band). Several hundred commands were transmitted to each spacecraft, and
good telemetry and two-way Doppler was obtained fiom each spacecraft. The opera-
tional advantages of this mode of operation were clearly demonstrated, particularly those
related to switching the uplink from one frequency to another, and subsequent acqui-
sition of individual spacecraft receivers. The DSN planned to develop &is technique for
future use as the Mars Exploration initiative would provide more opportunities for effect-
ing economy in usage of DSN antennas.

For many years, it had been the practice to make regular, biweekly Earth motion and
clock synchronization measurements between complexes by using two 70-m antennas
simultaneously to observe a celestial radio source. Comparison of time off-sets in the
observables provided a measure of the differences in time and frequency between the
complexes. In 1997, with the addition of updated VLBI equipment and precision data
from GPS receivers at the compIexes, it became possible to accomplish the same biweek-
ly measurements with rwo 34-m HEF antennas rather &an two 70-m antennas.

In November 1997, the Future Programs Council at JPL heard a briefing on the future
of deep space communications. The scope of the briefing included optical as well as
conventional radio communications, but in the context of existing DSN antennas, it
recommended that "to meet the increased needs of a growing mission set, NASA should
rapidly deploy Ka-band capability throughout the existing DSN as a low[-]cost path
which would allow future missions to double their data return while cutting their (anten-
na) contact time in hdE"

In time, by 1997, that statement seemed to indicate the direction of future develop-
ment for antennas in the Network.

Along with design concepts for beam waveguide antennas, the advantages of Ka-band
communication links and the application of these ideas to hture uplinks and downlinks
for the DSN were discussed by Joel G. Smith, Robert C. Clauss, and James W. Layland
in three companion reports in 1987.11These studies showed that a spacecraft-to-Earth
downlink on Ka-band (32 GHz) could carry telemetry data at 3 to 10 times the data
rate possible with an X-band downlink, given the same spacecraft transmitter weight,
antenna size, and power requirements. This enhancement in downlink performance derived
from increased antenna gain at higher frequencies (shorter wavelengths), but was reduced
somewhat by other factors such as atmospheric and antenna losses and susceptibility to
wet or foggy weather. A more indepth study of the tradeoffs associated with &-band
operation in tile DSN was carried out in 1988. This study described a scenario for estab-
lishing Ka-band as the primary Deep Space communications frequency of the future
and proposed a baseline plan for reaching that goal.I2
As the first two steps toward realization of Ka-band communication links in the future
DSN, it was proposed that a Ka-band receiving capability should be added to the new
research and development beam-waveguide antenna being built at DSS 13, and that a
Ka-band beacon experiment (KABLE) should be planned for the Mars Observer space-
craft due to be launched to Mars in September 1992.13

This methodology had been used to introduce the X-band into the Network in 1973
as an S-/X-band experiment on the Mariner 10 missions to Venus and Mercury. The
Ka-band transponder for the Mars Observer spacecraft was the first step in a parallel
development plan for spacecraft dwice and component technology proposed by Arthur
L. Riley.14

The BWG antenna at DSS 13 was completed by mid-1990, as described in the pre-
ceding chapter, and its Ka-band capability was evaluated over the period July 1990
through January 1991.15 The microwave feed package used on this antenna for KABLE
consisted of an WKa-band dichroic plate and two low-noise amplifiers (LNAs), one for
X-band and one for Ka-band. While the X-band LNA used a simple high-electron mobil-
ity transistor (HEMT) to achieve a system temperature of 37 kelvin, the ultra-LNA for
Ka-band required the development of JPCs first Ka-band cavity maser to achieve a sys-
tem temperature of 5 kelvin.16The Ka-band and X-band downlink signals, amplified by
the LNAs, were converted to intermediate frequency and delivered to separate digital
Advanced Receivers (ARX) for tracking, demodulation, and delivery to the telemetry
decoders. A data-handling terminal carried out the telemetry recording and delivery tasks.
Of the two advanced receivers, one was the unit that was being developed for the Galileo
application in the DGT; the other was specially built for use at DSS 13 and would
remain at the station. A special Doppler tuner was required for this unit to handle the
Doppler jerk (rate of acceleration) expected on the Ka-band downlink when MGS wodd
be in orbit around Mars.

The Mars Observer spacecrafi was equipped with a special 33-rnilliwatt Ka-band trans-
mitter and used the concave surface of the subreflector on the spacecrafts high-gain
antenna as an antenna for the Ka-band downlink. To keep the Ka-band downlink coher-
ent with the X-band uplink and downlink, its frequency, 33.66 GHz, was obtained by
simply multiplying the X-band downlink frequency (8.4150 GHz) by four. Telemetry
and ranging modulation were provided on both downlinks. At this time, fiber-optic
cables were being introduced for intra-site communications and transmission of highly
stable reference and timing signals between antennas on each complex. Full advantage
was taken of this new facility to connect DSS 13 to the SPC 10 Control Room some
20 miles away, where the existing ranging and Doppler equipment was located.
The major objectives for W L E included a telemetry demonstration, a ranging demon-
stration, and a tracking experiment. The telemetry demonstration would prove the feasibility
of a telemetry link at Ka-band by collecting a minimum of one million bits of data, at
250 bits per second, with minimal errors. The ranging demonstration would follow the
telemetry demonstration and illustrate the correct demodulation of DSN ranging codes
at Ka-band. The tracking experiment would develop a database of Ka-band downlink
statistics including weather, system noise, temperature, and antenna performance for use
in the future development of Ka-band capability in the Network.

By an unfortunate coincidence, the telemetry demonstration took place during some of

the worst weather conditions ever experienced in the Goldstone area. Heavy rains and
dense fog were daily occurrences. This type of weather has a major degradation effect
on Ka-band link performance and proved to be a limiting factor in reaching the data
collection goal for telemetry. Although the ranging demonstration was limited to two
passes only, due to various equipment problems and to the weather, it yielded a SUE-
cient number of good ranging points for favorable comparison with theoretical
performance. Equipment problems and bad weather also limited the amount of track-
ing data that was gathered. Nevertheless, sufficient Ka-band tracking data was retrieved
to show a potential improvement of 4 to 7 dB over X-band performance under equiv-
alent conditions.

Ironically, on 21 August 1993, just when the weather was improving and the equip-
ment problems were being resolved, uplink and downlink communications with the
Mars Observer spacecraft were lost. Later investigations led to the conclusion that the
spacecrafi had been destroyed during the engine pressurization sequence prior to the
maneuver for Mars orbit insertion.

Despite its abrupt ending, sufficient data was obtained to declare the KAELE effort par-
t i d y successfd. The data that had been collected was thorougMy analyzed, and it ultimately
provided a strong justification for a repeat experiment as soon as another oppormnity
presented itself.17

That opportunity would not be long in coming. In November 1996, when the Mars
Global Surveyor spacecrafi was launched on a mission similar to that of the ill-fated Mars
Observer, it carried with it a much-improved version of the original Ka-band technolo-
gy experiment called KABLE-11. AIthough W L E - I 1 was identical in concept to the
earlier experiment, it benefited from several changes to the spacecrafi and ground de-
ments of the link. The Ka-band transmitter power had been increased to approximately
1 watt, and a change had been made to the way in which the frequency of the down-
link Ka-band signal was derived in the spacecraft transponder. This change effectively
placed the downlink signal frequency in the center of the receiving pass-band, rather
than to one side of it, as was the case in Kable-I. Since the microwave components in
the DSS 13 BWG antenna, particularly the dichroic plate, were optimized to receive at
this band-center frequency, substantial loss in the downlink signal margin was avoided.
At DSS 13, a new seven-element Ka-band array feed was used to receive the downlink
and point the 34-m antenna with rnillidegree precision; Block V receivers replaced the
earlier ARX developmental versions; and numerous other improvements related to oper-
ational reliability were made.

When the MGS spacecraft was pointed to Earth in mid-January 1997, strong signals
were immediately received at DSS 13 and simultaneous X-/&-band tracking was car-
ried out on six occasions thereafter. The array feed tracked rhe spacecraft within 1 &degree
in angle, and both carriers were tracked simultaneously to generate Doppler data by an
experimental tone-tracker. Measurements were in good agreement with theory. The teleme-
try experiment began in February with an improved low-noise tracking feed which not
only incorporated a multimode coupler to generate pointing error signals for driving the
antenna pointing system, but also was cooled to reduce the system noise temperature.

On 21 March, the KABLE-I1 experimenters reported that a major objective had been
successfully accomplished. Telemetry data received over the 1 watt, Ka-band link at 2
kbps had been compared bit-by-bit with data received over the 25-watt X-band down-
link. All 12 million contiguous bits were in agreement. There were no errors. This event
marked the first error-fiee telemetry reception at Ki-band in the DSN.

At this time, the Ka-band array feed, which had been replaced by the multimode-cou-
pXer feed at DSS 13, was installed and tested on the DSS 14 antenna in preparation
for the start of Ka-band efficiency tests of the 70-m antenna in May. This capability
would be needed for Cassini in the years ahead.

In mid-April, real-time telemetry data received from MGS over the Ka-band and the
X-band links was presented to the MGS project operations team at JPL for evaluation.
Dual-frequency tracking and ranging measurements were also made at this time. The
experimental Ka-band downlink was received at DSS 13, while an operational X-band
uplink and downlink was maintained at DSS 15. Both downlinks had been converted
to 300-MHz IF frequency and transferred over fiber-optic circuits to SPC 10 for teleme-
try and radio-metric data processing and transmission to JPL.
The KABLE-I? telemetry demonstrations with MGS at DSS 13 continued as scheduled
through 8 May 1997. The successful demonstrations of Ka-band technology occasioned
by the two KABLE experiments lent impetus to its transfer to the operational Network
beginning with DSS 14 and DSS 25 at Goldstone.

The first two missions that would be carrying Ka-band downlinks to demonstrate an
operational rather than an experimental capability in the DSN were the New Millennium
mission DS-1 and the Cassini mission to Saturn.

DS-1 was a technology demonstration mission with secondary scientific objectives.

Primary communications would be conducted on X-band, and DSS 25 was to pro-
vide a Ka-band receive capability as an operational technology demonstration. DSS
13 would also provide some analytid support. DS-1 would not be launched before
the middle of 1998.

Cassini also planned to use X-band for its primary telemetry and radio metric support,
but, additionally, it planned an extensive project program that would be accomplished
using a combination of S-, X-, and Ka-band downlinks and X-band and Ka-band uplinks.
The Ka-band uplink and downlink were required to support the Cassini Gravitational
Wave Experiment in 200 1.

While the DSN telecommunications engineers had been pursuing a path leading to an
operational Ka-band capability in the Network, spacecraft telecommunications engineers
had been pursuing a somewhat parallel path in developing an X-bandlKa-band transpon-
der for future spacecraft. The NASA Deep Space Transponder (DST) had been in
development since 1991 and was intended for first use with Cassini, at that time planned
for launch in 1996. It would replace the existing S-band transponders, which, with an
external S-band-to-X-band down-converter to provide the X-band capability, had been
used on all previous planetary spacecraft. Using conventional spacecraft technology, the
DST contained an automatic phase-tracking receiver for X-band uplink only and an X-
band exciter to drive redundant downlink X-band transmitters. It provided appropriate
reference frequency signals to devices external to the DST for the generation of inde-
pendent Ka-band and S-band downlink signals.

The Cassini spacecraft was launched in October 1997, carrying the DST into space for
the first time. Cdssini would depend on the DST for its primary spacecraft communi-
cations links with the DSN; uplink and downlink on X-band for telemetry, command,
and radio metric data; and downlink only on Ka-band and S-band for project and grav-
ity-wave experiments.
The Ka-band transponder, also known as the Small Deep Space Transponder, had been
under development since 1995 and was designed specifically for Ka-band uplink and
downlink operation.l8 It included a selectable X-band capability that could be switched
in or out as required, as well as several state-of-the-art components, including sampling
mixers, a Ka-band dielectric resonator oscillator, and microwave monolithic integrated
circuits (MMICs) to perform the functions of up and down frequency conversion, mod-
ulation, and arnpliication. It was proposed for first use as a telecommunications evaluation
experiment on DS-1, to be launched in 1998.

Ka-band was definitely on the move in the DSN in 1997, but its arrival as an opera-
tional capability lay several years in the future.

ORBITING
VLBI SUBNEWORK

As discussed earlier in this chapter, Space Very Long Baseline Interferometry (SVEBI)
was a cooperative project among NASA, ISAS, and ASC, representing space agencies
from the United States, Japan, and Russia. ISAS provided the VSOP (HALCA) space-
craft, ASC was to provide the RadioAstron spacecraft, and the Russian tracking stations
and NASA provided the SVLBI subnet of four tracking stations. One of these stations
was the existing 14-meter NRAO Green Bank station in West Virginia. The other three
antennas were built by NASA specially for this cooperative project and, at their com-
pletion in 1996, became part of the Deep Space Network. One antenna, with its unique
electronics, was installed at each complex. Within the DSN, they were referred to col-
lectively as the Orbiting VLBI (OVLBI) 11-meter subnet and designated DSS 23, DSS
33, and DSS 53 for Goldstone, Canberra, and Madrid, respectively. However, from the
overall project viewpoint, they remained part of the SVLBI subnet.

The main functions of 11-meter stations were to automatically acquire and track the
spacecrafi X-band and Ku-band downlinks, generate and provide rwo-way integrated
Doppler data to the DSN Navigation Subsystem; provide uplink to the spacecraft with
a phase-stable reference frequency tied to the DSN Hydrogen Maser Frequency stan-
dard at each site; and receive, demodulate, and record the 144 Mbps downlink telemetry
data in special VLBA andlor S-2 data format. The spacecraft data would be delivered
to the project for later correlation with data fiom the co-observing ground telescopes.
These functions were to be performed autonomously, without the need for operator
intervention except to change recording tapes, at a frequency of 4 passes per day 7 days
per week. Interfaes with the existing DSN systems would provide the new antennas
with frequency and time references; monitor and control functions; and the necessary
schedule, Gequency prediction, and antenna pointing and orbital state vector informa-
tion to enable autonomous operation.19

Initial studies in 1990 had shown that considerable cost savings to NASA would result
from the use of commercidy available resources, rather than JPL "in-house" resources,
to provide the antennas. A comprehensive set of technical design requirements was pre-
pared in 1991, and after due process, a contract for the design and implementation of
three high-precision 11-meter antennas and associated elecuonics and software as a "turn-
key" package was awarded to Scientific-Adanta (S-A) Corporation in September 1991.

JPL was responsible for selecting and preparing the sites and providing access roads,
antenna foundation, buried cable conduits, air-conditioned buildings, power, water, and
related facilities.

Although the design was based on an S-A standard, the 11-meter production antenna,
a great deal of new development was necessary to meet the extremely tight design require-
ments for frequency and phase stability required for the intended OVLBI appli~ation.~~

The functional block diagram in Figure 6-13 shows the interfaces between the 11-m
antenna in the OVLBI subnet and various systems of the Network.

Each station in the new subnetwork contained two new subsystems: the OVLBI Tracking
Subsystem (OTS) and the OVLBI Data Subsystem (DTS).

The OVLBI Tracking Subsystem included the 11-meter-diameter antenna, on an

azirnuthlelevation mount, with a third "tilt" axis to allow for a direct overhead pass with-
out loss of tracking data due to the keyhole effect inherent in AZIEL mounts. Concentric,
five-horn monopulse feeds in a cassegrain subreflector configuration handled uplink and
downlink signals at X-band, while the center horn also acted as a common receiving
aperture for Ku-band. The Ku-band monopulse error signals were derived from a TE21
mode coupler at the base of the horn. Low-noise amplifiers, microwave components,
tracking receivers, antenna-control servos and data receivers and demodulators provided
for antenna pointing and downlink data reception and demodulation. Frequency con-
verters and local oscillators driven from computers in the Data Tracking Subsystem
provided the important Doppler compensation and signal coherence functions. The uplink
for Radiohtron was provided by a small 5-watt solid-state X-band transmitter, while
the uplink for VSOP was provided by a 0.5-watt Ku-band transmitter.
r \

DSS :
I

I
SPC j GCF
f
NOCC
1

I
PROJ

L /'

Figure 6-13. OVLBl Subnetwork functional block diagram. The following abbreviations are used
in this diagram:
FTS Frequency and Timing System
DMC DSN Monitor and Control
NOCC Network Operations Control Center
NMC NOCC Monitor and Control
NSS NOCC Support Subsystem
NAV NOCC Navigation
DSS Deep Space Station
SPC Signal Processing Center
GCF Ground Communications Facility
PROJ Flight Project Mission Operations Teams
S/S Subsystem

On these antennas, special measures were taken to minimize the effect of temperature
variations on the phase stabiliry of the X-band and Ku-band signals. Transmission of
these signals between the antenna and the Control Room was carried on fiber-optic
cables, and all freqllency conversion equipment was located in the temperature-regulat-
ed Control Room. The temperature of microwave and optical equipment mounted on
the antenna itself was maintained to within *1.0 degrees of normal by an independent,
proportional air-conditioning system. To fitrther enhance phase stability, the fiber-optic
cables between the antenna and the Control Room were buried six feet below ground
level. Except for routine maintenance and changing of recording tapes, the station was
fully operated by a ground station control computer as part of the OTS.

The OVLBI data subsystem contained various components that were peculiar to the
VSOP and RadioAstron missions. A Doppler compensation computer drove the local
oscillators in the Tracking Subsystem, which programmed out the expected Doppler fre-
quency shift on both the uplink and downlink. A Doppler extractor integrated the
Doppler frequency shift and provided this data to the DSN Navigation Subsystem for
calculating the precise spacecraft position. Afier decoding, the telemetry data containing
the VLBI observations from the spacecraft were recorded digitally in a special Very Long
Baseline Array (VLBA) format. An additional interface to provide the VLBI data to the
Canadian S-2 recording subsystem was added later.

In August 1995, four years after the award of the contract, factory acceptance testing
of the first two systems for Goldstone and Canberra was successfully completed at Scientific-
Atlanta. The hardware performed well, and performance against critical requirements
relating to antenna gain-to-system noise temperature (GIT), phase stability, Doppler
compensation, tracking accuracy, and telemetry bit error rate was we11 within specifica-
tion. There were, however, a significant number of software anomalies which would be
corrected over the following weeks. The Goldstone system was dismantled and prepared
for shipping, while the Canberra system was retained on the antenna test range to pro-
vide a test-bed for clearing the software anomalies. The Canberra system was shipped a
few weeks later, but the system for Madrid was delayed by shipping and other prob-
lems and did not reach the Madrid Complex until May 1996.

By then, the Goldstone system (DSS 23) was acquiring and tracking the SURFSAT
spacecraft automatically without operator intervention at X-band, as well as generating
two-way Doppler on a regular basis. The SURFSAT tracks were continued throughout
the year to gain experience, while the Doppler data was evaluated for quality and accu-
racy and the last of the software problems were worked on.

A photograph of DSS 23 at the Goldstone Apollo Site with the DSS 24 BWG anten-
na in the background is shown in Figure 6-14.

In May and june, while the Canberra system (DSS 33) was completing its onsite sys-
tem acceptance testing, assembly of the antenna for DSS 53 began at the Madrid site.
By the end of 1996, all three I l-m antennas had been completed and most major prob-
Figure 6-14. DSS 23, II-meter OVtBl antenna at Goldstone, 1996.

Iems with the DPRG and the Doppler and phase generation software had been solved.
Full end-to-end capability between a ground station (DSS 33) and the VLBI correlator
at Socorro, New Mexico, was demonstrated for the first time in December, using X-
band data from the SURFSAT spacecraft. At that time, none of the 11-m stations had
yet been exposed to a Ku-band downlink, and the VSOP launch, set for February 1997,
was then just over two months away.

In mid-January 1997, the SURFSAT spacecraft was commanded to switch its downlink
from X-band to Ku-band to give the new 11-rn stations some experience in routine
tracking at the frequency they would be seeing for the VSOP downlink. For the next
four weeks, all 11-m stations engaged in routine tracking of the SURFSAT Ku-band
downlink, while final s o b a r e deliveries were installed and preparations were made for
the start of VSOP inflight operations.

The VSOP launch took place on 12 February 1997, and the first Ku-band track of
HALCA, as it became known after launch, was made at DSS 23 on 20 March. With
minor exceptions, 11-m subnet support for HALCA operations in 1997 was satisfacto-
ry and is described earlier in this chapter. In the longer term flight operations environ-
ment, however, the I I-m subnet required signLficant operauons and engineering effort
from JPL and the Complexes to meet most of its design specifications.

Each of the facilities that compose the DSN had a unique set of policies and proce-
dures to cover emergencies and disasters that might occur in its own particular locality.
The response of each fscility to an emergency situation is determined by its own set of
standard emergency procedures. Should the capability of the overall Network be affect-
ed by the local emergency or the steps taken to deal with it, the Network situation
would be handled by the Operations Chief at the NOCC using whatever alternate
resources were at his disposal at the time. However, the loss of the NOCC at JPL, for
whatever reason, would critically impair the ability of the entire Network to continue
support for inflight spacecraf?.

Aware of this situation, the Galileo project had, in 1995, already developed a capabili-
ty to conduct the critical "insertion into Jupiter orbit" phase of its mission from Goldstone
in the event that a natural disaster precluded operations at JPL. A key feature of rhis
fairly basic emergency center was the ability to send commands to the Galileo spacecrafk
from any DSN 70-m antenna, independent of support from the regular Galileo mis-
sion control center at JPL.

As a consequence of the growing perception of Network vulnerability to the occurrence

of a disaster at JPL, NASA Headquarters soon requested that DSN develop and pub-
lish an oficial Network Disaster Plan. The Network Disaster Plan recognized the potential
for disablement of the NOCC due to earthquake, bomb threat, fue, or industrial dis-
pute, and established appropriate procedures to be followed by Network personnel. The
procedures involved the relocation of basic NOCC and mission control activities to an
alternate location at the Goldstone Deep Space Communications Complex (GDSCC),
located about 120 air-miles northwest of JPL. There, in 1996, a permanent new facil-
ity called the Emergency Control Center (ECC) was implemented, with voice and basic
data communication capabilities, specifically for the purpose of continuing Network and
mission control activities on a limited scale in the went of a disaster at JPL.2'

In its original concept, the ECC was intended to accommodate only the DSN per-
sonnel involved with Network control. It quickly became evident, however, that the
spacecraft and mission control facilities that were accommodated in the same Space
Flight Operations Facility (SFOF) building at JPL as the NOCC were equally vulner-
able to the effects of a disaster. Under the direction of the Telecommunications and
Mission Operations Directorate (TMOD), the design was broadened to include these
latter capabilities in addition to those required for Network control, and the facility
became the TMOD ECC.

In its basic form, the TMOD ECC consisted of a central router which could be con-
nected, on demand, to the Goldstone Signal Processing Center (SPC) via an ethernet
link and to the SPCs at Canberra and Madrid via permanently available and dedicated
voiceldata communication circuits. By design, the circuits to both of the overseas sites
bypassed the high-earthquake-risk Los Angeles area. In the event of an emergency, the
Complexes would be instructed to patch their communications circuits to alternate com-
munications routers which would redirect their communications traffic to the ECC at
Goldstone rather than to the NOCC at JPL.

W~thinthe ECC, an ethernet LAN connected the central router to several Hewlett-
Packard and Sun Microsystems high-capacity workstations arranged in a 10-base T-hub
configuration. Most of these were for the use of spacecraft and mission controllers in
carrying out navigation, command, telemetry data handling, and sequence generation
functions. Network control functions and the generation and dissemination of predic-
tion data products for the antenna pointing, radio metric, telemetry, and frequency
control subsystem were carried out at a large VAX workstation. A DSN data base was
kept current by frequent downloading of current data from the active Network support
subsystem in the NOCC. Likewise, flight project software for the spacecrafi and mis-
sion workstations was updated on a regular basis to reflect all changes and revisions.

Funding for implementation of the ECC task was in place by mid-1996, and respon-
sibilities for development, planning, and implementation were assigned. The ECC was
to be located at h e Echo site at Goldstone in the vacant control room of the DSS 12
antenna, which had been decommissioned from DSN service earlier in 1996. Under
pressure to be ready prior to the Cassini launch in October 1997, progress was rapid.
By early 1997, the workstations were installed in the ECC and testing had begun. Late
deliveries of the remaining equipment and a "freeze" on all new implementation activ-
ity in the Network during the Mars PathjnAr landing and Mars surface operations in
July delayed the completion of the work until mid-August, when Cassini and Ulysses
mission controllers were able to begin evaluation of the ECC for flight operations sup-
port. The first tests with Canberra and Madrid used 28-kilobits-per-second communication
circuits routed through JPL, since the special 5 12-kilobits-per-second direct circuits between
the ECC and the overseas sites had not yet been made available to JPL.
Meanwhile, the Cassini project had developed a new launch "constraint" requirement
for an independent back-up for the existing Cassini Launch Operations Center at JPL.
Fortunately, the 28-kilobits-per-second circuits codd be used to satisfj the newly imposed
the Cassini launch criteria in the event of further delay in the availability of the wide-
band 5 12-kilobits-per-second circuits. A fay weeks later, these circuits did become available,
just in time to fully test the system with the overseas sites and to satisfy the new Cassini
requirements.

The T M O D Emergency Control Center was reported to be operational for all JPL flight
projects on 6 October 1997. Cmsini was successfully launched nine days later, with the
ECC providing back-up for its prime Launch Operations Center at JPL.

The ECC was first used for full in-flight mission support by GaZiZeo at the end of
October 1997. A complete track, including planning tasks and spacecraft command
functions, was run successfully from the ECC by GaliIeo flight operations personnel.
Similar demonstrations were planned to familiarize Ulysses, Voyager, and Mars Global
Surveyor operations personnel with the capabilities (and limitations) of the new facil-
ity. In due course, these demonstrations came into regular scheduled use as a means
of keeping the ECC in a state of readiness and exercising its potential users in emer-
gency awareness.
OTHER ASPECTS

The years 19961997 brought to a close the traditional form of mission operations that
the DSN had employed to support all of NASA's planetary missions, and many of its
lunar and Earth-orbiting missions, for almost forty years. New, more efficient, and less
costly ways of conducting mission operations, including a move to Ka-band frequencies,
were being introduced in the Network to match NASA's call for "faster, better, cheap-
er" missions. In the opening years of the Cassini Era, Mars Global Surveyor and Mars
Pathfinder, the first planetary missions designed to new guidelines, were launched on a
fast track to Mars. Cmini, the last planetary spacecraft based on the old principles, began
its long mission to Saturn.

Whiie the proper business of the DSN was always the support of NASA's inflight space-
craft missions, "mission operations" represented only one aspect of DSN history over
the period covered by this book. Technology, science, and organization, although not
directly involved with "mission operations", were also essential to the growth, scientific
stature, and fiscal well-being of the Network during that time.

To this point, the narrative has emphasized "mission operations" in terms of five major
eras of planetary missions from 1957 though 1997. With that completed through the
Cassini Era, we turn now to consider other aspects of DSN history, beginning with the
growth of technology in the DSN.
Endnotes

1. Jet Propulsion Laboratory, Telecommunications and Mission Operations

Directorate, "Bridging the Space Frontier" (30 January 1998). http://deepspace.jpl.
nasa.gov/920/pu
bli~/922~stratplan/.

2. Voyager Web site: http://www.jpl.

nasd.gou/reledses/98/vgr21Z html.

3. NEAR Web site: http://nea~.jhuapl.

edul

4. Mars Global Surveyor Web site: http://mars.jpLnaa.gov/mgsl.

5. Mars Pathfinder Web sites: http://ww~.jpl.nasd.~ov/marsnews,

http://mpfiww.jpl.
nasa.gov/mpj7engineering.html.

6. N. R. Mysoor, J. D. Perret, and A. W. Kermode, "Design Concepts and Performance

of a NASA X-band (7162 MHz18415 MHz) Transponder for Deep-Space
Spacecraft Applications," TDA Progress Report PR 42-104, October-December
1990 (15 February 1991), pp. 247-56.

7. N. R. Mysoor, J. l? Lane, S. Kayalar and A. W. Kermode, "Performance of a Ka-band

Transponder Breadboard for Deep-Space Applications," TDA Progress Report PR
42-122, April-June 1995, (15 August 1995), pp. 175-83.

8. Cassini Web site: http://www.jpl.nasa.gov/cassinii

9. VSO P Web site: http://www.

vsop.isas.ac.jp.

10. J. S. Ulvestad, C. D. Edwards, and R. l? Linfield, "Very Long Baseline

Interferometry Using a Radio Telescope in Earth Orbit," TDA Progress Report PR
42-88, October-December 1986 (15 February 1987), pp. 1-10.

11. J. G. Smith, "Ka-band (32-GHz) Downlink Capability for Deep Space

Communications," TDA Progress Report PR 42-88, October-December 1986 (15
February 1987), pp. 96-103; J. W. Layland and J. G. Smith, "A Growth Path for
Deep Space Communications," TDA Progress Report PR 42-88, October-
December 1986 (15 February 1987), pp. 120-25; R. C. Clauss and J. G. Smith,
"Beam Waveguides in the Deep Space Network," TDA Progress Report PR 42-88,
October-December 1986 (15 February 1987), pp. 174-82.
12. J. W. Layland, "Ka-band Study-1988, Final Report," 890-212, JPL Document D-
60 15, (Pasadena, California: Jet Propulsion Laboratory, 15 February 1989).

13. J. G. Smith, "Proposed Upgrade of the Deep Space Network Research and
Development Station," TDA Progress Report PR 42-88, October-December 1986
(15 February 1987), pp. 158-63; A. L. Riley, D. M. Hansen, A. Mileant, and R. W.
Hartop, "A Ka-band (32 GHz) Beacon Link Experiment (KABLE) With Mars
Observer," TDA Progress Report PR 42-88, October-December 1986 (15 February
1987), pp. 14147.

14. A. L. Riley, "Ka-band (32 GHz) Spacecraft Development Plan," TDA Progress
Report PR 42-88, October-December 1986 (15 February 1987), pp. 164-73.

15. S. D. Slobin, T. Y. Otoshi, M. J. Britcliffe, L. S. Alvarez, S. R. Stewart, and M. M.

Franco, "Efficiency Calibration of the DSS 13 34-m Diameter Beam Waveguide
Antenna at 8.45 and 32 GHZ," TDA Progress Report PR 42-106, April-June 1991
(15 August 1991), pp. 283-89.

16. J. Shell and R. B. Quinn, 'A Dual-Cavity Ruby Maser for the Ka-band Link
Experiment," TDA Progress Report PR 42-1 16, October-December 1993 (15
February 1994), pp. 53-70.

17. T. A. Rebold, A. Kwok, G. E. Wood, and S. Butman, "The Mars Observer Ka-band
Link Experiment," TDA Progress Report PR 41-1 17, January-March 1994 (15 May
1994), pp. 250-82.

18. N. R. Mysoor, J. I? Lane, S. Kayalar, and A. W. Kermode, "Performance of a Ka-

band Transponder Breadboard for Deep-Space Applications," TDA Progress Report
PR 42-122, April-June 1995 (15 August 1995), pp. 175-83.

19. J. Ovnick, "DSN Orbiting VLBI Subnet, Task Implementation Plan," JPL
Document D-7787, 803-120, Vol. 1 (August 1990).

20. J. Ovnick, "DSN Orbiting VLBI Subnet, Design Requirements," JPL Document D-
9619, DM5 15606 (November 1993).

21. Jet Propulsion Laboratory, "Deep Space Network Standard Operations Plan," DSN
Document 841, Vol. 1, DSN Operations (31 July 1996), p. 14-1.
THE DSN TECHNOLOGY PROGRAM

No history of the DSN would be complete without a fdl appreciation of its contribu-
tion of advanced technology to the successful development of the Network. The wellspring
of new and innovative ideas for increasing the existing capability of the Network; for
improving reliabiity, operability, or cost effectiveness; or for enabling recovery from poten-
tial mission-threatening situations has resided, from the very beginning of the Network's
history, in a strong program of advanced technology, research, and development. Known
by various names through the years, the numerous elements of the advanced technolo-
gy program were formally identified as a complete entity and placed under the direction
of the newly established Technology Ofice of the Telecommunications and Mission
Operations Directorate (TMOD) in 1997. It was known as the TMOD Technology
Program. The program scope was further expanded to include technologies relevant to
the full end-to-end Deep Space Mission System (DSMS). This expansion included flight
components of the physical communications links (data services) and added mission
services activities like protocol developments, mission planning and execution tools, sci-
ence data visualization, and the merging of tools and autonomy.

These disciplines covered almost every aspect of DSN-related technology. Theoretical work
in any of them could always be complemented with experimental work in well-equipped
laboratories and machine shops at JPL and verified with field measurements at DSS 13,
which was maintained primarily for research and development (R&D) purposes.

Prior to 1997, the program was known most generally as the DSN Advanced Systems
Program and covered antennas, low-noise amplifiers, Network signal processing, frequency
and timing, radio metric tracking, navigation, Network automation, atmospheric prop-
agation. It also involved the evolution of the program research and dwelopment tracking
station, DSS 13, at Goldstone. These were all elements of gound-based systems. In
1985, a second Systems Development Program was added to cover telecommunications
systems analysis, spacecraft radio communication systems, and inflight demonstrations
of new uplink and downlink communications systems.

Due in no small part to the unremitting efforts of Nicholas A. Renzetti to ensure that
the results of these efforts were properly documented, a continuous record of the early
work carried out under the DSN Technology Program was published by JPL in a series
of progress reports. Starting in 1969 as Volume I1 of the JPL Space Programs Summary
37-XX series, it became the DSN Progress Reports in 1971, TDA-PRs in 1980, and
the TMO-PRs in 1998. In August 1994 the series began publication on the World Wide
Web. An index of all articles published after 1970 also became available online at that
time. (See the appendix at the back of the book.)

The technology underlying many of the major engineering changes, discussed earlier in
the context of DSN operations support for flight missions, was described by James W.
Layland and Lawrence L. Rauch in 1997 (see section 1 of the appendix). As those
authors made dear, most of the progress in antenna design; uplink and downlink per-
formance improvement; and the application of radio metric techniques to spacecraft
navigation, radio science, radio astronomy, and radar astronomy originated in the DSN
program of advanced research. These new technologies were later made available to the
DSN engineering groups for implementation into the operational Network.

This phase of DSN endeavor is illustrated by tracing the advancement of technology in

those key areas. The material that follows is derived from the work of Layland and
Rauch and is intended for the general reader. The technical reader is referred to the
appendix at the back of the book, which cites significant publications associated with
each topic.

THEGREATANTENNASOF THE DSN

A photograph of the NASA Deep Space Communications Complex near Canberra,

Australia, is shown in Figure 7-1. The largest antennas in the photo are quasi-parabolic
reflector antennas, one with a diameter of 70 meters; the others, 34 meters. These were
used for deep space mission support, while the other, smaller antennas, 26-meter or
9-meter antennas, provided tracking support for Earth-orbiting missions.

Each of the antennas has what is termed a Cassegrain configuration with a secondary
reflector mounted on the center axis just below the focal point of the primary reflector
"dish." The secondary reflector serves to relocate the focal point closer to the surface of
the main dish and thus establish a more convenient location for the low-noise ampli-
fiers, receivers, and powerful transmitters.

The efficiency with which parabolic antennas collect radio signals from distant spacecraft
is degraded to some extent-by radio noise radiated by the ~ a r t hterrain surrounding the
antenna. This form of radio noise is known scientifically as "black body radiation" and
is a physical characteristic of all material with a temperature above absolute zero (-273
degrees Celsius or 0 kelvin). The magnitude of noise power radiated by a material body
depends on its temperature. While it is incredibly small at the temperature of typical
Earth surfaces, it is enormous at the temperature of the Sun, for instance. All parabolic
Figure 7-1. The Deep Space Communications Complex in Canberra, Australia

radio antennas have sidelobes in the beam pattern, and the magnitudes of those side-
lobes can increase as the antennas deviate from the ideal shape. The shape of the beam
and its sidelobes is essentially the same whether the antenna is used for transmitting or
receiving signals. The sidelobes are analogous to the circles of light surrounding the main
beam of a flashlight when it is held close to a reflecting surface.

When an antenna is used for transmitting a strong signal to a spacecraft, the sidelobes are
of no great consequence. However, when the antenna is receiving a weak signal from a dis-
tant spacecraft, particularly at or near the horizon, the "Earth noise" picked up by the
sidelobes is sufficient to obscure the spacecraft signal in the extremely sensitive receivers used
on the DSN antennas. This interference produces errors in the data stream being delivered
to the spacecraft engineers and scientists, and it may cause the DSN receivers and anten-
nas to lose the spacecraft signal altogether, in which case the data stream is completely lost.

The earliest antennas in the DSN (Figure 7-2) were of commercial design and were par-
abolic in shape. Then, as now, the actual efficiency of the antenna represented a compromise
between maximum signal-gathering capability and minimum susceptibility to radio noise
picked up from the surrounding Earth.
Figure 7-2. An early parabolic design of a DSN antenna.

Improved technology that could reduce the sidelobes and increase the signal collection
capability (gain) of fitwe DSN antennas appeared in the Advanced Systems Program
in the early 1970s. The new technology was based on a "dual-shape" design wherein
the surface shapes of both the primary and secondary reflectors were modified to illu-
minate the slightly reshaped "quasi-parabolic" surface of the main reflector more
uniformly. However, it was not until the 1980s, when the first 34-meter high-efficien-
cy antennas were built, that the new "dual-shape" design saw operational service in the
Network (Figure 7-3).

These antennas were needed by the DSN for support of the Voyager spacecraft in their
tour of the outer planets. At the time, the DSN was in transition from the lower, less
capable S-band (2.3 GHz) operating frequency, for which the early spacecraft and anten-
nas were designed, to X-band (8.4 GHz), a higher, more capable operating frequency.
When X-band technology became available in the DSN, largely as a result of work in
the Advanced Systems Program, all the later spacecraft and DSN antennas were designed
to operate at X-band frequencies. These antennas were therefore the first to be opti-
mized for performance at X-band.
Figure 7-3.The quasi-parabolic 34-m high-efficiency (HEF) antenna, DSS 15.

As the Vyager 2 spacecraft headed outward toward Neptune, it was recognized that an
increased signal-collecting area was needed on Earth to effectively support this unique
science opportunity. The DSN's largest antennas at the time were 64-m parabolas of the
original design. Calculations showed that the best investment of scarce construction funds
would be to modify these antennas, using the dual-shape design, to expand their diam-
eter to 70 m. It was also apparent that the upgraded large antennas would benefit the
planned Galileo and Magellan missions.

Completed in time for support of Vyager 2 at Neptune, the 70-m enhancement proj-
ect (Figure 7-4) resulted in an increase of more than 60 percent in the effective collecting
area of these large antennas. Fully half of the increase was attributed to the dual-shape
design, a product of the DSN Advanced Systems Program.
Figure 7-4. A 70-meter antenna with dual-shape reflector design.
By the end of the century, several new 34-m antennas employing the dual-shaped reflec-
tor design in conjunction with beam waveguide (BWG) techniques had been constructed
for operational use in the Network. The dual-shaped reflector design enhanced the radio
performance of the antenna, while the beam waveguide codguration greatly facilitated
maintenance and operation of the microwave receivers and transmitters. Using a series
of additional secondary reflectors to relocate the focal point into a stationary room below
the main dish, the BWG design feature enabled these critical components to be mount-
ed in a f ~ e denvironment rather than in the more conventional type of moving and
tipping enclosure mounted on the antenna itself.

Beam waveguide antennas had been used for many years in Earth communications satel-
lite terminals where ease of maintenance and operation outweighed
- the consideration of
losses introduced into the microwave signal path by the additional microwave-reflecting
mirrors. For deep space applications, however, where received signal power levels were
orders of magnitude smaller, any losses in the signal path were a matter of great con-
cern, and the losses associated with BWG designs kept such antennas out of consideration
for DSN purposes for many years. Researchers in the DSN Advanced Systems Program
nevertheless pursued the idea of BWG antennas for the DSN and, by 1985, were ready
to conduct a collaborative experiment with the Japanese Institute for Space and
Aeronautical Sciences (ISAS) using its new 64-m beam waveguide antenna at Usuda,
Japan. Using one of the DSN's low-noise microwave receivers installed on the Usuda
antenna to receive a signal from the International Cometary Explorer (ICE) spacecraft,
the researchers made very precise measurements of the microwave losses, or degradation,
of the downlink signal.

The results of the experiment were very surprising. The measured losses, attributable to
the BWG design, were much smaller than expected, exhibiting similar performance at
zenith and better performance at low-elevation angles; they confirmed the efficacy of the
BWG configuration.

Encouraged by this field demonstration, researchers sponsored by the Advanced Systems

Program moved forward with the construction of a prototype BWG antenna for poten-
tial application in the Network.

This new prototype BWG antenna was built at the Venus site at Goldstone and replaced
the aging 26-m antenna that had served for many years as a field test site for technol-
ogy research and development (R&D) programs. The designers used microwave optics
analysis software, an evolving product of the Advanced Systems Program, to optimize
the antenna for operation over a wide range of current and future DSN operating fie-
Figure 7-5. A 34-meter beam waveguide antenna a t Goldstone Technology Development Site,
DSS 13.

quency bands. When completed, the antenna successllly demonstrated its ability to
operate effectively at S-band, X-band, and Ka-band (approximately 2, 8, and 32 GHz,
respectively). Figure 7-5 shows the completed BWG antenna, and Figure 7-6 shows the
interior of the equipment room below the antenna structure.
Figure 7-6. Stationary equipment room below the BWG antenna.

The various frequencies and modes of operation for the BWG antenna were selected by
rotating the single microwave mirror at the center of this room. Lessons learned by
Advanced Systems Program personnel in the construction and evaluation of this anten-
na were incorporated into the design of the operational BWG antennas for the Network,
with the result that the performance of these somewhat exceeded that of the prototype,
especially at the lower frequencies. A selection of technical references related to great
antennas of the DSN can be found in section 2 of the appendix at the end of the book.
The large antennas of the DSN are used for transmission of radio signals carrying instruc-
tions and data to the spacecraft, as well as for reception of signals. Getting data to distant
spacecraft safely and successfully requires that substantial power be transmitted from the
ground and directed in a narrow beam at the spacecraft. For most "normal" situations,
the compatible design of spacecraft and the DSN is such that power of about 2 kW to
20 kW is adequate. However, situations in space are not always normal. Unexpected
events can redirect a spacecraft's main antenna away from Earth, leaving only a low-gain
or omnidirectional antenna capable of receiving anything from Earth. Transmitter power
of up to 400 kW at S-band can be sent from the 70-m antenna during attempts to
regain contact with a spacecraft in such an emergency situation.

The initial design and evaluation of R&D models of the high-power transmitters and
their associated instrumentation was carried out under the Advanced Systems Program.
Much of the essential field testing was carried out as part of the planetary radar exper-
iments. This cooperative and productive arrangement provided a realistic environment
for testing without exposing an inflight spacecraft to an operational unqualif~eduplink
transmission. Later, DSN engineers implemented fully qualified operational versions of
these transmitters in the Network at all sites.

The pointing of the narrow forward link signal to the spacecraft is critical, especially
when making initial contact without having received a signal for reference, as is typical
in emergency situations. The beamwidth of the signal from the 70-m antenna at S-band
is about 0.030 degrees, while that of the 34-m antenna at X-band is about 0.017 degrees.
Achieving blind pointing to that precision requires a thorough understanding of the
mechanics of the antenna, including the effects of gravity and wind on the dish and
specifics of the antenna bearing and positioning mechanisms, as well as knowledge of
the spacecraft and antenna positions, atmospheric refraction, and other interferences.

Forward link data delivered to a spacecraft, if incorrectly interpreted, have the potential
for causing that spacecraft to take undesirable actions, including some that could result
in an emergency situation for the spacecraft. To guard against that possibility, the for-
ward link signal is coded with additional redundant data that allow the spacecraft data
system to detect or correct any corruption in that signal. Operating on the presump-
tion that it is always better to take no action than to take an erroneous one, the forward
link decoding accepts only data sets for which the probability of error is extremely small,
and it discards those that cannot be trusted. A selection of technical references related
to the forward commandldata link (uplink) can be found in the section 3 of the appen-
dix at the end of the book.

Throughout the Network, the stations use the same antennas for both the forward link
and the return data-link signals. Because the strength of a signal decreases as the square
of the distance it must travel, these two signals may differ in strength by a factor of
loz4in a single DSN antenna. Isolating the return signal path from interference by the

Figure 7-7. Dichroic (frequency-selective) reflector developed under Advanced Systems Program.
much stronger forward signal poses a significant technical challenge. Normally, these two
signals differ somewhat in frequency, so at least a part of this isolation can be accom-
plished via dichroic or frequency-selective reflectors. These reflectors (Figure 7-7) consist
of periodic arrays of metallicldielectric elements tuned for the specific frequencies that
either reflect or pass the incident radiation. These devices must not only be frequency-
selective, but they must also be designed to minimize the addition of extraneous radio
noise picked up from the antenna and its surroundings, which would corrupt the incred-
ibly weak signals collected by the antenna from the desired radio source in deep space.

The DSN Advanced Systems Program developed the prototypes for almost all the reflec-
tors of this type in current use in the Network. As an adjunct to this work, powerful
microwave analytical tools that can be used to & i design details for almost any con-
ceivable dichroic reflector applicable to the frequency bands of the DSN were also developed
under the Program.

Low-Noise Amplifiers

The typical return data link signal is incredibly small and must be arnplzed before it
can be processed and the data itself reconstructed. The low-noise amplifiers that reside
in the antennas of the DSN are the most sophisticated in the world and provide this
amplification while adding the least amount of noise of any such devices.

Known as traveling-wave masers (TWMs), the quietest (in terms of adding radio noise)
of these operation> devices amplify signals that are propagated along the length of a
tuned ruby crystal. Noise in a TWM depends upon the physical temperature of the
crystal, and those in operation in the DSN operate in a liquid helium bath at 4.2 kelvin.
(Zero kelvin is equivalent to a temperature of minus 273.18 Celsius. Therefore, the tem-
perature of the helium bath is equivalent to approximately minus 269 Celsius.) The
practical amplifiers for the DSN were invented by researchers at the University of Michigan,
and early development of these amplifiers was carried out under the DSN Advanced
Systems Program, as were many improvements throughout the Network's history. The
quietest amplifiers in the world today (Figure 7-8), which operate at a physical tem-
perature of 1.2 kelvin, were developed by the DSN Advanced Systems Program and
demonstrated at the Technology Development Field Test Site, DSS 13.
 Some of the low-noise amplifiers
in the DSN today are not
TWMs, but a special kind of
transistor amplifier (Figure 7-9)
using high-electron mobility
transistors (HEMTs) in amplifiers
cooled to a physical temperature
of about 15 kelvin.

Developed initially at the Uni-

versity of California at Berkeley,
such amplifiers were quickly
adopted by the scientific com-
munity for radio astronomy
applications. This, in turn,
spawned the JPL development
work that was carried out via col-
laboration invo!ving JPL and the
DSN Advanced Systems Program,
radio astronomers at the National
Radio Astronomy Observatory
(NRAO), and device developers at
General Electric. This work built
upon progress in the commercial
sector with uncooled transistor
amplifiers. In the 2-GHz DSN
band, the cooled HEMT ampli-
Figure 7-8. Ultra-low-noise amplifier (ULNA) at DSS 13. fiers are almost as noise-free as the
corresponding TWMs, and the
refrigeration equipment needed to
cool the HEMTs to 15 kelvin is much less troublesome than that for the TWMs. Primarily
for this reason, current development efforts in the DSN Advanced Technology area are
focused on improving the noise performance of the HEMT amplifiers for the higher DSN
frequency bands.

The first DSN application of the cooled HEMT amplifiers came with the outfitting of
the NRAO Very Large Array (VLA) in Socorro, New Mexico, for collaborative support
of the Voyager-Neptune Encounter. The VLA was designed for mapping radio emissions
from distant stars and galaxies and consists of 27 antennas, each 25 meters in diame-
Figure 7-9. High-electron mobility transistor (HEMT) low-noise microwave amplifier.

ter, arranged in a tri-axial configuration. Within the funding constraints, only a small
part of the VLA could be ouditted with TWMs, whereas HEMTs for the entire array
were affordable and were expected to give an equivalent sensitivity for the combined full
array. In actuality, technical progress with the HEMTs under the Advanced Systems
Program during the several years taken to build and deploy the needed X-band (8-GHz)
amplifiers resulted in better performance for the fully equipped VLA than would have
been possible with the VLA partially equipped with the more expensive TWMs. Since
that time, many of the DSN operational antennas have had cooled HEMT amplifiers
installed for the 2-GHz and 8-GHz bands. A selection of technical references related to
return telemetry (downlink) can be found in section 4 of the appendix at the end of
the book.
Phase-Lock Tracking

Once the first stages of processing in the low-noise amplifiers are completed, there are
still many transformations needed to convert the radio signal sent from a spacecraft into
a replica of the data stream originating on that spacecraft. Some of these transforma-
tions are by nature analog and linear; others are digital with discrete quantification. All
must be performed with virtually no loss in fidelity in order for the resultant data stream
to be of practical use.

Typically, the downlink signal consists of a narrow-band "residual carrier" sine wave,
together with a symmetric pair of modulation sidebands, each of which carries a
replica of the spacecraft data. (Specifics of the signal values vary greatly, but are not
essential for this general discussion.) If this signal is cross-correlated with a pure iden-
tical copy of the residual carrier, the two sidebands will fold together, creating a
low-frequency signal that contains a cleaner replica of the spacecraft data than either
sideband alone. Of course, such a pure copy of the carrier signal does not already
exist; it must be created, typically via an adaptive narrow-band filter known as a
phase-locked loop. The recreated carrier reference is thus used to extract the side-
bands. The strength of the resultant data signal is diminished to the extent that this
local carrier reference fails to be an identical copy of the received residual carrier.
Noise in the spectral neighborhood of the received residual carrier and dynamic vari-
ations in the phase of the carrier itself limit the ability to phase-lock the local reference
to it.

These dynamic variations are due predominantly to the Doppler effect in play between
a distant spacecraft and the DSN antenna on the surface of a spinning Earth. The vari-
ations interfere with the return data link process, but they themselves provide for a radio
location function. Over the years, the DSN Advanced Systems Program has contributed
significantly to the design for the phase-locked loops and to the knowledge of phase-
coherent communications, and thus to the performance of the operational DSN. A
selection of technical references related to phase-lock trachng can be found in section
5 of the appendix at the end of the book.

Synchronization and Detection

Further steps in converting a spacecraft signal into a replica of the spacecraft data stream
are accomplished by averaging the signal over brief intervals of time that correspond to
each symbol (or bit) transmitted from the spacecraft and by sampling these averages to
create a sequence of numbers, often referred to as a "symbol stream." These averages
must be precisely synchronized with the transitions in the signal as sent from the space-
craft so that each contains as much as possible of the desired symbol and as little as
possible of the adjacent ones. Usually, a subcarrier, or secondary carrier, is employed to
shape the spectrum of the spacecraft signal, and it must be phase-tracked and removed
prior to the final processing of the data itself. The Network contains several different
generations of equipment that perform this stage of processing. Designs for all of these
have their roots in the products of the DSN Advanced Systems Program. The oldest
current equipment is of a design developed in the late 1960s by a partnership between
the DSN Advanced Systems Program and the DSN implementation programs. This
equipment is mostly analog in nature and, while still effective, is subject to component
value shifts with time and temperature, and thus requires periodic tending and adjust-
ments to maintain desired performance.

As digital devices became faster and more complex, it became possible to develop dig-
ital equipment that could perform this stage of signal processing. Digital demodulation
techniques were demonstrated by the Advanced Systems Program in the early 1970s in
an all-digital ranging system. Similar techniques were subsequently employed for data
detection in the second generation of the Demodulator-Synchronizer Assembly. A selec-
tion of technid references related to data synchronization and detection can be found
in section 6 of the appendix at the end of the book.

A Digital Receiver

Rapid evolution of digital technology in the 1980s led researchers to explore the appli-
cation of digital techniques to various complex processes found in receiving systems such
as those used in the Network. The processes of filtering, detection, and phase-lock car-
rier tracking, formerly based on analog techniques, were prime candidates for the new
digital technology.

In this context, the Advanced Systems Program supported the development of an all-
digital
- receiver for Network use. Known as the Advanced Receiver (ARX),the
developmental model embodied most of these new ideas and demonstrated capabili-
ties far exceeding those of the conventional analog receivers then installed throughout
the Network.

Encouraged by the performance of the laboratory model, an engineering prototype was

built and installed for evaluation in an operational environment at the Canberra, Australia,
Complex. Tests with the very weak signal from the Pioneer I0 spacecraft, then approach-
ing the limits of the current DSN receiving capability, confirmed the designer's
 performance and, incidentally, significantly extend-
ed the working l i e of that spacecraft.

As a result of these tests, the DSN decided to pro-

ceed with the implementation of a new operational
receiver for the Network that would be based on
the design techniques demonstrated by the ARX.
The new operational equipment, designated the
Block V receiver (BVR), would include all of the
functions of the existing receiver in addition to sev-
eral other data processing - functions, such as
demodulation and synchronization, formerly car-
ried out in separate units.

As the older generation receivers were replaced with

the all-digital BVR equipment, the Network
observed a general improvement in weak signal track-
ing performance and operational reliability. The
receiver replacement program was completed
throughout the Network by 1998. A selection of
technical references related to digital receivers can
be found in section 7 of the appendix at the end
of the book.

Encoding and Decoding

The data generated by science instruments must be

reliably communicated from the spacecraft to the
ground, despite the fact that the signal received is
extremely weak and that the ground receiver cor-
rupts the signal with additive noise. Even with
optimum integration and threshold detection, indi-
vidual bits usually do not have adequate signal energy
to ensure error-free decisions. To overcome this prob-
lem, structured redundancy (channel encoding) - is
Figure 7-10. The Advanced Receiver added to the data bit-stream at the spacecraft.
(AFH) as used to track the Pioneer Despite the fact that the individual ''symbols" result-
10 spacecraft. ing from this encoding have even less energy at the
receiver, the overall contextual information, used
properly in the decoding process on the ground, results in more reliable detection of
the original data stream.

High-performance codes to be used for reliable data transfer from spacecraft to DSN
were identified by research performed under the Advanced Systems Program and adopt-
ed for standard use in the Network while the search for even more powerful and efficient
codes continued. New, more eficient block codes, which made better use of the lirnit-
ed spacecraft transmitter power by avoiding the need to transmit separate synchronizing
signals, were developed under the DSN Advanced Systems Program and first demon-
strated on the Mariner 6 and 7spacecraft in 1969. By putting the extra available spacecraft
transmitter power into the data-carrying signal, the new block code enabled the return
of the Mars imaging data at the astonishing (for the time) rate of 16,200 bits per sec-
ond, an enormous improvement over the 270-bps data rate for which the basic mission
had been designed. Of course, conversion of the encoded data stream back to its orig-
inal error-free form required a special decoder. The experimental block decoder, developed
under the same program for this purpose, formed the basis for the operational block-
decoders implemented in the Network as part of the Multimission Telemetry System
shortly thereafter.

While the JPL designers of the Mariner spacecraft were pursuing the advantages of
block-coded data, the designers of the Pioneer 9 spacecraft at the Ames Research Center
(ARC) were looking to very complex convolutional codes to satisfy their scientists. The
scientists agreed to accept intermittent gaps in the data caused by decoding failure in
exchange for the knowledge that successllly decoded data would be virtually error-
free. In theory, a convolutional code of length k = 25 would meet the requirement,
but it had a most significant drawback. The decoding process was (at the time) extraor-
dinarily difficult. Known technically as "sequential decoding," this was a continuous
decoding operation rather than the "one block at a time" process used by the DSN for
decoding the Vyageer data.

The original plan was to perform the decoding operation for Pioneer 9 in non-real-time
at ARC, using tape-recorded data provided by the DSN. However, Pioneer engineers
working in conjunction with the DSN Advanced Systems Program explored and demon-
strated the potential for decoding this code in real time via a very-high-speed engineering
model sequential decoder. With the rapid evolution in capability of small computers, it
became apparent that decoding Pioneer's data in such computers was both feasible and
economical. Subsequent implementation of sequential decoding in the Network was done
via microprogramming of a small computer, guided by the knowledge gained via the
efforts of the Advanced Systems Program. The subsequent Pioneer I0 and I I spacecraft
flew with a related code of length k = 32 and were supported by the DSN in a com-
puter-based decoder.

The DSN standard code, flown on Vyageer and Galileo, consisted of a short convolu-
tional code that was combined with a large block-size Reed-Solomon code. The standard
algorithm for the decoding of convolutional codes was devised in consultation with JPL
researchers and demonstrated by simulations performed under the Advanced Systems
Program. Prototypes of the decoding equipment were fabricated and demonstrated at
JPL, also with the support of the Advanced Systems Program.

The application of coding and decodmg technology in the DSN was paced by the evo-
lution of digital processing capability. At the time of the Voyager design, a convolutional
code of length k = 7 was chosen as a compromise between performance and decoding
complexity, which would grow exponentially with code length. Equipment was imple-
mented around the DSN to handle this code from Viager and subsequently from Magelkzn,
Galileo, and others. Modern digital technology has permitted the construction of much
more complex decoders, and a code of length k = 15 was devised with the support of
the Advanced Systems Program. This code was installed as an experiment on the Galileo
spacecraft shortly before its launch. The corresponding prototype decoder was complet-
ed soon afterward. Though not used for Galileo because of its antenna problem, the
more complex decoder was implemented around the Network for support of the Cassini
and subsequent missions.

Efforts of the Advanced Systems Program provided the understanding of telemetry per-
formance to be expected with the use of these codes. Figure 7-1 1 displays the reliability
of the communication (actually, the probability of erroneous data bits) as it depends
upon the spacecraft signal energy atlocated to each data bit for uncoded communica-
tion and three diierent codes.

Research on new and even more powerful coding schemes, such as turbo codes, con-
tinued to occupy an important place in the Advanced Systems Program. Turbo codes
are composite codes made up of short-constraint-length convolutional codes and a data
stream interleaver. The decoding likewise consists of decoders for the simple component
codes, but with an iterative sharing of information between them. These codes, which
push hard on the fundamental theoretical limits to signal detection, can result in almost
a full decibel of performance gain over the best previous concatenated coding systems.
A selection of technical references related to data encoding and decoding can be found
in section 8 of the appendix at the end of the book.
Figure 7-11. Telemetiy communication channel performance for various coding schemes. The
first set of cutves shows the Voyager k = code, both alone and in combination with the Reed-
Solomon code. The second set of codes illustrates the k = 15 code, which was to be demonstrated
with Galilee's original high-rate channel, shown alone and in combination with the Reed-Solomon
code, either as constrained by the Galileo spacecraft data system (I = 2) or in ideal combination.
The third set shows the k = 14 code, devised by the Advanced Systems Program researchers for
the actual Galileo low-rate mission, both alone and in combination with the selected variable-redun-
dancy Reed-Solomon code and a complex four-stage decoder. The added complexity of the codes,
which has its greatest effect in the size of the ground decoder, clearly provides increased reliabili-
ty for correct communication.

Data Compression

Source encoding and data compression are not typically considered a part of the DSN's
downlink functions, but the mathematics that underlie coding and decoding are a coun-
terpart of those that p i d e the development of data compression. Simply stated, channel
encoding is the insertion of structured redundancy into a data stream, while data com-
pression is the finding and removal of intrinsic redundancy. Imaging data are often highly
redundant and can be compressed by factors of at least two, and often four or more,
without loss in quality. For Vqyager, two influences led to a factor-of-two increase in the
number of images returned from Uranus and Neptune. The first was the effecting of a
very simplified image-compression process constrained to fit into available onboard mem-
ory. The second improvement involved corresponding changes to the channel coding.

The success of data compression technology in enhancing the data return from the
Voydger missions firmly established the technique as an important consideration in the
design of all hture planetary downlinks. The original telecommunication link design for
the Galileo spacecraft used data compression to almost double the amount of imaging
data that the spacecraft could transmit from its orbital mission around Jupiter. The fail-
ure of the spacecraft's high-gain antenna prior to Galilee's arrival at Jupiter prompted an
intense effort to find even more complex data compression schemes that would recov-
er some of the Jupiter imaging data that otherwise could not have been returned. A
selection of technical references related to data compression can be found in section 9
of the appendix at the end of the book.

The technique of antenna arraying, as practiced in the Deep Space Network, made use
of the physical fact that a weak radio signal from a distant spacecraft that is received
simultaneously by several antennas at different locations is degraded by a component of
radio noise that is independent of each receiving station. By contrast, the spacecraft sig-
nal itself is dependent, or coherent, at each receiving site. In theory, therefore, the power
of the signal, relative to the power of the noise, or signal to noise ratio, (SNR) could
be improved by combining the individual antennas in such a way that the coherent
spacecraft signals were reinforced, while the independent or non-coherent noise com-
ponents were canceled out.

In practice, this involved a complex digital process for compensating for the time, or
phase, delays caused by the different distances between each station and the spacecraft.
It also called for compensating- for differing
- distances between the various antenna loca-
tions and the common station where the combining function was carried out. This
technique became known as antenna arraying, and the digital processing h c t i o n that
realized the theoretical "gain" of the entire process was called "signal combining."

By 1970, conceptual studies had described and analyzed the performance of several lev-
els of signal combining and two of these schemes, carrier and baseband combining, were
of potential interest to the Network. Both techniques involved compensation for the
 phase delays caused by the various locations of the arrayed antennas. The difference lay
in the frequency at which the combining function was performed. "Carrier" combining
was done at the carrier frequency of the received signal, while "baseband" combining
was carried out at the frequencies of the subcarrier and data signal that modulated it.
Each had its advantages and disadvantages, but baseband combining proved easier to
implement and was, obviously, tried first.

1
The "arraying and signal combining concept was first developed and demonstrated in
1969 and 1970 by J. Urech, a Spanish engineer working at the Madrid tracking sta-
tion. Using signals from the Pioneer 8 spacecraft and a microwave link to connect two
26-m stations located 20 km apart (DSS 61 and DSS 62), he succeeded in demon-
strating the practical application of the principle of baseband combining in the Network
for the first time. Because of the low baseband frequency of the Pioneer 8 data stream
(8 bits per second), as well as the close proximity of the antennas, no time-delay com-
pensation was necessary.

Within the bounds of experimental error, this demonstration confirmed the R&D the-
oretical estimates of performance gain and encouraged the Advanced Systems researchers
to press forward with a more complex form of baseband combining at a much higher
data rate (117 kilobits per second) in real time.

The demonstration took place at Goldstone in September 1974, using the downlink
signals from the Muriner-Waw-Merntry (MVM) spacecraft during its second encounter
with the planet Mercury. Spacecraft signals from the two 26-m antennas, DSS 12 and
DSS 13, were combined in an R&D combiner with signals from the DSS 14 64-m
antenna in real time at 117 kbps. The less-than-predicted arraying gain obtained in this
demonstration (9 percent versus 17 percent) was attributed to small differences in per-
formance between key elements of the several data-processing systems involved in the
test. Although this experience demonstrated both the practical difficulty of achieving full
theoretical gain of an antenna arrayed system and the critical effect of very small vari-
ations in the performance of its components, it also established the technical feasibility
of baseband arraying of very weak high-rate signals.

In 1977, with the lessons learned from these demonstrations as background, the DSN
started to develop an operational arraying capability for the Network. The Vyuger 1 and
2 Encounters with Saturn in 1980 and 1981 would be the first to use the arraying in
the Network. A prototype baseband real-time combiner (RTC), based on the analysis
and design techniques developed by the earlier R&D activity, was completed in the fall
of 1978. Designed to combine the signals from DSS 12 and DSS 14 at Goldstone, it
was used with varying degrees of success to enhance the signals from the Vyugers at
Jupiter in March and July of 1979 and the Pioneer 11 Encounter of Saturn in August
and September of that year.

Like the previous demonstration, this experience emphasized the critical importance of
having all elements of the array-receivers, antennas, and instrumentation operating pre-
cisely according to their specified performance capabilities. With this very much in mind,
the DSN proceeded to the design for operational versions of the RTC for use at all
three complexes to support the Vyuger l a n d 2 Encounters of Saturn. The operational
versions of the RTC embodied many improvements derived from the experience with
the R&D prototype version. By mid-August 1980, they were installed and being used
to array the 64-m and 34-m antennas at all three complexes as Vyuger I began its far-
Encounter operations. During this period, the average arraying gain was 0.62 dB, about
15 percent greater than that of the 64-m antenna alone. 'While this was good, improve-
ment came slowly as more rigorous control and calibration measures for the array elements
were instituted throughout the Network. By the time Vyuger 2 reached Saturn in August
1981, these measures, supplemented with additional training and calibration procedures,
had paid ofE The average arraying gain around the Network increased to 0.8 dB (approx-
imately 20 percent), relative to the 64-m antenna alone. This was clearly a most satisfactory
result and the best up to that time. Antenna arraying had become a permanent addi-
tion to the capability of the Network.

While researchers working within the Advanced Systems Program continued to explore
new processes for arraying antennas, engineers within the DSN took advantage of the
long flight time between the Voyager Saturn and Uranus Encounters to refine the
existing RTC configuration. Over the next five years, the formerly separate data-pro-
cessing functions of combining, demodulation, and synchronization were integrated
into a single assembly. This integration facilitated improvements in performance, sta-
bility, and operational convenience. By the time Vyuger 2 approached Uranus in 1985,
the new Baseband Assemblies (BBAs), as they were called, had been installed at all
three Complexes. In addition, a special version of the basic four-antenna BBA was
installed at the Canberra Complex. This provided for combining the Canberra array
of one 64-m and two 34-m antennas with signals from the 64-m Parkes Radio Telescope,
200 krn distant (Figure 7-12).

In January 1986, this arrangement was a key factor in the successll return of Voyager
imaging data from the unprecedented range of Uranus. But even greater achievements
in antenna arraying lay ahead.
Figure 7-12. The 64-m antenna of the Radio Astronomy Observatory, Parkes, Australia.

In 1989, the DSN used a similar arrangement with great success to capture the Voyager
imaging data at a still greater range-from Neptune. This time the Goldstone 70-m and
34-m antennas were arrayed with the 27 antennas of the Very Large Array (VLA) (Figure
7-13) of the National Radio Astronomy Observatory at Socorro, New Mexico.
Figure 7-13. The Very Large Array (VLA) of the National Radio Astronomy Observatory (NRAO)
at Socorro, New Mexico.

DSN support for the Voyager Encounter of Uranus was further augmented by the Canberra-
Parkes array in Australia, which the DSN had reinstated with the addition of new BBAs,
new 34-m antennas, and the upgraded 70-m antenna.

The success of these applications of the multiple-antenna arraying technique provided

the DSN with a solid background of operational experience. The DSN drew heavily on
this experience a few years later, when it was called upon to recover the science data
from GaLileo after the failure of the spacecraft's high-gain antenna in 1991. Together
with the data compression and coding - techniques discussed earlier, the Network's
Canberra/Parkes/Goldstoneantenna arrays succeeded in recovering a volume of data that,
according to the Galiieo project, was equivalent to about seventy percent of the origi-
nal mission.

With time, arraying of multiple antennas within Complexes, between Complexes, or

between international space agencies came into general use as a means to enhance the
downlink capability of the Network. In the latter years of the century, most of the
enhancements to the arraying in the DSN were driven by implementation and opera-
tional considerations rather than by new technology, although the Advanced Systems
Program continued to explore the boundaries of performance for various alternative array-
ing architectures and combining techniques. A selection of technical references related
to antenna arraying can be found in section 10 of the appendix at the end of the book.

In addition to being able to exchange forward and return link data with an exploring
spacecraft, it is equally important to know the precise location of the spacecraft and its
velocity (speed and direction). Information about the position and velocity of the space-
craft can be extracted from the one-way or two-way radio signals passing between the
spacecraft and the DSN. When these data are extracted by appropriate processing and
further refined to remove aberrations introduced by the propagation medium along the
radio path between spacecraft and Earth, it can be used for spacecraft navigation.

Radio metric techniques similar to those used for spacecraft navigation can also be used
for more explicit scientific purposes, notably radio science, radio astronomy, and radio
interferometry on very long baselines (VLBI).

Since its inception, the DSN Advanced Systems Program has worked to develop effec-
tive radio metric tools, techniques, observing strategies, and analysis techniques that
furthered the DSN pre-eminence in these unique fields of science. In more recent times,
the Advanced Systems program demonstrated the application of Global Positioning System
technology to further refinement of radio metric data generated by the DSN. A selec-
tion of technical references on radio metric tools can be found in section 11 of the
appendix at the end of the book.

Doppler and Range Data

If the Earth and the spacecraft were standing still, the time taken for a radio signal to
travel from the Earth to the spacecraft and back would be a measurement of the dis-
tance between them. This is referred to as the round-trip light time (RTLT). However,
since the Earth and the spacecraft are both in motion, the RTLT contains both posi-
tion and velocity information, which must be disentangled through multiple measurements
and suitable analysis. The precision at which such measurements can be obtained is lim-
ited by the precision of the time-tag marker to the radio signals, and by the strength
of the signal in proportion to the noise mixed with it, or by the signal-to-noise ratio
(SNR).
Precise measurements of changes to this light time are far easier to obtain via observ-
ing the Doppler effect resulting from the relative motions. Such measurements are
mechanized via the phase-locked loops in both spacecraft and ground receivers using
the spacecraft's replica of the forward link residual carrier signal to generate the return
link signal, and counting the local replica of the return link residual carrier against
the original carrier for the forward link signal. The raw precision of these measure-
ments is comparable to the wavelength of the residual carrier signal, e.g., a few centimeters
for an X-band signal (8 GHz). Numerous interesting error sources tend to corrupt
the accuracy of the measurement and the inferred position and velocity of the space-
craft, and they have provided significant technical challenge for work under the Advanced
Systems Program.

The observed Doppler contains numerous distinct components, including the very sig-
nificant rotation of Earth. As Earth turns, the position of any specific site on the surface
describes a circle, centered at the spin axis of the Earth, falling in a plane defined by
the latitude of that site. The resultant Doppler component varies in a diurnal fashion
with a sinusoidal variation, which is at its maximum positive value when the spacecraft
is first observable over the eastern horizon. Its corresponding negative value occurs at
approach to the western horizon. A full-pass Doppler observation from horizon to hori-
zon can be analyzed to extract the apparent spacecraft position in the sky, although the
determination is somewhat weak near the equatorial plane. Direct measurements of the
RTLT are usefd for resolving this dificulty.

Three distinct generations of instruments designed to measure the RTLT were devel-
oped by the Advanced Systems Program and used in an ad hoc fashion for spacecraft
support before a hybrid version was designed and implemented around the DSN. The
third instrument designed, the Mu-I1 Ranging Machine, was used with the V k n g Landers
in a celestial mechanics experiment, which provided the most precise test, up to that
time, of the general theory of relativity.

These devices function by imposing an additional "ranging" modulation signal on the

forward link, which is copied on the spacecraft (within the limits imposed by noise)
and then imposed on the return link. The ranging signal is actually a very long peri-
od-coded sequence that provides the effect of a discrete time tag. The bandwidth of
the signal is on the order of 1 MHz, giving the measurement a raw precision of a
few hundred meters, resolvable with care to a few meters. Among other features, the
Mu-I1 Ranging Machine included the first demonstrated application of the digital
detection techniques that would figure strongly in future developments for the DSN.
Timing Standards

The basic units of measurement for all radio metric observations, Doppler or range, derive
from the wavelength of the transmitted signal. Uncertainties or errors in knowledge of that
wavelength are equivalent to errors in the derived spacecraft position. The need for accu-
rate radio metric5 has motivated the DSN Advanced Systems Program to develop some of
the most precise, most stable frequency standards in the world. While the current suite of
hydrogen maser frequency standards in the DSN field sites was built outside of JPL, the
design is the end product of a long collaboration in technology development, with research
units being built at JPL under the DSN Advanced Systems Program and elsewhere.

Continued research under the Advanced Systems Program for improved frequency stan-
dards resulted in the development of a new linear ion trap standard (Figure 7-14) that

Figure 7-14. The new linear ion trap (LIT) standard.

offered improved long-term stability of a few parts in 10l6, as well as simpler and casi-
er maintenance than that required by the hydrogen masers.

Work was under way to implement the LIT standard in the DSN, while research efforts
continued for improvements that could be transferred to field operation in the future.
A selection of technical references related to timing standards can be found in section
12 of the appendix at the end of the book.

Earth Rotation and Propagation Media

Radio metric Doppler and range data enable the determination of the apparent location
of a spacecraft relative to the position and attitude of the rotating Earth. Earth, how-
ever, is not a perfectly rigid body with constant rotation, but contains fluid components
as well, which slosh about and induce variations in rotation of perhaps a few millisec-
onds per day. Calibration of Earth's attitude is necessary so that the spacecraft's position
in inertial space can be determined-a necessary factor in navigating the spacecraft toward
a target planet. Such calibration is available via the world's optical observatories and,
with greater precision, via radio techniques, which will be discussed further in the sec-
tions entitled "VLBI and Radio Astronomy" and "Global Positioning System."

The interplanetary media along the signal path between Earth and the spacecraft affect
the accuracy of the Doppler and range observations. The charged ions in the tenuous
plasma spreading out from the Sun, known as the solar wind, bend and delay the radio
signal. Likewise, the charged ions in Earth's own ionosphere and the water vapor and
other gases of the denser lower atmosphere bend and delay the radio signal. All of these
factors are highly variable because of other factors, such as intensity of solar activity, sea-
son, time of day, and weather. All factors must be calibrated, modeled, or measured to
achieve the needed accuracy; over the years, the DSN Advanced Systems Program has
devised an increasingly accurate series of tools and techniques for these calibrations. A
selection of technical references related to Earth rotation and propagation can be found
in section 13 of the appendix at the end of the book.

Radio Science

Radio science is the term used to describe the scientific information obtained from the
intervening pathway between Earth and a spacecraft by the use of radio links. The effects
of the solar wind on the radio signal path interfere with our efforts to determine the
location of the spacecraft, but if the relative motions of Earth and the spacecraft are
modeled and removed from the radio metric data, much of what remains is informa-
tion about the solar wind and, thus, about the Sun itself. Other interfering factors are
also of scientific interest.

In some situations, the signal path passes close by a planet or other object, and the
signal itself is bent, delayed, obscured, or reflected by that object and its surrounding
atmosphere. These situations provide a unique opportunity for scientists to extract infor-
mation from the signal about object size, atmospheric density profiles, and other factors
not otherwise observable. Algorithms and other tools devised to help calibrate and
remove interfering signatures from radio metric data for use in locating a spacecraft
often become part of the process for extracting scientific information from the same
radio metric data stream. The precision frequency standards, low-noise amplifiers, and
other elements of the DSN derived from the Technology Program are key factors in
the ability to extract this information with a scientifically interesting accuracy.
Occasionally, engineering models developed by the Program are placed in the Network
in parallel with operational instrumentation for ad hoc support of metric data-gather-
ing for some unique event.

The effects of gravity can also be observed by means of the radio link. Several situa-
tions are of interest. If the spacecraft is passing by or in orbit about an object that has
a lumpy, uneven density, that unevenness will cause a variation in the spacecrafts path-
way that will be observable via the radio metric data. If the radio signal passes near a
massive object such as the Sun, the radio signal's path will be bent by the intense grav-
ity field, according to the theories of general relativity. And in concept, gravitational
waves (a yet-to-be-observed aspect of gravity field theory) should be observable in the
Doppler data from a distant spacecraft. All of these possibilities depend upon the sta-
bility of the DSN's precision frequency standards for the data to be scientifically interesting.
A selection of references related to radio science can be found in section 14 of the appen-
dix at the end of the book.

VLBI and Radio Astronomy

The technical excellence of the current DSN is, at least in part, a result of a long and
fruitful collaboration with an active radio astronomy community at the California Institute
of Technology (Caltech) and elsewhere. Many distant stars, galaxies, and quasars are
detectable by the DSN at radio frequencies. The furthest of these are virtually motion-
less and can be viewed as a fixed-coordinate system to which spacecraft and other
observations can be referenced. Observations relative to this coordinate set help to reduce
the distorting effects of intervening material in the radio signal path and uncertainties
in the exact rotational attitude of Earth during spacecraft observations.
Little precise information can be extracted by observing these objects one at a time
and from a single site, but concurrent observation at a pair of sites will determine the
relative position of the two sires referenced to the distant object. The observing tech-
nique is known as very long baseline interferometry (VLBI) and was developed by the
research of many contributors, including substantial work by the DSN's Advanced Systems
Program. If three sites are used in VLBI pairs and multiple objects are observed, the
positional attitude of Earth and the relative positions of the observed objects can be
determined. If one of the observed is a spacecraft transmitting a suitable signal, its posi-
tion and velocity in the sky can be very accurately defined. A demonstration of this
technique via the Advanced Systems Program led to operational use for spacecraft such
as Vyager and Magellan.

VLBI can also be used in conjunction with conventional radio metric data types to pro-
vide the calibration for the positional attitude of Earth. Such observations can be made
without interfering with spacecraft communication except for the time utilization of the
DSN antennas. In addition to determining Earth's attitude, the observations measure
the relative behavior of the frequency standards at the widely separated DSN sites, and
thus help to maintain their precision performance. Again, demonstration of this capa-
bility via the Advanced Systems Program led to routine operational use in the DSN.

Design and development of the DSN equipment and s o h a r e needed for VLBI signal
acquisition and signal processing (correlation) was carried out in a collaboration involv-
ing the Advanced Systems Program, the operational DSN, and the Caltech radio astronomy
community. Tools needed to produce VLBI metric observations for the DSN were essen-
tially the same as those for interferometric radio astronomy. Caltech received funding
from the National Science Foundation for this activity, and both Caltech and the DSN
shared in the efforts of the design while obtaining products that were substantially bet-
ter than any that they could have been obtained independently.

Another area of common interest between the DSN and the radio astronomy commu-
nity is that of precision wideband spectral analysis. Development efforts of the Advanced
Systems Program produced spectral analysis tools that have been employed by the DSN
in spacecraft emergency situations and in examining the DSN's radio interference envi-
ronment, and they have served as pre-prototype models for equipment for the DSN.
Demonstration of the technical feasibility of the very-wide-band spectral analysis and
preliminary observations by a megachannel spectrum analyzer fielded by the Advanced
Systems Program helped establish the sky survey planned as part of the former SET1
(Search for Extraterrestrial Intelligence) Program.
j Another technique (one similar to the use of VLBI for a radio metric reference) is used
4
1
4
if two spacecraft are flown to the same target; the second can be observed relative to
the first, providing better target-relative guidance once the first has arrived at the tar-
get. Techniques for acquiring and analyzing such observations have been devised by the
Advanced Systems Program. A selection of technical references related to VLBI and radio
I
t astronomy can be found in section 15 of the appendix at the end of the book.

The Global Positioning System

i, The Global Positioning System (GPS) is a constellation of Earth-orbiting satellites designed

(initially) to provide for military navigation on Earth's surface. Research under the Advanced
Systems Program showed that these satellites could provide an excellent tool to calibrate
and assist in the radio metric observation of distant spacecraft. GPS satellites fly above
the Earth's atmosphere and ionosphere in well-defined orbits so that their signals can be
used to measure the delay through these media in a number of directions. With suit-
able modeling and analysis, these measurements can be used to develop the atmospheric
and ionospheric calibrations for the radio path to a distant spacecraft.

Additionally, since the GPS satellites are in free orbit about Earth, their positions are
defined relative to the center of mass of Earth, and not its surface. They provide anoth-
er method to observe the uneven rotation of Earth.

GPS techniques can also be used to determine the position of an Earth-orbiting space-
craft relative to the GPS satellites, as long as the spacecraft carries a receiver for the GPS
signals. The potential of this technique was initially demonstrated by the Advanced
Systems Program. GPS was subsequently used by the TOPEX/POSEIDON Project for
precise orbit determination and a consequent enhancement of its scientific return. A
selection of technical references related to the GPS can be found in section 16 of the
appendix at the end of the book.

The Goldstone Solar System Radar (GSSR) is a unique scientific instrument for mak-
ing observations of nearby asteroids, the surfaces of Venus or Mars, the satellites of
Jupiter, and other objects in the solar system. Although the GSSR makes use of the
DSN 70-m antenna for its scheduled observing sessions, its receiving, transmitting, and
data-processing equipment is unique to the radar program. The GSSR is a product of
many years of development by the DSN Advanced Systems Program. In the early days
of the DSN, the Advanced Systems Program took ownership of the radar capability at
the DSN's Goldstone, California, site and evolved and nurtured it as a vehicle for devel-
oping and demonstrating many of the capabilities that eventually would be needed by
the Network.

Scientific results abounded as well, but they were not its primary product. Timely devel-
opment of DSN capabilities was the major result. Preparations for a radar observation
at the DSN Technology Development Field Site bore many resemblances to those for
a spacecraft planetary encounter, since the radar observations could only be successful
during the few days when Earth and the radar target were closest together.

In the conventional formulation of the radar sensitivity equations, that sensitivity depends
upon the aperture, temperature, power, and gain of the system elements. Here, aperture
refers to the effective size, or collecting area and efficiency, of the receiving antenna;
temperature is a way of referring to the noise in the receiving system, where a lower
temperature means a lesser noise; power refers to the raw power level from the trans-
mitter; and gain is the effective gain of the transmitting antenna, which depends in turn
upon its size, its surface efficiency, and the frequency of the transmitted signal. Where
the same antenna is used both to transmit and to receive, the antenna size and e 6 -
ciency appear twice in the radar equations.

Significant improvements to the DSN's capability for telemetry reception were to come
from the move upward in frequency from S-band (2 GHz) to X-band (8 GHz) on the
large 64-m antennas. Performance of these antennas at the higher frequencies and the
ability to successfully point them were uncertain, however, and these uncertainties would
best be removed by radar observations before spacecraft with X-band capabilities were
launched. The radar had obvious benefit from the large antenna and the higher fre-
quency. The first flight experiment for X-band communication was carried out on the
1973 Mariner Venus Mars mission. Successful radar observations from the Goldstone
64-m antenna demonstrated that the challenge of operating the large antennas at the
higher X-band frequency could be surmounted.

High-power transmitters were needed by the DSN for its emergency forward link h c -
tions but were plagued by problems such as arcing in the waveguide path when power
densities became too high. High-power transmitters were essential for the radar to "see"
at increased distances and with increased resolution. Intense development efforts at the
DSN Technology Development Field Site could take place without interference or risk
to spacecraft support in the Network. Successful resolution of the high-power problems
for the radar under the Advanced Systems Program became the successful implementa-
tion of the high-power capability needed by the Network for uplink communications.
Low-noise amplifiers were needed by the DSN to increase data return from distant space-
craft. Low-noise amplifiers were essential for the radar to enable it to detect echoes !&om
increasingly distant targets or to provide for increased resolution of already detectable
targets. The synergistic needs of both the radar system and the Network led to the devel-
opment of the extremely low-noise maser amplifiers that became part of the standard
operational inventory of the DSN.

Digital systems technology was rapidly evolving during this period and would play an
increasing role in the developing DSN. Equipment developed by the Advanced Systems
Program for its radar application included 1) digital encoders to provide for spatial
resolution of parts of the radar echo, 2) computer-driven programmable oscillators to
accommodate Doppler effects on the signal path from Earth to target to Earth, and
3) complex, high-speed digital signal processing and spectrum analysis equipment.
Much of the digital technology learned this way would transfer quickly to other parts
of the signal processing work under the Advanced Systems Program and wentually
into the operational DSN. Some of the elements would find direct application, such
as the programmable oscillators, which became essential for maintaining contact with
the Byager 2 spacecraft following a partial failure in its receiver soon after launch.
And the signal analysis tools would be called on many times over the years to help
respond to spacecraft emergencies.

Some of the products of the early radar observations (see Figure 7-15) were both sci-
entific in nature and essential for providing information for the planning and execution
of NASAIs missions.

One notable "first" was the direct measurement of the astronomical unit. (One astro-
nomical unit (AU) is equal to 1.5 x 108 km,the mean distance between Earth and the
Sun.) It sets the scale size for describing distances in the solar system. The measurement
was made in support of preparations for sending Mariner 2 to Venus and provided a
correction of 66,000 km from conventional belief at that time. It also enabled correc-
tions that brought the mission into the desired trajectory for its close flyby of the planet.
The GSSR was also used in qualifying potential Mars landing sites for the Viking Landers,
and it continues to provide information about the position and motion of the planets,
which is used to update the predicted orbits for the-planets of the solar system. A selec-
tion of technical references related to the Goldstone solar system radar can be found in
section 17 of the appendix at the end of the book.
Figure 7-15. First high-resolution radar image of Venus,

The progress of deep space communications capability over the period of forty years
since the inception of the Network is illustrated in Figure 7-16.

In interpreting the data presented in Figure 7-16, it will be observed that the logarith-
mic scale that displays the data rate gives an impression that the early improvements are
more significant than the later improvements. This is because the steps represent frac-
tional or percentage increases, rather than incremental increases. The latter would show
the actual data rate increases, which are much larger in the later improvements. If the
value was proportional to the amount of data, then the display of the incremental increas-
es would be more meaningfd than the logarithmic display.
/ 3

PROFILE OF DEEP SPACE COMMUNICATIONS CAPABILITY

SPACE-TO-EARTH
EQUIVALENT IMAGING DATA RATE CAPABILITY AT JUPITER DISTANCE - 750 MILLIONKILOMETERS
10'0 1 1 1 1 1 1 1 1 1 1 1 ~ 1 1 ~ ~ i 1 ~ ~ ~ I I ~ ~ ( I ~ ~
109
- DSN POM +34M) + VLA(z7 l25M)
INTERAGENCVARRAY, VCR NEF'NNE(G) -
OPTICAL, 2W LASER

DSN (64M I34M) + PARKES(MM)

- HOME INTERAGENCYARRAY. VGR URANU R

IMPROVEDCODING(11.114) ( 0 6 W)
-
PERSONALCMIPUTER 48M ANTENNA, GU(SIC) -
70M + 2 x 34M ARRAY (G)
-
-
w .unw
D3" .IYa,PYI*T*DU
0 .omw -
a,
Ulw*"rw .I"%"
I .
.wLco
UmmwMYUISm"
rum ..uroumcanr".mn.m."k-
P,
%c. .8%E*3a,WCrn,m
,*Acccnm
-
..
WXW. .1111*1mm

-.
YU ~n-IUI(U,mIXI.IXI"IXIIXI

w
YDYAeSR
w*,,
.m"m
-

1960 62 64 66 66 70 72 74 76 78 80 82 84 86 88 90 92 91 96 98 20W 02 04 06 08 10 12 14 16 18 2020

CALENDAR YEAR
Thls chatt documents the (welve orders o l magnHude improvement o l deep space communfcations wpablilly slnce the beglnnlngs of deep space
exploatlon to the present. Another 3 orders of magnitude Improvement are forecast by 2020. The Increase ot pdormance lo due to a series of
i n both the spacecraft and ground. Key factors include higher operetlng trequency and improved coding
l n n o v a l i v ~~ o o p e r a t l vImpr~vements
~
technlquas, s p a u t e a n hlgher power and antenna alre, and ground system lower nolse empllfiers and increased antenna slze.

\ I

Figure 7-16. Profile of deep space communications capability. The timeline on the horizontal
axis of the figure covers the first forty years of actual Network operational experience through the
close of the century and extends for a further twenty years to forecast the potential for future improve-
ments through the year 2020. The vertical axis displays the growth in space-to-Earth downlink
capability of the Network. The downlink capability is given in units of telemetry data rate (bits per
second) on a logarithmic scale and represents the equivalent imaging data rate capability for a typ-
ical spacecraft at Jupiter distance (750 million kilometers). Significant events in the history of deep
space telecommunications and deep space exploration are appropriately annotated on both axes.

Presented this way, however, the figure clearly shows that, from inception though 1997,
the downlink capability of the Network grew from lo4 to lo6 bits per second, equiva-
lent to twelve orders of magnitude.

This remarkable progress is not, of course, solely due to improvements in the Network.
Many of the steps result from "cooperative" changes on the part of both the DSN and
the spacecraft. Coding, for example, is applied to the data on the spacecraft and removed
on Earth. A change in frequency has resulted in some of the larger steps shown by caus-
ing the radio beam from the spacecraft to be more narrowly focused. Such change
necessitates equipment changes on both the spacecraft and Earth.

Other steps represent advances that are strictly spacecraft-related, such as increases in
return-link transmitter power or increases in spacecraft antenna size, which improves per-
formance by more narrowly focusing the radio beam from the spacecraft.

Still other steps depict improvements strictly resulting from the DSN, such as reduction
in receiving system temperature, increase in the size of the ground antennas, or use of
arrays of antennas to increase the effective surface area available for collecting signal
power. A selection of technical references related to telecommunications performance can
be found in section 18 of the appendix at the end of the book.

With one exception, the program continued to pursue the same broad themes in this
period, December 1994 to December 1997, that had characterized its earlier work. The
new theme that began to appear in the program in 1994 was directed at the reduction
of Network operations costs.

A Network automation work area was set up to develop automated procedures to replace
the extremely operator-intensive work of running a spacecraft "pass" over a DSN track-
ing station. This effort soon produced demonstrable results. A fully automated satellite
tracking terminal that could reduce operations costs for near-Earth satellites was demon-
strated in 1994. A s o b a r e prototype that reduced the number of manual inputs for a
typical 8-hour track from 900 to 3 was installed at DSS 13 and used to support Ka-
band operations at that site. Eventually this technology would find its way into the
operational Network. A contract for a new, small deep space transponder offering lower
size and power needs, and most importantly lower production costs, was initiated with
Motorola in July 1995. This became an element in JPZs hture low-cost micro-space-
craft. Development of a space-borne micro-GPS receiver, which offered the promise of
low-cost orbit determination for most low-Earth orbiting spacecraft, was also initiated
in 1995.

Prototypes of a new class of low-cost, fully automated, autonomous ground stations that
would simplify implementation and operation and reduce the life-cycle cost of tracking
stations in the DSN were introduced in 1995, 1996, and 1997. The first of these ter-
minds was designed for tracking spacecraft in low-Earth orbit and was named LEO-T
It was enclosed in a radome and mounted on the roof of a building - at JPL, where it
accumulated over two years of unattended satellite tracking operations without prob-
lems. Prompted by the success of LEO-T, the program undertook a fast-track effort to
develop a similar automated terminal for deep space applications. It would be called
DS-T The prototype DS-T was to be implemented at the 26-m BWG antenna at
Goldstone. Automation technology was carried one step further into the area of Network
operations in 1997 with the introduction of Automated Real-time Spacecraft Navigation
(ARTSN). In addition to the antenna system, the DS-T included an X-band microwave
system, a 4-kW transmitter, and an electronics rack containing commercial-off-the-shelf
equipment to carry out all baseband telemetry downlink, command uplink, and Doppler
and ranging functions. It was planned to demonstrate DS-T with Mars Global Surveyor
early in 1998 and to use this technology (autonomous uplink and downlink) in the
Network with the New Millennium DS-1 spacecraft later that year. A selection of tech-
nology references related to cost-reduction initiatives can be found in section 19 of the
appendix at the end of the book.
KA-BAND DEVELOPMENT

In the past, the major improvements in the Network's deep space communications capa-
bilities were made by moving the operating frequency to the higher frequency bands.
Recognizing this, research and development at Ka-band (32 GHz) was started in 1980.
Initial efforts were directed toward low-noise amplifier development and system benefit
studies. However, it was also clear that the performance of existing antennas (which were
designed for much lower frequencies) would severely limit the improvement in per-
formance that could be realized from the higher operating frequency. Accordingly, in
1991, a new antenna specifically designed for research and development at Ka-band was
installed at the DSS 13 Venus site at Goldstone. It would be used as the pathfinder for
development of large-aperture beam waveguide antennas that would, in due course, be
implemented throughout the Network.

Small imperfections in the surface of an antenna cause larger degradations at Ka-band

than at lower frequencies, and, because of the narrower beam width, small pointing
errors have a much larger effect. In 1994, improvements in antenna efficiency and in
antenna pointing- were made on the research and development Ka-bmd antenna at DSS
13 and on the new operational antenna at DSS 24. These improvements were effected
by the use of microwave holography for precise determination of antenna efficiency,
along with a special gravity-compensation system to counteract the effect of gravitational
sag as a function of antenna elevation angle. In a search for further downlink improve-
ment, a new feed system consisting of a maximally compact array of seven circular
Ka-band horns was designed and tested at DSS 13. Each horn was connected to a cryo-
genically cooled low-noise amplifier, a frequency down-converter, an analog-to-digital
converter, and a digital signal processor. The signals from each horn were optimally com-
bined in a signal processor and presented as a single output with a quality equivalent
to that of a signal from an undistorted antenna. The measured gain in downlink per-
formance was 0.7 dB.

The technology program continued to develop operational concepts that would even-
tually lead to the adoption of Ka-band for deep space missions. Tradeoff studies between
X-band and Ka-band showed the overall advantage of Ka-band, taking due account of
the negative effects inherent in its use, to be about a factor of four, or 6 dB, in data
return capability. Obviously, end-to-end system demonstrations were needed to instill
confidence in the new technology. In 1993, the Mars Observer spacecraft carried a non-
linear element in its transmission feed to produce a fourth harmonic of the X-band
signal. The demonstration (called KABLE for Ka-band Link Experiment) provided a
weak Ka-band signal for the tracking antenna at DSS 13. A second demonstration
(KABLE 11) was conducted using the Mars Global Surveyor mission in 1996. This exper-
iment would be used to characterize Ka-band link performance under real f l i t conditions
and validate the theoretical models derived from the studies mentioned above.

KABLE-I1 required additions to the MGS spacecraft radio-transponder to generate a

modulated Ka-band downlink from which the improvements that had been made to
the DSS 13 antenna to reduce pointing errors and improve performance while tracking
a spacecraft at Ka-band could be evaluated. While the main objective of KABLE-I1 was
to evaluate Ka-band for future operational use, it also served as a testbed for new Ka-
band technology applications in both the flight systems and the ground systems.

In the course of transition to simultaneous X-/Ka-band operation, the DSN needed the
capability to support various combinations of X- and Ka-band uplinks and downlinks.
These included Ka-band receive only; X-/%-band simultaneous receive, with or with-
out X-band transmit; and full X-band transmitlreceive simultaneously with Ka-band
receiveltransmit. The technology program developed new microwave techniques using
frequency-selective surfaces and feed junction diplexers to provide the frequency and
power isolation necessary to realize the performance required by a practical device. In
late 1996, a demonstration at DSS 13 succeeded in showing that these four different
modes of operation could coexist on a single beam waveguide antenna within accept-
able performance limits. This work provided a viable solution to the problem of
simul&eous X-/Ka-band operation on a single antenna; this solution would be need-
ed by the operational network to support the Cassini radio science experiment (search
for gravitational waves) in 2000.

Recognizing the need for a cheaper, smaller, less power-consuming radio-transponder to

replace the existing device on future deep space missions, the advanced development
program embarked on a joint program with other JPL organizations to develop the Small
Deep Space Transponder (SDST). The concept employed Microwave Monolithic
Integrated Circuits in the RF circuits and Application Specific Integrated Circuit tech-
niques to perform digital signal-processing functions, with a IUSC microprocessor to
orchestrate overall transponder operation. The transponder would transmit coherent X-
band and Ka-band downlinks and receive an X-band uplink. Besides minimizing- -production
costs, the principal design drivers were reduction in mass, power consumption, and vol-
ume. The Small Deep Space Transponder was flown in space for the first time aboard
the DS-1 New Millennium spacecraft in July 1998.

With a view to providing a better understanding of the performance of Ka-band links

relative to X-band links from the vantage point of a spaceborne radio source, the DSN
technology program engaged in the development of a small low-Earth-orbiting space-
craft called SURFSAT-1. Launched in 1995, the experiment provided an end-to-end test
of Ka-band signals under all weather conditions and DSS 13 antenna elevation angles
as the spacecraft passed over Goldstone. The SURFSAT data was also used for com-
parison with the KABLE data received from MGS. Later, the SURFSAT X-band and
Ka-band downlinks were used to great advantage to test and calibrate the DSN's new
1l-meter antennas prior to their support of the VSOP (HALCA) Orbiting VLBI mis-
sion in 1996. A selection of technical references related to Ka-band development can be
found in section 20 of the appendix at the end of the book.

While reduction of Network operating costs and the introduction of Ka-band into the
Network were important features of the Technology Program in the late 1990s, they
were not the only areas of activity. Interagency agreements allowed Wide-Area Differential
GPS techniques, originally developed under the technology program for NASA space-
craft orbit determination and DSN Earth platform calibrations, to be made available for
applications within the Department of Transportation and Federal Aviation Administration.

In addition to these other technologies, further advances were made in the areas of pho-
tonics and optical communications. A selection of technical references related to other
technologies can be found in section 21 of the appendix at the end of the book.

Beginning in 1980, the Technology Program supported theoretical analyses that pre-
dicted, under certain system and background light conditions typical of deep space
applications, the ability to communicate at more than 2.5 bits of information per detect-
ed photon at the receiver. Laboratory tests later confirmed these theoretical predictions.
However, detection power efficiency was only one of the many factors that needed to
be studied to bring optical communications to reality. Others included laser transmitter
efficiency, spatial beam acquisition, tracking and pointing, link performance tools, flight
terminal systems design, definition of cost-effective ground stations, and mitigation of
Earth's atmospheric effects on gound stations.

Several system-level demonstrations were carried out as this work progressed. The first,
carried out in December 1992, involved the detection of a ground-based pulsed laser
transmission by the Galileo spacecraft during its second Earth fly-by. The second demon-
stration was carried out over the period November 1995 to May 1996 with the Japanese
Earth-orbiting satellite ETS VI.

Both experiments yielded important observational data in support of theoretical studies

and encouraged the further development of optical communications technology with
supportive flight demonstrations. A selection of technical references related to optical
communications development can be found in section 22 of the appendix at the end
of the book.

DSN SCIENCE

Science and technology have always been closely coupled in the DSN. Since the very
beginnings of the DSN, its radio telescopes had provided world-class instruments for
radio astronomy, planetary radar, and radio science. Many technology program achieve-
ments were of direct benefit to these scientific endeavors, and DSN science activities
frequently resulted in new techniques that eventually found their way into the opera-
tional Network.

In the period reported here, the program supported radio astronomy investigations relat-
ed to the formation of the stars and to the study of microwave radio emissions from
Jupiter, as well as radio science measurements of the electron density in the solar plas-
ma outside the plane of the ecliptic (Ulysses). It also supported a program of tropospheric
delay measurements which would be of direct benefit to the Cassini gravitational waves
experiment. The Goldstone Solar System Radar (GSSR) continued its highly successful
series of Earth Crossing Asteroid (ECA) observations,which began with images of Toutatis
(asteroid 4179) in 1992 and continued with Geographos (asteroid 1620) in 1994 and
Golevka (asteroid 6489) in 1995. This work was expected to increase in the years ahead
as new and improved optical search programs enabled the discovery of more ECAs. A
selection of technical references related to DSN science can be found in section 23 of
the appendix at the end of the book.
THE TDA SCIENCE OFFICE

In addition to their principal function of tracking and acquiring data from distant space-
craft, the unique capabilities of the Deep Space Network were also used for scientific
research in the fields of radio science, radio astronomy, radar astronomy, crustal dynam-
ics, and the Search for Extraterrestrial Intelligence
- (SETI). the completion of the
first 64-m antenna at Goldstone in 1966, radio astronomy, which explores the origin
and location of natural sources of radiation in the galaxies, and radar astronomy, which
investigates the surface and shape of planetary bodies by studying the characteristics of
a reflected radar return, became of much -greater significance
- in the overall function of
the Network.' Originally, both radio astronomy and radio science were included in the
term "radio science." As they commanded more attention and more resources from the
DSN, it became necessary to identify and manage them independently of each other.
They are discussed separately in later sections of this chapter.

From 1967 through 1995, the vigorous efforts of Nicholas A. Renzetti brought the
potential of the DSN as an instrument for scientific research to the attention of the sci-
entific community throughout
- the United States, Germany, France, Italy, Australia, and
Japan. It was his advocacy that led to the negotiation of international cooperative exper-
iments in radio science, as well as radio and radar astronomy, and to funding for U.S.
interagency support agreements for technology development in the fields of crustal dynam-
ics and SETI.

While the work in radio science, radio astronomy, and radar astronomy directly involved
the DSN in development, engineering, and the use of the DSN antennas and facilities,
the work in crustal dynamics, except in its very earliest stages, was independent of the
mainstream of DSN activity. Other than its need for observing time on the 70-m anten-
nas, the SETI program was also independent of the DSN. It was the strong cross-coupling
between the technologies used by both these latter programs and those of the DSN that
justified their residence at JPL and their association with the DSN. As the following
paragraphs will show, DSN association with the crustal dynamics and SETI programs
was relatively short-lived, while radio science, radio astronomy, and radar astronomy
became permanent features of the DSN mission set.

In 1983, Peter T Lyman, then Director of Telecommunications and Data Acquisition at

JPL, created an office to coordinate and properly manage the several streams of science
activity with which the DSN was becoming involved. The TDA Science Office would
guide and coorhnate all of JPfi efforts on behalf of NASA in the fields of radio science,
SETI, crustal dynamics, radio astronomy, and planetary radar astronomy (Goldstone Solar
System Radar). Nicholas A. Renzetti was appointed as its first manager.

Nicholas Renzetti had already established a successful career in research and develop-
ment when he came to JPL in 1959. Reared in the Bronx, New York, as the son of an
Italian immigrant family with a will to succeed, Renzetti had, by 1940, developed those
humble beginnings into a Ph.D. from Columbia University, New York During World
War 11, Renzetti was engaged in research on countermeasures for acoustic and magnet-
ic mines for the Navy's Counter-Mine Warfare Service. In the decade that followed,
Renzetti was involved with research on the ballistics of rocket projectiles and develop-
ment of test facilities and optical and electronic instrumentation for missiles at the Naval
Ordnance Test Station, China Lake, California. The years 1954 to 1959 saw Renzetti
working as a physicist, conducting research on the physical and chemical processes of
the Los Angeles atmosphere under the auspices of the Air Pollution Foundation, chaired
by President Lee A. Dubridge of Caltech.

When Eberhardt Rechtin hired Renzetti in 1959 to manage the actual operation of the
new Network, he brought a wealth of in-the-field experience to the position. As man-
ager of the ncw DSIF Communications and Operations Section in Rechtin's
Telecommunications Division, Renzetti's experience served him well in the years to fol-
low, when the Network expanded and the spacecraft it was called upon to support
increased in number, complexity, and institutional affiliation.

Stocky in build, Renzetti always dressed in a sport shirt and bolo tie (thin leather cord
fastened with a silver or turquoise bolo). He made an exception on formal occasions,
when he favored a bright bow tie. A cigar or curved Sherman pipe completed the pic-
ture. He was a gruff man, loud of voice, and inclined to shout for emphasis when he
was excited or agitated. Renzetti's phone conversations or exchanges at staff meetings
were frequently audible to people in nearby offices or corridors. Behind that rather intim-
idating exterior lay an extraordinary gentleness that became evident when, on occasion,
he invited his engineers to join his family Christmas celebrations.

From the outset, Renzetti determined to leave a written record of the work and accom-
plishments of the DSN. He insisted that his people produce publishable accounts of all
of the spaceflight operations in which they were involved, and he saw to it that all
research and development work associated with the DSN, even though it was performed
in areas beyond his responsibility, was likewise properly reported. The formal documents
that he established for publishing these reports were known as the DSN (or later, TDA)
Progress Reports. They remain a testament to his success and a permanent record of the
accomplishments of the Network and those associated with it.

Always promoting the potential of the DSN for purposes other than tracking planetary
spacecraft, Renzetti left the area of DSN "operations" in 1983 to head the DSN Science
Ofice. There, he led the DSN effort in radio astronomy, radar astronomy, geodynam-
ics, and the Search for Extraterrestrial Intelligence. The value of his continuous support
for and promotion of the Goldstone Solar system Radar for Planetary Radar Astronomy
was recognized by the International Astronomical Union in the naming of Asteroid
Renzetti in 1997.

He retired from JPL in 1996 and died at his home in San Marino, California, in 1998.

The combination of sound scientific leadership and NASAs liberal policy for support
of ground-based radio science in the Deep Space Network proved very productive. A
quantitative measure of its productivity is depicted in Figure 8-1, which shows the num-
ber of papers published in the scientific literature by investigators using the facilities of
the DSN for radio astronomy, radio science, and radar astronomy for over thirty-five
years, from 1962 to 1997.

Note: In the absence of a better estimate, the data shown in Figure 8-1 for Goldstone
radar astronomy represents fifty percent of the total number of publications in the lit-
erature for the years 1962 to 1987.2

At the end of 1995, Renzetti moved to a technical st& position in the TDA director's
office, and management of the TDA Science Ofice passed to M. J. Klein. By that time,
the scope of the effort in the ofice had diminished somewhat. Two of the major sci-
ence programs described later in this chapter, crustal dynamics and the Search for
Extraterrestrial Intelligence Project (SETI), were no longer under the purview of the
TDA Science Ofice. The work on crustal dynamics had moved out of JPL to the
National Geodetic Survey, and SETI had been canceled. A third program, radio science,
became more the province of the flight projects as Galileo arrived at Jupiter and com-
menced an intense two-year campaign of Radio Science experiments with multiple Jovian
satellite encounters.

Nevertheless, radio astronomy conrinued to flourish under the strong scientific leadership
of Thomas B. Kuiper, while radar astronomy, which had been brought to prominence
in the scientific world by Richard M. Goldstein, expanded the scope of its scientific inves-
tigations under the direction of Steven J. Ostro.
 DSN Radio Science
Publications: 1965-1 995

30T

INumber Published

DSN Radio Astronomy

Publications: 1965-1995

INumber Published

DSN Radar Astronomy

Publications: 1962-1997

I INumber Published

Figure 8-1. DSN science-related publications, 1962-97.

In a major reorganization of the entire TDA organization the following year, the TDA
Science Office was incorporated into a new Plans and Commitments Program Office and
its activity was identified collectively as "DSN Science." Klein remained as manager.

"The DSN is recognized internationally as a world[-]class instrument for radio astron-

omy," said Klein in 1997, "and allows astronomers to take advantage of the very latest
technology in Earth[-] and space-based interferometry to produce first[-Irate science. The
close symbiotic relationship that has developed between technology in the DSN and sci-
ence in the DSN serves as a valuable stimulant to the benefit of both. In addition, a
strong constituency for science in the DSN promotes improvements in both technical
and cost performance for the DSN, and it enhances the image of NASA in the inter-
national scene by virtue of the significance of its contributions to the sciences of radio
and radar astronomyn3
RADIO SCIENCE

By the time the DSIF was established in 1958, radio astronomy had been in existence
as a scientific discipline for many years. Radio astronomers, supported largely by the
NSF, generally directed their attention to extragalactic radio sources, leaving the explo-
ration of the bodies and media of the solar system to optical astronomers.

Thus, the availability of radio-equipped spacecraft traversing the immediate vicinity of

the inner planets encouraged scientists to explore ways in which the downlinks from
these spacecraft might be used to yield new data about the planets, the Sun, and the
interplanetary medium. Furthermore, this type of science required no additions to the
spacecraft science payload and, for the most part, made use of the signals already gen-
erated by the DSN for its own use in spacecraft navigation. The new science became
known as radio science and was always associated with a particular spacecraft and the
total science inventory of the flight project to which it bel~nged.~ Radio science inves-
tigations for the flight projects formed the major science activity in the DSN for the
first half of the 1960s. Using only the Doppler and range data generated by the track-
ing system, a great deal of significant new planetary science was generated in this period.

In 1981, Remetti and Allen L. Berman published the first well-documented description
of radio science in the DSN.5 Later, Sami W Asmar, then manager of radio science in
the Network, expanded this work and brought it up to date through 1993.6

NOTE The DSNpublished Lutev versions of the Radio Science System online at http://radio-
science.jpl.nasa.gov.

It was Sami Asmar who explained the basic technique of radio science. "By examining
small changes in the phase and/or amplitude of the radio signal received from a plane-
tary spacecraft, radio science investigators are able to study the atmospheric and ionospheric
structure of planets and satellites, planetary gravitational fields, shapes and masses, plan-
etary rings, ephemerides of planets, solar plasma, magnetic fields, cometary dust, and
several aspects of the theory of general relativity."

The early radio science investigators relied on the Doppler and range data generated by
the DSN Tracking system, primarily for spacecraft navigation purposes, to provide the
minute changes in phase and amplitude of the radio signal that were clues to the sci-
ence they were seeking from a distant planet or satellite.
Doppler data represents the relative motion (velocity) between the spacecraft and the
Earth station. It is derived from the measured frequency difference between the trans-
mitted uplink signal and the received downlink signal. Range data represents the distance
(range) between the spacecraft and the Earth station. It is derived from the measured
time delay between a precision range code transmitted on the uplink and its delayed
image received from the spacecraft on the downlink. Special radio transponders carried
on the spacecraft receive the uplink from Earth, add telemetry data, and retransmit the
signal back to Earth without introducing any unwanted changes in the phase charac-
teristics of the signal. Such a downlink frequency is said to be "coherent" with the uplink
frequency; therefore, any phase changes observed by the receiving station are attributed
to the "relative velocity" of the spacecraft.

The spacecraft also carries a very stable radio-frequency oscillator of known frequency
that can be used to supply the downlink frequency when no uplink is present. In this
case, "one-way" Doppler is derived from the measured difference in frequency between
the known frequency transmitted by the spacecraft and the actual frequency received by
the Earth station.

Before it can be used by radio scientists or spacecraft navigators, Doppler and range data
must be refined (to remove ~erturbationsintroduced by the interplanetary media though
which it has traveled) and calibrated (to remove errors introduced by the radio circuit-
ry in the spacecraft, ground receivers, and transmitters).

The tracking system is the element of the overall DSN architecture that generates Doppler
and range data from the coherent uplink and downlink between spacecraft and tracking
station. It provides these data, along with appropriate time tags and ancillary data, in a
suitable format for use in spacecraft navigation, orbit determination, and radio science.

To generate radio metric data of great precision and stability, the tracking system requires
an ultra-stable clock as a reference, or yardstick, for its uplink and downlink frequency
measurements. In the early years, this was provided by an atomic clock at each Complex.
These were rubidium and cesium frequency standards with stability characteristics of
one part in 1OI2 (rubidium) or one part in 10" (cesium).

NOTE To appreciate the significance of these numbers in the time domain, a simple cal-
culation willshow that one part in ten raised to the thirteenthpower (I x I P ) is approximately
equivalent to three second per one million years.
The rubidium and cesium atomic clocks were eventually superseded by wen more sta-
ble clocks driven by hydrogen masers (I x lo-'*). Equally stringent requirements for
phase and frequency stability applied to the spacecraft radio system where the range and
Doppler "turnaround" functions were performed.

Radio metric data generated by the tracking system was used for many applications in
radio science. Doppler data from spacecraft on planetary flyby or orbiting missions pro-
vided information regarding the orbit, mass, gravitational, and physical properties of the
planetary bodies for celestial mechanics investigations. It also provided planetary atmos-
pheric information as the downlink signal was refracted through planetary neutral
atmospheres. When dual-frequency downlinks (S-band and X-band) became available
(first on Mariner 10 (VenuslMercury) in 1973), it enabled measurements of columnar
electron density changes in the solar wind and planetary ionospheres.

Range data was used to measure planetary distances, as well as the signal time delay due
to the solar wind (electron density) and gravitational (relativistic) bending near the Sun.

In 1976, the various components (such as open-loop receivers, stable oscillators, and
recorders) that had been used for the previous decade or so to make radio science obser-
vations were incorporated into a new DSN system designed to handle the unique needs
of radio science for all flight projects. Known as the Radio Science System, it generat-
ed radio science data as a data product for delivery to the Flight Project Science Team
in the same way as telemetry or radio metric data was considered to be a "deliverable."
Together with the tracking system and ancillary calibration data, the Radio Science System
generated all of the DSN radio science data for the planetary missions after this time.

As described by Berman, the essential feature of the new DSN Radio Science System
was its ability to reduce the amount of radio science data to be recorded (digitally) for
each occultation by a factor of 10 to 100, compared to the f ~ e dtuned analog record-
ing technique previously employed by the Network. This was accomplished by
automatically tuning the open-loop receivers in conformance with a predicted Doppler
frequency profile based on precise knowledge of the planet and spacecraft orbits, along
with a model of the planetary atmosphere. Depending on the accuracy of the "predicts"
and the nature of the data itself, the output of the open-loop receiver would then become
a new "baseband frequency lying within a few kHz of the predicted frequency and
could easily be digitized and written on computer-compatible tape at the observing sta-
tion. In addition to the "baseband" frequency, the digital recording contained a copy of
the frequency profile used to generate the "baseband" data with appropriate timing infor-
mation. This information was then shipped to the experimenter, transmitted to JPL via
/ \
^SHIPPED

4
RADIO SCIENCE
EXPERIMENTOR
\ J

Figure 8-2. Real-time bandwidth reduction in the radio science Occultation Data Assembly.

the GCF, or both. By combining the "predict" frequency with the "baseband" frequen-
cy, the experimenter could recover the actual received frequency. The new system also
provided real-time spectral data for display on Signal Spectrum Indicators (SSIs) at the
tracking station and in the Radio Science Team area at JPL. The SSIs were used for
monitoring performance, calibration, and analysis purposes, as well as for quick-look sci-
ence evaluation.

The new capability, referred to as "real-time bandwidth reduction," was delivered to the
Network for first operational use with the Pioneer Venus mission in December 1978.
The Occultation Data Assembly, a part of the Radio Science Subsystem in which the
bandwidth reduction function described above was performed, is illustrated in functional
form in Figure 8-2.

As an integral part of the DSN, the Radio Science System was upgraded from time to
time to reflect the introduction of new technology, processes, and capability in the over-
all DSN. In 1996, the Radio Science System acquired a capability that allowed the Radio
Science Team at JPL to share control of the Radio Science System at the Complexes
with the station operators. This placed the burden of operating the complicated Radio
Science System back on the user group and reduced the responsibility and workload of
the DSN tracking stations. At last, the Radio Science Teams would have the same flex-
ibility and control of their instrument as the other science experiment teams enjoyed
with instruments carried by the spacecraft. Remote control for radio science was first
used for Galileo orbital operations in late 1996 and, after some initial integration prob-
lems, proved to be of great value to both the investigators and the DSN in dealing with
the heavy load of radio science through the end of the Galileo Mission.

Renzetti and Berman suggested that the DSN (then DSIF) radio metric capabilities
should be first used for radio science purposes by Anderson and Warner for celestial
mechanics experiments with the 1962 Mariner 2 mission to Venus. Using only the
Doppler navigational data from this flight, the experimenters were able to make order-
of-magnitude improvements in the determination of the masses of the Moon and Venus.
All through the Mariner Era, the Mariner 4 (Mars), Mariner 5 (Venus), and Mariner 9
(Mars) missions were used for celestial mechanics experiments to determine the masses
of Venus and Mars.

Doppler measurements from the Lunar Orbiter and Apollo missions led to the discovery
of mass concentrations (mascons) on the Moon, and analysis of the Mariner 10 Doppler
data in 1973 and 1974 enabled the determination of the mass and shape of Mercury.

In the Voyager Era, celestial mechanics investigators used the DSN radio science data
from the Pioneer and Voyager missions to Jupiter and Saturn to produce a wealth of
new knowledge on the physical properties of those planets and their satellites.

Venus had received intense study by celestial mechanics investigators since the time of
the Mariners. Data from long-term orbiting spacecraft, most notably Pioneer Venus Orbiter
and Magellan Radur Mapper, produced a data volume equivalent to thousands of single
flybys, making the gravity-field of Venus almost as well understood as that of Earth.

Celestial mechanics studies of Mars also had the benefit of orbiting spacecraft, starting
with Mariner 9 in 1971 and the KWO Ekings in 1976. These studies would have been
extended with Mars Observer in 1993 had the spacecraft not been lost just prior to
entering Mars orbit. They were, however, resumed with the Mars Global Surveyor mis-
sion in 1997.
Prior to the Galileo mission to Jupiter in 1995, knowledge of the gravity fields of Jupiter
and its satellites was based on the combined results of celestial mechanics experiments
conducted during the flybys of Pioneers 10 and I I and Vyagers I and 2. This had pro-
vided a good estimate of the Jupiter mass but contributed little to existing knowledge
of the satellites. Although the GaliIeo orbits of Jupiter were too high to contribute much
to gravity field measurements of the planet itself, the close Encounters of 10, Ganyrnede,
Callisto, and Europa provided an immense amount of data for analyses of the internal
structure of these Jovian satellites.

During the period July 1996 to December 1997, investigators carried out twenty-three
celestial mechanics experiments while the Galileo spacecraft was in orbit around Jupiter.
The purpose of this campaign was to refine the knowledge of the position of Jupiter in
the plane of the sky which, in turn, led to a significant improvement in knowledge of
the ephemeris of Jupiter. To obtain the increased accuracy required for these observa-
tions, the DSN generated ranging data from several stations simultaneously during the
spacecraft overlap view period between Complexes, with the spacecraft in the "one-way"
mode. Reduction of these data to extract the simultaneous spacecraft-to-station, path-
length differences from the known station locations yielded data from which a very
precise determination of the spacecraft orbit could be made. Investigators were then able
to refine their knowledge of the position of Jupiter from its effect on the very precise-
ly known orbit of the spacecraft. The range differencing technique used in these experiments
was known as Delta Differenced One-way Ranging (Delta DOR).

The Vyuger I, Vqyager 2,and Pioneer I 1 Encounters of Saturn contributed radio sci-
ence data for celestial mechanics estimates of the masses and gravity fields of Saturn and
its satellites Rhea, Titan, Tethys, and Iapetus. A previously unknown moon, S13, was
first discovered by its gravitational perturbations of the Doppler data from the Vqyger
2 Saturn Encounter. Its presence was later verified by optical imaging from the space-
craft.

Further celestial mechanics investigation of Saturn must await the arrival of Cmsini with
the Huygens Titan Probe in 2004. Then radio science data at X-band and Ka-band
from multiple Cassini orbits through the Saturnian system will provide investigators with
additional information on masses, densities, and interactions between the planet and its
satellite Titan.

Celestial mechanics information on Uranus was derived from the radio science data taken
during the Vyuger 2 Encounter of the planet in 1986. Combined with imaging data of
the satellites, the Doppler perturbations enabled great improvement in determination of
the masses of five of the major satellites and of the planet itself.

The two-way Doppler data obtained by the DSN Radio Science System during the
Vyager 2 Close Encounter of Neptune in 1989 provided celestial mechanics investiga-
tors with basic data from which the masses of Neptune and its satellite Triton could be
estimated to an accuracy of 0.0003 percent and 0.3 percent respectively.

In 1965, as the Mariner 4 (Mars) spacecraft passed within 0.6 degrees of the solar disk,
Richard M. Goldstein used the not-yet-fully-operational 64-m antenna (DSS 14) at
Goldstone with an "open-loop" receiver to observe the spectral spreading of the space-
craft signal as it passed through the solar corona. This marked the first use of the DSN
for Solar Corona research. Berman, and later Woo, were able to interpret the observed
results in terms of the relationship between spectral broadening and integrated coronal
electron density. Subsequently, similar experiments with the Helios and Pioneer space-
craft would confirm and expand the original measurements and the underlying theories.

In addition to spectral broadening, the radio signal from a planetary spacecraft experi-
ences a second effect as it passes through the Solar Corona. The RF signal experiences
a small, but measurable, rotation of its plane of polarization. Known as "Faraday Rotation,"
this effect was linked to both electron density and the solar magnetic field in an exper-
iment carried out at DSS 14 by Charles T Stelzried in 196fL7To support this experiment,
DSS 14 had been equipped with a rotatable, linear microwave feed system driven by a
closed-loop polarimeter that would automatically track the orientation of the polariza-
tion of the received signal. Using the linearly polarized signal from the Pioneer Gspacecraft
as it was occulted by the Sun in November 1968, these observations yielded valuable
information on both transient and steady-state effects in the solar corona.

Similar experiments were carried out at the other 64-m stations with Helios and Pioneer
spacecraft in 1975 and 1977. An ambitious Faraday Rotation experiment, planned for
the Galileo mission to Jupiter in 1995, was eliminated in 1991 when the failure of the
Galileo high-gain antenna limited the downlink to S-band right circular polarization only.
However, successful Faraday Rotation experiments using the Magellan spacecraft at Venus
were carried out by Asmar and Bird in 1991 at both S-band and X-band.

Range data generated by the DSN tracking system also provided a powerful tool for
determination of electron density in the solar corona, provided that all other sources of
error could be properly accounted for. Duane 0. Muhleman, John D. Anderson, and
others first used this method during the solar conjunctions of Mariners 6 and 7 in 1970,
using ranging data acquired at DSS 14.

During the 1975 solar conjunctions of Pioneer 10, Pioneer 11, and Helios I , M e n L.
Berman and Joseph A. Wackley of JPL, working in collaboration with Jose M. Urech
at the Madrid complex, were able to demonstrate that the fluctuations observed in the
Doppler signature of a single S-band downlink provided an excellent measure of tur-
bulence in the solar wind. Since Doppler fluctuations (or "noise") were automatically
computed for monitoring the performance of the DSN tracking system, it represented
a "free" source of scientific data and was compared favorably with the more accurate
dual-frequency methods. Electron density measurements based on dual-frequency S- and
X-band Doppler were not degraded with many of the sources of error present in sin-
gle-frequency measurements. Measurements with dual-frequency Doppler downlinks on
the Mariner 10 spacecraft in 1974 and dual-frequency ranging on the Eking spacecraft
in 1977 and 1979 yielded excellent results.

Observations of Doppler
-- scintillation from Pioneer Venw were compared with inde-
pendent observations of solar plasma transients from Helios 1 by Woo and Schwen in
1991. They showed that several Doppler transients, observed by Pioneer Venw at 15 to
18 solar radii, were observed about 48 hours later as solar plasma shocks by Helios I at
170 to 174 solar radiL8

Solar wind scintillation experiments played a large part in the Galileo radio science pro-
gram. The experiments were designed to gather data for the study of the solar wind
within 0.3 AU of the Sun. The scintillation experiments were originally
- .
intended to
complement solar corona experiments but were eliminated by the loss of the HGA and
the dual-frequency downlinks. Five experiments were conducted successfully, on S-band
only, during superior conjunctions between December 1991 and February 1997.

Almost every planetary mission, past and future, is concerned with the investigation of
planetary atmospheres, ionospheres, rings, and magnetic fields. Such investigations are
based on the perturbation of the phase and amplitude of the spacecraft downlink by
the extended atmosphere of the planet as the radio signal passes through it during peri-
ods of occultation. These perturbations are contained in the data captured by the open-loop
receivers of the Radio Science System and can be converted into a refractivity profile
for the planetary atmosphere after the data are corrected for the contribution due to the
geometrical fight path of the spacecraft. Information regarding electron distribution in
the ionosphere, the temperature-pressure profile in the neutral atmosphere, or particle
size distribution in the ring material can be derived from the refractivity profile.

According to Asmar and Renzetti, the method was first proposed by V. R Eshelrnan in
1962. Independently, D. L. Cain, leading a JPL team engaged in analyzing the effect
of refraction in Earth's aunosphere on the accuracy of Doppler data, recognized the pos-
sibility of applying the sensitivity of Doppler phase measurements to the study of
atmospheres and ionospheres of other planets. The theoretical analysis was developed by
G. Fjelbo in 1964.

The occultation experiments required special receivers to extract the Doppler data in addi-
tion to the normal DSN receivers. The normal receivers in the DSN tracking system
were of the "closed-loop" type; that is, they employed a "phase-locked tracking loop" to
keep them tuned to the frequency of the downlink carrier signal. As the frequency of the
downlink changed as a result of the Doppler effect of the spacecraft's motion relative to
Earth, the "phase-lock loop" caused the receiver to follow the changes and thereby "main-
tain lock." Doppler data was generated by comparing the received downlink frequency
with the uplink transmitter frequency (two-way), or a precise estimate of the frequency
of the Ultra Stable Oscillator (USO) carried by the spacecraft (one-way).

When a spacecraft entered occultation by a planet, the downlink signal was cut off
and the receivers dropped lock. When the spacecraft exited occultation and the sig-
nal reappeared, valuable time was lost while the receivers reacquired the downlink.
Thus, the important radio science data was only available for a short period imme-
diately before entry and immediately after exit as the signal transited the thin planetary
atmosphere. Therefore, in 1964, special wide-bandwidth "open-loop" receivers were
added to the DSN tracking system to ensure that the downlink would appear in the
"open-loop" receiver passband the instant it emerged from occultation. The "open-
loop" receivers were tuned by a programmable local oscillator, which was driven by
a computer program based on the spacecraft orbit and other pertinent parameters.
The open-loop output was recorded on analog magnetic tape recorders at the station
and subsequently shipped to JPL. There it was digitized and delivered to the exper-
imenters for analysis.

Occultation experiments using open-loop receivers in the DSN were made for the first
time in 1965, when Mariner 4 flew by Mars. These experiments were highly success-
ful; they showed that the Martian atmosphere was predominantly carbon dioxide and
that the surface pressure was less than one percent of that of Earth, an order of mag-
nitude less than had previously been thought.' Measurements of the electron density in
the Martian ionosphere were also obtained. During the late 1960s and early 1970s,
occultation experiments were carried out by the Mariner and Viking missions to the
inner planets (Mercury, Venus, and Mars).

The Voyager Era was prolific in producing occultation radio science for the outer plan-
ets. This began in 1973 with the Pioneer 10 flyby of Jupiter, repeated a year later by
Pioneer 11. Pioneer I1 went on to make the first occultation measurements of the
atmosphere and ionosphere of Saturn in 1979. Occultation experiments were conducted
with Vqyagers 1 and 2 at Jupiter in 1979 and at Saturn in 1980 and 1981, with Vqyager
2 at Uranus in 1986 and at Neptune in 1989. These experiments determined atmos-
pheric pressure levels, temperatures, and constituents, as well as the characterized
ionospheres of Jupiter, Saturn, and Triton.I0The theory of radio occultation techniques
was extended in 1980 to include ring occultation experiments during the Nyager 1
Encounter of Saturn.

Ulysses experimenters used occultation radio science during their spacecraft gravity assist
flyby of Jupiter in Februay 1992 to derive the columnar electron density of the 10 plas-
ma torus. This experiment was enhanced by the presence of a dual-frequency (S- and
X-band) downlink from the spacecraft.

In 1991 and 1992, Magellan experimenters obtained highly accurate ~rofilesof atmos-
pheric refractivity and absorptivity of Venus using occultation radio science. These profiles
were used to deduce ionospheric density, atmospheric temperature, pressure, and sulfu-
ric-acid abundance on Venus. The dual frequency combined with the higher power of
the Magellan downlinks, enhanced the end result by eliminating errors and making it
possible to penetrate the atmosphere to greater depth than would otherwise have been
possible.

The orbital part of the Galileo mission to Jupiter provided many opportunities for occul-
tation radio science. In 1996 and 1997, twenty-two occultations were observed. These
included a single occultation by Callisto and multiple occultations by Jupiter, 10, Gaymede,
and Europa. The information derived from these data substantially increased the exist-
ing body of knowledge about the ionospheric characteristics of all of those bodies.

The Cassini mission to Saturn in 2004 will include several occultation flybys of Saturn
and Titan, which will be observed with S-band, X-band, and Ka-band downlinks. Data
obtained from observations in each of these bands has advantages in the study of dis-
persive media like the atmospheres of Saturn and Titan and also makes it possible to
determine the particle size distribution in Saturn ring structure.

Depending on the Earth-Sun-spacecraft geometry, range measurements fiom Earth to a

planetary spacecraft passing through solar conjunction can enable a test of a portion of
the General Theory of Relativity. Einstein's theory predicts that electromagnetic waves
passing close to massive bodies, such as the Sun, are influenced by the Sun's gravita-
tional field in such a way that the signal path between spacecraft and Earth is "bent"
by the intense gravitational force of the Sun. The effect of the "bending" is to intro-
duce a time delay along the signal path between spacecraft and Earth. This time delay
can be measured by the DSN tracking system. Using the measured signal time delay
and knowledge of the Earth-Sun-spacecraft geometry at the time the measurement was
taken, radio scientists are able determine a value for a key parameter in the mathemat-
ical equations by which the theory is described. This parameter, known as the
"Schwartzchild metric," is denoted by the Greek symbol gamma. In the theory, the value
of gamma is 1.0. For a relativistic experiment, therefore, the closer the results approach
a value of 1.0 the better its agreement with theory.

To measure signal time delay, the DSN used the ranging system which forms a part of
the DSN tracking system described earlier.

The first major attempt to measure the relativistic bending of a spacecraft signal was
made by John D. Anderson and others during the solar conjunction of Mariners 6 and
7 in the mid-1970s." The results obtained from this experiment verified the value of
gamma to about 6 percent, i.e., gamma = 1.00 + 0.06. The major sources of error were
attributed to uncertainty in the accuracy of the spacecraft orbit and in the measurement
of time delay due to scattering effects of free electrons in the solar corona. In time, these
errors were reduced, and a later experiment with Mariner 9 brought the uncertainty in
the value of gamma down to 2 percent.

A very significant improvement in accuracy resulted from the relativistic experiments

carried out in 1976 and 1977 by Shapiro, Cain, Reasonberg, and others with the Viking
Mars Orbiter and Lander spacecraft. Benefiting from significant improvements in the
calibrations of the tracking system, the use of dual-frequency downlinks, and precise
knowledge of the position of the landed spacecraft, independent experimenters at MIT
and JPL determined the value for gamma of 1.00 + 0.002 the most accurate test of the
theory to that time. Repeating the experiment again during the Vyager 2 solar con-
junction in December 1985, Anderson and others obtained a value of 1.00 + 0.03 for
gamma.

Improved accuracy is expected from two relativistic experiments to be conducted dur-

ing the solar conjunction periods of the Cassini mission to Saturn in 2003 and 2002.
The greater accuracy will derive from the use of Ka-band, which, with its wavelength
of less than one centimeter, is less affected by interplanetary media than the longer wave-
length signals used previously.

The General Theory of Relativity predicted that the violent collapse of stellar bodies
into supermassive black holes would generate gravitational waves in the form of spatial
strains propagating through the solar system at the speed of light. These waves were
expected to slightly alter the distance between two separated free masses, which would
be detectable as a proportionate fractional-frequency shift in ultra-precise Doppler data
from a space-based detector using Earth and a planetary spacecraft as free test masses.
Because gravitational waves are very weak and propagate at very low frequencies, space-
based systems offered the best chance for their detection.

The use of ultra-precise two-way Doppler from a distant spacecraft (greater than 1 AU)
to detect very-low-frequency gravitational waves was proposed by Estabrook and Wahlquist
in 1975.12Driven by an ultra-stable frequency standard, the DSN tracking system gen-
erates precision two-way Doppler from a comparison of the uplink and coherent downlink
frequencies from the spacecraft and, in doing - so, continuously measures the fractional
frequency shift due to relative Earth-spacecraft motion. A gravitational wave passing
through the system causes a unique "three-pulse" signature to appear in the Doppler
data due to a combination of buffeting effects on the Earth and the spacecraft and
speed-up of the tracking station clocks (frequency and timing systems). The character-
istics of the "three-pulse" signature depend on the relative positions of Earth and the
spacecraft and the direction of the gravitational wave. These effects are illustrated very
simply in Figure 8-3.

Experiments conducted through 1990 using Wking, Vyageer, and Pioneer spacecraft failed
to detect the existence of gravitational waves. However, the sensitivity of these tests was
limited by noise due to plasma propagation.

The advent of hydrogen masers as frequency standards for the DSN and the addition
of X-band uplinks to the existing S- and X-band downlinks improved the sensitivity of
I \

Gravitabonal
Wave Spacecraft

Earth

Earth buffet~ng Spacecraft

buffet~ng

v
Earth buffetmg,
transponded
(1 c cos €3) uc
\ J

Figure 8-3. Characteristic three-pulse signature of a gravitational wave in two-way Doppler

data.

the tracking system for detection of the gravity waves by a factor of about 10. With the
launch of Galileo in 1989, carrying X-band uplink in addition to the standard S-band
uplink and S- and X-band downlinks, expectations were high for the detection of grav-
ity waves in the Galileo Era. A particularly interesting coincidence experiment involving
Mars Observer, Galilea, and Ulysses was planned for early 1993. The sensitivity of the
tracking system had been set at 5 x 10-l5as offering the best chance for successful detec-
tion of the elusive gravitational waves. This experiment afforded several unique
opportunities for immediate confirmation of the detection of a gravity wave and deter-
mination of the direction of the incident wave from the Doppler signatures of the three
spacecraft. The joint experiment ran for three weeks in the spring of 1993. Two further
searches for gravitational waves with the Galileo spacecraft were planned for May 1994
and June 1995 during the solar opposition periods.

Unfortunately, the loss of the X-band uplink and downlink, as a consequence of the fail-
ure of the Galiieo High Gain antenna in April 1991, reduced the sensitivity of the radio
science system for gravity wave detection to that of the residual S-band mission, deemed
marginal at best. None of the experiments succeeded in detecting the presence of grav-
itational waves; nevertheless, interest in pursuing the search for gravity waves persisted.
The launch of Cassini in 1997, with Ka-band and X-band uplinks and downlinks, as
well as more stable ground and spacecraft elements in the tracking system, offered the
hope of observations at a much higher sensitivity level. It was estimated that the sensi-
tivity could be as high as 10-l6for about one month of tracking near each of three solar
oppositions in 2001, 2002, and 2004, provided that tropospheric effects could be suf-
ficiently removed with water vapor radiometer techniques.

Looking to the future for gravitational wave searches, S. W. Asmar believed that sen-
sitivity improvements beyond this level would require moving the tracking station
itself into Earth orbit so that both test masses would be in space, unperturbed by
the many unmodeled geophysical effects that are associated with an Earth-based sta-
tion. In fact, this concept is to be employed in future spaceborne, laser-based,
gravitational wave experiments.
RADIO ASTRONOMY

The origins of the DSN owe much to the science of radio astronomy. By the time of
its establishment in 1958, radio astronomy had been in existence as a scientific disci-
pline for many years. As discussed in the early chapters of this book, the original
26-m-diameter antennas of the DSIF were adaptations of existing, commercially avail-
able radio astronomy dishes and mounts. The original 64-m-diameter antenna built at
Goldstone in 1966 was based on the design of a radio telescope of similar size operat-
ed by the Australian Scientific and Industrial Research Organisation at Parkes, Australia.

It came as no surprise when the radio astronomy community showed early interest in
the substantial new global facility then being implemented by NASA for exploration of
the solar system by means of planetary spacecraft. However, the National Science
Foundation, which was the principal sponsor of radio astronomy research in the United
States, tended to promote research in regions of the cosmos beyond the solar system. It
was more interested in sponsoring studies associated with astrophysics, the galaxies, and
stars than those related to hrthering the existing knowledge base of the solar system.
Thus, the early capabilities of the DSN, designed for in situ observations of the inner
solar system with planetary spacecraft, offered little of interest to radio astronomers depend-
ent on the NSF for their sponsorship.

In 1966, the first of the DSN 64-m antennas was put into operation at Goldstone.
With that event, radio astronomy, as a scientific endeavor distinct from that associated
with planetary spacecraft, became a new capability for the DSN. The new antenna, over
six times more sensitive than the existing 26-m antennas, more than doubled the range
of the Network and, in doing so, provided the best instrument available for radio astron-
omy at that time. With such an instrument, radio astronomers could now direct their
research to regions beyond the solar system with a capability that was comparable to,
and in many ways better than, that of the other large radio telescopes throughout the
world. Prompted by a strong advocacy from Nicholas A. Renzetti, who was managing
tracking operations for the DSN at that time, the NASA Office of Tracking and Data
Acquisition (OTDA) approved the allocation of a limited amount of time on the new
antenna for radio science observations on a noninterference basis with the antenna's
prime task of tracking planetary spacecraft. Ofkial notice of the availability of the new
facility brought an immediate response from the radio astronomy community.

In 1967, the DSN was approached by David S. Robertson of the Australian Space Research
Group with the first proposal for a long baseline interferometry measurement of extra-
galactic radio sources. This was to be done using the baseline between Goldstone and
DSN antennas at Woomera and Tidbinbilla in Australia, which would make a resolving
power of 2 x 10" arc-second possible. W ~ t hM e n T. Moffet as a coexperimenter, the first
observations were made in 1967 and continued at intervals for several years thereafte~.'~

In 1968 and 1969, DSS 14 was used extensively by Moffet of Caltech and Ekers and
Downs of JPL to study pulsar emissions. By 1969, the volume of requests for observing
time on the antenna had increased to the point where it became necessary to evaluate and
select the most appropriate and worthy non-NASA proposals to compete for the limited
amount of available antenna time. A Radio Astronomy Experiment Selection Panel (RAES),
composed of distinguished radio astronomers nationwide, was established for this purpose,
and a DSN radio astronomy support group was created to facilitate the technical and logis-
tical interfaces between the astronomers and the operational DSN. These organizations were
announced to the radio science community in the technical literature at the end of 1969.

The next two years saw a substantial increase in the demand for radio astronomy time
at DSS 14 and DSS 13 as pulsars, very long baseline interferometry, Jupiter radio emis-
sions, brightness, temperature, and rotation period continued to attract the attention of
radio astrononier~.'~

Management policy for support of "Ground-Based Radio Science" in the Deep Space
Network was formalized by NASA in 1971 with an appropriate NASA Management
Instruction setting forth the requirements for submission of requests for experiment sup-
port and the levels of approval required. A level of support was determined in terms of
percentage of available antenna time and assigned to the RAES panel, and its subse-
quent equivalent for distribution amongst approved experimenters. To distinguish
between the radio experiments involving spacecraft and those involving celestial bodies,
NASA adopted the following definitions.

"Radio science" would refer to the acquisition and extraction of information from space-
craft-originated signals that had been affected by celestial bodies or the propagation
media. "Radio astronomy" would refer to the acquisition and extraction of information
from signals that had been emitted by or reflected from natural (non-spacecraft) sources.

In the past, all experiments had been loosely referred to as "radio science," but from
this point forward, the DSN would observe these new conventions in its management
and support for radio science and radio astronomy. Although at that rime the impor-
tant planetary radar observations being made at Goldstone came within the definition
of radio astronomy, it would later evolve into a separate activity as the Goldstone Solar
System Radar (GSSR).
For the next decade and a half, radio astronomy continued to play an important role in the
total sum of DSN activity. As the DSN capability increased with the construction of 64-m
antennas in Australia and Spain, and later with their enlargement to 70-m diameter, so did
the capability for radio astronomy, and the scientists were quick to use them to advantage.

The level of radio astronomy activity in the DSN over the period 1971 through 1987
is summarized below:

Radio Astronomy in the Deep Space Network: 1967-87*

1967-7 1
Various Sponsors, NSF, CIT, JPL, Australia
Twenty experiments, various subjects
1972-87
RAES-Sponsored Research
Sixty discrete experiments in various subjects
NASA-Sponsored Research
Five continuous experiments in the following:
Interstellar microwave spectroscopy
Planetary radio astronomy
Hipparcos VLBI
Quasar and galactic nuclei VLBI
Southern Sky Survey with Tidbinbilla Interferometer

While individual radio astronomers in the United States continued to carry out their
work through the scheduling and interface organization in the DSN Operations Office
at JPL, associations of a more international nature were also formed during this period.

In 1983, the Goldstone 70-m antenna cooperated with the National Radio Astronomy
Observatory 43-m antenna and the 22-m antenna of the Center for Astrophysics in the
Crimea, USSR, in an interferometric network to study extragalactic sources at a wave-
length of 3 cm with high resolution. Thus was forged the DSN's first close cooperative
association with radio astronomers from NRAO, Cornell, Caltech, and the Soviet Union.
This association would continue to be productive in the field of radio astronomy research
in the years that followed.
In 1972, Samuel Gulkis of JPL, in association with David L. Jauncey and Michael J.
Yerbury, then of Cornell University, put forward a proposal that would use the two exist-
ing DSN antennas at the Canberra Deep Space Communication Complex in Australia to
create a high-sensitivity, phase-stable interferometer that could be operated in near-real
time. The two antennas (26 m and 64 m in 1972, later enlarged to 34 m and 70 m)
were separated by 195 meters on a north-south baseline and, together with available low-
noise maser amplifiers and a common operating frequency at 2.3 GH5 provided a unique
opportunity to implement a high-sensitivity interferometer with a positional capability of
approximately 2.0 arc-seconds at minimal cost. It would be the only instrument in the
Southern Hemisphere capable of carrying out rapid measurements of weak radio source
positions with arc-second accuracy and flux densities to the millijansky level. These data
would be used for unambiguous identification of radio sources with cataloged infrared and
optical celestial objects. Earlier, whole (southern) sky surveys made with the Parkes Radio
Telescope at 2.7 GHz had listed over 12,080 radio sources, but these data did not have
sufficient positional accuracy to permit confident identification with visible objects and left
unanswered many questions of interest to radio astronomers. The proposed interferome-
ter would be used to supplement these data and provide a complete southern sky survey

The Tidbinbilla two-element interferometer was constructed and tested in the years 1977
to 1979 and began radio astronomy observations at 2.3 GHz in 1980.15It was upgrad-
ed to operate simultaneously at 2.3 GHz and 8.4 GHz in 1986.

By 1988, over 1,500 of the radio sources listed in the Parkes survey had been measured,
a d many had been identified optically A great deal of attention was paid to the study
of quasars. The identification in 1982 by Jauncey and others of the most distant quasar
discovered to that time was attributed solely to the accuracy of the Tidbinbilla interfer-
ometer.l6 A sensitive search of the brighter stars in the Southern Hemisphere for possible
radio emissions was initiated in 1986;depending on the narrowness of the interferometer
beam to reduce codhion with the background galactic sources of radiation. Data like
these would be used by the Space Telescope and other astronomical spacecraft (Hipparchus)
as well as for navigation of conventional spacecraft and station location purposes.

The Tidbinbilla interferometer continued to operate as a significant capability for radio

astronomy in Australia under the Host Country program described below.
ORBITING
VLBI

The feasibility of using a spacecraft in Earth orbit as one element of a very long base-
line interferometer for radio astronomy observations was demonstrated by Gerry Levy
and his team in 1983.17 Building on these ideas, James Ulvestad, Roger Linfield, and
Chad Edwards applied these principles to make successll observations of Extra Galactic
Radio Sources (EGRSs) in August 1986. The enormous increase in baseline length (sev-
eral Earth diameters) thus obtained would provide a proportionate increase in resolving
power, greatly exceeding that of an interferometer constrained by Earth-based antennas.
The experiment was conducted at a frequency of 2.3 GHz using the 4.9-m antenna of
the geosynchronous NASA Tracking and Data Relay Satellite System (TDRSS) as the
space-borne element. The DSN 64-m antenna at Canberra, Australia, and the ISAS 64-
m antenna at Usuda, Japan, formed Earth-based elements. In January 1987, 23 EGRSs
were detected on a baseline as long as 2.15 Earth diameters.

These successful experiments demonstrated that a ground-based frequency reference could

be accurately transferred to an orbiting spacecraft and that observations of radio sources
acquired by an orbiting spacecraft could be transmitted and recorded on Earth with the
necessary precision to achieve correlation with similar data from Earth-based antennas. The
success of these early experiments had positive implications for the future of orbiting VLBI.

Within the next ten years, the DSN would become a participant in the Space VLBI
cooperative program involving NASA (U.S.A.), ISAS (Japan), and ASC (Russia). The
first of two space VLBI missions, VSOP (HALCA),would be launched in 1997 and a
special subnetwork of 1l-m tracking antennas, operating at Ka-band, would be added
to the Network. The DSN 70-m antennas would also participate as co-observers in the
Space VLBI project with many other radio astronomy observatories around the world.

The original intergovernmental treaties that were executed between the United States,
Spain, and Australia for the implementation and operation of Deep Space Communications
Complexes in those countries made explicit provision for the conduct of independent
radio astronomy research on these facilities by Host Counuy agencies on a noninter-
ference basis with the prime NASA business of spacecraft tracking. In the DSN resource
planning process, "Host Country" programs were allotted a specified amount of anten-
na time on an annual basis, and it was left to the local complex directors to assign the
time to the individual investigators as appropriate.
In Spain, the prime cooperating agencies were the Instituto Geografico National and the
Instituto Nacional Tecnica Aeroespacial, whiie in Australia, the prime cooperating agency
was the Commonwealth Scientific and Industrial Research Organisation (CSIRO). In
both countries, the DSN antennas participated in interferometric geodetic surveys of the
continents and observation of pulsars under the Host Country program. In 1992, the
Tidbinbilla interferometer was brought into the Host Country program and continued
to share antenna time with other Australian radio astronomy investigations at the CDSCC.

The important role played by the Parkes Radio Astronomy antenna in supporting the
Voyager Neptune Encounter as an element in the Parkes Canberra Telemetry Array
(PCTA) was discussed in an earlier chapter. The NASAICSIRO negotiations for this
support in September 1986 resulted in a Memorandum of Understanding. This docu-
ment provided time on the DSN 70-m antenna for the use of CSIRO radio astronomers
on a quid pro quo basis for the time provided by Parkes for the Voyager Encounter.
This "cross-support" agreement was a productive source of radio astronomy science,
although the over-subscription of the DSS 43 antenna for spacecraft support in later
years made it difficult for the DSN to meet the original expectations for "payback."

Under the Guest Observer Program, DSN facilities were made available to non-JPL sci-
entists for radio astronomy research, subject to peer review and approval of their proposals
by the U.S. National Radio Astronomy Observatory (NRAO). The purpose of the review
was to establish the scientific merit of the proposal and verifi its 111 dependence on
the unique capabilities of the Deep Space Network for success. Similar review process-
es applied to the use of DSN facilities for cooperative VLBI experiments with other
agencies in the United States and Europe.

The numerous agreements, directives, and requirements for radio astronomy support,
accumulated through the intervening years, were brought up to date by Larry N. Dumas,
director of the TDA office at JPL in 1990. With the agreement of the program office
at NASA Headquarters, Dumas established new limits for the DSN antenna time that
would be applied to radio astronomy throughout the Network.
NASA- and RAES-approved experiments would not exceed 3 percent of the antenna
time available at any operational DSN station. NASAICSIRO cross-support experiments
using DSS 43 would receive a minimum of 108 hours per year through August 1999,
with a 34-m antenna-hour equivalence of two for one. Host Country experiments would
not exceed 3 percent of the time available at any one of the stations available at each
complex (approximately 250 hours per year).

The new guidelines formed the basis for scheduling DSN resources for radio astrono-
my support thereafter.

A summary of antenna utilization for the ensuing ten years, given in the table below,
reveals how well the guidelines were followed in the operational environment that pre-
vailed through the Network in those years.

The figures represent hours of observing time on the 70-m and 34-m DSN antennas
and represent the fulfllment of the 1988 guidelines discussed above.

Radio Astronomy in the Deep Space Network: 1987-97

(Observing Antenna Hours)

Year JPLINASA Guest Other* Total
1997 1,999 253 712 2,964
1996 2,735 129 4,691 7,555
1995 1,884 115 1,201 3,200
1994 1,196 531 1,316 3,043
1993 1,379 234 1,086 2,699
1992 1,947 217 1,496 3,660
1991 1,733 128 887 2,748
1990 1,189 147 573 1,909
1989 1,491 57 199 1,747

" Included Host Count?y Space Geode9 CSIRO Cross-Suppon

The unusually large amount of time devoted to other institutions and JPLINASA pro-
grams in 1996 reflects a substantial JPLICSIRO effort devoted to the development of
a new wideband spectrum analyzer for radio astronomy at Canberra and Goldstone, in
addition to normal observation time.
THEDSN ALSOBENEFITS

Without question, the science of radio astronomy has benefited enormously from the
resources made available to it by the Deep Space Network since the advent of the first
64-m antenna in 1966.

However, radio astronomy was not the only beneficiary of this association with the DSN.
There was a less obvious, but no less substantial, benefit that accrued to the DSN from
its association with radio astronomy.

In a 1994 white paper on the role of radio astronomy in the DSN, Thomas B. Kuiper,
then DSN Program Manager for radio astronomy in the DSN Science Office, addressed
this subject.'' He cited the following examples of technology, methodology, and opera-
tional innovation that, currently used in various forms by the DSN, were attributable
to the use of the DSN by radio astronomers:

* digital spectrometers (Goldstein, 1963),

very long baseline interferometry (Moffet, Ekers, Robertson, 1967),
antenna calibration with natural radio sources (Klein, 1972),
connected element interferometry (Gulkis, Jauncey, 1979),
antenna pointing models (Peters, Ruis, 1982),
* K-band masers (Gulkis, Kuiper, Jauncey, 1982), and
Internet service to deep space Complexes (Kuiper, 1993).

In addition to its scientific merit, radio astronomy, by its association with other prestigious
scientific institutions and the technical challenges it brought with it, stimulated the inter-
est of many talented professionals in all areas of the DSN. DSN capabilities were extended;
seldom-used functions were exercised and maintained; and new developments were field-
tested in the less demanding risk environment of radio astronomy. It provided an intellectually
rewarding, nonoperational environment in which to explore new ideas and new techno-
logical development. As Kuiper pointed out, "Radio astronomers would accept lower reliability
to achieve performance improvement, thus giving the DSN the valuable opportunity of
evaluating and developing experience with new technologies in noncritical operations."

At that time (1994), Kuiper continued, "the key DSN technologies of antennas and
receivers and radio astronomy were nearing maturity, and the next dramatic innovations
would come from other technologies, such as information management, automation,
and networking. As applied to radio astronomy in the DSN, these technologies could
lead to unattended observations, remote operation and monitoring, and nonintederence
with tracking operations. Because of lower reliability requirements, these new techniques
could be rapidly integrated into the radio astronomy capability of the DSN for evalu-
ation, before adapting them for operational use." That appeared to be the pathway to
the h r e for radio astronomy in the DSN, and so it proved to be.

By 1997, the DSN had created a world-class facility for radio astronomy research.'"
When operated at their highest potential frequencies, the DSN antennas equaled or
exceeded the capabilities of the best radio telescopes of the time. A well-documented
site on the World Wide Web encouraged potential investigators worldwide with a pletho-
ra of technical descriptive material, along with interactive instructions to propose experiments
using the radio astronomy resources of the Deep Space Network.

NOTE The DSN resources avaibble to potential investigators in radio drtronomy were desmbed
online at http://dsnra.jpl.nasa.gov/.

As an instrument for radio astronomy research, the DSN was highly regarded by the
international science community and had become a visible tribute to NASAls continu-
ing commitment to international science. Encouraged by the radio astronomers that used
it, the DSN continued to improve in the knowledge that radio astronomy research would
quickly adapt new opportunities in deep space communications technology to further
its scientific objectives.

In the role of a scientific instrument for radio astronomy research, the DSN contributed
much to the field over the years 1967 to 1997. As a measure of this contribution, the
DSN lead radio astronomer, Thomas Kuiper, selected the following published papers as
representing significant advances in science over the period 1967 to 1997." Incidentally,
these examples reflected not only advances in scientific knowledge, but advances in the
DSN's capability to collect and process the data on which such achievements were based.

Significant advances in radio astronomy using DSN antennas include the following:

1967: First intercontinental very long baseline interferometry (VLBI).

Fine details in the structure of radio sources can be resolved by combining the
signals from distantly separated radio telescopes (VLBI). The greater the separa-
tion, the finer the details. Gubbay and others performed VLBI measurements
using baselines of intercontinental dimensions. Prior to this, VLBI had been l i i -
ited to telescopes in North
1969: Pulsar observations at 2,295 MHz (13-cm wavelength), highest frequency to date.
Moffet, Ekers, and others used the Goldstone 64-m antenna at S-band (2,295
MHz) to determine the polarization characteristics of a number of pulsars. At
higher frequencies, pulsar signals are less affected by the interstellar medium through
which they pass on their way to Earth."

1969: Detection of the first jump in the period of the Vela pulsar.
Reichley, Downs, and others detected a sudden increase in the period of the Vela
pulsar while monitoring the period of a set of pulsars with the Goldstone 64-m
antenna. They attributed the jump to the settling of the matter of the neutron
star, similar to an earthq~ake.'~

1971: Discovery of superluminal motions in 3C273 and 3C279.

By comparing the images made at two different times using the VLBI technique
(see 1967), Clark and others showed that matter in these objects appeared to be
moving with a speed greater than the speed of light. Later, this was understood
as an illusion due to the orientation of matter flow and the effect of relati~ity.'~

1973: Best upper limit to date on cosmic background small-scale anisotropy.

The currently dumpy distribution of matter in the universe, in stars and galax-
ies, and clusters of galaxies evolved from a very smooth initial distribution according
to radio astronomer Carpenter. Explaining this remained (in 1997) a challenge
for the c~smologists.~~

1982: Most distant quasar discovered with Tidbinbilla interferometer.

Using a radio interferometer consisting of two antennas at NASAls Canberra
(Tidbinbilla) Complex, Batty, Jauncey, and others determined precise positions
for a large number of radio sources. This allowed these sources to be associated
with optical objects which were then identified. One of them turned out to be
the most distant quasar known to that time.26

1986: First space-based VLBI with TDRSS antenna.

As a proof of concept, Linfield, and others used the DSN antennas with a NASA
Tracking and Data Relay Satellite to make VLBI (see 1967) observations of three
quasars. In addition to obtaining finer detail than ever before, this demonstrat-
ed that VLBI baselines could be larger than the diameter of
1992: First speckle hologram of the interstellar plasma.
Desai, Gwinn, and others used an array of Southern Hemisphere antennas, includ-
ing two at NASA's Canberra Complex and two former DSN antennas
(Hartebeesthoek, South Africa, and Hobart, Australia) to observe the Vela pulsar.
Correlations in the time variation of the Vela data from the various stations were
analyz~dto obtain information about the structure of the interstellar medium.29

1995: First images of motion in the nearest active galactic nucleus.

Over a period of eight years, Tingay, Jauncey, and others carried out a series of
VLBI observations using telescope arrays including NASA 70-m antennas.
Through these observations it was learned that jets of plasma flow out of the
core of Centaurus A. The behavior of this core was thought to be typical of other,
more distant active galactic nuc1eie3O

1995: First image of infall of cloud material onto a protostar.

A protostar is a star that is still growing by accreting matter from the cloud out
of which it formed. Vellusamy, Kuiper, and Langer, by combining data from the
Goldstone 70-m antenna with data from the Very Large Array, made images of
the material falling onto the protostar B335. Neither telescope by itself could
have produced such images.31

1996: First evidence for the formation of a preprotostellar core by coagulation.

By comparing Goldstone 70-m data with VLA data, and comparing the results
with those obtained with other telescopes, Kuiper, Langer, and Vellusamy found
the cloud core L1498 to be growing by the process of accreting material. The
cloud was not collapsing like B335 (see above), but was thought to be growing
towards a stage when collapse would occur. Prior to this, it was not known
whether prestellar cores grew by slow accretion of surrounding material.32

1996: Nano-arcsecond resolution of the Vela pulsar.

Using interstellar scattering data derived from further analysis of the speckle data
obtained in 1992 (see above), Gwinn and others determined the size (approxi-
mately 500 km)and approximate shape of the emitting region of the Vela pulsd3

1997: First VLBI astrometric detection of an unseen companion star.

Guirado, Reynolds, and others used the 70-m antenna at NASA's Canberra
Complex, combined with the (former DSN) 26-m radio astronomy antenna at
Hobart, Australia, and the 64-m radio astronomy antenna at Parkes, Australia,
to make precise measurements of the position of the star AB Doradus. A wob-
 bling in the positions of the star revealed the presence of a dark companion with
about a tenth of the mass of the Sun."

1997: First images from the HALCA Space VLBI mission.

The Japanese HALCA space telescope was the first dedicated orbiting VLBI anten-
na. Its orbit was chosen to provide the longest baselines to that date. The NASA
70-m antennas, because of their size and geographic locations, played an impor-
tant role as a ground element in the HALCA mission to produce extremely sensitive
and detailed images of compact radio sources.
T H E SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE (SETI)

In the late 1970s, against a background of serious scientific interest in the detection of
extraterrestrial intelligence dating back to the early 1960s and strong advocacy by sci-
entists at Jet Propulsion Laboratory and the Ames Research Center, NASA initiated a
formal program to search for evidence of extraterrestrial intelligence. Under the acronym
SETI (Search for Extraterrestrial Intelligence), the program was to be managed by the
Astrophysics Division of the NASA Headquarters Office of Space Sciences (OSS).

SETI was to be a research and development program aimed at demonstrating the nec-
essary technology to search a well-defined volume of sky and frequency space, within a
finite time, for evidence of intelligent signals of extraterrestrial origin. It was to use exist-
ing antennas augmented with new spectrum analysis and data-processing systems.

In June 1980, TDA progress reports by Robert E. Edelson and Gerry S. Levy35describ-
ing the telecommunications technology associated with SETI, and by Berman3bdescribing
the SETI observational plan, reflected the growing involvement of the TDA Office in
the new program. Both of these reports drew attention to the dependence of the pro-
posed SETI program on the use of existing large antennas, ultra-sensitive microwave
receiving systems, identification of terrestrial sources of radio frequency interference (RFI),
and wideband spectrum analysis and data management technology, most of which was
to be found in the existing Deep Space Network. Although not explicitly stated, there
was also perceived to be a symbiotic relationship between the new technology that would
be needed for SETI and that which the DSN would use for wideband signal acquisi-
tion, and RFI and spectrum management. These were the principal factors that
established a basis for the DSN association with the SETI program.

Supported by funding from the NASA Office of Space Science and, to a lesser extent
by funding fiom the NASA Office of Tracking and Data Acquisition, the SETI pro-
gram at JPL was carried forward as a parallel but independent program to that of the
Deep Space Network. Within JPL, the SETI program was managed by the TDA Science
Office under the direction of Nicholas A. Renzetti.

In the overall plan, two parallel but complementary strategies were to be used to conduct
the SETI observations-a "targeted survey" and an "all-sky survey" The targeted survey
would target approximately 800 to 1,000 candidate stars within 100 light-years of Earth.
It would search for pulsed or continuous signals in the 1- to 3-GHz frequency band and
would require high-sensitivity, narrow-bandwidth data acquisition systems that would be
used with the 64-m DSN antennas and the NRAO 305-m antenna at Arecibo, Puerto
Rico. The "all-sky survey" would trade lower sensitivity for the ability to survey 99 percent
of the sky that was not covered by the targeted search, and it would span a wider frequency
range, from 1 to 10 GHz. It would be conducted on the DSN 34-m antennas.

Working within these broad objectives and guided by a SETI Science Working Group
that had been set up by NASA at the outset of the SETI program, engineers at JPL
began to develop the elements of hardware and software based on Fast Fourier Transform
(FFT) techniques that would be required for both strategies." It should be made clear
at this point that the engineering groups doing this work for the SETI program were,
at the same time, doing similar work for the DSN. Although the funding sources were
different, the organizations
- were one and the same, and the benefits of the DSN tech-
nology base obviously accrued to the SETI program.

The objective of this effort was to be a field demonstration at Goldstone, using the DSS
13 antenna and known sources of radiation such as NASA planetary spacecraft.

The first step in this direction was to assess the radio frequency interference (RFI) envi-
ronment in which the sky survey would be carried out. The resulting data base would
reveal the incidence of false signals and the presence of interfering signals and noise-
information that would be essential to the design of the final signal-detection, analysis,
and data-processing equipment for SETI.

For this purpose, a Radio Spectrum Surveillance System was built, mounted in a transportable
van with a-1-meter-diamefer antenna, and deployed at Goldstone in 1984.38 he- system
comprised seven low-noise Field Effect Transistors (FETs) to cover the range of frequencies
1-10 GHz; a commercially available, programmable, swept-Gequency spectrum analyzer; and
a floppy disc storage device. These were all controlled by a BASIC software program. In
addition to its primary use by the SETI program, the RFI van proved to be a remarkably
effective &cility that was subsequently used for many years by DSN spectrum management
engineers to continuously monitor and idenufy RFI in the deep space bands at Goldstone.

The initial field demonstrations for SETI, using FFT-based hardware and software, began
at DSS 13 in March 1985.39The field-test instrumentation consisted of the DSS 13 26-
m antenna with low-noise receiving systems for the 2,200- to 2,290-MHz and 8,400-
to 8,500-MHz bands. It also included a 65,000-channel FFT spectrum analyzer, designed
and built by JPL and housed in the RFI van, which was parked nearby. A 72,000-chan-
nel Multichannel Spectrum Analyzer, designed and built for ARC at Stanford University
for the targeted search program, was also part of the field demonstration. The detection
strategy and codguration used for the sky survey field tests is shown in Figure 8-4.
f >

PATTERN
+ ------iI tL -&ONREMnTIO_N-- 1
I
I
I
I

FREQ v --+---, 1-1 EVENT FILE

* DISCARD RFi
CONVOLUTIONAL THRESHOLD
FILTER

Sky survey detection sbategy

-----------

I
I
-
HARD
DISKS
r----
SEll RSSS - I
I
I
ANALYZER-
I
SYSTEM I
I
L---,-, I
I

System configuration for Sky Survey field tests (1985)

i 1

Figure 8-4. Detection strategy and system configuration for the SET1 sky survey field tests,
1985.
/ \

I
300 - PN 12
256 Hz
35

200 -
Y Y

?- Z
51 33 51

, ,
31 I
-75 -50 -25 0 +25 +50 +75 -100 -50 ,0 +50 +I00
A FREQUENCY,KHz A FREQUENCY, KHz',
I
\\ 32Hz
200 .
SYSTEM: /
DSS 13 26-METER ANTENNA /
DSN FFT SPECTRUMANALYZER \
SET1 CONFIGURATION AT 2295 MHz /
100
SPACECflARDISTANCES (MARCH 1985) - /' 1 !
PIONEER 11.....280 X 10' KM I
\
* PIONEER 12.........6 X 10' KM
0 !.+-A- -- L..4..+-2
\
\

A > * # I # , , L
-8 -6 -4 -2 0 +2 +4 +6 +8
A FREQUENCY, KHz
\ 1

Figure 8-5. SET1 detection of spacecraft signals, 1985.

The tests were based on the detection of narrow-band signals from the Vqyager and
Pioneer spacecraft, which, in well-defined celestial locations, provided low-level signals
with known characteristics perfectly suited to the evaluation of the basic SETI system.
Typical spectrum analyzer displays for the detection of Pioneer I 1 and Pioneer 12 space-
craft signals are shown in Figure 8-5.

For the next two years, the field-test -program

- was used to verify hardware designs,
- fur-
ther develop signal-detection strategy and automated procedures, and characterize the
RFI environment at the site over the range of 1 to 10 MHz.

In April 1987, a program plan proposing separate but complementary search strategies
was presented to NASA by representatives of the SETI program from ARC and JPL.
The two Field Centers jointly proposed to develop and operate observing systems using
high-speed signal-processing equipment that could be used in conjunction with existing
ground-based radio astronomy and deep space communication facilities. The targeted
search was to be the responsibility of ARC, while JPL would be responsible for the all-
sky survey. ARC would act as the Lead Center for the program, which would be managed
by the Life Sciences Division of OSS with important technical and financial support
from the Office of Space Communications (OSC, formerly OTDA). The DSN would
provide support on its 34-m antennas for the all-sky search and time on its 70-m anten-
nas for those parts of the targeted search that could not be provided by the Arecibo
antenna or the NRAO antenna. With this decision made, the JPL team focused its
efforts on the development of equipment and techniques that were optimized for the
all-sky survey. Henceforth, the newly directed program would be known as the High
Resolution Microwave Survey (HRMS).

Building on search strategies and FFT techniques that had been developed while the
initial demonstration field tests were in progress at Goldstone," a more refined signal
detection strategy for an operational HRMS all-sky survey evolved, and a much larger,
wide-band, high-resolution spectrum analyzer (WBSA) had been designed.41These and
the observing procedures derived from the field tests experience at Goldstone would be
assembled into a prototype system and used for a limited proof-of-concept search pro-
gram to be conducted in the early 1990s. Experience gained from these proof-of-concept
tests would be used to improve the design of the full-scale operational system, which
was then scheduled for completion in 1996.

The objective of the all-sky survey was to search the entire sky over the frequency range
from 1.0 to 10.0 GHz for evidence of narrow band signals of extraterrestrial origin.
Frequency resolutions as narrow as 20 Hz were required. The objectives of the survey,
as presented by Coulter Klein and others are summarized in the table below.

Characteristics of the All-Sky Survey for a 34-m Antenna

Spatial coverage 4 pi steradians (full sphere)

Frequency coverage 1-10 GHz with higher spot bands
Duration of survey 5-10 years ('-/-yeargod)
Dwell time per source 0.5 to 10.0 seconds
Frequency resolution 10-30 Hz
Sensitivity, (allowing for spatial Approx. 10-23wattslmeter?.
uniformity and variation with
frequency)
Polarization Simultaneous dual circular
Signal type Continuous Wave (CW)
To accommodate the K i t e d observing time that would be available on the DSN 34-m
antennas for HRMS, an antenna scan pattern of "racetrack" form was developed. The
racetrack scan pattern utilized the available antenna time efficiently, maintained maxi-
mum sensitivity with uniformity across each sky h e , and facilitated the signal detection
process. Specific features of this pattern could also be used to discriminate against cer-
tain types of RFI.

In collaboration with the DSN Advanced Systems Program, an engineering development

model of a wideband spectrum analyzer W S A ) was built to serve as a prototype for
the final operational model. This machine was key to the design of the HRMS system.

The WBSA featured a pipelined FFT architecture that transformed 40 MHz of input
bandwidth into 20-Hz wide bins in the frequency domain. To accommodate the wide
bandwidth in real time, the WBSA performed 4.5 billion operations per second and
used two input channels to search simultaneously for left and right polarized signals.
The output of the WBSA was passed to a signal-detection module that performed base-
line estimation, filtering, and thresholding functions. Although the prototype system
bandwidth was limited to 20 MHz, plans were made to increase this to 320 MHz in
the final operational design in order to complete the survey in 7 years. It would have
taken 100 years to complete the survey with a 20-MHz bandwidth.

Progress in developing the two parallel systems at JPL and at ARC was such that on 12
October 1992, the observational phase of HRMS was inaugurated with simultaneous,
coordinated observations of a selected sky frame by the all-sky system at DSS 13, and
a star within that frame by the targeted-search system at Arecibo. The occasion was
marked with appropriate pomp and circumstance, and keynote addresses were delivered
by SET1 pioneers Philip Morrison (at Arecibo) and Carl Sagan (at Goldstone). The use
of the recently completed 34-m BWG antenna at DSS 13 added to the significance of
the event.

A view of the eastern sky as seen from Goldstone at noon on that day is shown in
Figure 8-6. The shaded areas define the boundaries of the first two sky frames of the
HRMS Sky Survey. The location of the first star observed by the Targeted Search Team
at Arecibo is shown in sky frame number 1.

The initial surveys were conducted near 8.5 GHz to characterize the system operating
performance and determine the RFI environment and the system's ability to discrimi-
nate against it. Later, repeated observations of a set of three sky frames at other, lower
frequencies provided data of interest to the radio astronomy community, as well as the
F \

64ST SE

View of the Eastern sky as seen from Goldstone at noon (1900 UT), October
12,1992. The shaded areas define the boundaries of the first two sky frames of the
HRMS Sky Survey. The location of the first star observed by the Targeted Search team
at Arecibo is shown in sky frame Number 1.
L J

Figure 8-6. The first HRMS sky survey, Goldstone, 12 October 1992.

HRMS program, and allowed further improvement in the efficiency of the signal-pro-
cessing algorithms that reduced the effects of RFI. During the first year of observations,
no signals from beyond our solar system were detected by either the sky-survey or the
targeted-search systems. However, many valuable lessons had been learned and were
about to be incorporated into the full bperational system. Confidence in and enthusi-
asm for the HRMS program were high.

Suddenly, in October 1993, the U.S. Congress unexpectedly terminated the program on
the grounds of financial expediency. Apart from preserving the advanced signal-process-
ing systems for potential future use, no fiuther work was possible and all HRMS obser-
vations with the sky-survey equipment were terminated by the end of 1993.42

Subsequently, the prototype sky survey system was redirected for further development
and use in the operational Deep Space Network. The SETI initiative was continued by
the SETI Institute of Mountain View, California, a corporation devoted to the search
for extraterrestrial intelligent life and to research relevant to the disciplines of life in the
universe. Its purpose included the conduct and promotion of public information and
education as it related to SETI. To this end, it sought private donations to continue the
development and use of the targeted survey system at the Arecibo Observatory in Puerto
Rico.
CRUSTAZ. DYNAMICS

By the early 1970s, very long baseline interferometry (VLBI) was well established in the
DSN as a technique for spacecraft navigation, radio astronomy observations of the struc-
ture and position of Extra Galactic Radio Sources (EGG), and the determination of
baseline vectors between fured reference points on Earth (DSN tracking stations). Success
in these areas gave rise to an increasing interest by the science community in applying
VLBI techniques to accurate measurements in the fields of geodynamics and geodesy.

However, the fured antennas of the DSN were not suited to the conduct of an exten-
sive geodetic monitoring program that would require frequent and accurate measurements
along many reference baselines. For this purpose, a mobile facility, which employed the
same VLBI techniques but possessed an added capability for rapid deployment to numer-
ous selected sites of geodetic interest, was required.

To this end, in 1971, the NASA Office of Applications funded a research program to
demonstrate the capabilities of portable, independent-station radio interferomeuy. Because
of its relationship to well-established DSN technology and its dependence on the 64-m
DSN antenna at Goldstone as a reference point, the program was assigned to JPL and was
managed by the Telecommunications and Data Acquisition Office,with Nicholas A. Renzetti
as the Program Manager. It was named Astronomical Radio Interferometric Earth Surveying
(ARIES) and was led by l? E MacDoran.

By refurbishing a surplus 9-meter transportable antenna and implementing the neces-

sary instrumentation in trailer-mounted vans, a transportable ARIES facility was rapidly
assembled and, by the end of 1973, was ready to start field operations at Goldstone.

The initial series of experiments were performed by Ong, MacDoran, and others, on a
short 307-meter baseline over the period from December 1973 to June 1974, using
ARIES as one observing station and DSS 14 as the other. The arrangement is illustrat-
ed in Figure 8-7.

Use of a short baseline for the initial tests enabled a comparison of conventional survey
methods with few-centimeter accuracy to minimize interferometry errors due to trans-
mission media effects, source locations, and Earth orientation parameters. The results,
representing approximately 28 hours of data, were in excellent agreement with the sur-
vey baseline in all dimensions within the formal uncertainty of 3 cm.43Over the next
five years, ARIES was moved to various locations in the western United States to refine
c \

MmAGAlAcllC
RANDOM RADIO
SIGNALS

mc

VLCTOR
WWNE
IGEN PURPOSE
UIMPUTW)
- - SYNTHESIS
(GEN WRWSE
CIMWIER)
CROSS CORRELATION
SIGNAL DmCTlON
(MINI-COMPUTER)
\ J

Figure 8-7. Project ARIES transportable antenna interferometer system.

and demonstrate the feasibility of the mobile VLBI concept and to initiate studies of
crustal motion on a regional scale.

The success of this program eventually led to a requirement for two additional mobile
VLBI units and a dedicated base station for the Crustal Dynamics project. The design
and construction of the mobile units and implementation of the Mojave Base Station
at Goldstone were carried out by Brun, Wu, and others as part of the ORION project
by the DSN in 1982.44, 45

The Mojave Base Station utilized a redundant 12.2-meter antenna and Control Room
built in 1962 to support a communications satellite program operated by the Goddard
Space Flight Center. In June 1983, after major refurbishment by a joint JPL and GSFC
engineering team, the Mojave Station, operated by Bendix contractor personnel, began
to support the Crustal Dynamics project. The completion of this station relieved DSN
stations DSS 14 and DSS 13 from the burden of supporting the Crustal Dynamics field
observing program. The two mobile VLBI (MV) units, which used much smaller anten-
nas and were more readily transportable than the original ARIES, started observing in
1980 and 1982, respectively.

Between 1980 and 1985, the three MV units, in conjunction with Mojave and sev-
eral other fixed base stations in the western United States, were operated as part of a
surveying program to determine the relative motions and regional strain fields near
tectonic plate boundaries in California and Alaska. The results of this effort, and the
diff~cultiesencountered in realizing the necessary accuracy to deliver viable data, are
described by J. M. Davidson and D. W. Trask, who led the effort for the TDA Ofice
at JPL.46They reported that "baseline measurements utilizing the current Mobile VLBI
systems had attained an accuracy of 2 cm, or better, in the horizontal plane" and
observed that "since average geological rates of horizontal motion are on the order of
5 cmlyr across the plate boundary regions being studied, it was likely that crustal
motion would be detected within the next few years, provided [it was] occurring at
the geological rates."

The mobile VLBI project at JPL had reached the point where it could be transferred
to its ultimate user, the National Geodetic Survey (NGS)/National Oceanic and
Atmospheric Administration (NOAA) and Goddard Space Flight Center (GSFC). The
transfer was completed in 1984 and 1985.

Even as the more economical, portable VLBI stations of the ORION task were
under construction, technologists were experimenting with a new, potentially low-
cost method for measuring Earth crustal deformation. This method utilized the
L-band signals transmitted by the Global Positioning System (GPS), a constella-
tion of 24 Earth-orbiting satellites under development in the 1980s by the
Department of Defense. Unlike the weak quasar signals recorded by VLBI, the GPS
signals broadcasted at 1.2 and 1.5 GHz were relatively strong and did not require
9-m antennas and low-noise receivers for their acquisition. A system based on GPS
technology, therefore, offered the promise of high performance, low cost, and much
greater mobility than the MV system. It would, in addition, have very significant
application to the existing DSN for spacecraft navigation and other purposes relat-
ed to station locations. With this in mind, the Geodynamics Program funded JPL
for an experimental demonstration of the new GPS technique. It started with the
Satellite Emission Range Inferred Earth Survey (SERIES) project in 1979, also under
the leadership of Peter F. Ma~Doran.*~

Under the SERIES project, two GPS receivers were built at JPL for proof-of-concept
measurements on short baselines. These receivers measured the differential group delay
of signals transmitted at two different frequencies from the GPS satellites. This infor-
mation could be used not only to determine the location of the receiver on Earth's
surface to a level of precision approaching that of the mobile VLBI system, but also
to calibrate the ionosphere in terms of the total electron content (TEC) along the
line of sight between the receiver and the satellite. The former Eunction and other
data derived from it was of particular interest to the Geodynamics Program for short-
baseline geodecics measurements, while the latter function was of particular interest
to the DSN for adding media calibrations to the radio metric data used for space-
craft navigation.

While the SERIES project successfully demonstrated the potential of the new GPS tech-
nology, it also showed the need for a more compact receiver with enhanced capability
for deployment at multiple field sites. At that time, receivers that had a capability for
parallel processing of signals from four (or more) satellites simultaneously and were suf-
ficiently portable for field applications were not available commercially. It was this situation
that led to the development of the SERIES-X receivers at JPL, starting in 1982 under
the direction of William G. Melbourne.

The SERIES-X design was also very promising as an instrument for continuous, precise
ionosphere calibrations at each DSN Complex. The Faraday Rotation satellites (ATS-6),
which had been used for this purpose for many years, were rapidly being decommis-
sioned. The GPS technique offered a cost-effective, more accurate means of calibrating
radio metric tracking data. The DSN proposed to implement GPS receivers at each
Complex and to dwelop the data retrieval system and software required to produce the
calibrations for the inflight projects. An upgrade to the SERIES-X design was developed
for this purpose. The receivers that were actually installed at the three complexes were
built by a commercial partner, Alan Osborne Associates (AOA), and were referred to as
the "Rogue Receivers."

Under the auspices of the NASA Geodynarnics program, the Rogue Receivers were used
for geodetics measurements and, as part of the JPL Crustal Dynamics project, to demon-
strate the concept of a Fiducial the DSN, they were used specifically
to determine the total electron content of the ionosphergOfor radio metric data cali-
bration purposes.

Following the implementation of the Rogue Receivers in the DSN, further development
of GPS receivers shifted to the Geodynamics program. The Rogue Receivers were pro-
duced commercially by AOA and were used in a number of field campaigns in the mid-
to late 1980s. The first campaigns to use the GPS system were conducted in 1985 with
baselines in Mexico and the Caribbean regions. Data from the DSN Rogue Network
were used to establish a terrestrial reference frame for these experiments. It was these
experiments which clearly demonstrated the economics of GPS Geodesy and sparked
further development of GPS technology for the Geodynamics program.

As the SERIES-X program got under way, another sponsor for GPS technology emerged
in the form of the TOPEX project. Managed by JPL and planned for launch in 1989,
TOPEX was an Earth-~rbitin~ satellite dedicated to the study of ocean topography.
TOPEX carried, amongst other scientific instruments, a highly accurate radio altime-
ter for the measurement of ocean surface height (sea altitude). To meet the specified
goal of 10-cm altitude accuracy, a very precise knowledge of the spacecraft orbit was
required, and the use of the system then available would have resulted in an accu-
racy of +14 cm.

In a proposal to the TOPEX Project in July 1984, the TDA Science Ofice described
a system by which the accuracy could be increased to 22.4 cm using GPSS5' The
proposal called for a flight-qualified GPS receiver and antenna aboard the TOPEX space-
craft, a SERIES-X (Rogue) gound receiver at each DSN Complex, and at least three
ground receivers at other remote locations. It also included a data communications net-
work and a central data processing facility to provide the TOPEX precision orbit
determination functions. The underlying concept of the demonstration was that an inflight
GPS receiver could, in conjunction with precisely located GPS ground stations, be used
for highly accurate determination of orbiting satellites in an Earth-based reference frame.

Orbit determination for TOPEX, replacement of the Mobile VLBI stations for Crustal
Dynamics, and ionosphere calibration for the DSN became the three principal drivers
for hrther GPS-based development, and funding was shared accordingly by the respec-
tive program offices.

GPS-based technology was formally introduced into the operational capability of the
DSN in November 1986 with a Functional Design Review for a new Media Calibration
Subsystem for the DSCCS.~'At each Complex, the new DSCC Media Calibration
Subsystem (DMC) would perform continuous measurements of local ground weather
parameters in addition to generating differential P-code group delay and carrier phase
from the GPS satellites. These data were to be recorded and transmitted from the Complexes
to the NOCC at JPL for the use of the Navigation Team in calibrating the radio met-
ric data delivered by the DSN Tracking system. A description of the GPS system is
beyond the scope of this book, but the essential elements as they applied to the DMC
in 1986 are shown in Figure 8-8.
 . 18 SPACE VEHICLES PLUS 3 SPARES
SPACED 60" (LONGKUDO APART
INCUNATION 55P
6 ORBITAL PLANES
0 120" BETWEENSPACEVEHICLESIN P M E

Figure 8-8. Global Positioning System, 1986. The GPS functions of the DMC were carried out by SERIES-X (Rogue) receivers, which
m
acquired and simultaneously tracked the two GPS L-band carriers from at least four satellites, extracted the ephemeris data from each, 9

fR
and measured the differential P-code group delay and the carrier phase. Because of the sensitivity of the military to the use of the NAVS-
TAR system for other than classified purposes, these operations were performed without an explicit knowledge of the classified P-codes
themselves. 0
I \

@llbWSMmEn

TAANSMmER OFFSET

IONOSPHERIC DELAY
PLUS TRANSMWER

Figure 8-9. Schematic view of GPS-based ionospheric calibration.

The basic principle of GPS-based ionospheric calibration is illustrated schematically in

Figure 8-9.

The remarkable performance of the receivers is reflected in the following estimates for
their accuracy in determining TEC:

Performance of the DMC for Ionosphere Calibration Using GPS

Differential delay Precision (el/mz) Accuracy (el/m2)

P-code groups 1 x 1OI6 5 x 1016
L2-LI carrier phase 0.02 x 1016 0.02 x 1016
y,z.<, .<:.,.<.,.<.,.,<,.:,.,.,<.,.,.~,
~ .,..,*7<.,...,..,.<,,. -.., <<
.., <,<.,... ..,....
",~
.................... <-,<
.; <
..- .-.-./,-<< ..,.*>..>.,.>.,.<..><.',.",. ?.<. >.>-. .. .:: -,:-.
9.- n ? : . ~ < . ~ . - ~ . . > . ? : > ~ . ~ ~
,<<.:~.-~,.~~~>~.~-<~.~.-<~.~3

NOTE Tonl Ekmon Content (TEC) of the wnophere ir measured in uniu of ekmom per square meter (end).

The DMC was put into operation at each of the three complexes in 1989 and imme-
diately became the DSN standard source for ionosphere calibration data. Eventually, the
original Rogue Receivers were replaced by smaller, laptop computer-size receivers, known
as the Turbo-Rogues. The first of the Turbo-Rogues to be produced were installed at
the DSN Complexes, and as they became available, the Turbo-Rogues gradually replaced
the original receivers within the NASA GPS Global Network also.

By 1988, GPS technology sponsored by JPL had become a significant new tool for
Geodynamics research on short baselines and as a complementary technique to VLBI
for measurements on long (intercontinental) baselines. At the same time, the Crustal
Dynamics project at GSFC was operating a Satellite Laser Ranging (SLR) System and
a VLBI System (using DSN antennas as well as international VLBI stations) and had a
large budget for that purpose. By comparison, the JPL budget for GPS applications to
Crustal Dynamics was very small. With time, it became apparent that GPS could not
only determine short baselines almost as accurately as VLBI, but was also approaching
VLBI performance on the long baselines, and some reassessment of the relative value of
GPS to the NASA Geodynamics program was called for.

To this end, a conference of all parties involved in the Geodynamics program was con-
vened by the NASA program manager at Coolfont, West Virginia, in July 1989. For
Geodynamics Solid Earth studies, the Coolfont conference recommended the establish-
ment of a backbone consisting of a multicountry, intercontinental network of VLBI
stations colocated with fured GPS stations. The VLBI network would maintain a strong
inertial reference for Earth rotation studies and provide precise measurements over inter-
continental distances for the study of tectonic motion. The VLBI stations would also
be complemented with a multicountry network of roving GPS stations for making meas-
urements on short baselines for crustal motion studies.

Thus, Coolfont represented a major turning point for the Geodynamics program in gen-
eral and the Crustal Dynamics project at JPL in particular. As a consequence of the
agreements reached there, the Crustal Dynamics project at JPL was ended and GPS
became a permanent and major part of the NASA Geodynamics program with JPL as
the Lead Center for GPS technology. The VLBI technology was pursued by GSFC and
involved JPL only to the extent that the VLBI program needed support from the DSN
antennas.53

Under the cognizance of the NASA International Geodynamics Program, the field observ-
ing program thus evolved into two components, one based on VLBI and managed by
GSFC, the other based on GPS and managed by JPL. The international VLBI network
(which included but was not limited to the DSN 70-m antennas) comprised a set of
permanent sites, each of which had a large radio astronomy antenna colocated with a
fured GPS receiver installation. The international GPS network was made up of a Network
of portable GPS receivers installed and operated by the observing scientists themselves
in areas of their specific interest.

So ended the Crustal Dynamics Project, the techniques for which had been founded and
nurtured by the TDA Science Office for fieen years. Out of this work had grown a bur-
geoning GPS technology, which, in addition to its proliferation in the Geodynamics program,
was finding increasing applications in the operations of the DSN. The synergy that had
developed between the Crustal Dynamics project and the DSN had given riie to sigrufi-
cant improvements in the orbit determination processes on which the DSN, and users of
the DSN, depended for spacecraft navigation purposes. Ionosphere and troposphere cali-
brations, Earth-orientation measurements, and intercomplex time synchronization were areas
in which VLBI techniques, complemented by GPS techniques derived from the Crustal
Dynamics work, had been most effective.

In recent years, the extension of GPS technology to DSN- and spacecraft-related prob-
lems, and to non-NASA applications, was sponsored by the TDA Technology Office
and is reported in the previous chapter.
PLANETARY RADAR

In the context of science in the DSN, it can be stated unequivocally that "in the
beginning, there was planetary radar." Planetary radar came before radio astronomy,
before radio science, and before SET1 or crustal dynamics became initiatives in the
Deep Space Network. As a matter of fact, planetary radar preceded planetary space-
craft in the DSN. As we shall see, the first planetary radar experiment (Venus) took
place on 10 March 1961, while the first successful planetary spaceflight (Mariner 2)
took place on 27 August 1962.

Early in 1960, plans were being made at JPL for spaceflights to the inner planets, begin-
ning with the Mariner mission to Venus. The DSIF, as it was known at that time,
compriied four antennas with diameters of 26 meters (85 feet): two at Goldstone, California;
one at Woomera, Australia; and one under construction at Johannesburg, South Africa.
Teletype, telephone, and undersea cable communications links tied each of the antenna
sites to a rudimentary Control Center at JPL.

The plmetary missions resented many new challenges to existing technology, among
which was a requirement for hitherto unprecedented accuracy in spacecraft navigation.
To a large degree, spacecraft navigation depends for its accuracy upon the quality of the
Doppler data provided by the tracking stations and a knowledge of the ephemeris of
the planet and the value of the Astronomical Unit (AU). The DSIF was working on
the Doppler data issue, but the accuracy of the Venus ephemeris and the value for AU,
based on optical astronomy observations, left much to be desired for planetary naviga-
tion purposes. W ~ t hthis situation in mind, Eberhardt Rechtin, Walter K. Victor, and
Robertson Stevens proposed to carry out a bold experiment during the 1961 conjunc-
tion of Venus, using existing facilities at Goldstone to refine the value of the Astronomical
Unit and improve the ephemeris of Venus. Using radar techniques, the round-trip time
required for a radio signal transmitted from Earth and reflected back from the surface
of the planet to return to Earth would be measured very precisely. From these and ancil-
lary data, the desired ephemeris and value for AU could be determined.

The experiment began on 10 March 1961, in the bistatic mode with one of the Goldstone
26-m antennas transmitting (Echo site) and the other receiving (Pioneer site). On 16
March, the Director of the DSIF, Eberhardt Rechtin, announced the success of the
Goldstone Venus Radar Experiment in a message to his colleagues at the MIT Lincoln
Laboratory, which had for several years previously been attempting to do the same thing,
"HAVE BEEN OBTAINING REAL TIME RADAR REFLECTED SIGNALS FROM
VENUS SINCE MARCH 10 USING 10 KW CW AT 2388 MC AT A SYSTEM
TEMPERATURE OF 55 DEGREES." The reply from Lincoln Laboratory offered hearty
congratulations but reported no success with their own efforts to that time. Once unam-
biguous daily contacts with Venus were well established with the ranging system turned
on, measurements of the Earth-Venus distance began and continued through mid-May
1961. Based on these results, Victor and Stevens announced their results to the scien-
tific world in Science, July 1961.54The value for the AU, based on data from the first
planetary radar experiment, was determined to be 149,599,000 with an accuracy of
+1,500 km. At that time, this represented a major improvement in accuracy compared
to that obtained from optical and other radar-based experiments. After further refine-
ment by D. 0. Muhleman and others, the uncertainty was reduced to k500 km by
August and to +250 km in the following year.

NOTE: The International Astronomical Union adopted the vilue of 149,600,000 km in

1964, based on the rahr methodjrst demonstrated succes$l~ at Gohtone in 1961. For
a detailed account of this event, and ofplanetary rahr in general, the reader is referd to
To See the Unseen by Andrew /. Butrica."

The new value of the AU and the refined ephemeris resulting from the Venus radar
experiment were immediately put to use for Mariner orbit determination. Later, it was
estimated that without the improved value for the AU and ephemeris improvement,
Mariner 2 would have passed Venus at too great a distance to have returned any useful
data. (Mariner 1 failed on launch.)

In the opinion of historian Andrew Butrica, "The JPL experiment succeeded because
it did not depend on . . . prior knowledge of the AU. On the other hand, Lincoln
Laboratory and Jodrell Bank based their experiments on an assumed, yet commonly
accepted, value for AU and consequently for the distance between Earth and Venus
during inferior conjunction." The experiment also owed much of its success to the sub-
stantial improvements in uplink and downlink performance that were made to the
existing equipment built to support the Echo Balloon experiment in August the pre-
vious year. Because the power in the return signal from Venus was expected to be 50
dB to 60 dB (100,000 to 1,000,000 times) weaker than had been received from the
Echo balloon, an improved receiving system working with maximum transmitter power
would be required to detect the Venus signal.

To meet these requirements, an S-band, low-noise maser and parametric amplifier was
added KO the receiver at the Pioneer antenna.s6The single-cavity, three-level ruby maser,
the first of its kind, had been developed by Walter Higa at JPL specifically for this appli-
cation.57Together with an improved microwave antenna feed, this combination reduced
the system noise temperature from the value of 1,570 K, which had been used for Echo,
to about 60 K for Venus. When installed at Goldstone and brought into operation for
the Venus experiment, the Pioneer receiver was believed to be the most sensitive oper-
ational receiving system in the world. At the time, nothing could be done to increase
the output of the existing S-band transmitter at Echo. However, it was operated for the
Venus experiments at its maximum power output level of 13 kW.

With this extraordinary instrument, and supported by the strong technical motivation
of Rechtin, Victor, and Stevens, the two young JPL astronomers Duane 0. Muhleman
and Richard M. Goldstein had succeeded where two other distinguished scientific insti-
tutions, Lincoln Laboratory and Jodrell Bank, which had conducted similar experiments
on Venus, had not.

Between 10 March and 10 May 1961, some 238 hours of scientific data were record-
ed during radar experiments conducted almost daily.

So began the DSN association with planetary radar.

Soon after the Venus radar experiment, Muhleman left JPL for Harvard, and Rechtin,
Victor, and Stevens turned their attention to the rapidly evolving Network and its sup-
port for the NASA planetary spacecraft missions. Goldstein, however, continued his close
association with the Goldstone radar and its further development for many years.

The first Venus radar experiment had been conducted amid growing concern at JPL
about the need for two operational antennas at Goldstone to support the rapidly expand-
ing lunar and planetary programs, which would need tracking support from the DSIE
Added to this concern was the urgent need for an antenna which could be dedicated
to research and development activities for the engineering side of the DSIF. The Echo
antenna, being of the Azimuth/Elevation (Az-El) type, was not entirely suitable for track-
ing planetary spacecraft where horizon-to-horizon passes at a constant declination were
generally required. It was therefore decided to build a second, 26-m polar-mount anten-
na at the Echo site and to move the existing Az-El antenna to a new site at Goldstone,
where it would be used principally for engineering development for the Network and,
incidentally, for planetary radar. The second operational antenna was completed in May,
and in a monumental feat of transportation, the 26-m Az-El antenna and its pedestal
were moved as a complete unit six kilometers across the desert to a new site to be
named, obviously, Venus, in June 1962.
As the new polar-mount antenna at Echo prepared to support the Mariner 2 flight to
Venus in August 1962, as part of the operational Network, the Az-El antenna was
refurbished at the new Venus site and resumed Venus observations as a research and
development station.

The 1961 Venus radar had been a bistatic radar, that is, it used two antennas, one for
transmitting (Echo) and another for receiving (Pioneer). The 1962 Venus radar was a
monostatic radar using a single antenna (the former Echo antenna) for both transmit-
ting and receiving. This required the addition of a high-powerllow-loss duplexer to
accommodate both the transmitting and receiving b c t i o n s on the one antenna. The
transmitter could be operated in a continuous-wave (CW) mode, an amplitude modu-
lation (AM) mode, or a keyed (onlo@ mode. A CW receiver and an AM receiver
provided open-loop or closed-loop reception for any transmitter mode. Signal level and
power spectra data were obtained using the open-loop receive configurations, while the
Earth-Venus velocity and Earth-Venus range data were obtained using the closed-loop
configurations. Several other changes were made to the Doppler and range systems to
accommodate the various modes of ~peration.~'

A functional block diagram of the 1962 Venus radar is shown in Figure 8-10.

With this powerfid new instrument, Goldstein and others continued further successful
radar observations of Venus during the 1962 and 1964 conjunctions and made the first
detections of Mars and Mercury in 1963.

Despite these scientific successes, however, the radar depended for its continued exis-
tence on the research and development program which the DSN maintained to support
its operational Network. It benefited from the development work that the DSN carried
out under its Advanced Systems Development Program but was never formally estab-
lished and funded as a NASA program in its own right. Because much of the technology
required for a planetary radar was the same as that required for deep space communi-
cations, the Goldstone radar received, in its early years, a great deal of support from the
DSN research and development teams. For them it offered a less critical environment
in which new technology could be evaluated before being adapted to meet the (mis-
sion) critical needs of the operational Network. There were many areas in which this
commonality of technology was used to the mutual advantage of the Goldstone radar
and the DSN research and development programs. These areas include:

very high-power transmitters,

very low-noise microwave amplifiers,
 'NORM& POURQ4nON
OTHER P o w m n o N s AVAI~ABLE

LBIClRCUlARTWWSMII, LEFTCIRCUUW RECUYE

MREXPERIMENIAL PURPOSES
RIGHT CIRCULAATWWSMII, RIGHT URCULAA RECUM

L B I C I R W U W T W N S M I I , RIGHT URCUUR RECENE

LINUW W S M I T . ROTATABLE UNUW RECEIVE
RECORDING

EQUIPMENT
PERFORMANCE
Lrl RECORDING

ANGLES DOPPLER RANGE

SIGNALCH~W\C~RI~C~''

1 J
Figure 8-10. Functional block diagram of the Venus radar, 1962.
* high-efficiency microwave antenna feed systems,
advanced ranging1Doppler systems,
* data acquisition and signal processing, and
* planetary ephemeris prediction.

In addition to providing a realistic demonstration or test bed for new technology in the
DSN, the Goldstone radar, under the guidance of R. M. Goldstein, continued to pro-
duce good scientific data and to develop a presence in scientific literature.

Under this arrangement, the Goldstone radar developed rapidly. A Rubidium Frequency
Standard and improved closed-loop ranging system were added in 1962. High-power
transmitter development for the Network led to a 100-kW S-band transmitter for the
radar in 1963. Continuing improvement in antenna feed systems and low-noise ampli-
fiers gave the radar a system noise temperature of 35 kelvin in 1964. Signal processing
and ranging improved again in 1966 with the addition of a 9-channel correlator and
narrower and more stable range gate.

With these improvements, the capability of the radar was extended to include planets
at greater distances from Earth. Mars, Mercury, and even Jupiter were soon detected,
and scientifically significant observations were reported.59Venus was revisited in 1963,
and its rotation and large bright surface features were observed."

Within those first few years, half of the planets in the solar system were probed by the
Goldstone radar.

The Goldstone radar, like all other science in the DSN, was affected in a major way by
the availability of the first DSN 64-m antenna at Goldstone in 1966. Goldstein lost no
time in taking advantage of the ten-fold increase in capability provided by the 64-m
antenna, compared with the 26-m antenna he had used previously at the Venus site.

The rationale for the DSN adoption of the 64-m Az-El antenna design as the second
generation of large antennas for the Network has been discussed in earlier chapters.
While none of the rationale was concerned with the application of the antenna for plan-
etary radar, the same argument applied insofar as the application of radar technology to
Network problems was concerned. What was good for the radar was (in most cases)
good for the Network. So the radar program continued on the 64-m antenna on the
same basis as before, with two very significant exceptions. First, the radar had to com-
pete with the flight projects for observing time on the antenna. The second exception
was that all the radar-related experimental hardware and software had to be completely
separate from the tightly controlled operational equipment used to support the inflight
missions. Time on the antenna generally had to be justified on the grounds of relevance
of the radar data to one or more of the current missions. To keep the radar equipment
separate from the operational systems, it was located in a suitable area of the antenna
pedestal structure, with entirely separate interfaces with the antenna itself. A separate
antenna feed cone, which could be illuminated by the subreflector when needed, was
allocated to radar and other research and development activity. T h s arrangement remained
substantially unchanged through the life of the radar reviewed here and is still in use
today. The new 64-m antenna site became the Mars site, and in due course the radar
became the Goldstone Solar System Radar (GSSR).

Using the new Mars antenna and its enhanced radar capability which now included
range-Doppler mapping, Goldstein and others observed Venus again during the 1967
conjunction. They discovered more distinct rugged sections of surface terrain and stud-
ied a region named Beta from earlier observations in more detail.

In June 1968, the Goldstone radar detected Icarus, its first of many Earth-crossing
asteroids. It was diKcult to detect a target of such small size. Fortunately, a develop-
mental 450-kW S-band transmitter was then available at the Venus site, and this, together
with the Mars antenna in the bistatic mode for receiving, brought the reflected signal
above the detection threshold. The bistatic arrangement also circumvented the prob-
lem of rapid changeover from transmit to receive states necessitated by the short time
interval between transmission and reception of the reflected signal from close-range tar-
gets like asteroids.

The results of the observations of Icarus were the first (Goldstone) radar work published
in the distinguished journal for planetary science, Icarus, which subsequently became
the primary publication for planetary radar resear~h.~~

In 1970, the first of an ambitious series of joint programs with the Lincoln Laboratory's
Haystack antenna was carried out. It was estimated that with Haystack as the trans-
mitting antenna and Mars as the receiving antenna, a very significant increase (up to
ten times) over the capability of either one could be achieved. An X-band maser tuned
to the frequency of the Haystack radar transmitter was installed on the Mars antenna,
and, after several frustrating delays due to 64-m-antenna scheduling problems, Goldstone
receivers made a successll detection of the Jupiter satellites G a n p e d e and Callisto in
May and June 1970. The gaseous structure of Jupiter prevented the return of a detectable
signal from the planet itself.
By 1969, the 400-kW S-band capability had been installed at the Mars site and was
used for further studies of the Alpha and Beta regions of Venus. In December 1972,
and again in January 1973, Goldstein and Morris detected the rings of Saturn with this
powerful new capability. The returned signals were unexpectedly strong, 5 to 10 times
stronger than those received from Venus and Mercury. This surprising result triggered
an upsurge of interest in Saturn rings amongst astronomers and led to many and var-
ied interpretations of the data.62

In August 1974, Richard Goldstein and George Morris succeeded in detecting Ganymede
on six occasions with the Mars radar alone, using its 400-kW S-band transmitter and
35-K receiving system. They interpreted their results to indicate a considerable degree
of roughness on the surface, possibly due to rocky material from meteorites embedded
in a matrix of ice.

Progress on the development of a high-power X-band transmitter for the Goldstone

radar was reported by C. I? wig gin^^^ in 1972. Operation at X-band would provide an
increase of 6.5 dB in antenna gain, resulting in a reduction in observation time by a
factor of 20. At the same time, the wider bandwidth of an X-band system would per-
mit the use of faster range codes for more precise range measurement. The radar transmitter,
receiver, microwave components and the antenna feed were to be assembled into a new
Cassegrain cone that would be mounted in the Research and Development (R&D) posi-
tion of the tricone on the Mars antenna. Excessive transmission losses would be avoided
by operating the transmitter close to the antenna feed. The transmitter would operate
at a center frequency of 8.495 MHz and be fixed tuned with a bandwidth of 50 MHz
at the -1 dB points. It would consist of two 250-kW klystrons with a four-port hybrid
to combine the two outputs into one waveguide port, and it would use the high-volt-
age power supply and cooling system that normally powered the operational DSN S-band
400-kW transmitter.

This new development was justified on the basis of a study of the rings of Saturn as a
possible hazard to spacecraft navigation for the future Voyager missions to Jupiter and
Saturn. The first opportunity to probe the Saturnian system with radar would start in
December 1974, just three years before the two Eyagers were scheduled to launch and
six years before the first wager would arrive at Saturn.

Despite delays caused by unavailability of 64-m antenna time for installation work, the
X-band transmitter was completed and began observations on Saturn's rings in December
1974. The data from these observations confirmed the previous Goldstone results and
also more recent results from the Arecibo radar, regarding the high radar cross-section
(reflectivity for radar signals) and the high linear and polarization ratios. A p n , the inter-
pretation of these results was the subject of a great deal of discussion in the scientific
community. Various models, from a thick cloud consisting of irregularly shaped, cen-
timeter-size chunks of water-ice to large, meter-size metallic boulders covered with frost,
were proposed. Voyager opted for the former model and continued planning its mis-
sion. The subject was not closed, however, and the rings of Saturn were observed again
in 1977, 1978, and 1979.

The question of the composition and structure of the Saturn rings and the role of radar
astronomy in addressing it led Andrew Butrica to observe, very astutely,

The case of Saturn's rings resulted in radar astronomers contributing to plan-

etary science, in contrast to their studies of the Galilean moons. Those studies
. . . had been limited to epistemological studies, namely, what caused the Galilean
moons' strange radar signatures? Radar contributed to Saturn science, on the
other hand, by focussing less on such questions of radar technique and more
on scientific questions, such as the size of the ring particles and the number
and thickness of the ring layers. Although the solution of technical problems
was a prerequisite for any radar astronomy problem solving, the lack of obvi-
ous relevance to planetary science was a serious matter; the ability to solve
scientific problems, especially those relating to NASA space missions, was the
basis on which scientists judged the value of radar astronomy and on which
funding decisions were made.64

As subsequent events would show, the issue raised by Butrica in the preceding para-
graph-justification, and its effect on funding for the continued operation of the Goldstone
radar-would be a growing concern for the DSN in the years ahead.

However, in the immediate future, the Viking mission to Mars was concerned with the
selection of a suitable landing site for its two Landers. Surface roughness and slope were
key parameters in the landing site selection process, and these were ideally suited to
radar observation. Although Mars observations had been made in 1969, they covered
latitudes that were of no interest to Viking, and in any case, the results were too noisy
to be of much use for landing site selection. New data was needed, particularly of the
Syrtis Major and Sinus Meridiana regions which included potential Viking landing sites.

The radar group at JPL had expanded by then and included Richard R Green, Howard
C. Rumsey, George S. Downs, Paul E. Reichley, and Ray F. Jurgens, in addition to
Goldstein and Morris. Using hnds provided by the Viking Project office, members of
this team used the Goldstone radar to make X-band observations of the Syrtis Major
and Sinus Meridiana regions from October 1975 through April 1976."5 The extent to
which the Goldstone results, along with similar observations from the Arecibo and Stanford
radars, influenced the final landing site decision remains an open question. It is clear,
however, that the radar data, presented in the form of degrees of rms slope, was not
readily understood by the scientists making the decision, who lacked an adequate under-
standing of the new radar techniques. For them, the high-resolution images from the
Viking Orbiter were a more persuasive source of information.

Whatever the role of radar in the final decision, the site that was eventually selected
proved to be very satisfactory, as was shown by the subsequent landings and the Lander's
long period of successll operation on the Mars surface.

Following the radar detection of Icarus in 1968, interest in asteroids increased amongst
radar astronomers, at first in simply detecting them, but later in inferring their shapes
and nature of the surface material. In 1972, Goldstein observed 1685 Toro, and in
1975 the asteroid 433 Eros was probed at S-band and X-band and with left and
right circular polarization. These data allowed Goldstein and Jurgens to characterize
the size, shape, and surface properties of the asteroid with better resolution than had
been possible to that time. This was an important step toward future progress in the
techniques of asteroid characterization based on dual-frequency, dual-polarization, and
radar data. Although it was not apparent at the time, asteroid research was eventu-
ally to become the principal activity for the Goldstone radar, but that situation lay
twenty years into the future.

Continuing the support for the Voyager mission now approaching Jupiter, Goldstein and
Green observed Ganyrnede and Callisto in December 1977 at X-band, and with left
and right circular polarization. The unexpected radar signature they obtained confirmed
earlier Arecibo results but led to an interpretation of the surface of the satellites which
were at variance with the photographic images returned from Vqyager during the actu-
al Jupiter Encounter. The reason for this apparent contradiction remained unclear. Several
theories were advanced to account for it, some of which involved a better understand-
ing of the radar properties of planetary surfaces rather than more general scientific issues.
The questions of scientific worth versus technological value were again pertinent to the
continued interest of NASA in the radar.

For the next few years, the Goldstone radar received only sporadic use. The resources of
the DSN were oversubscribed with flight projects like Eking Orbiters I and 2, Eking
Lander I , Helios I and 2, Pionem 6 through 9, and Vqugers I and 2. These missions were
reluctant to relinquish 64-m antenna time at Goldstone for radar experiments, no matter
how scientificallyinteresting rhey claimed to be. Furthermore, Rechtin, Victor, and Goldstein
had by then moved on, and within the DSN there was no longer the strong constituen-
cy for radar support as there had been in former years. Budget constraints in the TDA
Office eliminated funding for maintenance of the Goldstone radar equipment, so that even
when astronomers were able to obtain observing time on the antenna, precious time was
ofien taken up with simply getting the radar to work. At NASA Headquarters, OTDA
was the source of finding for the DSN. ODTA declared that all further planetary radar
support by the DSN should be conducted in response to a formal requirement from OSSA
in the same manner as the DSN responded to OSSA requirements for tracking and data
acquisition support for flight missions. The necessary NASA Management Instruction was
issued (NMI 8430.1B) to this effect, and the TDA Office at JPL developed the lower
level documentation to cover radar requirements and support agreements.

While these changes were slowly evolving, and despite the difficulties of working with
the radar, Jurgens and his team were able to refurbish the radar in time to make their
first successful radar observation of a comet, IRAS-Araki-Alcock, in May 1982. Attempts
to observe comets on subsequent occasions were not successful until 1996, when J. K.
Harmon and others succeeded in detecting Comet Hyakutake in 1996.

In 1984, in response to persistent efforts by Jurgens and Downs, the NASA Ofice of
Space Science made funds available for the procurement of a new Digital Equipment
Corporation VAX 111780 computer to be used for data acquisition and processing func-
tions in the Goldstone radar. The computer formed the heart of a new High Speed
Data Acquisition System (HSDAS), whose primary mode of operation was the collec-
tion and processing of radar data. It supported CW and ranging radar experiments,
which would be required for the terrestrial and outer planet targets planned in the h r e
years. These included Mars, Mercury, Venus, asteroids, comets, Galilean satellites, and
the rings of Saturn.

As a subsystem of the planetary radar, the HSDAS collected the 15-MHz bandwidth
RJ?signal from the radar receivers and performed all of the necessary data rate com-
pression (filtering, phase correction, and correlation) functions necessary to get the radar
data on to a general-purpose computer for further processing and analysis. These finc-
tions had been performed in the past by an ever-increasing array of hardware devices
that were inadequate for the high-bandwidth radar experiments planned for the future.
The extreme flexibility designed into the HSDAS and the ability to control it from sim-
ple high-level programming languages made it a usefid tool for many other, nonradar
purposes as well, such as radio astronomy, advanced receiver development, real-time
 -
T >

FROM GENERAL PURPOSE VAX iinso

BASEBAND m DATA ACQUISITION COMPUTER
HARDWARE

I '
v I
-
1 -
IF CORRELATOR FPS
FROM PROCESSOR c ARRAY
RECEIVERS ASSEMBLY )
PROCESSOR
A 4
f 1,

TO TRANSMITTER PN
TRANSMITTER DRIVER GENERATORS
L J

Figure 8-11. High Speed Data Acquisition System for the Goldstone Radar, 1984.

arraying, and SETI. It was installed with the other radar equipment in the DSS 14
antenna pedestal area and, with periodic improvements, served the radar well for the
next decade. An overall block diagram of the High Speed Data Acquisition System is
shown in Figure 8-1

In 1983, Peter T. Lyman, the Assistant Laboratory Director for Tracking and Data
Acquisition, placed responsibility for direction of the Goldstone radar in the TDA Science
Office under the management of Nicholas A. Renzetti. It was Renzetti's goal to convince
NASAs Ofice of Space Science that the Goldstone radar warranted funding on the basis
of its value to NASA as a scientific instrument. With the support of Steven J. Ostro,
who had recently arrived at JPL from Cornell, where he had worked with Arecibo radar,
they redefined the Goldstone Solar System Radar (GSSR) in terms of a "scientific instru-
ment" on which planetary radar science could be performed. This closely paralleled the
approach that had been used successfidly to support radio astronomy, SETI, and crustal
dynamics in the DSN. To implement this approach, and to provide an interface between
the radar and the scientific community, Renzetti also created the concept of "Friend of
the Radar," a position filled first by Thomas W. Thompson. Martin A. Slade Med the
position in 1988.

It was expected that requests for GSSR support would be received from OSS and fund-
ing to meet the requirements would come from OTDA, just as was done with the flight
projects. But this never happened, and despite the endeavors of Renzetti and others to
sell the idea of the GSSR as a national resource for planetary radar research, the key
issue of funding for the GSSR remained unsolved.

In 1987, the Director of the Space Science and Instruments Office at JPL obtained a
modest level of &ding from NASA for planetary radar research and appointed Ostro
to manage its disposition in support of an observing program with the GSSR. Under
Ostro's influence, science finally supplanted technology as the justification for the GSSR's
existence, and excellence in its unique field of science became its goal.

Then, as had happened frequently in the past, events external to its own program over-
took planetary radar and resulted in a very significant advance in the capabilities of the
instrument. Over the period October 1987 through May 1988, the DSS 14 antenna
was upgraded and extended from 64 meters to 70 meters to meet the requirements of
the expanding planetary spacecraft program.

Driven initially by the tracking and data acquisition requirements of the Voyager Neptune
Encounter, described in an earlier chapter, the 64-meter to 70-meter upgrade was a
Network-wide implementation task which had been several years in the planning. DSS
14 was the last of the three antennas to be completed.

As Butrica viewed it, "the Voyager upgrade had a profound impact on the practice of
Radar Astronomy at JPL; it provided the GSSR with the sensitivity needed to carry out
research on a whole new set of targets (and to begin solving new sets of problems). Not
only did the GSSR gain the ability to undertake asteroid research but, when linked to
the Very Large Array in New Mexico, it became a new radar research tool."

The immediate benefit to the GSSR is illustrated in Figure 8-12 and shows the rela-
tive sensitivity of the 64-m and 70-m antennas for radar observations at X-band as a
function of declination. The improvement of about 2 dB corresponds to the increase
in antenna gain resulting from the increase in antenna diameter and other related
improvement^.^'

The figure also shows the improvement that would have resulted from replacing the
400-kW transmitter with a super-high-power, 1,000-kW transmitter. This latter initiative
was promoted by Renzetti in 1989 as part of a M e r upgrade to the GSSR which includ-
ed changes to the radar's transmitlreceive system to improve its capability for tracking
asteroids. Although the megawatt radar transmitter proposal was not approved, the aster-
oid modification to the radar feed system was eventually implemented. From 1992 through
Figure 8-12. Relative sensitivity for the Goldstone Radar Antenna.

1993, a low-noise waveguide switch was developed and installed. This device enabled trans-
mit and receive through the same microwave horn, with a switch-over time of two seconds.
It thereby obviated the cumbersome sub-refleaor switching procedure, which, with a switch-
over time of 25 to 30 seconds, had always limited the GSSR ability to observe targets
close to Earth. In 1996, the waveguide switch was superseded by an elegant arrangement
of two movable mirrors which, mounted external to the feed horns, did not compromise
either the radiated power or the sensitivity of the low-noise receiver.

By a fortuitous circumstance (for planetary radar), the so-called Voyager upgrade in the
DSN included the addition, at DSN expense, of low-noise X-band amplifiers to all 27
of the 25-meter antennas of the Very Large Array radio telescope operated by the NRAO
in New Mexico. Long affiliated with activities in the DSN, the VLA had participated
in the Voyager Neptune Encounter as an element of a Goldstone-VLA array for recov-
ery of downlink telemetry. The Voyager images of Saturn's moon Titan revealed a body
covered with an opaque layer of cloud, and the instruments suggested there might be a
global ocean of liquid ethane beneath the cloud cover.

Prior to the encounter, Caltech professor Duane 0. Muhleman had suggested that the
new VLA capability might be used effectively in the bistatic mode with GSSR for plan-
etary radar to further study Titan, with the objective of characterizing its surface features
and composition. Previous attempts to probe Titan with either the VLA or the GSSR
had not been successful, but the enhanced sensitivity of both combined offered the best
chance of success.
As a first demonstration of the GSSRNLA bistatic system, and with GSSR transmit-
ting and VLA receiving, the rings of Saturn were successfdy observed in early 1988.
The following year, Muhleman, Butler, and Slade observed Titan for the first time on
the nights of 3-6 June. The echoes from Titan were similar to those that had been
received from the Galilean satellites and suggested an icy surface, but more observations
were needed. Further observations were made jointly with VLA in 1992 and again in
1993.68The results of these experiments, although somewhat at variance with current
models of the Titan surface, were of great interest to the Cassini Saturn mission plan-
ners and were factored into the design for the Huygens Probe radar, scheduled to descend
to the surface of Titan in 2004.

With minor changes to the VLA configuration, and using different polarizations, the
VLA-Goldstone bistatic arrangement was used by the same experimenters to observe
Mars, first in 1988 and again, several times, during the 1992-93 opposition, and to
further investigate the presence of ice on the surface of Mercury in 1991.@

Amongst planetary radar astronomers, interest in Venus had diminished during the late
1980s for a number of reasons, some of them having to do with availability of resources
and the attraction of more distant targets. With the launch of Magellan and the extraor-
dinary success of its orbital radar mission in 1990, scientific investigation of Venus by
planetary radar essentially came to an end. There were, however, several investigations
of anomalous radar signatures from Venus conducted in subsequent years.

The Voyager upgrade to the Goldstone radar antenna, along with the addition of the
new HSDAS, gave the GSSR the sensitivity and resolution needed for the detection of
asteroids. The arrival of Steven J. Ostro provided the TDA Science Office with the expe-
rience and motivation required to engage in this new field of research in a major way.
An opportunity to combine the new capability with the addition of Ostro's experience
in a way that would change the course of the GSSR forever was not long in coming.

Three-dimensional modeling of asteroid shapes based on range-Doppler radar images

was still a relatively new technique when the Earth-crossing asteroid 4179 Toutatis was
discovered in January 1989. Toutatis was predicted to pass very close (9.4 lunar dis-
tances) to Earth on 8 December 1992. This would make it an ideal target for range-Doppler
imaging with the GSSR. Ostro proposed an elaborate observing program to Renzetti
that involved the GSSR, Arecibo, VLA, and the new DSS 13 BWG antenna working
as an interferometer with the GSSR. The outcome would be the highest resolution three-
Figure 8-1 3. Goldstone Solar System Radar images of Asteroid Toutatis, 1992.

dimensional images of an Earth-crossing asteroid ever produced. Renzetti accepted, and

the planning went forward.

Observations were carried out at the various sites over the period from 27 November
through 19 December 1992, and a good data return was obtained.'O Although the sub-
sequent reconstruction of the images was complicated by the slow tumbling rotational
state of the asteroid, they revealed the general shape and surface topography of Toutatis
at the highest resolution ever obtained of an asteroid. Typical images from this experi-
ment are shown in Figure 8-13 and reveal the presence of several craters on the two
(apparently connected) components of the irregularly shaped main body of the asteroid.
Using similar techniques, Ostro and Hudson successfully observed the next Earth-cross-
ing asteroid, 1620 Geographos, in 1994 and produced equally impressive imaging of
the asteroid body. In 1995, observations of 6489 Golevka included delay-Doppler imag-
ing, GSSR-VLA astrometry, and the first intercontinental planetary radar experiments
from Goldstone to Russia and Japan. Less than 600 meters across, Golevka was the
smallest solar system object imaged to that time.71

As scientific interest in radar observation of asteroids increased, it gave rise a heightened

awareness of the hazards posed by possible Earth encounters with asteroids and raised
questions of how such threats might be handled in the best interests of the public. The
role that the GSSR might play in continued evolution of this intriguing subject is dis-
cussed in the relevant scientific literature."

Astrometry is a branch of astronomy that deals with the measurement of the positions,
motions, and distances of planets, stars, and other celestial bodies, including asteroids.
In recent times, when Ostro and others employed radar techniques to study and pre-
dict the motions of asteroids, the term "radar astrometry" was coined to describe this
aspect of radar astronomy.

Thus, data generated by the GSSR were used, not only for science data for imaging and
surface characterization experiments, but also to refine the orbits of Earth-crossing aster-
oids. This extension of the GSSR capability was proposed originally by Jurgens and
carried forward by Ostro, who enlisted the support of two of JPCs orbit determination
specialists, Donald K. Yeomans and Paul W. Chodas. Supported by this powerful group
of experts, radar astrometry gined prominence in the period 1990-92 using data from
the GSSR and the Atecibo radars.73Together
- with like advances in radar imaging,
- - radar
data were recognized by the International Astronomical Union (IAU) as a data source
in its circulars to announce the discovery and orbit of newly discovered asteroids.

It seemed most likely that postdiscovery MU circulars for new asteroids in the near
future would be composed of computer files containing high-resolution video images of
the objects and highly precise ephemerides, based on data taken by the GSSR and
Arecibo planetary radars.
The renaissance of the Goldstone Solar System Radar dates from about the early 1990s.
By that time, the radar was no longer perceived primarily as a test bed for new DSN
technology, but was recognized in its own right as a scientific instrument for solar sys-
tem research. The remarkable success in the first detection of the Earth-crossing asteroids
did much to popularize the new approach to planetary radar. Over rhe period from
1993 to 1997, most GSSR research was conducted under the aegis of NASKs mission
to the solar system. Research topics lay within the scope of the Solar System Exploration
or Structure and Evolution of the Universe programs as shown in the table below.

NASA Theme: Structure and Evolution of the Universe

GSSR Experiment:
Mercury Tests of Gravitational Theories-
Ongoing experiments
NASA Theme: Formation and Dynamics of Earth-like Planets
GSSR Experiments:
Mercury Surface Properties-
Ongoing analysis of new and existing data
Mercury Polar Studies-
X-band 111-disk imaging planned for 199811993
Venus Surface Properties-
Observations planned for 199811999
Venus Geophysics-
Observations planned for 199811999
Mars Surface Properties-
Observations planned for 1999 Mars opposition
Mars Polar Studies-
Observations planned for 1999 Mars opposition
Mars Geophysics-
Analysis of existing topographic maps in progress
Galilean satellites-
No new observations currently planned
NASA Theme: Mission to Planet Earth
GSSR Experiments:
Earth Stratospheric Studies-
No new observations currently planned
Lunar Polar Observations-
Data from 1997 observations being analyzed
 NASA Theme: Mars Exploration Support
GSSR Experiment:
Mars Landing Site Certification-
Radar data aided landing site certification for 1997
Mars Pathfinder landing74
Observations in 1999 will aid hture 2001 landings
NASA Theme: Building Blocks and Our Chemical Origins
GSSR Experiments:
Asteroids, Imaging and Orbit Refinement-
Several asteroid observations ~lannedeach year
Radar Data Crucial for Orbit Determination of Near-Earth
Asteroids
Titan Surface Properties-
Bistatic observations with VLA planned for
October 1998 to aid Cassini Huygens-Probe
mission
Comet studies-
Comet Hyakutake observed in 199675
Observations to continue as opportunities arise
..* 7c* ' / ii I t i i i i l iir,i ",,TI-C"C i"'ii" lid
iiir ii'i"ii, i -. Ll.2//,"ri(,z ii<i-".T.'r<<.r"<z i.-*.%A"".-<" T
, X" ,i " ' .' .' ,"V

The outcome of all this activity was a plethora of publications in the scientific litera-
ture. In the ten-year period from 1988 to 1997, nearly 100 papers based on data from
the GSSR were published in highly respected journals such as Science, Icam, Bulletin
of the American As~onomicalSociety, and Lunar and Planetary Science. The productivi-
ty of the Goldstone Radar, in terms of yearly scientific publications for the most recent
ten- year period, is shown in Figure 8-14.

The data clearly show the revival of interest in radar astronomy following the Voyager
upgrade and the start of the asteroid detection program. The data also suggest that the
level of scientific interest, when constrained by the amount of antenna time available
for radar observations, results in the publication of 10 to 12 papers per year. It does
not lead to any conclusion regarding the scientific value of the papers so pr0duced.7~77

As we have seen, the constant efforts of a relatively small group of dedicated scientists
produced a steady stream of scientific knowledge throughout the lifespan of the Goldstone
radar reviewed here. Sometimes the stream slowed to a trickle and sometimes it was in
full flood, but it never dried up. The number of publications that resulted from these
efforts attests to the productivity of the radar.
I \

DSN Radar Astronomy

Publications1988-1997
18
16
14
12
10
8
8
4
2
0
1988 1990 1992 1994 1996 1998
= Number Published
\ I

Figure 8-14. Goldstone Radar astronomy publications, 1988-97.

To better appreciate the overall scope of the work done with the radar over the first 35
years of its active life, the reader may wish to consult the following table of significant
events in the scientific life of the radar.

Significant Events in the DSN Planetary Radar Program

r , < ~ , < ~ ~ ~ - ~ z < ~ ~ ~ ~ ~ ~ ~ ~ , ~ < ~ , c < ~ < ~ ~ r < ~ - ~ z ~ ~ ~ ~ ~ ~ ~ ~ Y r m ~ ~ m < ~ ~ ~ < ~ ~ ~ 2 ~ r ~ < ~ ~ ~ f l , r < < < R < < ? ~ , r 2 ~ m

1961 Venus First detection, determined value for AU to accuracy of 500 km,
refined ephemerides

1962 Venus Determined slow retrograde motion

1963 Mars First detection, measured albedo vs. longitude at 13 degrees lat-
itude

1963 Mercury First detection, measured albedo and range

1963 Jupiter Statistically significant detection, not repeated

1964 Venus Discovered large bright surface features

1965 Mars Measured albedo and roughness vs. longitude, Ma IV camera
directed to image Trivium Charontis, a very bright smooth radar
target

1966 Venus Located of a number of new surface features, rotation found to

be approx. 243.0 days

1967 Venus Accurate range measurements led to improved knowledge of Venus

orbit and radius, and also AU

1968 Icarus First detection of asteroid, verified ephemeris and estimated size
and roughness

1969 Venus Range measurements, large map with highest resolution, north-
south ambiguity resolved

Mars Determined accurate ( 4 . 2 km)altitude profile of equatorial region

IVIercury Surface features detected (not Venus-like)

Mars Detailed surface mapping

Jupiter Observed satellites Ganpede, Callisto for Voyager with Haystack

in the bistatic mode

Mars Detailed surface mapping

Toro Asteroid 1685 Toro detected

Saturn Saturn rings observed

Ganymede Observed surface of Ganyrnede

Mars Surface roughness studies for V i n g Lander site selection

Eros Asteroid 433 Eros detected

Ganymede Ganyrnede observed right and leh circular polarization returns

Saturn Saturn rings observed

Saturn Saturn rings observed

Comet Comet IRAS-Araki Alcock detected

Jupiter Determined polarization ratios and radar cross sections of Galilean

satellites, noted strange radar signatures

Saturn Observed satellite Titan for Cassini with Haystack in bistatic mode

Mercury Observed Mercury with Haystack in bistatic mode

Earth Observed Earth orbital debris

Mars Observed Mars with VLA in bistatic mode

Asteroid Observed first Earth-Crossing Asteroid (ECA) 4179 Toutatis with

DSS 13 in bistatic mode

Mars Observed Mars with VLA in bistatic mode

Mercury Full-disk radar images, detection of ice at North Polar region

Asteroid Three-dimensional modeling of Earth-Crossing asteroid 1620

Geographos

Mars Pathfinder landing site characterization

Comet Successful observation of Comet Hyakutake

Moon Search for ice in shadowed areas of the Polar Regions

Endnotes

1. M. S. Reid, R. C. Clauss, D. A. Bathker, and C. T. Stelzried, "Low-Noise Microwave

Receiving Systems in a Worldwide Network of Large Antennas," Proceedings of the
IEEE, vol. 61, no. 9 (September 1973), p. 1330-35.

2. Andrew J. Butrica, To See the Unseen: A History of Phnetary Raddr, The NASA
History Series, NASA SP-4218 (Washington, DC: NASA History Office, 1996).

3. M. J. Hein, oral interview with author, 9 April 1998.

4. N. A. Renzetti, "Oral History," J. Alonso, 17 April 1992.

5. N. A. Renzetti and A. L. Berman, "The Deep Space Network as an Instrument for

Radio Science Research," JPL Publication 80-93 (15 February 1981).

6. S. W. Asmar and N. A. Renzetti, "The Deep Space Network as an Instrument for

Radio Science Research," JPL Publication 80-93, Rev. 1. (15 April 1993).

7. C. T. Stelzreid, 'RParaday Rotation Measurement of a 13cm Signal in the Solar

Corona," JPL Technical Report TR 32-1401 (1970), p. 83.

8. R. Woo, and R. Schwenn, "Comparison of Doppler Scintillation and In-Situ

Spacecraft Plasma Measurements of Interplanetary Disturbances," Journal of
Geophysical Rasearch 96 (199 1): 2 122744.

9. A. Kliore, D. L. Cain, and T. W. Hamilton, "Determination of Some Physical

Properties of the Atmosphere of Mars from Changes in the Doppler Signal of a
Spacecraft on an Earth-Occultation Trajectory," JPL Technical Report TR 32-674
(1964).

10. G. L. Tyler, "Radio Propagation Experiments in the Outer Solar System with
Voyager," Proceedings of the IEEE, vol. 75, no. 10 (1987).

1I. J. D. Anderson, E B. Exposito, W. Martin, C. L. Thornton, and D. 0.Muhleman,

"Experimental Test of General Relativity Using Time-Delay Data from Mariner 6
and Mariner 7,"Awophysical Journal 200 (1975): 221-33.
12. F. B. Estabrook and H. D. Walhquist, "Response of Doppler Spacecraft Tracking to
Gravitational Radiation," General Rehtivity and Gravitation 6 (1975): 439-47.

13. N. A. Renzetti et al., "The Deep Space Network-An Instrument for Radio
Astronomy Research," JPL Publication 82-68, Rev. 1 (1 September 1988).

14. K. W. Linnes, "Radio Science Support," JPLTechnical Report TR 32-1526, Vol. 111,
The Deep Space Network, Progress Report, March-April 1971 (15 June 1971):
46-5 1.

15. M. J. Batty, D. L. Jauncey, S. Gulkis, and M. J. Yerbury, "The Tidbinbilla

Interferometer," PR 42-58, May-June 1980 (15 August 1980): 6-8.

16. M. J. Batty, D. L. Jauncey, l? T. Rayner, and S. Gulkis, "Accurate Radio Positions

with the Tidbinbilla Interferometer," TDA Progress Report PR 42-58, May-June
1980 (15 August 1980), pp. 1-5.

17. G. S. Levy et al., "Orbiting Very Long baseline Interferometer Demonstration Using
the Tracking and Data Relay Satellite System," Proceedings of International
Astronomical Union Symposium, no. 110, VLBI and Compact Radio Sources
(Bologna, Italy, 27 June-1 July 1983), pp. 405-06.

18. T. B. H. Kuiper to N. R. Haynes, "On the Role of Radio Astronomy in the DSN,"
internal memorandum, TBK 94-007, 6 February 1994.

19. T. B. H. Kuiper and M. R. Wick, "Radio Astronomy Observations Program:

Detailed Mission Requirements," Jet Propulsion Laboratory, JPL Document No. D-
12940, DSN 870-348 (1 February 1996).

20. T. B. H. Kuiper to D. J. Mudgway, "Significant Advances in Radio Astronomy

Using the DSN," personal communication, August 1999.

21. J. S. Gubbay et al., "Trans-Pacific Interferometer Measurements at 2300 MHz,"

Nature: 222, 730.

22. G. S. Downs et al., "Average Pulsar Energies at Centimeter Wavelengths," Nature

222 (1969): 1257.
23. I? E. Reichle~and G. S. Downs, "Observed Decrease in the Period of Pulsar PSR
0833-43," Nature 222 (1969): 229.

24. B. G. Clark et al., "Variations in the Fine Structure of Quasars 3C279 and 3C273,"
Bulletin of the American Astronomical Society 3 (197 1): 383.

25. R. L. Carpenter et al., "Search for Small-scope Anisotropy in the 2.7 Cosmic
Background Radiation at a Wavelength of 3.56 Centimeters," AstrophysicalJournal
182 (1973): L61.

26. M. J. Batty et al., "Tidbinbilla Two-element Interferometer," AstronomicalJournal

87 (1982): 938.

27. J. S. Ulvestad, C. D. Edwards, and R. Linfield, "Very Long Baseline Interferometry

Using a Radio Telescope in Earth Orbit," TDA Progress Report PR 42-88, October-
December 1988 (15 February 1986): 1-10.

28. R. Linfield et al., "VLBI Observations of Three Quasars Using an Earth-orbiting

Radio Telescope," Bulletin of the American Astronomical Society 18 (1986): 993.

29. K. M. Desai et al., "First Speckle Hologram of the Interstellar Plasma," Astrophysical
Journal 393, no. 2 (10 July 1992): L75-L78.

30. S. J. Tingay et al., "The Subparsec-Scale Structure and Evolution of Centaurus A.

The Nearest Active Radio Galaxy," AstronomicalJournal 115, issue 3 (March 1998):
960-74.

31. T. Velusamy T. B. H. Kuiper, and W D. Langer, "CCS Observations of the

Protostellar Envelope of B335," AstrophysicalJournal45 1 (October 1995): L75.

32. T. B. H. Kuiper, W. D. Langer, and T. Velusamy, "Evolutionary Status of the Pre-

protostellar Core L1498," AstrophysicalJournal 468 (September 1996): 76 1.

33. C. R. Gwinn et al., "Size of the Vela Pulsar's Radio Emission Region: 500
Kilometers," AstrophysicalJournal483 (July 1997): L53.

34. J. C. Guirado et al., "Xstrometric Detection of a Low-Mass Companion Orbiting

the Star AB Doradus," AstrophysicalJournal 490 (December 1997): 835.
35. R. E. Edelson and G. S. Levy, "The Search for Extraterrestrial Intelligence:
Telecommunications Technology," TDA Progress Report PR 42-57, March-April
1980 (15 June 1980), pp. 1-8.

36. A. L. Berman, "The SETI Observational Plan," TDA Progress Report PR 42-57,
March-April 1980 (15 June 1980), pp. 9-15.

37. B. K. Levitt, "SETI Pulse Detection Algorithm: Analysis of False Alarm Rates,"
TDA Progress Report PR 42-74, April-June 1983 (15 August 1983), pp. 149-58;
S. Gulkis and E. T. Olsen, "Gain Stability Measurements at S-band and X-band,"
TDA Progress Report PR 42-74, April-June 1983 (15 August 1983), pp. 159-68;
B. Crow, "SETI Downconverter," TDA Progress Report PR 42-67, November-
December 1981 (15 February 1982), pp. 1-5; E. T. Olsen and A. Lokshin, "The
SETI Interpreter Program-A Software Package for the SETI Field Tests," TDA
Progress Report PR 42-74, April-June 1983 (15 August 1983), pp. 169-82.

38. B. Crow, A. Lokshin, M. Marina, and L. Ching, "SETI Radio Spectrum

Surveillance System," TDA Progress Report PR 42-82, April-June 1985 (15 August
1985), pp. 173-84.

39. S. Gulkis, M. J. Klein, E. T. Olsen, R. B. Crow, R. M. Gosline, G. S. Downs, M. l?

Quirk, A. Lohhin, and J. Solomon, "Objectives and First Results of the NASA
SETI Sky Survey Field Tests at Goldstone," TDA Progress Report PR 42-86, April-
June 1986 (15 August 1986), pp. 284-93.

40. J. Solomon, W. Lawton, M. l? Quirk, and E. T. Olsen, 'X Signal Detection Strategy
for the SETI All Sky Survey," TDA Progress Report PR 42-83, July-September 1985
(15 November 1985), pp. 191-201; M. l? Quirk, H. S. Wilk, and M. J. Grimm, 'R
Wide-Band, High Resolution Spectrum Analyzer," TDA Progress Report PR 42-83,
July-September 1985 (15 November 1985), pp. 180-90.

41. M. J. Klein, S. Gulkis, E. T. Olsen, and N. A. Renzetti, "The NASA SETI Sky
Survey: Recent Developments," TDA Progress Report PR 42-98, April-June 1989
(15 August 1989), pp. 218-2-26.

42. G. R. Coulter, M. J. Klein, l? R. Backus, and J. D. Rummel, "Searching for

Intelligent Life in the Universe: NASA's High Resolution Microwave Survey," in
Sjoerd L. Bonting, ed., Advances in Space Biology and Medicine, Vol. 4 (Greenwich,
Connecticut: JAI Press, 1992), pp. 189-224.
43. K. M. Ong, I? F. MacDoran, J. B. Thomas, H. F. Fliegel, L. J. Skjerve, D. J.
Spitzmesser, l? D. Batelaan, S. R Paine, and M. G. Newsted, "A Demonstration of
Radio Interferometric Surveying Using DSS 14 and the Project Aries Transportable
Antenna," DSN Progress Report PR 42-26, January-February 1975 (15 April
1975), pp. 41-53.

44. D. L. Brun, S. C. Wu, E. H. Thom, F. D. McLaughlin, and B. M. Sweetser, "Orion

Mobile Unit Design," TDA Progress Report PR 42-60, September-October 1980
(December 1980), pp. 6-32.

45. C. G. Koscielski, "Mojave Base Station Implementation," TDA Progress Report PR

42-78, April-June 1984 (15 August 1984), pp. 216-24.

46. J. M. Davidson and D. W. Trask, "Utilization of Mobile VLBI for Geodetic

Measurements," TDA Progress Report PR 42-80, October-December 1984 (15
February 1985), pp. 248-66.

47. L. A. Buennagel, l? F. MacDoran, R. E. Neilan, D. J. Spitzmesser, and L. E. Young,

"Satellite Emission Range Inferred Earth Survey (SERIES) Project: Final Report on
Research and Development Phase, 1979 to 1983, JPL Publication 84-16 (Pasadena,
California: JPL, 1984).

48. J. M. Davidson, C. L. Thornton, S. A. Stephens, S. C. Wu, S. M. Lichten, J. S.

Border, 0. J. Sovers, T. H. Dixon, and B. G. Williams, "Demonstration of the
Fiducial Concept Using Data From the March 1985 GPS Field Test," TDA Progress
Report PR 42-86 (April-June 1985), pp. 301-06.

49. J. M. Davidson, C. L. Thornton, T. H. Dixon, C. J. Vegos, L. E. Young, andT. l?

Yunck, "The March 1985 Demonstration of the Fiducial Network Concept for GPS
Geodesy: A Preliminary Report," TDA Progress Report PR 42-85, January-March
1986 (15 May 1986), pp. 212-18.

50. G. Lanyi, "Total Ionospheric Content Calibration Using Series GPS Satellite Data,"
TDA Progress Report PR 42-85, January-March 1986 (15 May 1986), pp. 1-12.

5 1. GPS Flight Demonstration on Topex, JPL proposal, July 1984.

52. DSCC Media Calibration Subsystem, DMD 1987-1992, Functional Design

Review, November 1986.
53. Catherine L. Thornton, interview with Douglas J. Mudgway (April 1998).

54. W. K. Victor and R. S. Stevens, "Exploration of Venus by Radar," Science 134

(1961): 46.

55. Andrew J. Butrica, Ta See the Unseen: A History of Planetary Radar, The NASA
History Series, NASA SP-4218 (Washington, DC: NASA History Ofice, 1996).

56. M. H. Brockman, L. R. Malling, and H. R. Buchanan, "Venus Radar Experiment,"

JPL Research Summary No. 36-8, Vol. 1 (FebruaryApril 1961), pp. 65-73.

57. W. Higa, 'RMaser System for Radar Astronomy," JPL Technical Report T R 32-103
(1 March 1961).

58. R. M. Goldstein, R. Stevens, and W. K. Victor, "Goldstone Observatory Report for

October-December 1962," JPL Technical Report TR 32-396 (1 March 1965).

59. R. M. Goldstein, "Mars: Radar Observations," Science CL (24 December 1965):

1715-17.

60. R. M. Goldstein and R. L. Carpenter, "Rotation of Venus: Period Estimated from

Radar Measurement," Science CXXDC (8 March 1963), pp. 9 10-1 1.

61. R. M. Goldstein, "Radar Observations of Icarus," Icarus no. 10 (1969): 430-3 1.

62. R. M. Goldstein and G. A. Morris, "Radar Observations of the Rings of Saturn,"

Icarus no. 20 (1973): 260-62.

63. C. I? Wiggins, "X-band Radar Development," JPL Technical Report TR 32-1526,

Vol. X I , September-October 1972 (15 December 1972), pp. 19-21.

64. See endnote 2, p. 21 5.

65. G. S. Downs, R. R. Green, and I? E. Reichley, "Radar Studies of the Martian Surface
at Centimeter Wavelengths: The 1975 Opposition," I c a m no. 33 (1978):
441-53.
66. L. J. Deutsch, R F. Jurgens, and S. S. BroM, "The Goldstone Research and
Development High-Speed Data Acquisition System," TDA Progress Report PR 42-
77, January-March 1984 (15 May 1984), pp. 87-96.

67. N. A. Renzetti, T. W. Thompson, and M. A. Slade, "Relative Planetary Radar

Sensitivities: Arecibo and Goldstone," TDA Progress Report PR 42-94, April-June
1988 (15 August 1988), pp. 287-93.

68. B. J. Butler, D. 0 . Muhleman, and M. A. Slade, "Results From 1992 and 1993
VLAJGoldstone 3.5 cm Radar Data," Bulletin of the American Astronomical Society
25 (1993): 1040.

69. M. A. Slade, B. J. Butler, and D. 0. Muhleman, "Mercury Goldstone-VLA Radar:

Part I," Bulletin of the American Astronomicul Society 23 (1991): 1197.

70. S. J. Ostro et al., "Radar Imaging of Asteroid 4179 Toutatis," Bulletin of the
American Astronomical Society 25 (1993): 1126.

71. S. J. Ostro, "Goldstone Radar Research of Near-Earth Asteroids," DSN Technology

und Science Program Akws issue 7 (February 1997): 13.

72. S. J. Ostro, "The Role of Ground-Based Radar in Near-Earth Object Hazard

Identification and Mitigation," in T. Gehrels and M. S. Matthews, eds., Hazards Due
to Comets and Asteroids (Tucson, Arizona: University of Arizona Press, 1994), pp.
259-82.

73. D. K. Yeomans et al., "Asteroid and Comet Orbits Using Radar Data," Astronomical
]ournal103 (1992): 303-17.

74. M. A. Slade, R. F. Jurgens, A. F. C. Haldeman, and D. L. Mitchell, "Hazard

Evaluation for the Mars Pathfinder Prime Landing Site from Synthesis of Radar
Ranging and Continuous Wave Observations During the 1995 Mars Opposition,"
Bulletin of the American Astronomical Society 27, no. 3 (1995): 1103.

75. J. K. Harmon et al., "Radar Detection of the Nucleus and Coma of Comet
Hyakutake," (Cl1996 B2), Science 278 (1997): 1921-24.
76. Radar astronomy online Web sites: A comprehensive listing of published
scientific work based on all aspects of GSSR observations is available at
http://wireless.jpl. nasa.gou/RADARl.

77. Papers dealing with radar observations of asteroids are listed at http://echo.jplnasa.gou/
publications~ubs.hrml.
IN THE BEGINNING (1958 TO 1963)

Following the ARPA decision to launch several lunar probes by the end of 1958 using
the Army JUNO-I1 launch vehicle, the Army turned to its long-time contractor, JPL,
for the payloads and the tracking and data acquisition facilities required for its portion
of the new Pioneer Program.'

The flight paths of the lunar probes required a tracking station at Cape Canaveral to
cover the launch phase, another in the vicinity of Puerto Rico for the near-Earth phase,
and a deep space station in the general vicinity of JPL to track the probe on its way to
the Moon. Time was short. There were only eight months to select a site, build an
antenna, and set up the requisite receivers, recorders, and communication circuits.
Fortunately the Army owned a considerable amount of property in the Mojave Desert,
and it was on this property at Goldstone Dry Lake, about 100 air miles from Los
Angeles, that a suitable site for building the deep space antenna was found.

Construction of the site began at Goldstone in June 1958; assembly of the antenna
commenced in August; and by December, the antenna, together with electronics, receivers,
and communications, was complete. Pioneer 3 was launched on 6 December 1958. This
station, appropriately named Pioneer, was the first station of what was to become the
Deep Space Instrumentation Facility (DSIF), and eventually the Deep Space Network
(DSN) described in earlier chapters.

As these events began to unfold in 1958, the engineering organization within JPL reflect-
ed the technologies that had been appropriate to the missile guidance and telemetry
support provided as a contractor to the U.S. Army. The three principal engineering
departments were Aerodynamics and Propellants, Design and Power Plants, and Guidance
and Electronics. The Guidance and Electronics Department, headed by Robert J. Parks,
consisted of three Technical Divisions: Guidance Analysis, led by Clarence R. Gates;
Guidance Development, led by Walker E. Giberson; and Guidance Research, led by
Eberhardt Rechtin. Rechtin's Guidance Research Division consisted of an Electronics
Research Section under Walter K. Victor and a Guidance Techniques Research Section
led by Robertson st even^.^ Design and implementation of the Pioneer DSIF station was
carried out under the technical management of engineers from these two sections, with
overall responsibility shared by Rechtin, Victor, and Stevens.
In addition to its work for the Army, JPL also supported a substantial aerodynamics
research facility known as the Southern California Cooperative Wind Tunnel. This facil-
ity was funded and used, on a cooperative basis, by the many aerospace companies
flourishing in the southern California region at that time. Although the Wind Tunnel
was in no way associated with the DSIF, its Chief of Facilities and Mechanical Equipment
would soon play a prominent role in the future history of the DSN. His name was
Wiliam H. Bayley.

A mechanical engineering !graduate (1952) from the University of Southern California

(USC) in Los Angeles, Bill Bayley worked on facilities engineering for the Lockheed
Aircraft Corporation in Burbank, California, before coming to JPL in 1956 to manage
the facilities of the southern California Cooperative Wind Tunnel. At that time, and for
several years afterward, this major research kcility reflected JPCs association with Caltech
and aeronautical research.

After JPG transfer to NASA, Rechtin's network of tracking stations, communications,

and data-processing systems, as well as the staff needed to run it, expanded rapidly. He
needed help. The addition of an assistant to interact with NASA on issues of staff and
budget and to deal with financial and administrative matters at JPL allowed Rechtin to
address the more technical issues associated with the future world network. He turned
to Bill Bayley, a perfect choice as it turned out, and made him General Manager in
1960. When Rechtin left JPL in 1967, Bayley assumed Rechtin's position as Assistant
Laboratory Director for Tracking and Data Acquisition, a position he held with dis-
tinction until he retired in 1980. During his term of ofice, the Deep Space Network
developed into a major NASA facility of 26-m, 34-m, and 64-m antennas in three coun-
tries around the globe. That achievement was due, in no small measure, to his skill in
dealing with both NASA Headquarters and JPL and his ability to optimize the rela-
tionship between them in the best interests of the DSN.

A pleasant man to talk to and a gentle man to deal with, both at work and outside the
work environment, Bill Bayley was popular with his colleagues at JPL and NASA
Headquarters, as well as his counterparts at the government agencies in Spain, Australia,
and South Africa, on whom NASA depended for support of its tracking stations in those
countries. His quiet sense of humor-he was a master of the "bon mot" and "double
entendrev-served to enhance his connectivity with the people he met. Bill Bayley intro-
duced "all hands" meetings to the DSN as a way of bringing top management into
direct contact with atl members of the organization for a frank and open discussion of
problems, policies, and procedures. Such meetings, led by Bayley himself, were always
held away from the normal workplace to create a less formal and more productive aunos-
phere for the free expression of ideas and opinions.

He was a devotee of healthy living and exercise, so it was ironic that Bayley's untimely
death in 1981 was due to a heart-related problem.

In 1958, while the Pioneer antenna was being rushed to completion at Goldstone to
meet the deadline for the first Pioneer Lunar Probe launch in December, momentous
events were occurring elsewhere in the U.S. space effort. Later that year, Congress passed
the National Aeronautics and Space Act to create the National Aeronautics and Space
Administration (NASA). President Eisenhower immediately signed the new Act into law,
and NASA began operating under the leadership of its first Administrator, T. Keith
Glennan, on 1 October 1958. The transfer of JPL from the Army to NASA followed
almost immediately. Instead of working under a U.S. Army contract, JPL now worked
under a NASA contract. NASA inherited not only JPCs experienced personnel and its
facilities, including Goldstone, but also JPCs vision of a worldwide tracking network for
deep space probe^.^ N M s plans for JPL were in keeping with this vision, and there
was no interruption to the work in progress as a result of this essentially political change.

However, within NASA and JPL, the change raised serious questions regarding the sta-
tus of JPL. NASA wanted to operate JPL like other Agency Centers, rather than, as JPL
desired, an outside contractor managed by Caltech. Describing this situation, historian
Clayton Koppes wrote, "Insider or outsider. Vast amounts of time and energy would be
consumed in resolving the question of insider or outsider throughout the next decade."4
Despite these internal disturbances, essential work moved rapidly forward.

When the new NASA Headquarters organization first came into being, responsibility
for NASA-wide tracking and data acquisition services and facilities rested in Abe Silverstein's
Office of Space Flight Development. Within that office, the position of Assistant Director
for Space Flight Operations was filled initially by Edmond C. Buckley.

In this capacity, Buckley interacted directly with JPL on all matters related to NASA
policy, guidelines, and budget for the original construction and operation of the DSIF.

Referring to Buckley's early association with the DSN, historian Cargill Hall said, 'R
graduate of Rensselaer Polytechnic Institute, personable and articulate, he brought to
NASA many years of [NACA] experience in the development of the Wallops Island
 \
OFFICE OF THE ADMINISTRATOR
ADMINISTRATOR
T. I(Eml GLENNAN
DEPUTYADMINISTRATOR
H. L DRMEN

OFFICE OF SPACE FLIGHT DEVELOPMENT

DIRECTOR
A SILVERSTUN

ASSISTANT DIRECTOR
SPACE FLIGHT OPERATIONS
E. C. BUCKLEY

I I
SR SCIENTIFIC EXECUTIVEASSISTANT
REPRESENTATIVE PLANS AND BUDGETS
IN AUSTRALIA
M. E. GREENSPON
E. P. HARTMAN
-
CHIEF CHIEF
ADVANCED DEVELOPMENT OPERATIONS
F. B. S M ~ G. M. TRUSZYNSKI

AND FAClLiTlES FREQUENCY

R. D. BRISKMAN

\ 1

Figure 9-1. NASA Office of Space Flight Development, 1959-61.

Launch Range, where he had been responsible for the development of tracking and
instrumentation associated with free flight research."

Fortunately for the fledgling DSN, Ed Buckley at NASA Headquarters and Eb Rechtin
at JPL enjoyed a great deal of mutual respect and worked well together to further their
common goal, the establishment of a worldwide network for NASA. Observed Cargill
Hall, "No major disagreements (between NASA and JPL) marred the planning and cre-
ation of the Deep Space Network."

Responsibility for tracking and data acquisition in the NASA Office of Space Flight
Development at this time is shown in Figure 9-1.

Within a very short time of its transfer to NASA, the organization of JPL was restruc-
tured to better meet its new responsibilities in the changing world. No longer was its
primary business that of missiles and rockets and the tracking and data acquisition tech-
nology to support them as a contractor to the U.S. Army, but lunar and planetary
r \

TELECOMMUNICATIONSDMSION
E. RECHTIN,
CHIEFAND PROGRAM DIRECTOR
W. H. BAMEY, ASST. CHIEF

COMMUNICATIONS COMMUNICATIONS COMMUNICATION COMMUNICATIONS ENGINEERING

RESEARCH ELEMENTS RESEARCH SYSTEM AND OPERATIONS
W. K. VICTOR R. STEVENS DEVELOPMENT N. A. RENZETTI
J. KOUKOL
-
JPL Resident Engineers TECH. STAFF
S. Africa, P. Jones R Goodwin
Australia, R. Fahenstock W. Spaulding
R Tardani

DAT SYSTEMS TRACKING DATA SYSTEM TEST SYSTEMS GOLDSTONE

ENGINEERING ANALYSIS AND OPERATIONS ENGINEERING OPERATIONS
C. W.Johnson J. R Fearey R. K Mallis R. Z. Toukdarlan W. Larkin
L J

Figure 9-2. Telecommunications Division at JPL, 1961-62.

spacecrafi and the technology to support them as a contractor to NASA. Seven new
technical divisions and six administrative divisions were created to deal with JPCs new
responsibilities in NASA's expanding Lunar and Planetary Program. All seven technical
divisions were directly responsible to JPL Director William H. Pickering, while the admin-
istrative divisions reported to the Assistant Laboratory Director (ALD) of the Business
Administration Offrce, Victor C. Larsen. Amongst the technical divisions, the
Telecommunications Division (Division 33), with Rechtin named as Chief and Bayley
as Assistant Chief, contained the essential technological expertise that was to become the
foundation stone of the emerging worldwide DSIF. In addition to Chief of the
Telecommunications Division, Rechtin was also designated as Program Director for the
DSIF, an indication of the growing importance of the DSIF in the JPL organization,
planning, and budget processes.

The Telecommunications Division, under Rechtin's leadership, comprised four sections

to provide the state-of-the-art technical and operational resources needed to transform
the DSIF from a single antenna in California to a worldwide network with tracking sta-
tions on three continents. The organization of the DSIF during this period, 1961-62,
is shown in Figure 9-2.

Working within this structure, Nicholas A. Renzetti led the engineering, implementa-
tion, and operations activity supported by an aggressive and very successful research and
development program led by Victor, Stevens, and Koukol. In 1961 and 1962, the total
engineering and technical st& for both functions totaled about 200 to 250 persons,
with a budget of 15 to 20 million dollars. Three years later, reflecting the demands of
an expanding network, the st& had increased to 300 to 400, while the budget had
increased to 50 to 60 million dollars. In addition to developing the DSIF, these same
technical resources were heavily engaged in supporting the first of the NASA lunar and
planetary missions, the Rangers and Mariners.

Although the Mariner 2 mission to Venus was very successfid and attracted a great deal
of scientific attention, the first five Ranger missions were not. In the aftermath of a
NASA inquiry into problems with the Ranger program, JPL restructured some parts
of its organization that were associated with the lunar and planetary program^.^ These
changes would have a far-reaching effect on the relationship between the organization
of the DSIF and that of the flight projects. The JPL structure which followed is shown
in Figure 9-3.

Of most significance to the future development of the DSIF was the creation of two
new offices headed by assistant laboratory directors who reported directly to the JPL
director. These were a Lunar and Planetary (Flight) Projects Office headed by Robert
J. Parks and a DSIF Office headed by Rechtin. These two offices would provide the
programmatic direction for 1) the Mariner, Surveyor, and remaining Ranger missions
and 2) the DSIF, respectively. Technical and administrative support would be pro-
vided by the seven technical divisions and the six administrative divisions. Although
initially the Technical Divisions were in direct line to the JPL director, an addition-
al Office for Technical Divisions was established within the next year with Fred H.
Felberg as ALD. This office coordinated and directed the activities of all seven tech-
nical divisions at the program level. With the exception of moving Victor to the
Chief position, with Joseph F. Koukol as deputy, the Telecommunications Division
remained intact.

JPL had arrived at an organization within which the DSIF, later to become the DSN,
would coexist with the flight projects and the technical divisions for the next sever-
al decades. As we shall see, fairly major modifications were made from time to time,
but with these basic offices in place, the DSN moved rapidly forward to meet the
demands for tracking and data acquisition support for the expanding NASA Lunar
and Planetary program.

Equally important to the evolution of the DSN in these early years were the changes
taking place in the newly formed NASA organization. As William R. Corliss explained,
 JET PROPULSION LABORATORY
W. H. PICKERING, DIRECTOR
B. SPARKS, DEPUTY

I
STUDY GROUPS FOR
L
-
ASSISTANT DIRECTORS FOR
RESEARCH, DEVELOPMENT,
AND ADMlNlSTRATlON

I I
E. RECHTIN
ASSISTANT DIRECTOR
DEEP SPACE INSTRUMENTATIONFACILITY

Ranger
Surveyor
Mariner

I
DIVISION31
SYSTEMS
C. R. GATES, CHIEF
!--= TECHNICAL DIVISIONS

TELECOMMUNICATIONS
W. K VICTOR, CHIEF

Systems Analysis System Test and Communications Systems Communications

T. W. Hamilton Elements Research
M. F. Easterling

Communications R.F. Systems Development

M. S. Johnson Engineering and Operations L W. Randolph
R. K. Mallis

System Design LaunchVehicle Systems SpacecraftTelemetry and

W. Downhower H. Magraf Command Spacecraft Radio
R. B. Harker R. P. Mathison

I
Figure 9-3. JPL organization, 1964.

Rather unexpectedly in the early 1960s, the tracking and data acquisition
function was assuming more and more importance in NASA's budget and,
consequently, in its organizational structure. Space flight turned out to be
not just all launch rockets and spacecraft, it depended very heavily upon
ground facilities for testing, launching, and, of course, tracking and com-
munication. The early literature of space flight does not foresee these
developments at all. Management practicalities soon forced NASA to recog-
/ -I

OFFICE OF THE ADMINISTRATOR

ADMINISTRATOR
JAMES E. WEBB
DEPUTYADMINISTRATOR
H. L. ORYOEN

OFFICE OF TRACKING AND

DATA ACQUISITION
DIRECTOR
E. C. BUCKLEY
DEPUTY DIRECTOR
G. M. TRUSZYNSKl
I
I I I I
NETWORK OPERATIONS 000 PROGRAM COORDINATION PROGRAM SUPPORT
AND FACILITIES COORDINATION AN0 RESOURCES AND ADVANCED SYSTEMS
AND MANAGEMENT
H. R. BROCKETT F. B. BRYANT T. V. LUCAS C. R. MORRISON

FACILITIES AND STATION COMMUNICATION AN0 NETWORK ADVANCED

IMPLEMENTATION FREQUENCY OPERATIONS SYSTEMS
MANAGEMENT
I N.PODNSKY I P.A. PRICE J. C. BAVELY R. R. STEPHENS
\ J

Figure 9-4. NASA organization for the Office of Tracking and Data Acquisition, 1961

nize the importance of tracking and data acquisition by placing this func-
tion on a par with space science, manned space flight, etc. O n 1 November
1961, a new Office of Tracking and Data Acquisition (OTDA) was created
at NASA Headquarters. Edmond C. Buckley, who had been in charge of
Space Flight Operations, was named director of the new office. In effect, the
entire tracking and data acquisition function was elevated a notch in the
NASA Headquarters hierarchy.

The new Office of Tracking and Data Acquisition in NASA Headquarters at the end
of 1961 is shown in Figure 9-4.

The organization remained in this form until OTDA moved up another step within the
NASA hierarchy when the position of Director, previously filled by Ed Buckley, was ele-
vated to Associate Administrator status in 1966. Buckley held that position until he
retired in 1968. He was replaced by Gerald M. Truszynski.
Truszynski, a native of Jersey City, New Jersey, came to NASA Headquarters in 1960
from Edwards Air Force Base in California, where he had been chief of the instru-
mentation division involved in developing tracking systems for jet- and rocket-powered
aircraft. A year later, he was named deputy in the Office of Tracking and Data
Acquisition and eventually followed Edmond Buckley to become the second Associate
Administrator of that office in 1968. In that position, he was responsible, at the
NASA Headquarters level, for the planning, development, and operation of global
tracking networks, including the DSN, facilities for NASA Communications
(NASCOM), and data acquisition and processing for all NASA spaceflight programs.
His term of office covered the Mariner and Viking Eras and paralleled the Bayley
period in the Tracking and Data Acquisition Office at JPL. This was a period of
enormous growth and change in the DSN. It was, perhaps, the period when the
Network passed from youth to maturity. Gerry Truszynski and Bill Bayley worked
well together as heads of the teams at Headquarters and JPL that brought about
those changes.

In addition to his involvement with NASA's Deep Space Network, Truszynski played a
major role in the provision of tracking, data, and supporting services for the Apollo 8
and Apollo 11 lunar missions.

Although the DSIF Office at JPL was not made an official program office until 1963,
interactions between it and OTDA had been conducted as if it were. This situation
changed in October 1963, when Rechtin's title was elevated from Program Director for
the DSIF to Assistant Laboratory Director for Tracking and Data Acquisition (ALDITDA)
and the former DSIF Office became the official TDA Program Office, a title it retained
for the next thirty years. At last, the organization at JPL paralleled that at NASA
Headquarters, as far at TDA was concerned. There remained one more event to finish
the story of the formation of the modern DSN. That final event occurred in December
1963 with the formal establishment of the Deep Space Network by the JPL director.
The historic interoffice Memo 218 that made the announcement, dated 24 December
1963, is reproduced in Figure 9-5.

In addition to the former DSIF, the tracking stations at Goldstone, California; Woomera,
Australia; Johannesburg, South Africa; and Cape Canaveral, Florida, the official DSN
included the intersite communications now called the Ground Communications Facility
(GCF) and the mission-independent portions of the new Space Flight Operations Facility
(SFOF) at JPL.
 J E T PROPULSION LABORATORY OFFICE OF THE DIRECTOR
Interoffice Memo 218
December 24. 1963

To: Senior Staff

Section Chiefs
Section Managers

From: W. H. Pickering

Subject: Establishment of the Deep Space Network

Effective immediately, the Deep Space Network i s established by combining

the Deep Space Instrumentation Facility, Interstation Communications. and the
mission-independent portion of the space Flight Operations Facility. Development
and operation of this network is the responsibility of the Assistant Laboratory
Director for Tracking and Data Acquisition by extension of the role statement for
this Assistant Director (Office of the Director IOM 200, October 2, 1963).

Funding sources a r e unchanged for Fiscal 1964 and for the budget sub-
mitted by J P L for Fiscal 1965. However, J P L will endeavor to have OTDA and
OSSA agree on a single source of funding a s quickly a s possible.

The interface with mission peculiar facilities and organizations will be

worked out between the Assistant Laboratory Director for Tracking and Data
Acquisition and the Assistant Laboratory Director for Lunar and Planetary Projects
using a s a guideline the definition of "mission-independent" as:

1. Required for two o r more flight projects.

2. Best handled by J P L and not outside flight project organizations

(ARC Pioneer, GSFC-BISFN, LeRc Centaur, etc.).

This change i s made in order to accommodate efficiently the increasing

number of outside flight projects for which the Jet Propulsion Laboratory has
been tasked to supply tracking and data acquisition support. The change should
also assist in closer integration of the previously separate facilities.

W. H. Pickering
Director

I
Figure 9-5.The Deep Space Network established, December 1963.
The events just described were of fundamental importance to the shaping the DSN dur-
ing its formative stages. These and subsequent events are summarized in the table below.

Top-Level Management of the DSN at JPL, 1960-97

Year Event or Deputy/Program JPL Director

Head of DSN: Title Manager
1958 NASA established Pickering
(October);
JPL moves to NASA
(December)
1960-62 Rechtin: Chief, Div. 33' Bayley Pickering
and DSIF Director
(January 1961)
1963 Rechtin: ALDZ/DSIF Bayley Pickering
(March)
Victor: Chief, Div. 33,
DSN established
(December)
1964 Rechtin: ALD/TDA3 Bayley Pickering
1967 Bayley: ALDITDA Victor Pickering
1980 Lyman: ALDITDA Johnson4 Murray
1987 Dumas: ALDITDA Johnson Men
1992 Haynes: ALDITDA Westmoreland Stone
1994 TMOD5 Established Stone
1996 Wesunoreland: Director, Coffin4 Stone
TMOD (June)
1997 Squibb: Director, Coffin Stone
TMOD (June)

1 Telecommunications Division.
2 Assistant Laboratory Director.
3 Tracking and Data Acquisition; changed to Telecommunications and Data Acquisition in 1982 to avoid
confusion with NASA's new Space Network using the Tracking and Data Relay Satellite System (TDRSS),
planned for operation the following year.
4 The tide of Deputy was superseded by Program Manager in 1980 and again in 1996.
5 Telecommunications and Mission Operations Directorate.
After the DSN was formally established at the end of 1963, the basic organizations
involving the DSN at NASA Headquarters and at JPL changed very little until 1994,
except for personnel and some vacillation concerning responsibility for the SFOF. Within
the DSN organization, however, many changes were made in the interest of improving
working relationships with OTDA and with the Flight Projects. The growth and devel-
opment of the DSN organization during this thirty-year period, 196494, is the subject
of our next discussion.
THE FORMATIVE YEARS ( 1964 TO 1994)

A DSN MANAGER FOR FLIGHT

PROJECTS

Within a few short months of the formalization of the DSN, further changes were made
to the organizational structure under which the DSN operated at JPL. A JPL organiza-
tion chart from 1964 showing the relative positions of the TDA and flight projects
offices, including their supporting technical divisions, is reproduced in Figure 9-6.

All of the technical divisions were directed by a single ALD, Fred Felberg. A technical
staff element was added to the TDA Office, and Richard K. Mallis took over the
Communications Engineering and Operations Section formerly managed by Renzetti.
Renzetti moved to a Technical Staff position in the TDA Office as the first DSN Manager
for the Ranger, Mariner, and Apollo missions. He was soon joined by DSN Managers
for the Lunar Orbiter, Surveyor, and Pioneer missions. In the Systems Division, Thomas
S. Bilbo replaced Marshall S. Johnson as Chief of Space Flight Operations. Thus, Mallis
and Bilbo were independently responsible to their division managers for the operation
of the SFOF and the DSN, respectively. They received technical support from other sec-
tions within the same divisions, but in a programmatic sense, their authority derived
from completely separate sources. The SFOF traced its funding and program direction,
via the Lunar and Planetary Projects Office at JPL, to OSSA at NASA Headquarters,
while the DSN traced its funding and program direction, via the Tracking and Data
Acquisition Office at JPL, to OTDA at NASA Headquarters.

The appearance of the DSN Manager function in the TDA organization needs some
explanation at this point, since it became a key factor in determining the future work-
ing relationships between the DSN and its client users, the flight projects.

As an inevitable consequence of the creation of separate offices at NASA Headquarters

for flight projects and tracking and data acquisition, a separate system of accountabili-
ty for the resources required by the one to support the other came into being. Based
somewhat on a long-existing system used at the military test ranges for receiving and
responding to range users' requirements for instrumentation to cover their tests, the sys-
tem was introduced to the DSN by a formal NASA Management Instruction (NMI
2310.1) dated March 1965. In this system, flight project requirements for DSN track-
ing and data acquisition services were identified in a Support Instrumentation
Requirements Document (SIRD). Upon receipt of a SIRD, the DSN would respond
with a NASA Support Plan that committed its resources as it deemed necessary to meet
the requirements in the SIRD. The SIRD and NSP were formal documents signed by
r \

JET PROPULSION LABORATORY

W. H. PICKERING, DIRECTOR

SERVICES, SUPPORT AND

STUDY GROUPS FOR -
ASST. LAB. DIRECTORS FOR
DIRECTOR'S OFFICE RESEARCH, DEVELOPMENT
AND ADMINISTRATION
I

R. J. PARKS F. H. FELBERG E. RECHTIN

ASST. LAB. DIR. ASST. LAB. DIR. ASST. LAB. DIR.
LUNAR AND PLANETARY TECHNICAL DIWSONS TRACKlNGAND DATAACQUISITION
PROJECTS
W. H. BAYLEY
GENERAL MANAGER
DEEP SPACE NETWORK

FLIGHT PROJECTS
Ranger OTHER TECHNICAL STAFF
Surveyor DSN MANAGERS
Mariner 64
Mariner 69
Pioneer, J. W. Thatcher
DIVISION 29, DSN RESIDENTS;
Project Englneerlng S. Africa, (Johannesburg)
Spain, (Madrid)
Australia, (Canberra, Woomera)

DIVISION 31
SYSTEMS TELECOMMUNlCAllONS
C. R. GATES, CHIEF W. K. VICTOR, CHIEF

Communications Systems Communications Elements

Systems Analysis
T. W. Hamilton
-- ComputerAppl~cations
F. Lesh Research -- Research
M. F. Easterling R. Stevens

Space Flight Operations Data Systems Communications Engineering R.F. Systems Development
T. S. Bilbo A. Accord and Operations -- L. W. Randolph
R. K. Mallis

SpacecraftTelemetry and Spacecraft Radio

Command - R. P. Mathison
R. 8. Harker
\ J

Figure 9-6. Organization for Tracking and Data Acquisition support at JPL, October 1964.

the program managers at OSSA and OTDA, respectively, and were intended to signify
approval for the expenditure of the TDA resources involved.

In earlier times, when one or two tracking stations and simple facilities for control of
flight operations were all managed more or less by a single entity, verbal agreements or
internal memos between participating groups were all that was needed to conduct mis-
sions for a single JPL-managed spacecraft. No longer was that the case. By the mid-1960s,
the DSN had expanded to two 26-m Networks and large 64-m antennas were under
construction. Spaceflight missions were conducted from a large, completely new hcility
called the Space Flight Operations Facility (SFOF), and several major, non-JPL projects,
including Apollo, were demanding its services.

To deal with this situation, Rechtin assigned a technical staff position to a DSN Manager
for each flight project. The DSN manager h c t i o n e d as single point of contact between
his flight project and the DSN, which at that time included the SFOF, for conveying
and negotiating requirements and commitments for tracking and data acquisition, mis-
sion operations, and data-processing support. The documentation system by which these
agreements were consummated was the SIRD and NSI? In actuality, the role of the
DSN Manager extended beyond the DSN to include responsibility for negotiating sup-
port for his flight project from other entities such as the NASA Communications Division
(NASCOM) at GSFC for communications, the Air Force Eastern Test Range (AFETR)
for launch support tracking coverage from land stations, Department of Defense (DOD)
for ships and aircraft, the Lewis Research Center (LeRC) for launch vehicle data, and
so on. Although the DSN manager did not have an inline operational role in flight
operations, he was at all times held accountable to the flight project manager for the
performance of the DSN in support of mission operations, according to the negotiated
agreements.

It had become apparent during the early part of the Ranger Program that the limited
facilities available at JPL for conducting flight operations and mission control would not
be adequate to support the greatly expanded space programs of the future.

With this in mind, JPL made a recommendation to NASA for the construction of a
new building at JPL that would be entirely dedicated to accommodating the personnel,
equipment, and facilities needed to manage all of the elements involved in flight oper-
ations on a continuous, round-the-clock basis. These elements included the DSIF, ground
communications, data processing and display, and mission control, with all the internal
communications needed to make them work together. NASA approved the recommen-
dation in mid-1961. Under the aegis of the NASA Office of Space Science and Applications
(OSSA), construction of what was called the Space Flight Operations Facility (SFOF)
began
- shortly afterward. The SFOF was completed in October 1963. The IBM 7094
and 7040 computers and 1301 disk files were moved in immediately and began oper-
ating just before the launch of the Mariner 4 mission to Mars in November 1964. Thus,
Mariner 4 was the first flight project to be supported by the new SFOF operating with
a combination of real-time and non-real-time computing systems.

In December 1964, however, programmatic responsibility for the SFOF was transferred
from OSSA to OTDA on the basis of the interrelated functions of mission control in
the SFOF and Network control in the DSN. This responsibility had been anticipated
in Pickering's announcement establishing the DSN in December 1963, and it remained
a major element of the DSN for almost the next decade? While the TDA Ofice assumed
programmatic responsibility for the SFOF, its operational and technical support was pro-
vided by the Space Flight Operations and Data Systems Sections of the Systems Division
3 1, in a somewhat analogous way to that in which the Telecommunications Division
33 supported the DSN. (See Figure 9-6.)

The years following the transfer of the SFOF to the DSN were marked by a great
increase in the number and complexity of the flight missions that the DSN was called
upon to support. Many of these missions were managed by institutions other than JPL.
The Lunar Orbiter and Pioneer missions (and, to a lesser extent, the Surveyor and Apollo
missions) were typical of this period.

At the same time, the DSN itself was expanding. The second 26-m subnet and the LIS-
band conversion were completed in 1965; the first 64-m antenna became operational
in 1966; and the multi-mission concept was introduced in 1967.

Toward the end of the 1960s, improvements in communication control, data distribu-
tion and display, and computer-based switching of intersite circuits were made in the
SFOF. Intersite communications were carried on high-speed and wideband data lines at
speeds of 4,800 bps and 50 kbps, respectively, by the DSN Ground Communications
Facility, which was also part of the SFOF. Paralleling the increased data-handling capa-
bility of the Mark I11 DSN, the data-handling capability of the SFOF data processors
was also expanded by the replacement of the old IBM computers with two IBM 360-
75 and two Univac 1108 machines with appropriate interface switches for flexibility and
redundancy.

While all of this development was taking place in the DSN, GCF, and SFOF, all three
facilities were simultaneously conducting an intense program of lunar and planetary mis-
sions operations. The success of all these missions attests to, amongst other things, the
adequacy of the organization under which they were performed. With the exception of
several personnel changes, notably the resignation of Rechtin and the appointment of
Bayley to replace him as ALDITDA, the organization remained as shown in Figure 9-6.
Nevertheless, despite the apparent suitability of the existing organization to the task it
was being called upon to perform, all was not well at the higher levels of management.
As William Corliss explains,

The deep space missions of the late 1960s and early 1970s brought with them
substantial increases in the SFOF's data[-]processing - load. Much of this data
processing was strictly scientific and unrelated to DSN operations. Yet, the
tracking and data acquisition function was obligated to provide for this com-
puter time without the authority to review requirements. The flight projects
were, in essence, requesting and getting large blocks of computer time and were
neither financially nor managerially accountable for them. It was a bad man-
agerial situation. NASA Headquarters recognized the situation and, in October
1971, Gerald Truszynski (OTDA) and John Naugle (OSS) reviewed the prob-
lem and decided to transfer the SFOF functions from OTDA back to OSSA.
In this way the responsibility for review and validation of requirements and
the associated costs of scientific data processing would be borne by the flight
projects themselves.

The change at Headquarters was immediately reflected at JPL, where responsibility for
the SFOF was moved from the TDA Office to a newly created Office of Computing
and Information Systems (OCIS). This new organization reported not to OTDA, but
to OSS. The transfer became effective in July 1972, and the organization of the tech-
nical divisions at JPL which resulted from or gave rise to (depending on the point of
view) the separation of the SFOF from the DSN is shown in Figure 9-7.

The Ofice of Computing

- - and Information Systems, with Man Finerman as manager,
directed the Data Systems Division, which, under the management of Glen E. airm more,
provided technical support and operational direction for the SFOF and the GCF.

The Telecommunications Division and Mission Analysis Division provided engineering

development and navigation-related analysis for the independent DSN.

The TDA Office, under the ALD (Bayley), assigned a manager to each of the three
major areas of effort in the Network for which it was responsible. Both implementation
of new hardware and software in the Network and Network Operations functions were
carried out by Kurt Hefunan, the DSN Engineering and Operations manager. Interfaces
between the DSN and all flight projects for the negotiation of requirements and com-
mitment of DSN resources were provided by Mission Support Manager Nicholas A.
Renzetti and his staff of five DSN managers. The integration of the various elements of
I \

JET PROPULSION LABORATORY

DIRECTOR
W. H. PlCKERlNG
--- I
I -----
UQ
o ~ ASSISTANT LABORATORY DIRECTOR ASSISTANT LABORATORY DIRECTOR ~
SYSTEMS FOR TECHNICAL OMSIONS ~
~ ~ ~
FOR TRACKING AND DATA
~ ~ G
A.FINERMAN J. N
. JAMES, MANAGER ACQUISITION
MANAGER - - - -w..5 z'!A%F"' - - -
I DEPUTY ASSISTANT LAB DIRECTOR
W.IC VICTOR
4ul 402

ENGINEERINGAND OPERAllONS DSN RESIDENTS

MANAGER, IC HEFTMAN -- W' E' LARKIN'AuSTRAUA
M. S. GLENN, SPAIN
SYSTEMSAND PROJECT ENGINEERS R.J. FAHNESTOCK, S.AFRICA

I I
OPERATIONS F A C I W 420 430 440
MISSION SUPPORT DSN SYSTEMS TDA PROGRAM CONTROL
N.& RENLmJ TDs_MANAGJRR
. w. mArcsw W. P. SPAULDING
DSN MANAGERS
P. S. GOOOWIN, HEUOS
R. P. LAESER. MARINER MARS 1971
K W. UNNES, MARINER 1969
MARINERVENUSI
MERCURY
D. J. MUDGWAY, WKING
A J. SIEGMEM, PIONEER
\ 1

Figure 9-7. Organization of the technical divisions after the transfer of the SFOF to OCIS,
1972.

the DSN into coherent unified systems was led by John W. Thatcher. The TDA Office
also maintained a representative at each overseas site to deal directly with local govern-
ment officials on matters pertinent to the operation of the tracking stations in their
countries. Program Control (budgetary matters) were handled by Wallace l? Spaulding.

Although the separation of the SFOF from the DSN may have appeared to be merely
a "paper exercise," it was not accomplished without considerable disruption to the care-
l l l y crafted interface agreements already in place between the DSN and the Pioneer
and Viking flight Schedules, interface agreements, and capabilities had been
negotiated with various elements of the flight projects and had been formally docu-
mented and approved, in accordance with current practices. These schedules, agreements,
and capabilities, of course, included the SFOF as well as the DSIF. When the separa-
tion took place, new interfaces between the DSN and the flight project, and the DSN
and OCIS, had to be developed and documented.
/ \
I
DSN I VIKING
FLIGHT
I PROJECT
I
I

I
I - - - - - - - - - -
I

I CENTRAL
COMPUTERS
I
I
I NON-OPERATION-
I DRNEN DATA
PROCESSING
I
\ J

Figure 9-8. Functional interfaces between the DSN and the SFOF for Viking, 1972.

A typical example of the functional interfaces between the DSN and the SFOF h e r
the separation of responsibility is shown in Figure 9-8.

The V i g Mission Control Center (VMCC), which included the SFOF central comput-
ing system, the mission support areas, and the V i g mission simulation system, was the
joint responsibility of the OCIS and the Viking Mission Operations System. The DSN was
responsible for the deep space stations, which included the 6 4 m and 26-m subnets, and
transport of data to and from the VMCC via the high-speed and wide-band data lines of
the GCF. Control and monitoring of Network performance and validation of the data
streams flowing between the VMCC and the deep space stations was to be accomplished
by a separate data-processing capability that would be independent of the mission-related
computers in the SFOE These functions would be accommodated in a new Network
Operations Control Center (NOCC), which was being designed at the time (1972).
As the SFOF came into use for multi-mission support in the early 1970s, common ter-
minology was changed to better reflect the way in which the facility was actually being
used. The building, JPL Building 230, and its support facilities retained the title SFOF.
The SFOF was eventually designated as a Historical Landmark by the U.S. Department
of the Interior in 1986. The SFOF housed two facilities that were controlled by the
DSN, namely the GCF (in the basement) and the NOCC (on the first floor). The
remainder of the building accommodated the flight controllers and spacecraft analysts
for the various flight projects and the data-processing equipment required to support
their missions. That ensemble of accommodations and equipment was called the Mission
Control and Computing Center (MCCC). It was occupied in turn by one or more
Flight Projects for the duration of each mission, after which it was vacated and recon-
figured for the next flight project. In some cases, non-JPL flight projects elected to
conduct flight operations from Mission Control Centers at their own institutions rather
than from the SFOF at JPL.

The first flight project to use the MCCC in this way was Pioneer, managed by the
Arnes Research Center (ARC) in Moffett Field. Similar arrangements for remote Mission
Control Centers came into greater use in later years, as the DSN began supporting
greater numbers of non-NASA spacecraft.

The Tracking and Data Acquisition Office at JPL (Office 400 in the JPL hierarchy) was
renamed Telecommunications and Data Acquisition Office and reached maturity under
the direction of Bill Bayley over the span of his tenure as ALD between 1967-80. From
a relatively small organization of fourteen in 1972 (Figure 9-7), the TDA Office had
grown to an organization of seventy-four people, as shown in Figure 9-9, by 1978. Two
offices--one for technology development, the other for long-range planning-had been
added. Because of the tremendous growth in DSN flight operations and the expansion
of the Network needed to meet the increasing number of flight missions, it was neces-
sary to split the former Engineering and Operations Section into separate offices. One
of these new offices (430) was managed by Renzetti and focused on developing the engi-
neering systems of the DSN. Interfacing with the flight projects and scheduling antenna
time, configuration control, and all other aspects of flight mission support except the
actual hands-on operations and maintenance became the responsibility of the DSN Mission
Support Office (440), under Spaulding.

In 1980, W H. Bayley retired and the extant JPL director, Bruce C. Murray selected
Peter T. Lyman to replace Bayley as ALDJTDA. Lyman retained that position until
r 1
400
TELECOMMUNICATIONSAN0 DATA ACQUISITION
W. H. BAYLEY
Asst. Lab. Dir.

OSN RESIDENTS
Spain, Australia

SPECIAL PROJECTS
P
M. Easterling
TDA PROGRAM
C. Johnson

I
430 410 420 440
TDA ENGINEERING TDA PLANNING TDATECHNOLOGY TDA MISSION SUPPORT
N. Renzetti E. Posner T. Thornton W. Spaulding

\ J

Figure 9-9. TDA Office 400 organization at JPL, 1978.

1987. Bruce Murray was succeeded as JPL Director by Lew M e n (1982), and the TDA
Office continued to expand to meet the demands placed upon it by an ever-increasing
number of NASA and non-NASA deep space and Earth-orbiter missions, as well as
numerous nonflight, scientific programs.

In Figure 9-10, a snapshot of the TDA organization, near the end of the Lyman years
in 1986, reveals the extent of the remarkable growth of the organization in that period.

Most noticeable in this organization was the addition of the TDA Science Office (450),
created by Lyman in 1983 for the reasons described in the previous chapter. Renzetti
managed that new office, which included a Geod~namicsprogram, the SET1 program,
the Goldstone Solar System Radar program, and several other special research projects.

Paul T. Westmoreland was appointed manager of the TDA Engineering Office (430),
which was by then responsible for interagency arraying, compatibility and contingency
planning, and implementation of new engineering capability into the Network and GCF,
in addition to its ongoing task of DSN System Engineering. The TDA Planning and
Technology Development ofEices appeared under differenr managers, and they, too, had
< >

JET PROPULSION LABORATORY

L ALLEN, DIRECTOR
R. J. PARKS, DEPUTY

400
TELECOMMUNlCAnONS AND DATA ACQUISITION
P. T. LYMAN
Asst. Lab. Oir.
-

TOA PROGRAM
C. W. Johnson

I I I I
430 410 420 440
TDA ENGINEERING TDA PLANNING TDA TECHNOLOGY TDA MISSION SUPORT AND
P. T. Westmoreland J. W. Layland J. G. Smith DSN OPERATIONS
R. J. Amorose
I I
TECHNICAL STAFF 450 TECHNICAL STAFF
System Engineering, Burt TDA SCIENCE OFFICE Control Center Opns., Wyatt
GCF Engineering, Evans N. A. RenzetIi Opns. and Eng. Support, Winn
Implementation,Walkice
Compatibility, Bryan Geodynamics, Renzetti DSN MANAGERS
Agency Arraying, Brown GPS Systems, Melbourne Galileo, Mudgway
Methodology Eng., Garcia TDA Observatory, Levy Mars Observer, Tmxler
SETI, Klein Magellan,Berman
GSSR, Renzetti Ulysses, Enari
ICE, Fanelli
Voyager, Cox
Radio Astronomy, Fahenstock
GSSR, SETI, Berman
GOES, Johnson
Special Projects, Goodwin

Figure 9-10. TDA Office 400 organization at JPL, 1980-87.

expanded in scope. Under the management of Raymond J. Arnorose, Office 440 includ-
ed management responsibility for the large Maintenance and Operations (M&O) Contract,
by which an industrial aerospace contractor provided engineering and operations serv-
ices to the DSN for the Goldstone Complex, the NOCC, and related activities in the
Pasadena area. Three DSN engineers managed the M&O contract, the Network Control
Center at JPL, and the Complex Control Center at Goldstone. Eight DSN managers
attended to the interests of over twenty flight projects that were active in the DSN at
that time. Finally, a program manager, chief engineer, and chief technologist provided
direct support for the ALDITDA in establishing budgetary, policy, and long-term devel-
opment goals.
At this point, a brief digression is in order to explain the organizations by which the
three Deep Space Communications Complexes (DSCCs) that composed the DSN were
related to the TDA Office.

From the outset, it was never intended that the various stations in the Network would
be operated by JPL personnel. In fact, the international agreements that permitted NASA
to establish tracking stations on foreign soil stipulated that the facilities would be oper-
ated by foreign nationals of the host countries. To this end, comprehensive facilities for
technical training in DSN-related technologies were set up at JPL and at Goldstone. At
these facilities, visiting foreign national engineers and U.S. national engineers received
theoretical and hands-on instruction and training in the hardware and s o b a r e which
they would subsequently be required to operate and maintain at their home sites. In the
early life of the DSN (DSIF), when new stations were completed, a few JPL technical
personnel remained at the site until the site could sustain continued operations with its
own personnel. Later, the sites sent key personnel to the U.S. for training so that upon
their return to their home bases, they could implement the new antennas or new capa-
bilities themselves. There was a constant flow of technical personnel between the sites
and Goldstone or JPL for this kind of information exchange and dissemination.

In the United States, the staff to operate the tracking stations and DSN-related services
at Goldstone and the Network Control Center and Ground Communications Facility at
JPL were provided by a technical services contract to NASAIJPL. The Maintenance and
Operations (M&O) Contract was managed by Office 440, as discussed above, and through
the years passed through the hands of various major U.S. aerospace companies-Bendix
Field Engineering Corporation in 1960, Philco-Ford in 1970, and Bendix Field Engineering
Corporation in 1978 and 1988. Bendix Field Engineering Corporation became Allied
Signal Aerospace in 1992 without interruption to the contract.

Under the direction of JPL personnel, the M&O contractor was responsible to the DSN
for the supporting services listed below (in greatly abbreviated form):

1. At JPL:
a. Operate and maintain the Network Operations Control Center (NOCC), the
Ground Communications Facility (GCF), the Compatibility Test Facility (CTA
21), and (occasionally) the DSN Launch Support Test Facility (MIL 71) at
Cape Kennedy.
b. Manage operations support programs such as antenna scheduling, discrepancy
reporting, data control, radio frequency interference management, and Network
performance evaluation.
 c. Provide general support to all Network ficilities in the areas of logistics, tech-
nical training, documentation, engineering change and configuration control,
Network-level maintenance, and sustaining engineering.
d. Provide high-level technical and administrative support for specific tasks in the
TDA Office and Telecommunications Division.

2. At the Goldstone Deep Space Communications Complex (GDSCC):

a. Operate and maintain all of the tracking stations, their intrasite communica-
tions, and the research and development facility at the DSS 13 Venus site.
b. Maintain the established Complex Maintenance Facility, Network Standards
Laboratory, and other services.
c. Maintain all of the buildings, plant equipment, roads, lighting, power, and air-
conditioning, as well as fire, food, and security services.

The M&O Contract for Goldstone assigned complete responsibility for running the com-
plex to the contractor, and over the many years of its operation, the GDSCC technical
staff were always ready to respond to a call for extra effort, whether it was for an inflight
mission, an engineering research and development task, or a DSN science program.

The staff and infrastructure needed to provide engineering and operations support for
the Canberra Deep Space Communications Complex (CDSCC) were provided by
Australian industry under a contract with the Australian government. However, all costs
associated with operation of the CDSCC were borne by NASA under the terms of the
intergovernmental agreement. The Complex Director remained an employee of the gov-
ernmental department (agency) charged with administering the contract on behalf of
the Australian government and was responsible to that department head for routine man-
agement of the Complex, its staff, and its facilities. On behalf of NASA, the ALDITDA
at JPL was responsible only for providing the operational facilities and directing their
use in support of the NASA programs. The ALDITDA retained a representative resi-
dent in Canberra to facilitate the interchange of technical and administrative information
between the CDSCC and JPL. He played no part, however, in the direction or opera-
tion of the Complex, and the position was abolished in 1985.

International agreements between NASA and the Spanish government for the manage-
ment, engineering, and operations support for the Madrid Deep Space Communications
Complex (MDSCC) were essentially similar to those between NASA and the Australian
government for management and operation of CDSCC. There, however, both the direc-
tor and his staffwere employees of a government agency Institute Nacional Tecnica Aerospacial
(INTA), until 1992, when INTA, retaining the NASA contract, delegated the maintenance
and operation to its wholly owned company, Ingenieria y Servicios Aeroespaciales) (INSA).

In both situations, this seemingly awkward organizational arrangement worked well, in

general terms. Some initial administrative difficulties were cleared up as time went on, but
the technical staffs at both complexes were always prepared (and on many occasions were
called upon) to respond unreservedly to the exigencies of deep space mission operations.

When the position of Deputy to the Director of JPL became vacant at the retirement
of Robert Parks in 1987, Lyman was appointed to fill the position, and Larry N. Dumas
became ALDITDA. Apart from the inevitable personnel changes and some structural
changes within the five internal offices of the TDA (400) organization, little changed
in an organizational sense in the DSN duriig the Dumas years, 1987 to 1992. More
changes in upper management, however, were on the way.

In 1990, Edward C. Stone replaced Allen as director of JPL, and he was joined in 1992
by Dumas as deputy. The then vacant position of ALDITDA was filled by Norman R.
Haynes, who brought Westmoreland from the TDA Engineering Office to become his
deputy. The TDA organization had remained essentially unchanged since 1986, and,
except for changes in personnel, is well represented by Figure 9-1 1. This was the final
form of the TDA Office. There was no hint of the very dramatic changes that would
overtake the entire TDA (400) organization in the next several years.

When Haynes took over the TDA organization in 1992, it was one of several offices
headed by an Assistant Laboratory Director (ALD). These offices, under the JPL direc-
tor, composed the basic structure of the entire JPL organization. When he left in 1996,
the JPL structure had been changed to one based upon an ensemble of directorates,
each headed by its own director. The directorates were not simply renamed offices, but
were restructured to reflect the "policy for change" brought to JPL by the new director,
E. C. Stone.

To fully appreciate the latest changes in the TDA organization, it is necessary to under-
stand the organizations at NASA Headquarters and in the MCCC at JPL at that time.
Both influenced the shape of the future TDA organization.
Figure 9-11. National Aeronautics and Space Administration, 1991.
By 1991, the organization at NASA Headquarters under Administrator Richard H. Truly
included an Office of Space Science and Applications (OSSA, Code S) and an Office
of Space Operations (OSO, Code 0 ) as shown in Figure 9-1 1.

In the context of program direction for users and suppliers of deep space tracking and
data-acquisition services, OSSA directed the flight projects (users) and associated gound
data-processing facilities (MCCC) while OSO directed the TDA Office and its associ-
ated tracking and data-acquisition facilities (DSN).

For clarification, during the period of DSN history reviewed in this book, the NASA
office that began as the Office of Tracking and Data Acquisition (OTDA) in 1961
became the Office of Space Tracking and Data Systems (OSTDS) in 1983, the Office
of Space Operations (OSO) in 1988, and the Office of Space Communications (OSC)
in 1992. During almost that entire period, it was identified as "Code 0"and was respon-
sible for programmatic direction of the DSN through the TDA Office at JPL.

The selection in July 1989 of Charles T Force as Associate Administrator of the Off~ce
of Space Operations (OSO) brought immediate changes to the DSN relationship with
NASA Headquarters. His highly relevant technical background, considerable experience
in tracking-network operations, and strong motivation effectively streamlined the
unwieldy, outmoded, and by that time largely irrelevant system of documentation that
controlled agreements for providing and expending NASA resources on the DSN anten-
nas. The time span of his position in that office, 1989-96, placed him squarely in the
Galileo Era, where he was most effective, at the Headquarters level, during the crises
that arose in that period of DSN history.

The funding approved during Force's tenure enabled the DSN to upgrade the 64-m
antennas to 70-m, add more 34-m antennas to the Network, support recovery of the
Galileo mission from almost certain failure, introduce multiple antenna arraying on a
routine operational basis, move toward K-band operations, and consolidate and upgrade
the Signal Processing Centers. Together with the improvements in formal Headquarters
documentation, these made an impressive record of achievement that set the stage for
the major reorganization that swept through all the NASA Networks in the late 1990s.

Forceful in attitude and spare with words, "Charlie" Force came straight to the point
in a discussion, an aspect of his managerial style that appealed to the JPL managers with
whom he interacted on DSN-related matters. His Eull responsibilities included several
other NASA Networks in addition to the DSN, and he was seen as fair and impartial
Figure 9-1 2. Multimission Operations Systems Office (440), 1993.

when at times his decisions regarding the allocation of limited resources were at odds
with prevailing opinions at JPL.

Force joined NASA in 1965 as Director of the Guam Tracking Station and, except for
returning to industry for a couple of years in the early 1980s, held increasingly respon-
sible positions throughout his career at NASA. A native of Shoals, Indiana, and resident
of the state of Maryland, he held a B.S. degree in aeronautical engineering from Purdue
University. Force left NASA in 1996 to pursue commercial business interests.

By 1993, the organization at JPL formerly referred to as the MCCC had embraced the
concept of multi-mission operations and had evolved into the Multimission Operations
Systems Ofice (440) with the structure shown in Figure 9-12.

Because of its close association with the evolution of the DSN, the path by which the
MCCC evolved is of historical interest and is neatly described by Alazard9 as follows:

From the mid-1960s through 1970, the ground data system was comprised of
three IBM 7094 computers to support each of the then current (flight) proj-
ects. Between 1970 and 1977, the architecture and organization of computing
changed and moved toward a multimission orientation. The Mission Control
 and Computing Center (MCCC) was created and data processing was done
on IBM 360-75's. The flight projects shared this single system in a multipro-
cessing mode. Between 1973 and 1981, minicomputers were phased into the
MCCC for realtime processing. Mainframe computers were used for process-
ing (flight) project applications programs and data records[,] i.e.[,] non-realtime
processing. These computing capabilities were provided to the flight projects
by the Fight Projects Support Office (FPSO). The realtime processors were usu-
ally dedicated to a given project while the non-realtime processors were shared.

In an effort to reduce the cost of data processing for flight operations in the
mid-1980s, FPSO initiated development of an entirely new facility, called the
Space Flight Operations Center (SFOC). The SFOC performed the functions
of the MCCC with the newer technology of distributed processing using micro-
computers and local area networks in a workstation environment. When it was
completed in the early 1990s, existing flight project support was gradually con-
verted from the MCCC to the new system, and new flight projects were adapted
to the multimission capabilities and services. In due course the MCCC was
phased out and the SFOC became the core of an even more advanced data pro-
cessing system called the Advanced Multi-Mission Operations System (AMMOS).

The name FPSO was changed to Multimission Operations Systems Office (MOSO) in
1992 to better reflect the multi-mission nature of the organization.
- MOSO became the
overall organization responsible for development, operations, and maintenance of the
flight project support capabilities. AMMOS provided the direct operations support func-
tions using the hardware and software of the Multimission Ground Data System (MGDS).

It was the task of MOSO to establish the set of multi-mission data processing capabil-
ities and operational services that supported the flight projects in accomplishing their
mission objectives. The AMMOS was the set of hardware and software tools by which
this task was accomplished. Defined in this way, the MOSO task closely paralleled the
DSN Operations and Mission Support Office (440) task when tracking and data-acqui-
sition services were substituted for data-processing services and the DSN was substituted
for AMMOS. The MGDS was the counterpart of the DSN.

THEJPL STRATEGIC
PLAN:1994

Shortly &er the Clinton Administration took office in January 1993, all federal agen-
cies, including NASA, began to feel the effect of a groundswell of change toward a smaller,
more efficient, and less costly bureaucracy. A new NASA Administrator, Daniel S. Goldin,
had been appointed early in 1992, and he was quick to respond to the presidential ini-
tiative for cost reduction in government agencies. In the course o f the changes that followed,
the Headquarters organization was signhcantly reduced in size; many functions were
deleted or combined; and a new, streamlined NASA organization emerged. In an envi-
ronment o f economic constraints and redefined national goals, the NASA budget declined
in Fiscal Year 1995 for the first time since the end o f the Apollo program in the late
1960s. N e w approaches and revised priorities for the U.S. space program were developed
to ensure that space science and technology would continue to advance for the benefit
o f the nation. All o f these factors required NASA and, consequently, JPL to rethink their
ways o f doing business, adopt new strategies, and create a shared understanding o f how
to meet customer needs in the new environment.

This state o f affairs formed the driver for a Strategic Plan developed for JPL in 1994
by i t s new director, Edward C. Stone.lo I t defined the set o f NASA programs in which
\
I
NASA SPACE COMMUNICATIONS AND OPERATIONS PROGRAM
Provide affordable, world-class support to large, complex missions: Ulysses, Galileo,
Cassini, and EOS.
* Advance the state of the art in low-cost support of small and moderate missions.
Develop the capability to provide Ka-band services for future low-cost missions.
Upgrade mission operations concepts and support systems to enable simultaneous
operation of many missions at significantly reduced unit cost.
* Develop concepts for promising new services such as optical communications and
non-DSN tracking terminals.
Revolutionize ground system designs to reduce complexity and operating costs while
emphasizing new technology opportunities and controlled risk.
* Support many more missions in the next five years while reducing DSN tracking oper-
ations costs by 50 percent.
Advance the state of the art in deep space and near-Earth navigation and commu-
nications to meet expected future mission requirements.
Seek new concepts for low-cost support of small and moderate missions.
Increase the use and scientific impact of the DSN in radio science, astronomy, and
planetary radar studies.
* Lead in developing international science agreements and cross-support systems for
an affordable global space exploration program.
< /

Figure 9-13. NASA Space Communications and Operations Program.

JPL would engage to support "NASA's strategic enterprises and other national needs."
One of these programs was NASA Space Communications and Operations, and because
of its importance to the future long-term development of the DSN, it is reproduced in
Figure 9-13.

One of the overarching concepts embodied in the newly defined program was the com-
bining of space communications with space operations to create a single entity of reduced
complexity with lower operating costs. This was to be achieved without sacrificing per-
formance and was concomitant with a stimulating call to further advance the state of
the art in associated technologies.

BIRTHOF THE TMOD

The Telecommunications and Mission Operations Directorate (TMOD) was created in

1994 to support the NASA Space Communications and Operations program defined
in the Strategic Plan described above. As such, it represented one element of the Laboratory's
overall response to the NASA Strategic Plan.

The transition from the TDA organization to the TMOD organization took place dur-
ing the latter half of the Haynes years (1994); although there was a considerable disruption
of ongoing activities initially, by 1996, when Pad T. Wesunoreland assumed the posi-
tion of Director, the new arrangement had settled down and assumed the form shown
in Figure 9-14.

The rearrangement and consolidation of former TDA functions, which is apparent in

Figure 9-14, reflected a new and cost- effective approach to complex business manage-
ment systems made popular at that time (1992-93) by Dr. Michael Hammer, co-author
of the book Reengineering the Corporation." With institutional encouragement, JPL man-
agement personnel at all levels attended Hammer's courses and returned to JPL to apply
the new methodology to the workplace. Early in 1994, the TDA senior staff spent two
months developing a process map for the TDA organization. Two of these processes
were selected for actual reengineering, and process owners were named. The selected
processes and owners were Services Fulfillment (Raymond J. Arnorose) and Asset Creation
(Edward L. McKinley). Leslie J. Deutsch was assigned to lead the Reengineering Team
(RET) for Services Fulfillment-the process deemed most suitable for .the first engi-
neering effort. Although the RET was formed initially to reengineer only the TDA
organization, the scope of its work was expanded to include the data capture and nav-
igation functions of MOSO when the two organizations were merged to form TMOD.
f \
900
TELECOMMUNICATIONS AND MISSION
OPERATIONS DIRECTORATE

P. T. Westmoreland, Director
-
R. C. Coftin, Program Manager

L J. Oeutsch, Chief Systems Architect

900 - 910
BUSiNESSOPERAnONS
R. k? Methison. Chief Engineer OFFICE
R. G. Polansky,AssistantChief Engineer
6.F. Squibb, Chief Mis.ion OperationsArchitect C.T. Stevens, Manager
R. J.Wallace, NASA tiason -

920 940 970 950

TMO PLANS AND DSN DATA SERVICES MULTlMlSSlON GROUND GAULEO PROJECT
COMMITMENTSOFFICE OFFICE SYSTEMS OFFICE
W. J. O'Neil, Mgr. 9M1
R. a. Miller, Mor. E. L McKinley, Mgr. T 0. Unick, Mgr.
0.6. Griflith. Deputy Mgr. M. E. Wit Deputy Mgr. S. M. Huifman,DepiyMgr. U.S. SYLBl PROJECT
J. k WacWey, Deputy Mgr. ------ J. G. Smith, Mgr.

TMO LEVEL 0 SYSTEM . OSN (LVL 1) SYSTEM ENG'G AMMOS (LVL 1) SYSTEM ENG'G
980
ULYSSES PROJECT

.
ENGINEERING
STRATEGY DML
OSN DEVELOPMENT . N M O S DEVELOPMENT
MUL~MISS~ON GDS OEVL
- E. a. Massey, Mgr. 990
OSN OPERATIONS PROJECTGOS DEYEL
DSN TECHNOLOGY
DSN SCIENCE
. AMMOS OPERATIONS
-VOYAGER PROJECT
6. k? Textor, Mgr.

M U R E MISSIONS
PLANNlNOCOMMlTMEKl
LEGEND:
RESOURCEALLOCATION
MMos=MMncedMula~lonOpe~~syWm
DSN = D&D Space Neturon
TMO =TdemmmUnWons and M l $ s i M O ~ S n s
Ccca =Mminlmtiv~OrgmhPn
< J

Figure 9-14. Telecommunications and Mission Operations Directorate, 1996.

The RET issued an interim report in September 1994 containing some key items for
the future structure of TMOD. Among these were the three Services Fulfillment Processes
recommended for reengineering-namely, Data Capture, Activity Planning, and Resources
Scheduling, all defined later in this chapter. There was also a proposal that the DSN be
changed from an engineering-driven organization, as it had stood in the past, to an oper-
ations-driven organization for the future.

The arrangement shown in Figure 9-14 was the first step in the overall reengineering
process. Essentially, the former TDA organization was condensed into two offices: one
for planning, committing, and allocating DSN resources; the other for DSN opera-
tions and system engineering. DSN science and technology were incorporated in the
former, DSN development in the latter. In addition to these two offices, the Multimission
Ground Systems Ofice, the project offices of the four inflight missions (Galileo, Space
VLBI, Ulysses, and Voyager) and a new business office were added to create TMOD.
There could be little doubt that the TMOD was now operations-driven rather than
engineering-driven.
While these organizational changes brought the Deep Space Network and the
Multimission Ground Data System together under one directorate, they did not effect,
alone, the substantial reduction in gound operations costs that was the real objective
of the reengineering initiative. DSN operations, now redefined in reengineering terms
as the Data Capture Process, still involved a plethora of facilities, starting with the
DSCCs, passing through the GCF to the NOCC and to MOSO before eventually
delivering the data to the customer. Reduced cost would come from reduction in the
size, complexity, redundancy, and performance margin of the infrastructure needed to
operate these separate facilities. This, in turn, required a radical change in the way in
which future operations were performed, and in the way in which the individual ele-
ments of the Data Capture Process would be operated. It was the task of the RET to
show how this could be done and to provide a road map for making the transition to
the new Data Capture Process.
REINVENTING THE FUTURE (1994 INTO THE NEW MILLENNIUM)

REENGINEERINGTMOD

The RET completed its redesign of key TMOD subprocesses and submitted its final
report to DSN Mission Operations Manager Ray Amorose in March 1995.12 It suc-
ceeded in not only meeting but exceeding its initial cost and efficiency goals of reducing
the DSN Operations budget by $9M and doubling the available tracking hours by Fiscal
Year 1999. The new designs would cost about $16M to implement and were predict-
ed to result in a cumulative savings to TMOD Services Fulfillment of approximately
$35M for the period FY 1996 through FY 2000, as shown in Figure 9-15.

f >
40.0

35.0

-3 30.0

25.0
9
-
-u
2
20.0

15.0
€e
10.0

5.0

.
0.0
P/95 FY96 FY97 FY98 FY99 WOO
Year

H Cumulative lmpl. Cost ($M) Cumulative Savings ($M)

\ J

Figure 9-15. Predicted cost and savings for reengineering TMOD Services Fulfillment
Processes. The three Services Fulfillment Processes include the following:
1. Data Capture-the process that provided the services of telecommunications, target tracking,
and data processing to customers.
2. Activity Planning-the process that generated and consolidated the support data that was need-
ed by the Data Capture Process to perform its functions.
3. Scheduling-the process that allocated the resources of the DSN and MGDS for use in the Data
Capture Process.
Because the savings to DSN Operations were estimated to be low, the RET elected not
to reengineer the Scheduling Process at that time. It was, therefore, only the Data Capture
and Activity Planning Processes that, in reengineered form, contributed to the substan-
tial projected cost savings noted above.

By basing its design on standard (rather than customized) services, the new Data Capture
Process simplified mission interfaces, increased operational efficiency, and allowed full
use of automation to provide these services. At the same time, a "Connection Operator"
was made available to provide specially tailored services essential to the customer's needs.
The parts of the NOCC, GCF, and MGDS involved in real-time operations were brought
together in a single work area called Central Operations. New, more easily maintained
equipment allowed them to be located in a considerably smaller work space with reduced
facilities costs.

On the other hand, supporting data such as radio metric predicts, antenna pointing
instructions, and schedules for Deep Space Network operations were considered an Activity
Planning function and had, in past practice, been generated at the NOCC considerably
in advance of the needed time. The inputs were, of necessity, immature, and they invari-
ably led to the generation of many contingency or "just in case" versions, all of which
created extra work. In the new, more rapid and efficient "just in time" system, predicts
were generated at the site as needed and were based on the most up-to-date require-
ments and conditions. The resulting system was much more responsive to last-minute
- .
changes, more timely, more accurate, and less costly to operate.

Data Capture Process

The new Data Capture Process described above was critically dependent for its imple-
mentation on the availability of new hardware and software tools with the requisite
capabilities. To a large degree, these were to be provided out of similar studies carried
out by the Reengineering Team for the Asset Creation Process, which ran a parallel
course to that of Customer Services Fulfdment.

In this context, it was assumed that the operations load on the 26-ml9-m subnetwork
would be relieved by the introduction of a subnetwork of small unattended tracking sta-
tions to be known as Low-Earth Orbit (LEO) terminals. It was further assumed that
updated electronics would allow a large degree of automation for future operation of
the 26-m and 9-m antennas and would eliminate the existing need for roving operators
(rovers) to configure antenna and microwave equipment in front-end areas (FEAs). At
the DSCCs, the already-planned Advanced DSN Monitor and Control (ADMC) sub-
system would provide the high degree of automatic features required for the Connection
Operations concept. A new "Reliable Data Delivery System" and improved Wide Area
Network were already at the advanced design stage and were assumed to be available to
meet the proposed transition plan schedule. Findy, it was assumed that technology
already developed in MOSO for data processing and management would be extended
to incorporate the Data Systems Operations Team (DSOT) as a vital element of the
new Data Capture system.

A road map for making the transition to the new Data Capture Process is shown in
Figure 9-16.

The dependency on key new implementation described above is shown in the top half
of the figure, while the key changes in operations processes are shown in the lower half.
The plan attempted to minimize disruption to ongoing operations whiie it progressed
rapidly toward its final realization by 2000. Near-term savings accrued from the early
transition to Connection Operation in 1995, collocation of TMOD operations, and
reduction in operations documentation.

Activity Plan Process

The two principal fknctions performed via the Activity Plan process were as follows:

1. Identifying the services to be provided. This was called Service Plan Generation and
was based on a schedule showing the TMOD resources required and the time at which
they were to be available, as well as a project sequence of events (PSOE) that identified
what the customer would be doing during the time allocated in the schedule.

2. Generating the information, internal to the Network, that was required for operat-
ing the equipment to provide the needed services. This was called Predicts Generation
and was required principally for generation of radio metric or telemetry data. It was
based on spacecraft state and ephemeris and station-dependent data such as location,
horizon mask, and planetary ephemeris.

At the time that the new Activity Plan Process was being designed, a standard infrastruc-
ture for the packapg and manipulation of spacecraft navigation products called the "SPICE
system had been in use for some years at JPL. The SPICE data sets were called kernels
and contained the information necessary to assist users in planning and interpreting sci-
ence observations from spacecraft instruments. For example, the "S" data set contained
spacecraft ephemeris data, and the "P"data set contained planetary ephemerides.
In reengineering the Activity Plan Process, the RET extended this idea to create kernels
containing spacecraft and planetary ephemerides data, telemetry data, station information,
and time and Earth polar motion information needed to compute station locations. The
data bases of these kernels would be "shadowed" or automatically updated to the tracking
station's data management system and would therefore be available for use in generating
telemetry andlor metric predicts on demand. At the station, computation of predicts could
be initiated only minutes before the actual activity was scheduled to take place.

In addition to these innovations, the reengineered Activity Plan Process improved the
way in which the customer provided the PSOE by referencing it to a standard catalog
of prenegotiated services. The Activity Plan Process would then translate these service
requests into the operations procedures necessary to meet them, thereby creating the
actual Service Plan.

Although implementation of the new Activity Plan Process began in 1995, it depend-
ed on upgraded versions of the data management capability at the stations for fuU operation.
Delay in providing this new capability resulted in a longer term return on the imple-
mentation costs than was the case for the Data Capture Process.

At the time of this writing, the implementation plans for the Data Capture and Activity
Plan Processes were in progress along the lines discussed above. When fully completed in
1999, the reengineered TMOD Customer Services Fulfillment Process would embody the
very best principles of cost-effective, process-based management, rather than the facility-based
management and control principles that had been in place for the previous forty years.

After a very short term of one year in office, Wesunoreland retired in June 1997, and
Gael F. Squibb was appointed director of the Telecommunications and Mission Operations
Directorate (TMOD). Squibb was well qualified to guide TMOD into the future. He
had served as Manager of Data Services for NASA's Space Operations Management
Office since 1995 and had managed numerous scientific flight projects in thirty years
of previous service with JPL.

When Squibb took over TMOD, the reengineering effort was well under way and the
DSN and MGDS functions in mission operations had become recognized as contiguous
elements of the Customer Services Fulfillment Process. They were, from 1994, still sep-
arate offices (940 and 970) of the original TMOD organization. It was time to bring the
institutional organization more into line with the concepts of process-based management.
I \
900
TELECOMMUNICATIONS AND MISSION
OPERATIONS DIRECTORATE

G. F. Squibb, Director
R. C. Coffin, Program Manager
9W 910
P. T. Wesbnoreland,TransitionAdvisar BUSINESS OPERATIONS
R. P. MatMson, Chiel Engineer OFFICE
R. G. Polanshy, Assistant Chief Engineer
M. J. Rodngues,AEst Pmg. Mgr. far SOMOICSOC Operatmns C. T. Stevens, Manager
E. L McMnley, Mgr. SOMO Data Sewices
P

TMOD PLANSAND
-
TMOD ENGINEERING TMOD OPERATIONS TMOD TECHNOLOGY W I L E 0 PROJECT
COMMITMENTSOFFICE OFFICE OFFICE OFFICE
R. 8. Miller, Mgr. J. I.Staiman, Mgr. T. Dl. Unick, Mgr. L. J. Deukh, Mgr.
D. G. Griffith, Dep. J. C. Klase, Dep. J. AWacWey, Dep. C. D. Edwards, Dep. VWAGER PROJECT
L

. . .
P

ULYSSES PROJECT
TMO LEVEL ENG'G
. DSNIAMMOS DEVEL. DSNIAMMOS OPERATIONS FLTIGRNDTECHNOLOGY
DEVEL -
.
u
MOS SYSTEM ENG'G PROJ. GOS DEVEL. FUGHT MISSION OPS
DSN & AMMOSTECHNOLOGY
STRATEGY DEVEL. PROJECTMOS MGMT. DEVEL
. . SPACEVLBIPROJECT

.
DSN SCIENCE
FUTUREMISSIONS
PIANNING
CSOC MANAGEMENT
RESOURCEALLOCATION
TELECOM. CENTER OF
EXCELLENCE

PROJECTINTERFACV LTGWO:
COMMrrMENT
DSN = o e ~ s p a c ensaoh
TMO = T B I e r n m r n ~ n l ~and
~ uMISSIO~
~~ Opsrauons
Kal=AdminlmIvmsme OrgmIrati~n
(ClRl= @maor

Figure 9-17. Telecommunications and Mission Operations Directorate, 1997.

The restructured Telecommunications and Mission Operations Directorate, under the

direction of Gael Squibb in 1997, is shown in Figure 9-17.

The former DSN Data Services and Multirnission Ground Systems Offices were combined
to form the TMOD Operations Office, from which the Customer Services Fulfillment
Process, including the allocation of resources, would be managed effectively. To develop
the new system engineering functions for this process, a TMOD Engineering Office, which
included engineering elements of both the DSN and MGDS, was also created. It would
be here that the Asset Creation Process would reside. The enabling technology on which
the reengineering teams based their new process-oriented designs was to be provided by
the TMOD Technology Office for both the DSN and AMMOS. It was defined as a
"Center of Excellence" for telecomunications to focus attention on the particular tech-
nology in which the DSN held a unique position in the NASA sphere of influence. The
Plans and Commitments Office and flight project offices remained essentially the same.
There was another feature of the TMOD organization that, although not apparent on
the organization chart, affected the interaction between TMOD and its customers in a
significant way-the change in the role of the DSN manager. As mentioned earlier, it
was customary for the TDA Office to appoint a representative to work with each flight
project in negotiating and using the tracking and data acquisition services that were
required for its mission. The title of this position was "Tracking and Data System (TDS)
Manager for (Name of Customer)."

Customers included those that used the DSN as a scientific instrument as well as those
that used it as a tracking and data acquisition service. The TDS manager helped his
designated customers to understand relevant technical and operational aspects of the
DSN and acted as an authoritative single point of contact for the customers, dealing
with all aspects of the service being provided. But the service rendered extended only
to the interface between the DSN and the MGDS. It did not include the processing
and management of the data beyond that point, which was, in reengineering terms,
"facility-oriented." When TMOD was reengineered into a "process-oriented organiza-
tion, the former TDS manager positions, which were properly resident in the Plans and
Commitments Office, were expanded in concept and scope to include the entire Customer
Fulfillment Services Process. They became the TMOD version of the "empowered cus-
tomer service representatives" advocated by Hammer and Champy and were called
Telecommunications and Mission Services (TMS) Managers.

The extent to which a single person would be able to discharge this task in a mean-
in@ way at all levels throughout the now extremely complex Services Fulfillment process
remained to be seen. Hammer and Champy recognized this problem and observed, "To
perform this role-that is, to be able to answer the customer's questions and solve the
customer's problems-the case manager needs access to all the information systems that
the people actually performing the process use and the ability to contact those people
with questions and requests for further assistance when necessary." The TMS managers
of the future would need 111 access to tools and resources such as these to properly
carry out their important functions in the new process-oriented TMOD.

"Bridging the Space Frontier"

A few months after taking office, the new director of TMOD issued a statement that
presented his goals and vision for the TMOD of the future. It included a plan for their
realization and was called "Bridging the Space Frontier."13 The plan traced the origin of
the powerful forces throughout NASA that had influenced the organizational structure
of TMOD and would determine its course for the future. The environment in which
TMOD must plan for the future, said Squibb, "is characterized by three fundamental
trends: increasing demand for telecommunication and mission services and technology
advancements, highly constrained resources, and organizational change."

Squibb based his plan on NASA's response to the National Space Policy, issued &om
the White House in 1996.14The National Space Policy of 1996 directed NASA to focus
its research and development efforts in the four principal areas of Space Science, Earth
Observation, Human Space Flight, and Space Technologies and Applications. Furthermore,
it instructed NASA to "seek to privatize or commercialize its space communications
operations no later than 2005" and to "examine with DoD, NOAA[,] and other appro-
priate federal agencies, the feasibility of consolidating ground facilities and data
communications systems that cannot otherwise be provided by the private sector."

NASA began to address this latter policy by consolidating the management of all its
space operations under a new Space Operations Management Office (SOMO), located
at the Johnson Space Center in Houston. As the service provider, SOMO was given
responsibility for implementing NASA's space operations and managing the associated
space operations work process. The NASA Centers, in support of their respective imple-
mentation plans for space operations, were responsible for the execution of the space
operations work process.15

In his plan, Squibb discussed TMOD's relationship to SOMO. "Through this cooper-
ative arrangement with SOMO, TMOD (would) oversee operation of the Advanced
Multimission Operations System (AMMOS) and the Deep Space Network (DSN). Under
SOMO's leadership, efforts were underway to consolidate and streamline major support
contract services." Transition to a Consolidated Space Operations Contract (CSOC) with
a single, ten-year, cost-plus-award-fee contract was expected to begin in Fiscal Year 1998.
In the meantime, TMOD worked to understand its future, dramatically different role
as a partner with a CSOC contractor, and possibly with other United States or foreign
agencies; it also tried to identify any functional activities that would be suitable for com-
mercialization in the immediate future.

JPCs management responsibilities for the DSN and AMMOS were finally combined
under Squibb's direction in 1996, and TMOD began to operate as a single process-
oriented service to its customers. Joe Staunan, manager of the TMOD Engineering
Office, saw the new way of operating as a challenge to "create one cultute-no more
separate DSNIAMMOS cultures." One step remained, however, in the completion of
the DSNIAMMOS unification, and that had to do with the still separate sources of
programmatic direction at NASA Headquarters.
I \

NASA
ADMINISTRATOR
Daniel S. Goldin

- -- -
Codes H, P, Z Codes I,L, F Codes J, G, C
L
I
I ,
I
Codes K, X, U
- Code 0
SPACE
I

COMMUNICATIONS

Code Y Code M Code R

ENTERPRISE - ENTERPRISE - ENTERPRISE
MISSIONTO SPACE FLIGHT AERONAUTICS SPACE SCIENCE
PLANET EARTH

GODDARO JOHNSON AMES

SPACE FLIGHT - - - RESEARCH
CENTER SPACE CENTER CENTER
'JKIs a comacW-opera$d
KENNEDY LANGLW fan~lty
- - RESEARCH
SPACE CENTER
CENTER

SPACE FLIGHT

STENNIS DRYDEN FLIGHT

SPACE CENTER RESEARCH
CENTER
\ /

Figure 9-18. Organization of the National Aeronautics and Space Administration, 1996. In the
new organization, the responsibilities of Code 0 (the former Office of Space Communications) were
distributed among the Centers, the Office of Space Flight, and the Space Operations Management
Office (SOMO). It was SOMO that provided programmatic direction for the DSN-related elements
of TMOD. Code S, the home of the Space Science Enterprise, continued to provide the source of
funding for the MGDS-related elements of TMOD.

The directives of the 1996 National Space Policy, and its focus on NASA research and
development efforts, were eventually manifested in the NASA Strategic Plan.l6This called
for a restructuring of the NASA organization based on four grand "Strategic Enterprises."
The organizational structure took the form shown in Figure 9-18.

Plans held that by Fiscal Year FY 1999, when the Consolidated Space Operations Contract
(CSOC) would be in effect, CSOC would have responsibility for both the DSN and
AMMOS elements of TMOD. With that final step, the DSNIAMMOS unification
would be complete.

NOTE Each of the four Strdtegic Enterprises was affiliated with one or more of the NASA
Field Centers as shown in the chart. The Space Science Enterprise was afiliated with JPL.
However, it wa$om endeavors in support of the Space Science, Mission to Planet Earth,
and Space Flight Enterprises that requirements for the ddta and mission services that were
the prerogative of TMOD would be derived. Two of the four Enterprises were subsequently
renamed as their missions were revised. In 1997, the four Enterprises were named Space
Science, Mission to Planet Earth, Human Exploration and Development of Space, and
Aeronautics and Space Tranqortation Technology.

TMOD Primary Challenge: 1997

In the 1998 Strategic Plan, NASA established the following near-term, mid-term, and
long-term goals for the four Enterprises:

1991-2002: Establish a virtual presence throughout the solar system

* 2003-09: Expand the horizons

2010-23: Develop the frontiers

Commensurate with these goals, SOMO estimated that the Space Science Enterprise
would require data and mission services support for 86 missions through 2004; Mission
to Planet Earth would require support for 34 missions; and Human Exploration and
Development of Space would require support for 18 missions. A significant number of
these missions would involve TMOD and, with more complex investigations and instru-
ments characterizing these missions, would lead to a dramatic increase in the demand
for data and mission services over the next several years. This, said Squibb, would take
place against an environment of highly constrained, albeit level, budget resources. The
situation and the scope of the challenge it presented to TMOD were well illustrated in
the graphic shown in the 1997 Space Communications Budget Review and reproduced
in Figure 9-19.
Figure 9-1 9. TMOD primary challenge, 1997.

As 1997 ended, TMOD was challenged to provide "world-class" data and mission sup-
port for an ever-increasing number of missions in the face of the essentially fured budget
shown in the graphic.

To meet this challenge over the following three to five years, Squibb established a set
of five goals for TMOD:

increase data return capacity by a factor of 2.6,

accomplish a significant portion of TMOD work through interaction with at least 10

strategic partners within five years,

improve TMOD's performance by capturing all of the data available &om space missions,
 complete the transformation of TMOD into an organization that provides a fill Mission
Operations Services system, and

reinvest operations costs savings into technology and development that will yield fur-
ther cost and performance improvement.

Together with a metric to measure progress and a strategy for action, the goals would
form the core of TMOD implementation planning for the years ahead.

With these goals and their enabling technology accomplished, Squibb expected TMOD
to not only survive in a realistic new budget environment, but also to move forward,
as a vital element of the NASA Strategic Plan, into the new millennium. He put his
expectations this way: "In the foreseeable future, the world will witness technological
advances that will dwarf those of the past. Before long, our culture will embrace an
understanding of the universe that includes, among other things, its origins, evolution
and destiny, the distribution and character of planets around the stars, and occurrence
or prevalence of life in those environments. I fully expect the people of TMOD to be
among the leaders and innovators who will make these things happen." It was a clear
and ringing challenge to the people of the DSN, the engineers and scientists, techni-
cians and administrators, and supervisors and managers at all levels who composed its
vital essence. The measure of their response and the record of what they acheived was,
of necessity, a task for future historians of NASA's Deep Space Network.
Endnotes

1. William R. Corliss, "A History of the Deep Space Network," NASA CR- 151915
(Washington, DC: NASA, 1976).

2. J. Bluth, personal communication with author, 22 June 1998.

3. Homer E. Newell, Beyond the Atmosphere, NASA SP-4211 (Washington, DC:

NASA, 1980).

4. Clayton R. Koppes, JPL and the American Space Program; A History of the Jet
Propuhion Laboratory (New Haven and London: Yale University Press, 1982).

5. R. Cargill Hall, Lunar Impact: A History of Project Ranger, NASA SP-4210

(Washington, DC: NASA, 1977).

6. William R. Corliss, "History of the Deep Space Network," NASA CR-151915

(Washington, DC: NASA, 1976).

7. D. J. Mudgway, "Viking Mission Support," JPL Technical Report TR 32-1 526, Vol.
X (15 August 1972), pp. 22-26.

8. A. J. Siegmeth, "Pioneer 10 and G Mission Support," JPL Technical Report TR 32-

1526, Vol. X (15 August 1972), pp. 27-34.

9. M. J. Alazard, A. I. Beers, and T. D. Linnick, "Multimission Operations Systems

Office User Guide," JPL Document D-9661 (15 June 1993).

10. Jet Propulsion Laboratory, "JPL Strategic Plan," JPL Document 400-549 (April
1995); see also http://~ww.jpL,nasa.~ov/stratphn/.

11. Michael Hammer and James Champy, Reengineering the Corporation (New York:
Harper Collins, 1993).

12. R. J. Atnorose, "Reengineering Team for Services Fulfillment (Operations); Final

Report," TMOD internal memorandum, RJA: 95-008 (14 March 1995).
13. Gael F. Squibb, "Bridging the Space Frontier: TMOD Implementation Plan,
FY 98" (Pasadena: Jet Propulsion Laboratory, 30 January 1998); see also
http://deepspace.jpL. nm.g0v/92O/public/~22~strat~lan/.

14. National Space Policy 1996; see also http://www. whitehouse.gov~/EOP/OSTP/

NSTC/htmI/$$-5. h t d .

15. Space Operations Management Office, "Space Operations Implementation Plan-

Draft E," (Houston : JSC, August 1997).

16. "NASA Strategic Plan," NASA Policy Directive (NPD)-1000.1; see also
http:l/www. hq.nasa.gov/o~ce/mpINSPTOC.
html.
LIST OF FIGURES

Figures, Introduction

1. Figure Intro-1. TJyuger I outward bound for Jupiter, September 1977.

2. Figure Intro-2. I/quger I views Jupiter, February 1979.

3. Figure Intro-3. Mean distances of the terrestrial planets from the Sun.

4. Figure Intro-4. Mean distances of the Jovian planets from the Sun.

5. Figure Intro-5. HA-Dec and Az-El mounted antennas of the DSN.

6. Figure Intro-6. Injection into interplanetary orbit.

7. Figure Intro-7. Cassegrain focus antenna.

8. Figure Intro-8. Improvement in DSN downlink performance during the first

twenty years.

9. Figure Intro-9. An essential part of the answer; global map of the Deep Space
Network, 1992.

Figures, Chapter 1

1. Figure 1-1. View of the Pioneer tracking station site, Goldstone, California, 1978.

2. Figure 1-2. Prominent personalities of the Deep Space Network unveiling the
commemorative plaque at the Goldstone Pioneer Site, 28 October 1978.

3. Figure 1-3. The 26-meter antenna and tracking station (DSIF 11) at Pioneer site,
Goldstone, California, 1958.

4. Figure 1-4. Essential features of the Project Echo Experiment, 1960

 5. Figure 1-5. Az-El transmitting antenna for the Echo Balloon Experiment,
Goldstone, 1960.

6. Figure 1-6. The 26-meter antenna and tracking station (DSIF 41), Woomera,
Australia, 1961.

7. Figure 1-7. The 26-meter antenna and tracking station (DSIF 51),
Hartebeestpoort, South Africa, 1961.

Figures, Chapter 2

1. Figure 2- 1. World view of the Deep Space Instrumentation Facility, 1961.

2. Figure 2-2. Flight operations control center, JPL, 1961.

3. Figure 2-3. First close-up picture of Mars. Returned from Mariner 4 on 14 July
1965.

4. Figure 2-4. Network Operations Control Center (NOCC) at JPL, 1969.

5. Figure 2-5. Side elevation of the Goldstone 64-meter azimuth-elevation antenna,

1966.

6. Figure 2-6. The 64-meter antenna at Goldstone, 1966.

7. Figure 2-7. Composition of the Deep Space Network, 1974.

Figures, Chapter 3

1. Figure 3- 1. The Viking Era mission set.

2. Figure 3-2. Functional block diagram of the DSN, 1978.

3. Figure 3-3. Heliocentric orbit of Helios I.

4. Figure 3-4. Configurations of the Viking Orbiter (top) and Viking Lander (bottom).

5. Figure 3-5. Viking Orbiter, Lander, Earth Telecommunications System.

6. Figure 3-6. First Viking close-up picture of Mars surface.

7. Figure 3-7. First panoramic picture of the surface of Mars.

8. Figure 3-8. Voyager trajectories to Jupiter, Saturn, and Uranus.

9. Figure 3-9. Downlink performance estimates for Pioneers 10 and 11

10. Figure 3-10. Functional block diagram of the DSN Mark 111-75 configuration for
64-meter stations.

11. Figure 3-1 1. Functional block diagram of the DSN Mark 111-75 configuration for
26-meter stations.

12. Figure 3-12. Hydrostatic bearing and ball joint assembly.

13. Figure 3-13. Mark 111-77 Network systems and subsystems.

14. Figure 3-14. General configuration of the DSN Mark 111-77 Network.

15. Figure 3-15. DSN Mark I11 Data System implementation schedule, 1977.

16. Figure 3-16. Basic organization of the Mark I11 Data System.

17. Figure 3-17. Functional block diagram of the Mark 111-77 Telemetry System.

18. Figure 3-18. DSN Mark I11 Command System, 1978.

19. Figure 3-19. DSN-GCF-NASCOM ground communications, 1976-78.

20. Figure 3-20. Structural changes to the 26-meter antenna for extension to 34-meter
d'lameter.

21. Figure 3-21. Goldstone Station, DSS 12 antenna after conversion to 34-meter, S-
X-band configuration, October 1978.
Figures, Chapter 4

1. Figure 4-1. Eyager Era Deep Space mission set, 1977-86.

2. Figure 4-2. Vyager I image of Jupiter, 13 February 1979, at a distance of 20 million

kilometers, showing 10 (left) and Europa (right) against the Jupiter cloud tops.

3. Figure 4-3. Eyager I image of 10, 4 March 1979.

4. Figure 4-4. Eyager I trajectory through the Jupiter system, 5 March 1979.

5. Figure 4-5. Vqyager I image of Saturn and its rings, August 1980.

6. Figure 4-6. Eyager I image of Saturn and its rings, November 1980.

7. Figure 4-7. Parkes 64-m radio astronomy antenna.

8. Figure 4-8. Enhanced link performance for Vyager 2 at Uranus.

9. Figure 4-9. Functional block diagram of the Parkes-CDSCC telemetry array for
Voyager Uranus Encounter, January 1986.

10. Figure 4-10. Pictorial diagram of the Eyageer 2 mission.

11. Figure 4-1 1. Vyageer 2 image of the Uranus satellite Miranda.

12. Figure 4-12. Pioneer I 0 and I 1 heliocentric trajectories, September 1979.

13. Figure 4- 13. DSN configuration for ICE encounter of comet Giacobini-Zinner,
11 September 1985.

14. Figure 4-14. 34-meter high efficiency (HEF) antenna: DSS 15, Goldstone.

15. Figure 4-15. Rehabilitation of the DSS 14 hydrostatic bearing and pedestal
concrete, June 1983-June 1984.

16. Figure 4-16. Work in progress on the 60-m to 70-m antenna extension at DSS 14,
Goldstone, December 1987.
17. Figure 4-17. Elements of the DSN and GSTDN, 1982.

18. Figure 4-1 8. Canberra Deep Space Communications Complex, January 1985.

19. Figure 4-1 9. Deep Space Network, Mark IVA configuration.

20. Figure 4-20. DSN antenna operations costs, 1980-2000.

21. Figure 4-21. Signal Processing Center, Mark IVA configuration.

22. Figure 4-22. Network Operations Control Center, Mark IVA.

23. Figure 4-23. Principal features of the DSCC Mark IVA Telemetry Subsystem.

24. Figure 4-24. Functional block diagram of the Mark IVA Baseband Assembly (BBA).

25. Figure 4-25. DSN Mark IVA Telemetry System key characteristics.

26. Figure 4-26. DSN Command System Mark IVA 1985.

Figures, Chapter 5

1. Figure 5-1. Galileo Era Deep Space mission set. DJM files.

2. Figure 5-2. Phobos L-band feedhorn mounted on the 64-m antenna.

3. Figure 5-3. Magellan Venus Orbit mapping profile.

4. Figure 5-4. Diagram of Magellan spacecraft in cruise configuration.

5. Figure 5-5. Magellan SAR image of an impact crater in the Atalanta Region of Venus.

6. Figure 5-6. Encounter geometry for Vqyager 2 at Neptune.

7 . Figure 5-7. Vqyager 2 views three most prominent features of Neptune.

8. Figure 5-8. LGyager 2 views south polar cap of Triton.

 9. Figure 5-9. Wyager2 views Neptune and Triton.

10. Figure 5-10. The Galileo Venus-Earth-Earth-Gravity-Assist(VEEGA) trajectory.

11. Figure 5-1 1. The Galileo spacecraft cruise configuration.

12. Figure 5-12. The Galileo Orbiter and Probe arrival at Jupiter, 7 December 1995.

13. Figure 5-13. Network configuration for Galileo launch and cruise, 1989.

14. Figure 5-14. Galileo 1989 VEEGA mission: predicted X-band downlink
performance at Jupiter arrival: 7 December 1995.

15. Figure 5-1 5. Geometry of the Galileo first Earth flyby.

16. Figure 5- 16. Galileo image of Earth: Antarctica.

17. Figure 5-17. Earth and Moon from Galileo, December 1992.

18. Figure 5-18. Asteroid Ida and its moon.

19. Figure 5-19. The DSN in full array configuration for Galileo Jupiter orbital
operations.

20. Figure 5-20. Full DSN array operations for Galileo:typical downlink data rate profile.

21. Figure 5-2 1. Galileo image of icebergs on the surface of Europa.

22. Figure 5-22. Ubsses solar polar trajectory.

23. Figure 5-23. The Ubsses spacecraft inflight configuration.

24. Figure 5-24. Ubsses second solar polar orbit.

25. Figure 5-25. Cracked bearing rollers and inner race: DSS 63, December 1989.

26. Figure 5-26. Twenty years of telemetry arraying in the DSN.

27. Figure 5-27. Parkes-CDSCC telemetry array.

28. Figure 5-28. Enhanced downlink performance at Uranus.

29. Figure 5-29. Enhanced downlink performance at Neptune.

30. Figure 5-30. VLA-GDSCC telemetry array.

3 1. Figure 5-3 1. The concept of DSN telemetry for the Galileo LGA mission.

32. Figure 5-32. Functional diagram of the DSN Galileo telemetry subsystem for the
Canberra Deep Space Communications Complex.

33. Figure 5-33. Beam waveguide antenna in conceptual form.

34. Figure 5-34. Design for 34-meter beam waveguide antenna at DSS 13.

35. Figure 5-35. Design for operational 34-meter beam waveguide antenna.

36. Figure 5-36. DSS 34 beam waveguide antenna at Canberra Deep Space
Communications Complex, Australia, 1996.

37. Figure 5-37. Cluster of three 34-m beam waveguide antennas at Goldstone, 1995.

38. Figure 5-38. Signal Processing Center configuration, 1993.

39. Figure 5-39. DSCC Telemetry Subsystem, 1997.

40. Figure 5-40. DSCC Command System, 1996.

41. Figure 5-41. DSCC Tracking Subsystem, 1995.

42. Figure 5-42. Ground Communications Facility, 1992.

Figures, Chapter 6

1. Figure 6-1. Cassini Era mission set, 1997.

2. Figure 6-2. Mars Global Surveyor spacecraft in the mapping configuration.

 3. Figure 6-3. DSN configuration for Mars Global Surveyor.

4. Figure 6-4. Mars Global Surveyor image of Olyrnpus Mons.

5. Figure 6-5. Pathfinder Lander deployed on surface of Mars.

6. Figure 6-6. The MPF Lander camera views the Sojourner Rover in operation near
Yogi Rock on the Surface of Mars, July 1997.

7. Figure 6-7. Cassini interplanetary trajectory.

8. Figure 6-8. Cussz'ni spacecraft with Huygens Probe.

9. Figure 6-9. Location identifiers for DSN and related sites.

10. Figure 6-10. Goldstone Deep Space Communications Complex, 1997.

11. Figure 6- 1I. Canberra Deep Space Communications Complex, 1997.

12. Figure 6-12. Madrid Deep Space Communications Complex, 1997.

13. Figure 6-13. OVLBI Subnetwork functional block diagram.

14. Figure 6-14. DSS 23: SVLBI ll-meter antenna at Goldstone, 1996.

Figures, Chapter 7

1. Figure 7-1. Deep Space Communications Complex in Canberra, Australia.

2. Figure 7-2. An early parabolic design of a 26-meter antenna.

3. Figure 7-3. The quasi-parabolic 34-meter high-efficiency antenna, DSS 15.

4. Figure 7-4. A 70-meter antenna with dual-shape reflector design.

5. Figure 7-5. A 34-meter beam waveguide antenna at Goldstone Technology

DevelopmtLLL Site, DSS 13.
 6. Figure 7-6. Stationary equipment room below BWG antenna.

7. Figure 7-7. Dichroic (frequency-selective) reflector developed under Advanced

Systems Program.

8. Figure 7-8. Ultra-low noise amplifier (ULNA) at DSS 13.

9. Figure 7-9. High-electron mobility transistor (HEMT) low-noise microwave

amplifier.

10. Figure 7-10. Advanced Receiver (ARX) used to track the Pioneer 10 spacecraft.

11. Figure 7-1 1. Telemetry communication channel performance for various coding
schemes.

12. Figure 7-12. The 64-m antenna of the Radio Astronomy Observatory, Parkes,
Australia.

13. Figure 7-13. The Very Large Array (VLA) of the National Radio Astronomy
Observatory (NRAO) at Socorro, New Mexico.

14. Figure 7- 14. The new linear ion trap (LIT) frequency standard.

15. Figure 7-1 5. First high-resolution radar image of Venus.

16. Figure 7-16. Profile of deep space telecommunications capability, 1960-2020.

Figures, Chapter 8

1. Figure 8-1. DSN science-related publications, 1962-97.

2. Figure 8-2. Real-time bandwidth reduction in the radio science Occultation Data
Assembly.

3. Figure 8-3. Characteristic three-pulse signature of a gravitational wave in two-way

Doppler data.
 4. Figure 8-4. Detection strategy and system configuration for the SETI sky survey
field tests, 1985.

5. Figure 8-5. SETI detection of spacecraft signals, 1985.

6. Figure 8-6. The first HRMS sky survey, Goldstone, 12 October 1992.

7. Figure 8-7. Project ARIES transportable antenna interferometer system.

8. Figure 8-8. Global Positioning System, 1986.

9. Figure 8-9. Schematic view of GPS-based ionospheric calibration.

10. Figure 8-10. Functional block diagram of the Venus radar, 1962.

11. Figure 8-1 1. High Speed Data Acquisition System for the Goldstone Radar, 1984.

12. Figure 8-12. Relative sensitivity for the Goldstone Radar antenna.

13. Figure 8- 13. Goldstone Solar System Radar images of Asteroid Toutatis, 1992.

14. Figure 8-14. Goldstone Radar astronomy publications, 1988-97.

Figures, Chapter 9

1. Figure 9-1. NASA Office of Space Flight Development, 1959-61.

2. Figure 9-2. Telecommunications Division at JPL, 1961-62.

3. Figure 9-3. JPL organization, 1964.

4. Figure 9-4. NASA organization for the Off~ceof Tracking and Data Acquisition,
1961.

5. Figure 9-5. The Deep Space Network established, December 1963.

6. Figure 9-6. Organization for Tracking and Data Acquisition support at JPL,
October 1964.

7. Figure 9-7. Organization of the technical divisions after the transfer of the SFOF
to OCIS, 1972.

8. Figure 9-8. Functional interfaces between the DSN and the SFOF for Viking,
1972.

9. Figure 9-9. TDA Ofice 400 organization at JPL, 1978.

10. Figure 9-10. TDA Off~ce400 organization at JPL, 1980-87.

11. Figure 9-1 1. National Aeronautics and Space Administration, 1991

12. Figure 9-12. Multimission Operations Systems Ofice (440), 1993.

13. Figure 9-13. NASA Space Communications and Operations Program.

14. Figure 9-14. Telecommunications and Mission Operations Directorate, 1996.

15. Figure 9-15. Predicted cost and savings for reengineering TMOD Services
Fulfillment Processes.

16. Figure 9-16. Data Capture road map.

17. Figure 9-17. Telecommunications and Mission Operations Directorate, 1997.

18. Figure 9-18. Organization of the National Aeronautics and Space Administration,
1996.

19. Figure 9-19. TMOD primary challenge, 1997.

APPENDIX

The history of technology in the Deep Space Network is well documented in a continu-
ous series of technical articles published in the JPL Space Programs Summaries and its
successor journals. Most of these articles describe work that was funded exclusively by the
DSN Advanced Systems Program or was funded in cooperation with other DSN programs
related to engineering implementation or mission operations. This appendix will provide
the specialist reader with an overview of key developments in areas of specific interest.

Most of the citations that follow were extracted from the online index to "The Deep
Space Network Progress Reportn and its subsequent names, for issues from 1970 and
later, or from the author index to earlier reports. The list also includes a number of cita-
tions from refereed external technical journals and several case studies of advanced
technology in the DSN.

The early JPL Space Programs Summary (SPS) was published as a five-volume set des-
ignated as SPS 37-nn. One of these volumes, Volume IV, covered all Supporting Research
and Advanced Development activities at JPL, including the Deep Space Instrumentation
Facility (DSIF) Advanced Systems Program. Implementation and Operations activities
were included in Volume 111.

The Space Programs Summary became a four-volume set with SPS- 37-47. All Supporting
Research and Advanced Development activities are contained in Volume 111, and all
DSN activity is described in Volume 11.

Publication of the Space Program Summaries ceased in 1970 with SPS 37-66. However,
the publication of progress reports for the DSN (including the DSN Advanced Systems
Program) were continued in a series of JPL Technical Reports beginning with JPL-TR
32-1526, Volume I (15 February 1971). The DSN continued to report its progress in
this form through TR 32-1 526, Volume XIX (15 February 1974), when the name became
"The Deep Space Network Progress Report," DSN-PR 42-nn.

DSN progress reports under the new title began with DSN-PR 42-20 in April 1974,
and, except for the substitution of "Telecommunications and Data Acquisition" for "Deep
Space Network" in June 1980 (TDA PR 42-57), followed by "Telecommunications and
Mission Operations" in 1998, the series continued without hrther change through the
end of the century.
For ease of identification, the appendix refers to these various publications as JPL-SPS,
JPL-TR, and TDA-PR Copies of these documents may be obtained fi-om the Jet Propulsion
Laboratory, Pasadena, California. Articles that are referenced in issues TDA-PR 42-66
and later may be accessed online at http://tmo~p1.i~d~a,gov/progres~report.

1. The evolution of technology in the Deep Space Network.

Layland, J. W., and L. L. Rauch. "The Evolution of Technology in the Deep Space
Network, A History of the Advanced Systems Program," TDA Progress Report PR 42-
130, April-June 1997, 15 August 1997.

2. The great antennas of the DSN.

Potter, l? D. "Improved Dichroic Reflector Design for the 64-m Antenna S- and X-
band Feed Systems," JPL-TR 32-1526, Vol. XTX: November and December 1973, 15
February 1974, pp. 55-62.

Potter, l? D. "Shaped Antenna Designs and Performance for 64-m Class DSN Antennas,"
DSN-PR 42-20: January-February 1974, 15 April 1974, pp. 92-1 11.

Galindo-Israel, V., N. Imbriale, Y. Rahmat-Samii, and T. Veruttipong. "Interpolation

Methods for GTD Analysis of Shaped Reflectors," TDA Progress Report PR 42-80:
October-December 1984, 15 February 1985, pp. 62-67.

Goodwin, J. I? "Usuda Deep Space Center Support for ICE," TDA Progress Report PR
42-84: October-December 1985, 15 February 1986, pp. 186-96.

Fanelli, N. A., J. l? Goodwin, S. M. Petty, T. Hayashi, T. Nishimura, and T. Takano.

"Utilization of the Usuda Deep Space Center for the United States International Cometary
Explorer (ICE)." Proceedings of the Fifteenth International Symposium on Space
Technology and Science, Tokyo, Japan, 1986.

Otoshi, T. Y., and M. M. Franco. "Dual Passband Dichroic Plate for X-band," TDA
Progress Report PR 42-94: Aprii-June 1988, 15 August 1988, pp. 110-34.

Chen, J. C., l? H. Stanton, and H. F. Redly. "Performance of the X-IKa-IKABLE-band

Dichroic Plate in the DSS 13 Beam Waveguide Antenna," TDA Progress Report PR
42-115: July-September 1993, 15 November 1994, pp. 54-64.
Rafferty, Wfiam, Stephen D. Slobin, and Charles T. Stelzried. "Ground Antennas in
NASA's Deep Space Telecommunications." Proceedings of the Institute of Electrical and
Electronics Engineers, Vol. 82, No. 5, May 1994.

3. Forward commandldata link (uplink).

Benjauthrit, B., and T. K. Truong. "Encoding and Decoding a Telecommunication Standard

Command Code," DSN Progress Report PR 42-38: January and February 1977, 15 April
1977, pp. 115-19.

4. Return telemetryldata link (downlink).

Clauss, R C., and R B. Quinn. "Low Noise Receivers: Microwave Maser Development,"
JPL Technical Report TR 32-1526, Vol. DC: March and April 1972,15June 1972, pp. 128-36.

Clauss, R. C., and E. Wiebe. "Low-Noise Receivers: Microwave Maser Development,"

JPL Technical Report TR 32- 1526, Vol. X K November and December 1973,15 February
1974, pp. 93-99.

Ulvestad, J. S., G. M. Resch, and W. D. Brundage (National Radio Astronomy Observatory,

New Mexico). "X-band System Performance of the Very Large Array," TDA Progress
Report PR 42-92: October-December 1987, 15 February 1988, pp. 123-37.

Shell, J., and D. NefF. 'R32-GHz Reflected-Wave Maser Amplifier With Wide Instantaneous
Bandwidth," TDA Progress Report PR 42-94: April-June 1988, 15 August 1988, pp. 145-62.

Tanida, L. "An 8.4-GHz Cryogenically Cooled HEMT Amplifier for DSS 13," TDA
Progress Report PR 42-94: Aprii-June 1988, 15 August 1988, pp. 163-69.

Bautista, J. J., G. G. Ortiz, K. H. G. Duh (GE Electronics Laboratory, New York),

W. F. Kopp (GE Electronics Laboratory, New York), I? Ho (GE Electronics Laboratory,
New York), l? C. Chao (GE Electronics Laboratory, New York), M. Y. Kao (GE Electronics
Laboratory, New York), I? M. Smith (GE Electronics Laboratory, New York), and
J. M. Ballingall (GE Electronics Laboratory, New York). "32-GHz Cryogenically Cooled
HEMT Low-Noise Amplifiers," TDA Progress Report PR 42-95: July-September 1988,
15 November 1988, pp. 71-81.
Glass, G. W., G. G. Ortiz, and D. L. Johnson. "X-band Ultralow-Noise Maser Amplifier
Performance," TDA Progress Report PR 42-116, October-December 1993, 15 February
1994, pp. 246-53.

Shell, J., and R B. Quinn. 'X Dd-Cavity Ruby Maser for the Ka-band Link Experiment,"
TDA Progress Report PR 42-1 16, October-December 1993, 15 February 1994, pp. 53-70.

5. Phase-lock tracking.

J&, R., and E. Rechtin. "Design and Performance of Phase-Lock Circuits Capable of
Near-Optimum Performance over a Wide Range of Input Signal and Noise Level," IRE
Trans. Information Theory, Vol. IT-1, March 1955, pp. 66-76.

Tausworthe, R C. "Another Look at the Optimum Design of Tracking Loops," JPL-

SPS 37-32, Vol. IV,30 April 1965, pp. 281-83.

Lindsey, W C. "The Effect of RF Timing Noise in Two-Way Communications Systems,"

JPL-SPS 37-32, Vol. TV, 30 April 1965, pp. 28488.

Lindsey, W. C. "Optimum Modulation Indices for Single Channel One Way & Two-
Way Coherent Communications," JPLSPS 37-37, Vol. IV,28 February 1966, pp. 287-90.

Lindsey, W. C., and R. C. Tausworthe. "Digital Data Transition Tracking Loops," JPL-
SPS 37-50, Vol. 111, 30 Apii 1968, pp. 272-76.

Tausworthe, R. C. "Eaciency of Noisy Reference Detection," JPL-SPS 37-54, Vol. 111,

31 December 1968, pp. 195-201.

Tausworthe, R C. "Theory and Practical Design of Phase-Locked Receivers," JPL Technical

Report TR 32-819, 27 April 1971.

Layland, J. W. "Noisy Reference Effects on Multiple-Antenna Reception," DSN Progress

Report PR 42-25, November and December 1974, 15 February 1975, pp. 60-64.

6. Synchronization and detection.

Baumgartner, W S., W Frey, M. H. Brockman, R W Burt, J. W. Layland, G. M. Munson,

N. A. Burow, L. Couvillon, A. Vaisnys, R G. Petrie, C. T Stelzried, and J. K. Woo.
"Multiple-Mission Telemetry System," JPL-SPS 3746, Vol. 111, July 1967, pp. 175-243.
Tausworthe, R C., et al. "High Rate Telemetry Project," JPLSPS 37-54, Vol. 11, 30
November 1968, pp. 71-81.

Simon, M., and A. Mileant. "DSA's Subcarrier Demodulation Losses," TDA Progress
Report PR 42-85: January-March 1986, 15 May 1986, pp. 111-17.

Doerksen, I., and L. Howard. "Baseband Assembly Analog-to-Digital Converter Board,"

TDA Progress Report PR 42-93: January-March 1988, 15 May 1988, pp. 257-64.

7. Digital receivers.

Kurnar, R., and W J. Hurd. "A Class of Optimum Digital Phase Locked Loops for the
DSN Advanced Receiver," TDA Progress Report PR 42-83: July-September 1985, 15
November 1985, pp. 63-80.

Brown, D. H., and W. J. Hurd. "DSN Advanced Receiver: Breadboard Description and
Test Results," TDA Progress Report PR 42-89: January-March 1987, 15 May 1987, pp.
48-66.

Hinedi, S. 'X Functional Description of the Advanced Receiver," TDA Progress Report
PR 42-100: October-December 1989, 15 February 1990, pp. 131-49.

Hinedi, S., R. Bevan, and M. Marina. "The Advanced Receiver 11: Telemetry Test Results
in CTA 21," TDA Progress Report PR 42-104: October-December 1990, 15 February
1991, pp. 140-56.

Sadr, R., R. Bevan, and S. Hinedi. "The Advanced Receiver I1 Telemetry Test Results
at Goldstone," TDA Progress Report PR 42-106: April-June 1991, 15 August 1991, pp.
119-31.

8. Encoding and decoding.

Stiffler, J. J. "Instantly Synchronizable Block Code Dictionaries," JPL-SPS 37-32, Vol.

I v 30 April 1965, pp. 268-72.

Stiffler, J. J., and A. J. Viterbi. "Performance of a Class of Q-Orthogonal Signals for

Communication over the Gaussian Channel," JPL-SPS 37-32, Vol. IV, 30 April 1965,
pp. 277-81.
Green, R R. 'X Serial Orthogonal Decoder," JPL-SPS 37-39, Vol. IV, 30 June 1966,
pp. 247-52.

Lushbaugh, W. A., and J. W Layland. "System Design of a Sequential Decoding Machine,"

JPL-SPS 37-50, Vol. 11, 31 March 1968, pp. 71-78.

McEleice, R. J., and J. W Layland. "A Upper Bound to the Free Distance of a Tree
Code," JPL-SPS 37-62, Vol. 111, 30 April 1970, pp. 63-64.

Layland, J. W. "Performance of Short Constraint Length Convolutional Codes and a

Heuristic Code Construction Algorithm," JPL-SPS 37-64, Vol. 11, August 1970.

Lushbaugh, W. A. "Information Systems: Hardware Version of an Optimal Convolutional

Decoder," JPL Technical Report TR 32-1526, Vol. 11: January and February 1971, 15
Apfi 1971, pp. 49-55.

Butman, S. A., L. J. Deutsch, and R L. Miller. "Performance of Concatenated Codes

for Deep Space Missions," TDA Progress Report PR 42-63: March and April 1981, 15
June 1981, pp. 33-39.

Divsalar, D., and J. H. Yuen. "Performance of Concatenated Reed-SolomonNiterbi Channel

Coding," TDA Progress Report PR 42-7 1: July-September 1982, 15 November 1982,
pp. 81-94.

Yuen, J. H., and Q. D. Vo. "In Search of a 2-dB Coding Gain," TDA Progress Report
PR 42-83: July-September 1985, 15 November 1985, pp. 26-33.

Dolinar, S. J. 'RNew Code for Galileo," TDA Progress Report PR 42-93: January-
March 1988, 15 May 1988, pp. 83-96.

Dolinar, S. J. "VLA Telemetry Performance With Concatenated Coding for Voyager at

Neptune," TDA Progress Report PR 42-95: July-September 1988, 15 November 1988,
pp. 112-33.

Staunan, J., G. Zimmerman, F. Pollara, and 0. Collins. "A Long Constraint Length
VLSI Viterbi Decoder for the DSN," TDA Progress Report PR 42-95: July-September
1988, 15 November 1988, pp. 134-42.
Divsalar, D., M. K. Simon, and J. H. Yuen. "The Use of Interleaving for Reducing
Radio Loss: Convolutionally Coded Systems," TDA Progress Report PR 42-96: October-
December 1988, 15 February 1989, pp. 21-39.

Collins, O., F. Pollara, S. J. Dolinar, and J. Staunan. "Wiring Viterbi Decoders (Splitting
deBruijn Graphs)," TDA Progress Report PR 42-96: Oaober-December 1988, 15 February
1989, pp. 93-103.

Berrou, C., A. Glavieux, and l? Thitimajshima. "Near Shannon Limit Error-Correcting

Coding and Decoding: Turbo Codes," ICC, 1993, pp. 106470.

Divsalar, D. and F. Pollara. "On the Design of Turbo Codes," TDA Progress Report PR
42-123, July-September 1995, 15 November 1995, pp. 99-121.

Benedetto, S., Divsalar, D., et al. "Serial Concatenation of Interleaved Codes; Performance
Analysis, Design, and Iterative Decoding," IEEE Transactions on Information Theory,
Volume 44, 1998, p. 909.

9. Data compression.

Cheung, K M., F. Pollara, and M. Shahshahani. "Integer Cosine Transform for Image
Compression," TDA Progress Report PR 42-105: January-March 1991, 15 May 1991,
pp. 45-53.

Ekroot, L., S. J. Dolinar, and K M. Cheung. "Integer Cosine Transform Compression

for Galileo at Jupiter: A Preliminary Look," TDA Progress Report PR 42-1 15, July-
September 1993, 15 November 1994, pp. 110-23.

10. Arraying of antennas.

Urech, J. M. "Telemetry Improvement Proposal for the 85-foot Antenna Network,"

Space Programs Summary 37-63, Vol. 11, Jet Propulsion Laboratory, Pasadena, California,
31 May 1970.

Urech, J. M. "Processed Data Combination for Telemetry Improvement at DSS 62,"

JPL Technical Report TR 32-1526, Vol. 11: January and February 1971, 15 Aprii 1971,
pp. 169-76.
W~lck,H. "A Signal Combiner for Antenna Arraying," DSN Progress Report PR 42-
25, November and December 1974, 15 February 1975, pp. 111-17.

Winkelstein, R. A. "Analysis of the Signal Combiner for Multiple Antenna Arraying,"

DSN Progress Report PR 42-26, January and December 1975, 15 April 1975, pp.
102-18.

Stevens, R "Applications of Telemetry Arraying in the DSN," TDA Progress Report PR

42-72: October-December 1982, 15 February 1983, pp. 78-82.

Layland, J. W., and D. W. Brown. "Planning for W D S N Arrayed Support: to the

Voyager at Neptune," TDA Progress Report PR 42-82: April-June 1985, 15 August
1985, pp. 125-35.

Brown, D. W., W. D. Brundage, J. S. Ulvestad, S. S. Kent, and K. I? Bartos. "Interagency

Telemetry Arraying for Voyager-Neptune Encounter," TDA Progress Report PR 42-102:
April-June 1990, 15 August 1990, pp. 91-1 18.

Bartok, C. D. "Performance of the Real-Time Array Signal Combiner During the Voyager
Mission," TDA Progress Report PR 42-63: Match-April 1981, 15 June 1981, pp. 191-202.

Layland, J. W. "ICE Telemetry Performance," TDA Progress Report PR 42-84: October-

December 1985, 15 February 1986, pp. 203-13.

Hurd, W. J., F. Pollara, M. D. Russell, B. Siev, and I? U. Winter. "Intercontinental

Antenna Arraying by Symbol Stream Combining at ICE Giacobini-Zinner Encounter,"
TDA Progress Report PR 42-84: October-December 1985, 15 February 1986, pp. 220-28.

Foster, C., and M. Marina. "Analysis of the ICE Combiner for Multiple Antenna Arraying,"
TDA Progress Report PR 42-91: July-September 1987, 15 November 1987, pp. 269-77.

Brown, D. W., W. D. Brundage (National Radio Astronomy Observatory), J. S. Ulvestad,

S. S. Kent, and K. I? Bartos. "Interagency Telemetry Arraying for Voyager-Neptune
Encounter," TDA Progress Report PR 42-102: April-June 1990, 15 August 1990, pp.
91-118.

Rogstad, D. H. "Suppressed Carrier Full-Spectrum Combining," TDA Progress Report

PR 42-107: July-September 1991, 15 November 1991, pp. 12-20.
1 1. Radio metrics-tools and techniques.

Titsworth, R. C. "Optimal Ranging Codes," JPL Technical Report TR 32-41 1, 15 April

1963.

Hamilton, T. W., and W. G. Melbourne. "Information Content of a Single Pass of

Doppler Data from a Distant Spacecraft," JPL-SPS 37-39, Vol. 111, 31 May 1966, pp.
18-23.

Tausworthe, R. C. "Minimizing Range Code Acquisition Time," JPL-SPS 37-42, Vol.

IV, 31 December 1966, pp. 198-200.

Lushbaugh, W. L., and L. D. Rice. "Mariner Venus 67 Ranging System Digital Processing
Design," JPL-SPS 37-50, Vol. 11, 31 March 1968.

Goldstein, R. M. "Ranging with Sequential Components," JPL-SPS 37-52, Vol. 11, 31

July 1968, pp. 4 6 4 9 .

Martin, W. L. "A Binary Coded Sequential Acquisition Ranging System," JPL-SPS 37-
57, Vol. 11, 31 May 1969, pp. 72-81.

Martin, W. L. "Performance of the Binary Coded Sequential Acquisition Ranging System

of DSS 14," JPL-SPS 37-62, Vol. 11, 31 March 1970, pp. 55-61.

Otoshi, T. Y. "SIX Experiment: A Study of the Effects of Ambient Temperature on

Ranging Calibrations," DSN Progress Report PR 42-23, July and August 1974, 15
October 1974, pp. 45-51.

Otoshi, T. Y. "SIX Band Experiment: A Study of the Effects of Multipath on Two-Way

Range," DSN Progress Report PR 42-25, November and December 1974, 15 February
1975, pp. 69-83.

Melbourne, W. G. "Navigation Between the Planets," Scientific American 234, no. 6

(June 1976): 57-74.

Martin, W. L., and A. I. Zygielbaum. "Mu-I1 Ranging," JPL-TM 33-768, 15 May 1977.

Layland, J. W., and A. I. Zygielbaum. "On Improved Ranging-11," DSN Progress Report
PR 42-50: January and February 1979, 15 Aprii 1979, pp. 68-73.
Thurman, S. W. "Deep-Space Navigation With Differenced Data Types-Part I: Differenced
Range Information Content," TDA Progress Report PR 42-103: July-September 1990,
15 November 1990, pp. 47-60.

Thurman, S. W. "Deep-Space Navigation With Differenced Data Types-Part 11:

Differenced Doppler Information Content," TDA Progress Report PR 42-103: July-
September 1990, 15 November 1990, pp. 61-69.

Kahn, R. D., W. M. Fokner, C. D. Edwards, and A. Vijayaraghavan. "Position

Determination of a Lander and Rover at Mars With Earth-Based Dierential Tracking,"
TDA Progress Report PR 42-108: October-December 1991, 15 February 1992, pp. 279-93.

12. Timing standards.

Finnie, D. "Frequency Generation and Control: Atomic Hydrogen Maser Frequency

Standard," JPL Technical Report TR 32-1526, Vol. I: November and December 1970,
15 February 1971, pp. 73-75.

Dachel, l? R., S. M. Petty, R. F. Meyer, and R. L. Syndor. "Hydrogen Maser Frequency

Standards for the Deep Space Networks," DSN Progress Report PR 42-40, May and
June 1977, 15 August 1977, pp. 76-83.

Prestage, J. D., G. J. Dick, and L. Maleki. "New Ion Trap for Atomic Frequency Standard
Applications," TDA Progress Report PR 42-97: January-March 1989, 15 May 1989, pp.
58-63.

Prestage, J. D. "Improved Linear Ion Trap Physics Package," TDA Progress Report PR
42-113, January-March 1993, 15 May 1993, pp. 1-6.

13. Earth rotation and propagation media.

Winn, F. B. "Tropospheric Refraction Calibrations and Their Significance on Radio-

Metric Doppler Reductions," JPL Technical Report TR 32-1526, Vol. VII: November
and December 1971, 15 February 1972, pp. 68-73.

Von Roos, 0. H., and B. D. Mulhall. "An Evaluation of Charged Particle Calibration
by a Two-Way Dual-Frequency Technique and Alternatives to This Technique," JPL
Technical Report TR 32-1526, Vol. XI: July and August 1972, 15 October 1972, pp.
42-52.
Wmn, F. B., S. R. Reinbold, K. W. Yip, R. E. Koch, and A. Lubeley. "Corruption of
Radio Metric Doppler Due to Solar Plasma Dynamics: SIX Dual-Frequency Doppler
Calibration for These Effects," DSN Progress Report PR 42-30, September and October
1975, 15 December 1975, pp. 88-101.

Slobin, S. D., and l? D. Batelaan. "DSN Water Vapor Radiometer- Tropospheric Range
Delay Calibration," DSN Progress Report PR 42-49: November and December 1978,
15 February 1979, pp. 13645.

Roth, M., and T. Yunck. "VLBI System for Weekly Measurement of UTI and Polar
Motion: Preliminary Results," TDA Progress Report PR 42-58: May and June 1980, 15
August 1980, pp. 15-20.

Sovers, 0 . J.; J. L. Fanselow; G. H. Purcell, Jr.; D. H. Rogstad; and J. B. Thomas.

"Determination of Intercontinental Baselines and Earth Orientation Using VLBI," TDA
Progress Report PR 42-71: July-September 1982, 15 November 1982, pp. 1-7.

Scheid, J. A. "Comparison of the Calibration of Ionospheric Delay in VLBI Data by

the Methods of Dual Frequency and Faraday Rotation," TDA Progress Report PR 42-
82: April-June 1985, 15 August 1985, pp. 11-23.

Lindquister, U. J., A. l? Freedman, and G. Blewitt. "A Demonstration of Centimeter-

Level Monitoring of Polar Motion With the Global Positioning System," TDA Progress
Report PR 42-108: October-December 1991, 15 February 1992, pp. 1-9.

Linfield, R. l?, and J. 2. Wilcox. "Radio Metric Errors Due to Mismatch and Offset
Between a DSN Antenna Beam and the Beam of a Troposphere Calibration Insuurnent,"
TDA Progress Report PR 42-114, April-June 1993, 15 August 1993, pp. 1-13.

14. Radio science.

Levy, G. S., and G. E. Wood. "Voyager-Jupiter Radio Science Data Papers," TDA Progress
Report PR 42-58: May and June 1980, 15 August 1980, pp. 114-15.

Peng, T. K., and F. F. Donivan. "Deep Space Network Radio Science System for Voyager
Uranus and Galileo Missions," TDA Progress Report PR 42-84: October-December
1985, 15 February 1986, pp. 143-51.
Nelson, S. J., and J. W. Armstrong. "Gravitational Wave Searches Using the DSN,"
TDA Progress Report PR 42-94: April-June 1988, 15 August 1988, pp. 75-85.

Kursinski, E. R,. and S. W. Asmar. "Radio Science Ground Data System for the Voyager-
Neptune Encounter, Part I," TDA Progress Report PR 42-105: January-March 1991,
15 May 1991, pp. 109-27.

15. VLBI and radio astronomy.

Miller, J. K. "The Application of Differential VLBI to Planetary Approach Orbit

Determination," DSN Progress Report PR 42-40, May and June 1977, 15 August 1977,
pp. 8 4 9 0 .

Moultrie, B., T. H. Taylor, and l? J. Wolff. "The Performance of Differential VLBI Delay
During Interplanetary Cruise," TDA Progress Report PR 42-79, July-September 1984,
15 November 1984, pp. 35-46.

Ulvestad, J. S., and R. l? Linfield. "The Search for Reference Sources for VLBI Navigation
of the Galileo Spacecraft," TDA Progress Report PR 42-84, October-December 1985,
15 February 1986, pp. 152-63.

Satorius, E. H., M. J. Grimm, G. A. Zimmerman, and H. C. Wilck. "Finite Wordlength

Implementation of a Megachannel Digital Spectrum Analyzer," TDA Progress Report
PR 42-86, April- June 1986, 15 August 1986, pp. 24454.

Quirk, M. I? (Institute for Defense Analyses, New Jersey), H. C. Wilck, M. E Garyantes,

and M. J. Grimm. "A Wideband, High-Resolution Spectrum Analyzer," TDA Progress
Report PR 42-93, January-March 1988, 15 May 1988, pp. 188-98.

Treuhaft, R N., and S. T. Lowe. 'XNanoradian Differential VLBI Tracking Demonstration,"

TDA Progress Report PR 42-109, January-March 1992, 15 May 1992, pp. 40-55.

16. Global Positioning System.

Lanyi, G. "Total Ionospheric Electron Content Calibration Using Series GPS Satellite
Data," TDA Progress Report PR 42-85, January-March 1986, 15 May 1986, p. 112.

Lichten, S. M. "Precise Estimation of Tropospheric Path Delay GPS Techniques," TDA

Progress Report PR 42-100, October-December 1989, 15 February 1990, pp. 1-12.
 Freedman, A. I? "Combining GPS and VLBI Earth-Rotation Data for Improved Universal
Time," TDA Progress Report PR 42-105, January-March 1991, 15 May 1991, pp. 1-12.

Guinn, J., J. Jee, I? WolK E Lagattuta, T Drain, and V. Sierra. "TOPEXIPOSEIDON

Operational Orbit Determination Results Using Global Positioning Satellites," TDA
Progress Report PR 42-116, October-December 1993, 15 February 1994, pp. 163-74.

17. The Goldstone Solar System Radar.

Carpenter, R. L., and R M. Goldstein. "Preliminary Results of the 1962 Radar Astronomy
Study of Venus," JPL-SPS 37-20, Vol. W, 30 April 1963, pp. 182-84.

Goldstein, R. M. "Radar Observations of Icarus," JPL-SPS 37-53, Vol. 11, 30 September

1968, pp. 45-48.

Leu, R L. "X-band Radar System," JPL Technical Report TR 32-1526, Vol. XTX: November
and December 1973, 15 February 1974, pp. 77-81.

Downs, G. S., R. R. Green, and I? E. Reichley. 'RRadar Study of the Backup Martian
Landing Sites," DSN Progress Report PR 42-36, September and October 1976, 15
December 1976, pp. 49-52.

Bhanji, A. M., M. Caplan (Varian Associates, Inc.), R W. Hartop, D. J. Hoppe, W. A.

Imbriale, D. Stone (Varian Associates, Inc.), and E. W. Stone. "High Power Ka-band
Transmitter for Planetary Radar and Spacecraft Uplink," TDA Progress Report PR 42-
78, April-June 1984, 15 August 1984, pp. 24-48.

Bhanji, A. M., D. J. Hoppe, B. L. Conroy, and A. J. Freiley. ''Conceptual Design of a

I-MW CW X-band Transmitter for Planetary Radar," TDA Progress Report PR 42-95,
July-September 1988, 15 November 1988, pp. 97-111.

Renzetti, N. A., T. W. Thompson, and M. A. Slade. "Relative Planetary Radar

Sensitivities: Arecibo and Goldstone," TDA Progress Report PR 42-94, April-June 1988,
15 August 1988, pp. 287-93.

18. Telecommunications performance.

Yuen, J. H., ed. Deep Space Teleommunications Engineering. New York: Plenum Press,
1983.
I
Layland, J. X, and L. L. Rauch. "Case Studies of Technology in the DSN: Galileo,
Voyager, Mariner," published in "Evolution of Technology in the Deep Space Network,"
TDA Progress Report PR 42-130, April-June 1997, August 1997.

Posner, E. C., et al. "Voyager Neptune Telemetry." Session V of ITC.USAl87, papers

87-0819 through 87-0824 of the International Telemetering Conference, Vol. XIII, cospon-
sored by the International Foundation for Telemetry and Instrumentation Society of
America, 26-29 October 1987.

Posner, E. C., L. L. Rauch, and B. D. Madsen. "Voyager Mission Telecommunication

Firsts," IEEE Communications Magazine 28, no. 8 (September 1990): 22-27.

19. Cost-reduction initiatives.

Jet Propulsion Laboratory, Charles T. Stelzried, ed. TMOD Ethnology and Science Proram
News 2, 4, 7, 8, 9 (March 1995-April 1998). http://deeppace,ipl.na.ra.gou/technology

20. Ka-band development.

Jet Propulsion Laboratory, Charles T. Stelzried, ed. TMOD Technology and Science P r o p
News 1, 2, 3, 6, 7 (December 1994-Februq 1997). http://deeppa~e~pl.ma.gou/technology.

Morabito, D., R. C. Clauss, and M. Speranza. "Ka-band Atmospheric Noise-Temperature

Measurements at Goldstone, California, Using a 34-m Beam-Waveguide Antenna," TMO
Progress Report PR 42-132, February 1998.

Feria, Y., M. Belongie, T McPheeters, and H. Tan. "Solar Scintillation Effects on

Telecommunication Links at Ka-band and X-band," TMG Progress Report PR 42-129,
May 1997.

Morabito, D. D. "The Eficiency Characterization of the DSS 13, 34-meter Beam

Waveguide Antenna at Ka-band (32.0 GHz and 33.7 GHz) and at X-band (8.4 GHz),"
TMO Progress Report PR 42-125, May 1996.

Vilnrotter, Y A., and B. Iijima. "Analysis of Array Feed Combining Performance Using
Recorded Data," 42-125, May 1996.

Mysoor, N. R., et al. "Performance of a Ka-band Transponder Breadboard for Deep

Space Applications," TMO Progress Report PR 42-122, August 1995.
21. Other technologies.

Jet Propulsion Laboratory, Charles T. Stelzried, ed. TMOD Technolobyand Science Program
News 2, 3, 4, 5, 8, 9 (March 1995-April 1998). http://dppacejipl.nma.gov/technology.

22. Optical communications development.

Hemmati, H., et al. "Comparative Study of Optical and Radio Frequency Communication
Systems for Deep Space Missions," TMO Progress Report PR 42-128, February 1997.

Lesh, J. R., L. J. Deutsch, and C. D. Edwards. "Optical Communications for Extreme

Deep Space Missions," Review of Laser Engineering 24, issue 12 (December 1996).

Lesh, J. R. "Impact of Laser Communications on Future NASA Missions and Ground

Tracking Systems." CRL International Symposium on Advanced Optical Communications
and Sensing, Tokyo, Japan, 15-16 March 1995.

Wilson, K. E., J. R. Lesh, et al. "GOPEX: A Deep Space Optical Communications

Demonstration with the Galileo Spacecraft," TDA Progress Report PR 42-103, July-
September 1990, 15 November 1990, pp. 262-73.

Toyoshima, M., K. Araki, et al. "Reduction of ETS VI Laser Communication

Equipment Optical-Downlink Telemetry Collected during GOLD," TDA Progress
Report PR 42-128, 15 February 1997.

23. DSN science.

Jet Propulsion Laboratory, Charles T. Stelzried, ed. TMOD Technology and Science Program
News 1-9 (December 1994-April 1998). http://dep~pacejpl.nasd.~ov/technology.
ABOUT THE AUTHOR

Douglas J. Mudgway came to the United States in 1962 to work at NASAls Jet Propulsion
Laboratory in Pasadena, California, following a fifieen-year career in the field of guided
missile research in Australia. At JPL, he was involved in the development and operation
of the Deep Space Network from its infancy in the early 1960s to its maturity in the
early 1990s. He was the recipient of the NASA Exceptional Service Medal (1977) for his
conrribution to the Viking mission to Mars, and the Exceptional Achievement Medal
(1991) for his contribution to the Galileo mission to Jupiter. A mathematics and science
major from the University of New Zealand (1945), Mr. Mudgway retired in 1991 and
lives in the wine country of Northern California.
A
Active Magnetospheric Particle Explorer, AMPTE, 170
Activity Plan Process, 618, 619
Advanced DSN Monitor and Control, ADMC, 616
Advanced Multimission Operations System, AMMOS, 328, 610, 620
Advanced Receiver, ARX, 473, 474
Advanced Research Projects Agency, ARPA, 7 , 8, 10, 12, 18, 19, 25
Air Force, U.S., 4, 7
Allen, Lew, 602
Ames Research Center, NASA ARC, 33, 49, 51, 475
Amorose, Raymond J., 603, 612, 615
Anderson, John D., 514
Antennas
9-meter, 56, 169, 334, 336, 459, 541, 616
X-meter, 4, 10, 20, 21, 30, 37, 40, 43, 44, 50, 51, 56, 57, 58, 65, 66, 77, 84, 89, 97, 98, 100, 155,
156, 157, 159, 166, 169,265, 305,307, 324,328,334,335, 336, 383,459,495, 521, 597, 600, 616
34-meter, dvi, 84, 157, 158, 159, 166, 263, 265, 285, 290, 302, 304, 307, 324, 326, 328, 331, 334,
342, 344, 381, 439, 459, 462, 465, 467
64-meter, 41, 43, 46, 50, 51, 57, 58, 69, 70, 75, 77, 84, 89, 96, 98, 156, 172, 256, 263, 265, 278, 290,
299, 300, 304, 337, 341, 343, 344, 481, 490, 513, 521, 597, 600
70-meter, dvii, 285, 290, 302, 303, 304, 306, 307, 313, 329, 333, 337, 338, 339, 340, 341, 342, 343,
459, 462, 463, 467, 523, 525
ApolIo, 4, 35, 36, 43, 52, 54, 56, 57, 58, 73, 74, 597
Az-El, antenna, xvi, xx, xxiv, m, xxxiii, xxxiv, m,m i i i , m , 17
Army, U.S., 4, 5, 7 , 20
Asmar, Sami W., 507, 513
Astronomical Radio Interferometric Earth Surveying, ARIES, 541, 542
Astronomical Unit, AU, *, xxx, xxxii, 15, 16, 87, 88, 289, 306, 325, 330, 332, 333, 491, 550, 551
Atlantis, 286
Adas-Agena, 38
Automated Real-time Spacecraft Navigation, ARTSN, 495
Australian Department of Supply (DOS)/Weapons Research Establishment (WRE), 19, 20, 22, 64
Australian Government's Commonwealth Scientific and Industrial Research Organisation, CSIRO, 343, 526

B
Baker-Nunn, Smithsonian tracking camera, 20, 23
Baseband Assembly, BBA, 256, 385
Bayley, William H., 3, 583, 598
BayleylLyman Years, 601
Beggs, James, 276
Bell Telephone Laboratories, BTL, 12, 13, 15
Bendix Field Engineering Corporation, 21
Berman, Men L., 507, 514
Big Viterbi Decoder, BVD, 303, 309, 386, 387
Biibo, Thomas S., 594
Blaw-Knox Company, 9, 18, 19, 20
Block V Receiver, BVR, 314, 315, 316, 358, 474
Buffered Telemetry Demodulator, BTD, 368
Butcher, Lou, 3
Butrica, Andrew J., 173, 551

C
C-band, 281
California Institute of Technology, CIT, Caltech, 6, 8, 11, 25, 171, 487, 488, 523
California, University of, Berkeley, 341
California, University, Los Angeles, 583
Canberra, Australia, CDSCC, d v , xlvi, 18, 19, 21, 30, 39, 56, 66, 75, 77, 84, 102, 173, 290, 291, 303,
307, 313, 317, 324, 325, 328, 333, 337, 343, 368, 371, 379, 383, 437, 438, 459, 460, 482, 524, 526,
605
Cape Canaveral, Florida, Cape Kennedy, m,4, 5, 18, 23, 36, 37, 88, 96, 305, 328, 582
Carnegie Institute, 9
Carter, Jimmy, U.S. President, 173
Casani, John, 295, 302, 303
Cassegrain focus antenna, m i i i
Cassini, 274, 299, 398, 408, 409, 428, 429, 430, 437, 439, 453, 497, 512
Cebreros, 158
Celestial Mechanics, 5 1 1
Central Communications Terminal, 86
Centre National &Etudes Spatiales, CNES, 280, 327, 329, 336
Chalknger, 265, 283, 295, 301, 322
Clinton, W i i a m Jefferson, U.S. President, Administration, 610
Collins Radio Company, 20, 77
Comet Giacobini-Zinner, 276, 331
Comet Hdley, encounter, 274, 275, 276, 277, 278
Command and Data Console, CDC, 53
Command modulator assembly, CMA, 258, 263, 388
Command processor assembly, CPA, 258, 388
Command switching assembly, CSA, 388
Commonwealth Scientific and Industrial Research Organisation, CSIRO, 526
Compatibility Test Area 21, CTA21, 96, 97
Congress, U.S., 171
Consolidated Space and Test Center, CSTC, Air Force, 305
Corliss, William R., 42, 43, 43, 44, 45, 56, 73
Cross Support Agreements, 526
Crustal Dynamics, 541, 542, 548, 549

D
Data Capture Process, 616, 619
Data Compression, 478
Data Systems Operations Team, DSOT, 618
Dedination, d i i
Deep Space Communications Complex, DSCC, 77, 256, 258, 264, 313, 315, 386, 389, 391, 422, 525,
604, 616
Deep Space Instrumentation Facility, DSIF, 11, 15, 16, 21, 30, 31, 32, 33, 34, 36, 37, 38, 39, 40, 50, 61,
71, 507, 511, 521, 582, 591, 596, 599, 603
Deep Space Network, DSN, mcviu, xxx, xxxii, xxxiii, xxxiv, mmv, 3, 4, 5, 6, 8, 16, 17, 18, 25, 30, 33, 35,
36, 37, 38, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 54, 56, 57, 58, 61, 62, 63, 64, 65, 66, 69, 70,
71, 72, 73, 74, 75, 77, 78, 82, 83, 84, 87, 89, 90, 91, 92, 93, 96, 97, 99, 100, 101, 159, 166, 167,
168, 169, 170, 172, 173, 174, 176, 258, 263, 264, 265, 274, 275, 276, 277, 278, 279, 280, 281, 284,
286, 288, 289, 290, 291, 292, 293, 294, 295, 296, 297, 299, 301, 302, 303, 304, 305, 306, 307, 309,
310, 311, 312, 313, 315, 316, 317, 318, 319, 323, 324, 325, 326, 328, 329, 330, 331, 332, 333, 337,
338, 341, 343, 368, 371, 383, 398, 434, 437, 445, 450, 458, 467, 471, 472, 478, 483, 487, 488, 489,
493, 494, 498, 499, 502, 507, 508, 511, 515, 517, 521, 522, 526, 528, 550, 567, 591, 594, 597, 599,
600, 602, 603, 604, 606, 615, 620
Deep Space Network Monitor and Control, DMC, 447
Deep Space Station, DSS, 3
DSS 16, 336
DSS 11, Madrid, 3, 4, 56, 84
DSS 12, 47, 84, 89, 155, 157, 158, 278, 305
DSS 13, 47, 328, 375, 458, 494, 497, 534, 542
DSS 14, 45, 47, 57, 58, 77, 84, 89, 96, 98, 306, 314, 315, 317, 330, 337, 340, 341, 342, 513, 514,
542
DSS 15, 309, 314
DSS 23, 449
DSS 27, 336
DSS 34, 379
DSS 42, 56, 84, 89, 90, 158, 278
DSS 43, 77, 84, 98, 102, 238, 314, 315, 317, 337, 338, 340, 341, 342
DSS 44, 84, 89, 90, 158
DSS 46, 336
DSS 61, 56, 84, 158, 278
DSS 62, 84, 89, 158
DSS 63, 77, 84, 98, 238, 306, 317, 338, 339, 340, 341
DSS 65, 288
Deep Space Tracking System, DSTS, ESA, 279
Defense, U.S. Department of, 4, 7 , 8, 11, 18, 19
Deimos, 46
Delta Differential One-way Ranging, DDOR, 307, 512
Deutsch, Leslie J., 612
Differenced Range versus Integrated Doppler, DRVID, 71, 96,
Digital Equipment Corporation, DEC, 264
Digitally Controller Oscillator, DCO, 263
Doppler, mmviii, div, dvi, dvii, 5, 9, 42, 62, 70, 74, 96, 101, 174, 278, 281, 284, 287, 304, 306, 329, 330,
338, 422, 472, 484, 485, 486, 507, 508, 511, 514, 518
DSCC Galileo Telemetry, DGT, 312, 313, 314, 315, 316, 317, 318, 359
Dryden, Hugh, 21
Dubridge, Lee A., 6
Dumas, Larry N., 526, 606
DumaslHaynes Years, 606
E
Earth Crossing Asteroid, ECA, 499
Earth Orbiter, EO, 169
Echo, balloon experiment, 13, 14, 15, 16, 17, 42, 44, 66
Edwards, Chad, 525
Effron, Leonard, 276
Eisenhower, Dwight David, U.S. President, 7 , 10, 15
Ellis, Jordan, 276
Eros, asteroid, 274, 275
European Space Agency, ESA, 168, 276, 277, 278, 279, 319, 321, 325
European Space Operation Center, ESOC, 278, 279
Explorer, missions, 5
Extra-Galactic Radio Sources, EGRS, xlvi, 525

F
Faraday Rotation, FR, 513
Federal Aviation Administration, FAA, 498
Federal Republic of Germany, 170
Feedback Concatenated Decoder, FCD, 368
Felberg, Fred, 594
Finerman, Man, 598
Flight Operations Conuol Center, FOCC, 34
Flight Project Mission Operations Teams, PROJ, 447
Force, Charles T., 608
Ford, Gerald, U.S. President, 102
France, French, 169
Frequency and Timing System, FTS, 447
Full Spectrum Recorders, FSR, 368

G
Galdeo, 159, 166, 168, 265, 274, 275, 283, 286, 288, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303,
303, 304, 305, 306, 307, 308, 309, 310, 311, 312, 313, 314, 315, 316, 317, 318, 319, 322, 323, 329,
331, 337, 338, 339, 340, 342, 368, 398, 408, 410, 450, 477, 478, 498, 511, 512, 608
Gates, Clarence R, 582
German Space Operations Center, GSOC, 89, 90, 305
Giberson, Walker E., 582
Giotto,276, 277, 278, 279
Global Positioning System, GPS, 483, 486, 489, 494, 547
Goddard Space Flight Center, GSFC, 33, 56, 62, 278, 335, 591
Goldin, Daniel S., NASA Administrator, 610
Goldstein, Richard M., 16, 504, 513, 528
Goldstone, California, GDSCC, xxxii, xxxiv, xl, xlvi, 2, 10, 15, 18, 35, 37, 39, 41, 42, 43, 44, 45, 47, 52,
57, 66, 69, 70, 71, 75, 77, 84, 90, 101, 157, 173, 290, 291, 302, 303, 306, 307, 316, 325, 328, 333,
336, 337, 338, 341, 342, 381, 437, 438, 450, 458, 481, 482, 490, 513, 521, 523, 530, 534, 541, 542,
603, 605
Goldstone Solar System Radar, GSSR, 16, 489, 491, 499, 522, 563, 566, 567
Grand Tour, 171, 173
Greenwich Mean Time, GMT, mmv, 102
Gravitational Wave Experiment, GWE, 323, 518
Ground Communications Facility, GCF, 39, 75, 84, 86, 368, 394, 396, 447, 597, 600, 602, 604, 614, 615
Ground Operational Equipment, GOE, 50
Ground Reconstruction Equipment, GRE, 55
G&s, Samuel, 524, 528

H
Helios,86, 87, 88,89,90,91, 159, 513, 514
Hall, R Carg'i, 6, 36, 73
Halley, Edmond, ii
Halley Intercept Mission, HIM, 276
Hammer, Michael, 612
Haynes, Norman R., 606, 612
Heacock, Raymond L., 171
Heftman, Kurt, 598
Higa, Walter, 551
High-Earth orbiting, HEO, 169, 256, 275, 334
High-Efficiency antenna, HEF, 299, 304
High-electron mobility transistor, HEMT, 470, 471
High-Gain antenna, HGA, 172, 284, 285, 296, 297, 307, 308, 309, 310, 311, 312, 313, 322, 323, 325,
328, 332
High-Rate Telemetry System, HRT, 44, 45
High-Resolution Microwave Survey, HRMS, 537, 538, 539
Host Country Programs, 525
Hour-Angle Declination, HA-Dec, d i i , xxxiv, 14
Hubble Space Telescope, HST, 288, 335, 336
Hume, Alan, Australian Minister of Supply, 21

I
Inertial Upper Stage, IUS, 168, 295, 305, 324
Ingenieria y Servicios Aerospaciales, INSA, 606
Institute for Space and Astronautical Science, ISAS, 290, 445
Instituto Geografico National, Spain, 526
Instituto Nacional Tecnica Aeroespacial, INTA, Spain, 526, 605
Interface Control Documents, ICDs, 336
Interior, U.S. Department of, 4
Intercontinental Ballistic Missile, ICBM, 4
Intermediate Range Ballistic Missile, IRBM, 7
International Astronomical Union, IAU, 566
International Cometary Explorer, ICE, 169, 170, 275, 276, 277, 300, 331
International Geophysical Year, IGY, 4, 5, 7 , 13, 20
International Solar Polar Mission, ISPM, 319, 320, 321
International Solar Terrestrial Physics, ISTP, 383, 408, 434
Iowa, University of, 5
-
J a e , Richard, 5, 8
Jauncey, David L., 524, 528
Jet Propulsion Laboratory, JPL, xlvii, xlviii, 3, 4, 5, 6, 7 , 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20,
21, 23, 25, 30, 31, 33, 35, 36, 37, 38, 39, 41, 44, 45, 46, 47, 48, 49, 50, 52, 56, 60, 61, 63, 64, 65,
66, 69, 70, 72, 75, 77, 78, 83, 86, 89, 91, 93, 96, 98, 102, 171, 172, 173, 264, 276, 278, 279, 281,
295, 300, 303, 305, 306, 316, 317, 323, 329, 335, 338, 339, 368, 385, 393, 414, 450, 458, 475, 485,
502, 504, 522, 533, 582, 591, 594, 596, 597, 598, 601, 603, 605, 606, 610, 612, 618
Johannesburg, South Africa, DSIF 41, 20, 24, 25, 30, 37, 38, 39, 44, 62
Hartebeestpoort dam site, 23
Johnson, Lyndon B., U.S. President, 35
Johnson, Lyndon B., Space Center, NASA JSC, 299, 305
Johnson, Marshall S., 594
Jupiter, 173
Amalthea, 173
10, 173, 175
Europa, 173, 175, 299
Ganymede, 173, 175, 299
Callisto, 173, 175, 299
Jupiter Orbit Injection, JOI, 299, 315, 316
Jupiter Orbiter Probe, JOE 294, 295

K
Ka-band, m i i i , 440, 496, 497, 498, 512
Ka-band Link Experiment, KABLE, 328, 496, 497, 498
Kennedy Space Center, NASA KSC, Cape Kennedy, xxvii, 265, 295, 305
Klein, M. J., 504, 528
Koscielski, Charles, 3
Kuiper, Thomas B., 504, 528, 529

Laeser, Richard E, 172

Laika, 4
Lairmore, Glen E., 598
Langley Aeronautical Laboratory, 13
Langley Research Center, NASA LaRC, 33, 54, 83, 87, 92
Layland, James W, 303
Leslie, Robert A., 64, 65, 66
Levy, Gerry, 525, 533
Lewis Research Center, NASA LeRC, 299
Lincoln Laboratory, 15
linear ion trap, LIT, 485, 486
Linfield, Roger, 525
Links
One-way, 483
Two-way, 483
 Downlink, xxxix, xli, xlvi, xlviii, 23, 468, 477, 493, 551
Uplink, k, xli, xlvi, 467, 468, 497, 514, 515, 517, 551
Linnes, Karl W., 73
Lockheed Martin Astronautics, 414
Low-Earth orbiter, LEO, 334, 495, 616
Low-Gain Antenna, LGA, 297, 312, 313, 323, 330
Lunik, 34
Lyman, Peter T., 502, 561, 601

M
Madrid, Spain, MDSCC, b v , 18, 19, 23, 56, 75, 77, 90, 288, 291, 302, 303, 307, 325, 337, 338, 368,
437, 439
Magellan, 159, 274, 283, 285, 286, 288, 299, 322, 323, 488, 513
Mallis, Richard K., 3, 20, 21, 594
Manchester, University of, 15
Mariner, di, 16, 17, 25, 32, 33, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 50, 51, 55, 61, 62, 64,
66, 70, 71, 73, 75, 78, 82, 89, 93, 96, 172, 279, 475, 490, 491, 511, 512, 514, 515, 596
Mariner Jupiter-Saturn, 171, 172
Mark, lA, 111-77, N-A, 71, 155, 159, 166, 167, 256, 258, 264, 285, 299, 303, 304, 323, 334, 383, 386
Man Global Surveyor, MGS, 274, 408, 413, 414, 418, 420, 439, 453, 495,497, 511
Mars Observer, 326, 327, 328, 329, 414
Mars Orbit Insertion, MOI, 329
Mars Pathfinder, MPF, 274, 408, 421, 422, 423, 424, 425, 439, 453
Martin, James S., Jr., 92
Martin, Warren L., 70, 169
Massachusetts Institute of Technology, MIT, 15
Mathison, Richard l?, 342, 343
McElroy, Neil H., 5
McKee, Dick, 24
McKinley, Edward L., 612
Melbourne, University of, 65
Merrick, William, xx, 8, 9, 69, 70
Merritt Island, Florida, 90, 96
Messerschmicht-Bolkow-Blohm, MBB, 316
Metric Data Assembly, MDA, 263
Meyer, Don, 24
M i l i i , Robert A., 6
Mission Control and Computing Center, MCCC, 84, 86, 305, 601, 606, 608
Mobile Tracking Station, MTS, 37
Moffet, Allen T, 522, 528
Moon, m, 6, 7 , 9, 23, 31, 33, 34, 35, 52, 54, 55, 56, 57, 78
Mu, 70, 71, 90
Mudgway, Douglas J., mi, 73, 599
Muhleman, Duane O., 514, 563
Multimission Operations Systems Ofice, MOSO, 328, 610, 614
Murray, Bruce, 276, 602
Mutch, Thomas A, 176
N
NASCOM, NASA ground communications system, 52, 62, 77, 87, 90, 278, 305
National Academy of Sciences, NAS, 171
National Aeronautics and Space Administration, NASA, w i i , xliv, xlvi, xlviii, 2, 3, 8, 11, 15, 17, 18, 19,
20, 22, 23, 25, 33, 35, 36, 38, 39, 41, 42, 43, 46, 49, 50, 54, 56, 63, 64, 65, 66, 72, 77, 78, 89, 93,
102, 155, 159, 168, 169, 171, 265, 275, 276, 277, 283, 286, 301, 303, 310, 319, 320, 321, 327, 328,
329, 334, 335, 336, 343, 502, 521, 525, 594, 602, 605
National Bureau of Standards, NBS, 71
National Geographic, 293
National Radio Astronomy Observatory, NRAO, 290, 303, 342, 343, 435, 481, 482, 523, 526, 533
National Space Development Agency, NASDA, Japan, 277, 336
Naugle, John, 598
Naval Research Laboratory, NRL, 9
Navy, U.S., 4, 20
Near Earth Asteroid Rendezvous, NEAR, 274, 412
Neher, Victor, 6
Neptune, planet, xxix, mcxi, mcxii, 289, 290, 291, 292, 293, 294, 300, 303, 332, 333, 343, 349
Network Operations Control Center, NOCC, 86, 256, 263, 264, 323, 388, 393, 434, 447, 450, 600, 604,
614, 615
New Zealand, Wellington, 6
Nigeria, South Africa, 5, 18, 19, 23
NOCC Monitor and Control, NMC, 447
NOCC Support System, NSS, 263, 264, 447
NOCC Navigtion, NAV, 447

0
Occultation Data Assembly, 510
Office of Space Sciences, NASA HQ, 533
Office of Tracking and Data Acquisition, OTDA, 521, 597, 598
Olsen, Howard, 24
O'Neill, W i i a m J., 31 1
Opticd Navigation, OPNAV, 309, 310
Orbiting Very Long Baseline Interferometry, OVLBI, 445, 446, 447, 525
Ostro, Steven J., 504, 561, 562, 564, 566

Parkes Radio Astronomy Observatory, 256, 290, 291, 292, 313, 343, 346, 410, 481, 482, 521
Parkes-CDSCC Telemetry Array, PCTA, 343, 344, 526
Parks, Robert J., 582
Payload Assist Module, PAM, 324
Phase-lock tracking, 472
Phobos, 46, 279, 280, 281, 282, 288
Pickering, William H., xvii, 5, 6, 35, 61, 591, 599
Piloted Space Flight Network, MSFN, 56, 57
Pioneer, 2, 3, 4,7 , 8, 10, 11, 12, 14, 20, 43, 44, 48, 50, 51, 52, 53, 62, 66, 73, 78, 82, 84, 86, 89, 91,
155, 159, 166, 167, 168, 171, 275, 276, 277, 283, 293, 294, 300, 330, 331, 341, 408, 410, 412, 473,
475, 511, 513, 514, 518, 582, 597
Planetary Radar, 550
Pluto, xxix, xxxii, xxxiii
Programmed oscillator, POCA, 263
Project Operations Control Centers, POCC, 86
Project Sequence of Events, PSOE, 618, 619
Puerto Rico, 18, 23

R
Radar Astronomy, 566
Radio Astronomy, 521
Radio Frequency Interference, RFI, 534
Radio Frequency (RF) Radiation, m i i , xxxix
Radio Propagation and Occultation, 514
Radio Science, 507, 508, 509, 510, 513, 514
Radioisotope Thermal Generator, RTG, 51, 305, 332
Ranger, 17, 21, 23, 24, 25, 32, 33, 34, 35, 36, 38, 60, 61, 62, 64, 78, 596
Reagan, Ronald, U.S. President, 276
Rechtin, Eberhardt, 3, 5, 7 , 8, 9, 30, 61, 502, 582, 584
Redstone, arsenal, 5
Relativistic Time Delay, 5 17
Renaissance, 567
Renzetti, Nicholas A, iii, xix, 21, 73, 155, 458, 502, 598, 601
Rohledo, 84
Roentgen, satellite, 408
Rohr Corporation, 66
Round-trip light time, RTLT, 483

S-band, m i i , dii, diii, dvii, 16, 39, 47, 50, 61, 62, 63, 84, 87, 97, 98, 155, 156, 157, 166, 256, 263,
264, 277, 285, 286, 290, 297, 299, 301, 304, 306, 308, 312, 313, 315, 323, 324, 325, 330, 368, 388,
467, 513, 518
Sakigake, 276
San Diego, California, 5
Saturn V, 57
Search for Extraterrestrial Intelligence, SETI, 488, 502, 504, 533, 535, 536
Shoemaker-Levy Comet, 3 13, 3 16
Siegmeth, Alfred J., 73
Signal Processing Centers, SPC, 383, 384, 385, 388, 390, 394, 447
Signal-to-Noise Ratio, SNR, m i i , xlvii, 478, 483
Significant Advances, 529
Significant Events, 569
Venus, xxx, 6, 6, 15, 16, 31, 37, 38, 39, 42, 43, 46, 47, 48, 61, 78, 82, 155, 159, 167, 274, 277,
283, 285, 286, 305, 306, 331, 339, 490, 511, 550, 569
Mars, Red Planet, xxvii, xxix,mmi, h i , m , xxxvi, 6, 31, 39, 41, 43, 44, 45, 51, 57, 62, 66, 75, 78,
82, 91, 92, 93, 96, 97, 101, 102, 103, 274, 280, 281, 282, 294, 327, 328, 329, 330, 475, 490, 511
Mercury, xxix,xxxi, 78, 82, 569
Jupiter, m i i , xxix, xxxi, h i , 31, 46, 50, 51, 52, 74, 78, 82, 166, 167, 169, 171, 172, 173, 175, 265,
294, 295, 296, 299, 300, 301, 304, 310, 312, 313, 316, 317, 333, 478, 569
 Saturn, xxvii, xxk,&, 31, 52, 166, 171, 172, 173, 176, 276, 333, 571
E a A , k, =xi, e i , m , xliii, dvi, 4, 5, 7 , 8, 12, 13, 15, 16, 33, 37, 40, 47, 54, 57, 70, 74, 86, 87,
91, 92, 100, 170, 275, 277, 284, 288, 290, 296, 325, 331, 483, 511, 517
Singapore, 5
Small Deep Space Transponder, SDST, 497
Solar Corona, wind, 513
Solar Heliospheric Observatory, Soho, 434
South African Council for Scientific and Industrial Research (CSIR)/National Institute of Telecommunications
Research (NITR), 19, 23, 25
South African Department of the Postmaster General, 24
Soviet Union, 169, 283
Space Flight Operations Facility, SFOF, 39, 41, 42, 52, 60, 61, 62, 63, 72, 75, 84, 86, 596, 596, 597, 598,
599
Space Flight Operations Center, SFOC, 610
Space Operations Management Operations, SOMO, 335, 610, 612, 614, 619
Space Very Long Baseline Interferometry, SVLBI, 445
Spaceflight Tracking and Data Network, STDN, 90, 91
Spaulding, Wallace E, 599
Search for Extraterrestrial Intelligence, SETI, 488, 502, 504, 533, 535, 536
Sputnik, 4, 6, 78
Standard Format Data Unit, SFDU, 327
Stelzried, Charles, 276, 513
Stevens, Robertson S., 8, 9, 337, 582
Stoller, Floyd \V, 20
Stone, Edward C., 171, 606, 611
Sun, A, =xi, k i , e i i , xxxv, m i , m i i , 16, 3, 35, 41, 74, 78, 87, 100, 169, 280, 325, 330, 331,
507, 514
Sun-Earth-Probe, SEE 89, 90
Support Instrumentation Requirements Document, SIRD, 72, 594
SURE satellite, 408
Surveyor, 4, 32, 35, 36, 52, 53, 55, 64, 78, 597
Synchronization and Detection, data, 472
Synthetic Aperture Radar, SAR, 284
System Noise Temperature, xliii

T
Tau, 41, 42, 71
Tauswonhe, Robert C., 70
Telecommunications and Data Acquisition, TDA, 502, 503, 506, 526, 533, 541, 549, 594, 597, 598, 601,
602, 603, 605, 606, 612
Telecommunications and Mission Operations Directorate, TMOD, 334, 335, 451, 452, 458, 458, 492, 613,
618, 619, 620
Birth of the TMOD, 612
Reengineering TMOD, 6 15
Telemetry Modulation Unit, TMU, 303
Telemetry Processor Assembly, TPA, 257, 258, 385
Thatcher, John W., 73, 599
Thor-Able, 7
Tidbinhiia, Australia, 21, 39, 44, 64, 65, 66, 84, 522, 524, 530
Titan-Centaur, mrvii, 88, 92, 93, 100, 172, 294, 295
TOMS-ED, 408
TOPEX-POSEIDON, 489
Toutatis, asteroid, 565
Tracking and Communications Extraterrestrial, TRACE, 18
Tracking and Data Acquisition, TDA, 96
Tracking and Data Relay Satellite System, TDRSS, 169, 275, 288, 310, 335, 336, 525, 530
Trajectory correction maneuvers, TCMs, 324
Transfer Orbit Stage, TOS, 328
Travelling Wave Tube, TWT, 330
Traveling wave masers, TWM,469, 470, 471
Triton, 290, 292
Truly, Richard H., 608
Truszynski, Gerald, 589, 598

U
Ultra-stable Oscillator, USO, xli, xlviii, 515
Ulvestad, James, 525
Ulysses, 168, 169, 265, 285, 300, 316, 321, 322, 323, 324, 325, 329, 408, 410, 499, 613
United Kingdom, 170
United States, U.S., 78, 169, 170, 276, 283
Universal Time Coordinated, UTC, m,288
Uranus, xxk, xxxi, xxxii, 166, 171, 172, 265, 290, 333, 344
Usuda Observatory, Japan, 290, 291, 292
USSR, 283
USSR Space Research Institute, IKI, Russia, 279
Urech, Jose M., 514

v
Van Allen, James A., 5, 6, 12
Vanguard, 4, 5
Vega 1, 2, 274, 277, 279, 280, 281
Venus-Earth-Earth-Gravity Assist, VEEGA, 295, 299, 301
Venus Orbiting Imaging Radar, VOIR, 283
Venus Radar Experiment, VRE, 15, 16
Venus Radar Mapper, VRM, 283
Very Large Array, VLA, 290, 291, 303, 304, 348, 351, 471, 481, 482
Very Long Baseline Interferometry, VLBI, 263, 277, 281, 304, 341, 434, 434, 435, 445, 483, 486, 487, 488,
525, 526, 530, 542
Very Long Baseline Interferometry Space Observatory Program, VSOD, 434, 436, 445
Victor, Walter K, 8, 9, 550, 582
Viking, xli, 46, 51, 75, 78, 82, 83, 87, 89, 91, 92, 93, 96, 97, 98, 100, 101, 159, 166, 167, 172, 265, 274,
279, 294, 327, 51 1, 514
V i n g Mission Control Center, VMCC, 600
von Braun, Wernher, 6
Voyager, mii, m,d i 52, 78, 83, 157, 159, 166, 167, 168, 170, 171, 172, 173, 174, 175, 263, 264, 265,
275, 276, 278, 285, 286, 288, 290, 291, 292, 293, 294, 300, 315, 331, 332, 333, 343, 344, 408, 410,
475, 475, 477, 478, 488, 511, 512, 536, 613
Voyager Interstellar Mission, VIM, 331, 332

--
Waddey, Joseph A., 514
Washington, DC, 18
Weilheim, 91
Westmoreland, Paul T., 606, 612
White House, 35, 173
White Sands Missile Range, New Mexico, 4, 7, 69, 335
Woornera, Australia, DSIF 41, 20, 21, 22, 23, 25, 37, 39, 42, 66, 77, 522, 550
Island Lagoon, 20, 21
World Net, 19

Y
Yerbury, Michael J., 524
THENASA HISTORY
SERIES

THE NASA HISTORY SERIES

Reference Works, NASA SP-4000:

Grimwood, James M. Project Mercury: A Chronology. NASA SP-4001, 1963.

Grimwood, James M., and C. Barton Hacker, with Peter J. Vonimmer. Project Gemini Technology and
Operations:A Chronology. NASA SP-4002, 1969.

Link, Mae Mills. Space Medicine in Project Mercury. NASA SP-4003, 1965.

Astronautics and Aeronautics, 1963: Chronology of Science, TechnoloB and Policy. NASA SP-4004, 1964.

Astronautics and Aeronautics, 1964: Chronology of Science, T e c h n o h ~and Policy. NASA SP-4005, 1965.

Astronautics and Aeronautics, 1965: Chronology of Science, Technology, and P 0 l 4 NASA SP-4006, 1966.

Astronautics and Aeronautics, 1 9 6 6 Chronology of Science, Terhnology, and Policy. NASA SP-4007, 1967.

Astronautics and Aeronautics, 1967: Chronology of Science, Technology, and Policy. NASA SP-4008, 1968.

Ertel, Ivan D., and Mary Louise Morse. The Apollo Spacecraft: A Chronology, Volume I, Through November 7,
1962. NASA SP-4009, 1969.

Morse, Mary Louise, and Jean Kernahan Bays. The Apollo Spacecraft: A Chronology, Volume IL November 8,
1962-September30, 1964. NASA SP-4009, 1973.

Brooks, Courtney G., and Ivan D. Ertel. The Apollo Spacecraft: A Chronology, Volume III, October I, 1964-
January 20, 1966. NASA SP-4009, 1973.

Ertel, Ivan D., and Roland W. Newkirk, with Courtney G. Brooks. The Apollo Spacecraft:A Chronology, Volume
IY January 21, 1966Juuly 13, 1974. NASA SP-4009, 1978.
Astronautics and Aeronautics, 1968: Chronology of Science, Technology, and Policy. NASA SP-4010, 1969.

Newkirk, Roland W., and Ivan D. Ertel, with Courtney G. Brooks. Skylab: A Chronology. NASA SP-4011,
1977.

Van Nimmen, Jane, and Leonard C. Bruno, with Robert L. Rosholt. NASA Historical Data Book, Volume I.
NASA Resources, 1958-1968. NASA SP-4012, 1976, rep. ed. 1988.

Ezell, Linda Neuman. NASA Historical Data Book, Volume II: Programs and Projects, 1758-1968. NASA SP-
4012, 1988.

Ezell, Linda Neuman. NASA Historical Data Book, %Qme III: Programs and Projects, 1969-1978. NASA SP-
4012, 1988.
Gawdiak, Ihor Y., with Helen Fedor, compilers. NASA HistoricalData Book, VolumeA4 NASA Resources, 1969-
1978. NASA SP-4012, 1994.

Rumerman, Judy A., compiler. NASA Historical Data Book, 1979-1988: Volume V NASA Launch Systems, Space
Transportation, Human Spacejlight, and Space Science. NASA SP-4012, 1999.

Rumerman, Judy A., compiler. NASA Historical Data Book, Volume M: NASA Space Applications, Aeronautics
and Space Research and Technology, Tracking and Data Acquisition/Space Operations, Commercial Programs, and
Resources, 1979-1988. NASA SP-2000-4012, 2000.

Astronautics and Aeronautics, 1969: Chronology of Science, Terhnology, and Policy NASA SP-4014, 1970.

Astronautics andAeronautics, 1970: Chronology ofscience, Technoloa and Policy NASA SP-4015, 1972.

Astronautics and Aeronautics, 1971: Chronology of Science, Technology, and Policy NASA SP-4016, 1972.

Astronautics andAeronautics, 1972: Chronology of Science, Technology, and Policy NASA SP-4017, 1974.

Astronautics and Aeronautics, 1973: Chronology of Science, Technology, and Policy NASA SP-4018, 1975.

Astronautics and Aeronautics, 1974: Chronology of Science, Technology, and Policy NASA SP-40 19, 1977.

Astronautics and Aeronautics, 1975: Chronology of Science, Tecbnologp and Policy NASA SP-4020, 1979.

honautics and Aeronautics, 1976: Chronology of Science, Terhnology, and Policy NASA SP-4021, 1984.

Astronautics and Aeronautics, 1977: Chronology of Science, Technology, and Policy NASA SP-4022, 1986.

Astronautics and Aeronautics, 1978: Chronology of Science, Terhnology, and Policy NASA SP-4023, 1986.

Astronautics and Aeronautics, 1379-1984: Chronology of Science, Technology, and Policy NASA SP-4024, 1988.

Astronautics and Aeronautics, 1985: Chronology of Science, Technology, and Policy NASA SP-4025, 1990.

Noordung, Hermann. The Problem of Space Travel: The Rocket Motor. Edited by Ernst Stuhlinger and J. D.
Hunley, with Jennifer Garland. NASA SP-4026, 1995.

Astronautics and Aeronautics, 1986l990: A Chronology. NASA SP-4027, 1997.

htronautics andAeronautics, 1990-1995:A Chronology. NASA SP-2000-4028,2000.

Management Histories, NASA SP-4100:

Rosholt, Robert L. An Administrative Histoy ofNASA, 1958-1963. NASA SP-4101, 1966.

Levine, Arnold S. Managing NASA in the Apollo Era. NASA 9-4102, 1982.
 THENASA HISTORY
SERIES

Roland, Alex. Model Research: The National Advisory Committeefor Aeronautics, 1915-1958. NASA SP-4103,
1985.

Fries, Sylvia D. NASA Engineers and the Age ofApollo. NASA SP-4104, 199.

Glennan, T. Keith. The Birth of NASA: The Diary of 7: Keith Glennan. J. D. Hunley, editor. NASA SP-4105,
1993.

Seamans, Robert C., Jr. Aimingat Targets: The Autobiopphy ofRobert C. Seamam, Jr. NASA SP4106, 1996.

Project Histories, NASA SP-4200:

Swenson, Loyd S., Jr., James M. Grimwood, and Charles C. Alexander. This New Ocean: A History ofProject
Mercury. NASA 9-4201, 1966; rep. ed. 1998.

Green, Constance McLaughlin, and Milton Lomask. Vanguard:A History NASA SP-4202, 1970; rep. ed.
Smithsonian Institution Press, 1971.

Hacker, Barton C., and James M. Grimwood. On Shoulders of Etam: A History of Project Gemini. NASA SP-
4203, 1977.

Benson, Charles D., and William Barnaby Faherty. Moonport: A History of Apollo Launch Facilities and
Operations. NASA SP-4204, 1978.

Brooks, Courtney G., James M. Grimwood, and Loyd S. Swenson, Jr. Chariots$rApollo: A History ofManned
Lunar Spacema$. NASA SP-4205, 1979.

Bilstein, Roger E. Stages to Saturn:A TechnologicalHistory of the Apollo/Saturn Launch Vehicles. NASA SP-4206,
1980, rep. ed. 1997.

SP-4207 not published.

Compton, W. David, and Charles D. Benson. Living and Working in Space: A History of Skylab. NASA SP-
4208, 1983.

Ezell, Edward Clinton, and Linda Neuman Ezell. The Partnership: A History of the Apollo-Soyu Est Project.
NASA SP-4209, 1978.

Hall, R Cargill. Lunar Impact: A History ofProject Ranger. NASA SP-4210, 1977.

Newell, Homer E. Beyond the Atmosphere: Early Ears of Space Science. NASA 9 - 4 2 1 1, 1980.

Ezell, Edward Clinton, and Linda Neuman Ezell. On Mars: Exploration of the Red Planet, 1958-1978. NASA
SP-4212, 1984.

I
Pitts, John A. The Human Factor: Biomedicine in the Manned Space Program to 1980. NASA SP-4213, 1985.
3.
Compton, W. David. Where No Man Hm Gone B&re: A Histo y ofApollo Lunar Exploration Missions. NASA
SP-4214, 1989.

Naugle, John E. First Among Eqmb: The Selection of NASA Space Science Experiments. NASA SP42 15, 1991.

Wallace, Lane E. Airborne Failblazer: Two Decades with NASA Langlq? Boeing 737 Fhing Laboratoty NASA
SP-4216, 1994.

Butrica, Andrew J., editor. Beyond the Ionosphere: F t f i Ears of Satellite Communication. NASA 9-4217, 1997.

Butrica, Andrew J. To See the Unseen:A History ofPlanetay Radarhonomy. NASA SP-4218, 1996.

Mack, Pamela E., editor. From Engineering Science to Big Science: The NAG4 and NASA Collier h p h y Research
Project Winners. NASA SP-4219, 1998.

Reed, R. Dale, with Darlene Lister. Wingless Flight: The LzfZtng Body Story. NASA 9-4220, 1997.

Heppenheimer, T. A. The Space Shuttle Decision: NASA? Searchfor a Reusable Space Vehicle. NASA SP-4221,
1999.

Hunley, J. D., editor. TowardMach 2: The Dough 0-558 Program. NASA SP-4222, 1999.

Swanson, Glen E., editor. "Before this Decade is Out. . .:.' Personal Reflections on the Apollo Program NASA SP-
4223, 1999.

Tomayko, James E. Computers Take Flight: A History of NASA: Pioneering Digital Fly-by- Wire Project. NASA
SP-2000-4224, 2000.

Morgan, Clay. Shuttle-Mir: The US. and Russia Share History? Highest Stage. NASA SP-2001-4225, 2001

Center Histories, NASA SP-4300:

Rosenthal, Alfred. Venture into Space: Early Ears of Godhrd Space Flight Center. NASA SP-4301, 1985.

Hartman, Edwin l? Adventures in Research: A Histoy ofAmes Research Center, 1940-1965. NASA SP-4302,
1970.

Hallion, Richard P. On the Frontier: Flight Research at Dyden, 1946-1381. NASA SP- 4303, 1984.

Muenger, Elizabeth A. Searching the Horizon: A Histoy ofAmes Research Center, 1940-1976. NASA SP-4304,
1985.

Hansen, James R. Engineer in Charge:A Histoy of the Langley Aeronautical Laboratory,1917-1958. NASA SP-
4305, 1987.

Dawson, Virginia I! Engines and Innovation: Lewiz Laboratory and American Propubion Technology. NASA SP-
4306, 1991.
THENASA HISTORY
SERIES

Dethloff, Henry C. "Suddenly Tomorrow Came . . .:.' A History of the Johnson Space Center. NASA SP-4307,
1993.

Hansen, James R. SpacefightRevolution: NASA Langley Research Centerfrom Sputnik to Apollo. NASA SP-4308,
1995.

Wallace, Lane E. Flights of Discovery: 50 Years at the NASA Dyden Flight Research Center. NASA SP-4309,
1996.

Herring, Mack R. Way Station to Space: A History of the John C. Stennis Space Center. NASA SP-4310, 1997.

Wallace, Harold D., Jr. WallopsStation and the Creation of the American Space Program. NASA SP-4311, 1997.

Wallace, Lane E. Dreams, Hopes, Realities: NASAs Goddkrd Space Flight Center, The First Fory Ears. NASA SP-
4312, 1999.

Dunar, Andrew J., and Stephen P. Waring. Power to +lore: A Histoy ofthe MarshallSpace Flight Center. NASA
SP-4313, 1999.

Bugos, Glenn E. Atmosphere of Freedom: Sixy Ears at the NASA Ames Research Center Astronautics and
Aeronautics, 1986-1990: A Chronology. NASA SP-2000-4314,2000.

General Histories, NASA SP-4400:

Corliss, W~lliamR. NASA Sounding Rockets, 1958-1968:A Historical Summary. NASA SP-4401, 1971.

Wells, Helen T., Susan H. Whiteley, and Carrie Kategeannes. Origins of NASA Names. NASA SP-4402, 1976.

Anderson, Frank W., Jr. Orders of Magnitude: A Histoy of NAG4 and NASA, 1915-1980. NASA SP-4403,
1981.

Sloop, John L. Liquid Hydrogen as a Propulsion Fuel, 1945-1959. NASA SP-4404, 1978.

Roland, Alex. A Spacefaring Peopk Perspectiues on Early Spaceflight. NASA SP-4405, 1985.

Bilstein, Roger E. Orders ofMagnitude:A Histoy of the NAG4 and NASA, 1915-1390. NASA SP-4406, 1989.

Logsdon, John M., editor, with Linda J. Lear, Jannelle Warren-Findley, Ray A. Williamson, and Dwayne A.
Day. Exploring the Unknown: Selected Documents in the History of the US. Civil Space Program, Wume I,
Organizingfor Ejcploration. NASA SP-4407, 1995.

Logsdon, John M., editor., with Dwayne A. Day and Roger D. Launius. Exploring the Unknown: Selected
Documents in the Histoy of the U S . Civil Space Program, VolumeIL Relatiom with Other Organizations. NASA
SP-4407, 1996.

Logsdon, John M., editor, with Roger D. Launius, David H. Onkst, and Stephen J. Garber. &loring the
Unknown:Selected Documents in the Histoy of the US. Civil Space Program, VolumeZZL Using Space. NASA SP-
4407, 1998.
Logsdon, John M., general editor, with Ray A. Williamson, Roger D. Launius, Russell J. Acker, Stephen J.
Garber, and Jonathan L. Friedman. Exploring the Unknown: Selected Documents in the History of the US. Civil
Space Program, Ihlume N Accessing Space. NASA SP-4407, 1999.

Logsdon, John M., general editor, with Amy Paige Snyder, Roger D. Launius, Stephen J. Garber, and Regan
Anne Newport. Exploring the Unknown: Selected Documents in the History of the US. Civil Space Program,
Ihlume Exploring the Cosmos. NASA SP-2001-4407, 2001.

Siddiqi, Asif A. Challenge to Apollo: The Soviet Union and the Space Race, 1945-1974. NASA SP-2000-4408,
2000.

Monographs in Aerospace Histo y, NASA SP-4500:

Maisel, Martin D., Demo J. Giulianetti, and Daniel C. Dugan. The History of the XV-15 Elt Rotor Research
Aircraj: From Concept to Flight. NASA SP-2000-4517,2000.

Jenkins, Dennis R. Hypersonics Before the Shuttle: A Concise History of the X-15 Research Airplane. NASA SP-
2000-4518, 2000.

Chambers, Joseph R Partners in Freedom: Contributions of the Lungley Research Center to US. Military Aircraft
in the 1990s. NASA SP-2000-45 19,2000.

Waltman, Gene L. Black Magic and Gremlins:Analog Flight Simulations at NASAs Flight Research Center. NASA
SP-2000-4520, 2000.

Portree, David S. F. Humans to Mars: F& Ears ofMission Planning, 1950-2000. NASA SP-2001-4521, 2001.

Thompson, Milton O., with J. D. Hunley. Flight Research: Problem Encountered and What They Should Each
US.NASA SP-2000-4522,2000.

Tucker, Tom. The Eclipe Project. NASA SP-2000-4523,2000.