[169] Although not given much thought at the time, our opportunity to consider different ways of exploring a planet and to plan to do so from scratch was unique. Until our generation, the only planet men had been able to explore was Earth, and its exploration had begun from a local point of view, by walking in ever-widening circles. After a time, men viewed valleys from mountaintops and finally saw both mountains and valleys at the same time from aircraft, but never from the perspective that would have been afforded space travelers arriving from elsewhere in our solar system.

After several years of maturing experiences, I am amused to recall our blase approach to planning the first missions to the Moon and planets. As engineers and scientists, we had confidence in our abilities and the technologies available, and we simply set about planning to do the things that needed to happen if we were to achieve our goals. Occasionally it would occur to me that we were very lucky to belong to the first generation having such an opportunity, but these thoughts were always fleeting and replaced quickly by the demands of tasks at hand.

No matter how one addressed the question of exploring a distant planet, the first requirement was to get closer. The flyby mission was enough of a challenge at first, and that mode occupied us fully for a time. The benefits of orbiting to extend the period of observation and to increase coverage were recognized as next-generation extensions of the flyby mode: landings were ultimately needed to assess the nature of the surface and features of the planet

Making simultaneous observations from orbiting and landing spacecraft was obviously a desirable means of multiplying returns. The combination of synoptic views from orbit and the detailed information obtained at a specific site on the surface would allow broader interpretations of the "global" properties and provide insight for the interpretation of point information.

[170] The Surveyor program was originally conceived as combinations of orbiters and landers, but because the Moon is a relatively "dead" body, there was no great need for orbiting and landing missions to occur at exactly the same time. Conditions on the planet Mars, however, are dynamic, with frequent changes in atmospheric conditions, changing seasons, frequent dust storms, and frost-covered polar caps; thus, obvious benefits could accrue from concurrent observations. There was another powerful reason for the simultaneous launch of orbiter-lander combinations to Mars and Venus. Because of the roughly 2 years between launch opportunities, complete information would be obtained much sooner if orbiters and landers could be dispatched at the same time.

However desirable, the first opportunity to plan for such a concept came with a program named Voyager. Early studies for Voyager began in 1962; however, it was not until the 1965-66 period that program approval was obtained. The Voyager project plan envisioned the development and use of an orbiter-lander spacecraft combination having broad capabilities for conducting missions to Mars and Venus at several opportunities. The long-range objective was to allow systematic exploration of these planets with two launches per opportunity, using production-like hardware in a manner similar to that being applied to missions to the Moon. Initially the Voyager launch vehicle was to be a Saturn lb integrated with the Centaur upper stage, a development thought to be relatively straightforward, since both vehicles already existed and were seemingly compatible. As no other planned uses for this vehicle combination existed, it would have been dedicated exclusively to Voyager missions. After a while, the undesirable economics of this exclusive use situation contributed to an alternate decision to adapt Voyager to the Saturn 5 vehicle that was already being used for Apollo. It was a larger and more costly vehicle, but by this time it was well along in its development, and being advocated as a production vehicle for long-term use, making it more attractive for the long-term Voyager program than the Saturn lb/Centaur.

Those of you familiar with the 1980s achievements of the Voyager spacecraft that have successfully flown by the planets Jupiter and Saturn may be wondering how Voyager was transformed from a Venus-Mars program to an outer planets program. Perhaps at this juncture it is well to explain that the original Voyager program was canceled before it really got going, and that the name Voyager was later given to a Jupiter-Saturn flyby program that had been identified for a time as the Mariner-Jupiter-Saturn, or [171] MJS, project. A name change was ordered by NASA Headquarters to portray the increase in scope over earlier Mariner-class missions, and since Voyager had been dropped some 20 years earlier, it was decided that the name could be used again. In spite of some JPL concerns that the new program might be tainted with the handed-down moniker from an unsuccessful effort, the program turned out well, as described briefly in a subsequent chapter.

In reviewing the setting for Voyager mission planning, it is helpful to recall the impact of Mariner 4 results obtained during the 1965 flyby of Mars. Like its Mariner 2 predecessor, Mariner 4 was the second of a pair of modest spacecraft launched by an Atlas/Agena, with Mariner 3 suffering a fate similar to that of Mariner 1. This time it was the shroud atop the launch vehicle that caused the heartbreaking failure and not the vehicle itself, but the results were the same. Mariner 4 made the long trip to Mars in late 1964 and into 1965, gallantly photographing the planet with a television camera and returning the pictures at the painfully slow rate of 81/3 bits per second.

A most significant finding from the close-up Mariner 4 pictures-only 21 and a fraction were taken-was the fact that Mars appeared a lot more Moon-like than Earth-like. To the surprise of everyone, including the scientific community, Mars was found to be heavily cratered, with no evidence in any of the photos of the canal-like features that had been envisioned from astronomical observations using telescopes.

After this revelation by Mariner 4, reasons for the large number of craters were readily forthcoming, yet searches of the scientific literature revealed only a few brief inferences by scientists that such might be expected. During the planning for Mariner 4, I had never heard any suggestion from our scientific advisors that they expected Mars to be covered with craters. This incident made me a little more wary of the profound projections some scientists were prone to make; several of the investigators' reports gave the impression that they, too, were somewhat humbled by the oversight.

The dashed hopes of finding "little green men" was devastating to the support for Mars exploration-especially from administrators and members of Congress. While Mariner 4 results were also disappointing to those directly involved in the "business," some of us felt that the evidence was so scant that we surely ought to conduct a more thorough search before writing Mars off as a dull, lifeless planet. Accordingly, our determined pursuit for approval of more Mariner missions continued while planning began for the Voyager-type program.

[172] Three burning questions were formulated concerning the matter of life the planet:

Is there life of some form on Mars now?

If not now, has there been in the past?

Could the planet become habitable for life as we know it?

In addition to these life science questions, there were many valid scientific questions about the planet that had not been answered by the Mariner 4 findings-details concerning its body properties, its atmosphere, its mysterious polar caps, whether volcanic activities existed, and the like. As discussions about Mars were again stimulated, a new wave of excitement arose in the scientific community which also began to infect others who had lost interest in Mars after Mariner 4.

Voyager flights were not planned to begin until the early 1970s, so we boldly continued to work toward Mariner Mars missions in 1969 and 1971

The 1969 flights were to be more sophisticated flybys, and in 1971 we hoped to orbit the "red planet," producing maps and other data that would be useful for planning Voyager missions. As it turned out, these Mariner missions were extremely important, and a slight digression is warranted to explain why.

From my viewpoint, the evolutionary advances in mission capability afforded by the smaller Mariner-class spacecraft were more logical steps than the "order of magnitude," scaled-up efforts required to develop and operate Saturn-launched Voyager spacecraft. I was worried that we would find ourselves with all our eggs in one basket-with higher risks and with financial, management, and organizational challenges much harder to control. As a matter of fact, technological improvements in the Mariners had already made them appear capable of addressing the most immediate scientific questions, at least until after we had been able to conduct orbiting missions to observe and map most of the planet.

I made my views known, but the enthusiasm of senior NASA officials for proceeding to large spacecraft and Saturn-class vehicles overswept my more modest ambitions, and I found myself spending more and more time organizing the large-scale Voyager effort. From the outset it was obvious that the program would require a coordinated effort of several field centers. At this time JPL was very busy with the Ranger, Surveyor, and Mariner programs furthermore, for Voyager we were talking about launchings of multiple [173] spacecraft on a production basis. For a program of this scale, it was decided that the effort should be organized along the lines of Apollo, with a Headquarters program office and a NASA project manager reporting directly to a Headquarters program director. This decision disappointed JPL because it meant that their organization would not be eligible for the same type of lead role they had performed for Mariners, Rangers, and Surveyors. They clearly were to be involved as principals-no other center had as direct experience in planetary programs as JPL, plus direct responsibility for many of the key facilities. However significant the JPL role might have been, the new management concept for Voyager precluded JPL's being given project management responsibility.

The way the program office was finally established, my title was changed from Director of Lunar and Planetary Programs to Director of Voyager and Acting Director of Lunar and Planetary Programs. Donald P. Hearth, who had been Chief of Supporting Research and Technology for Lunar and Planetary Programs, was named Acting Project Manager and quickly began setting up a project office using a cadre of experienced JPL engineers and scientists. A floor of a new bank building was rented in downtown Pasadena, as JPL did not have space that could be dedicated to a new organization of the size required for Voyager. Don began to spend most of his time in Pasadena, but he "commuted" from Washington and never made a permanent move.

Although I became engrossed in Voyager, I was very pleased to be allowed to keep responsibility for directing all other lunar and planetary programs I was in the thick of things with the ongoing Lunar Orbiter, Surveyor, Mariner, Pioneer, Apollo Science, and related activities, and it would have been a major blow to give those up after so much had been put into them. Had I been required to make a choice between continuing to direct those programs or Voyager, I would have opted to stay with the several smaller programs, even though being Director of Voyager was more prestigious.

The initial plan to use the Saturn lb/Centaur vehicle for Voyager called for two launches at each opportunity of a 2000-pound orbiter and a 2200-pound landing capsule combination. The 1965 decision to change to the Saturn 5 meant that one launch could be made at each opportunity, with a single Saturn s carrying two Voyager orbiter-lander combinations having a gross weight of 62 700 pounds! Stacked on one Saturn would have been two orbiters, two landers, two surface science laboratories, and all the attendant [174] retro motors, entry capsules, and ancilliaries. Such a mission would have truly been an exploratory expedition, launched simultaneously on a single rocket. This would indeed have put a lot of eggs in one basket!

In beginning the Voyager program, Headquarters had to give the major field centers their project assignments. To ensure that proper attention was paid to this matter, I organized a Voyager Board of Directors with the blessings of Ed Cortright and Homer Newell. A general management plan had been roughed out, prescribing program direction by our management team at Headquarters and in Pasadena, with major center participation by JPL, Langley, Marshall, and Kennedy. The directors of these centers, William Pickering, Floyd Thompson, Wernher von Braun, and Kurt Debus, were very cooperative in agreeing to serve on the Voyager Board with Don Hearth and me.

Our understanding was that we would meet quarterly to establish organizational relationships and to develop guidelines for all major activities. I knew that such a beginning relationship with the center directors meant we would get the right kind of people assigned. With regularly scheduled quarterly meetings, we would also have a good means of reviewing progress and, if necessary, dealing with problems. The first Voyager directors meeting was held at NASA Headquarters on April 27, 1967, and the second was 3 months later at JPL. The board started off well, and it looked as if the Voyager program had everything going for it shortly after it was officially initiated.

A Lunar and Planetary Missions Board was also established at about the same time through the National Academy of Sciences to provide advice concerning the science activities related to the Voyager missions. The group was chaired by Harry Hess, a renowned geologist from Princeton, and included a "Who's Who" list of scientists from astronomy, life sciences, geology, radio astronomy, and biology from around the country.

Unfortunately, all this administrative and scientific support for the program was not enough. In the summer of 1967, shortly after the management and planning efforts were established, the Voyager program was dealt a death blow when Congress pared it completely from the NASA budget. The problem was not so much sentiment against Voyager per se as a generally perceived need to stop what some considered a runaway budget situation, making this large new program a target for a major reduction. Everyone involved fought to save the program, but by September it became clear that appropriations would not be forthcoming to sustain the momentum of.....

	.	Characteristic	Baseline	Typical growth potential
			1973	1975	1977/1979
		.
		Spacecraft bus/orbiter	2,500	2,500	2,500
		Capsule	5,000	6,000	7,000
		Propulsion	13,000	14,000	15,000
		Total (one planetary vehicle)	20,500	22,500	24,500
		Net injection weight (two planetary vehicles)	41,000	45,000	49,000
		Shroud/adapter	9,300	9,300	9,300
		Project contingency	5,000	3,700	2,700
		Gross injected weight	55,300	58,000	61,000

....Voyager. The political climate was such that support simply could not be mustered for the combined requirements of Apollo and a large planetary program. Continuing to fight for Voyager would clearly have compromised our support for other important commitments, with no guarantee of success. We bit the bullet, closed the project office at JPL, and disbanded the Board of Directors.

Our struggle to maintain continuity for planetary exploration through Mariner-class missions succeeded, with only one failure in the remainder of the series. Mariner 5 successfully performed a flyby mission to Venus in 1967, Mariners 6 and 7 made flybys of Mars in 1969, Mariner 9 orbited Mars in 1971, and Mariner 10 flew by both Venus and Mercury in 1974. These Mariners effectively bridged the gap that would have developed if they had been abandoned in favor of the Voyager effort. It is my view that the total planetary program turned out well, taking into account the fact that the Mariners and the Viking replacement for Voyager were complementary, affording the scientific community a meaningful basis for continuing study through several planetary opportunities.

Mariners 6 and 7, both successful flybys, clearly showed their superior technology over Mariner 4, although their design was begun with the ground rule that the spacecraft would be the same and only the scientific instruments upgraded. Unfortunately (or fortunately), the best way to upgrade the science return required technological advances in the spacecraft. The most [176] notable improvement was in the communications bit rate-increased from the paltry 8 1/3 bits per second of Mariner 4 to 16 200 bits per second. In addition, the data subsystem was upgraded markedly with two specially designed tape recorders to meet requirements 35 times as great as those of Mariner 4. Improvements were also made in the telemetry subsystem: the remarkable scan platform had the ability to adapt to changing requirements as well as to accommodate modifications in the second flyby mission, thus allowing the two spacecraft to perform complementary rather than repetitive roles.

The greatly improved images of Mars obtained by Mariners 6 and 7 provided many surprises. Our concept of Mars changed again, from a barren Moon-like planet that appeared lifeless to a more Earth-like body having many types of terrain, clouds, variations in atmosphere, and evidence of erosion, strongly suggesting that water had once been abundant. These findings led to a revitalization of interest in Mars as a place where life had been harbored at some time, if not in the present. This interest was quickly shared by scientists, with administrators and politicians becoming advocates as well. The result was continuing support for the orbiting missions of Mariners 8 and 9 planned for the 1971 opportunity. If the planet could not be surveyed concurrently from orbit and on the surface, at least the next most vital steps, conducting orbital surveys, preparing maps, and allowing more sophisticated planning for landing site selection, could proceed.

After a string of successes, Mariner 8 became just another statistic as a result of launch vehicle failure. The Atlas performed well, and powered flight proceeded normally until shortly after separation and ignition of the Centaur stage. At that point, a pitch control problem in the Centaur flight control system allowed the stage to tumble and shut down. This disheartening loss was followed by the usual reviews, modifications, and adjustments, but these were completed in time for Mariner 9 to be launched successfully.

In keeping with the general goals for planetary exploration, Mariners 8 and 9 were to provide detailed photographic surveys of the planet at much higher resolution than ever before. Special studies were to be made of the so-called "wave of darkening" along the edges of the polar caps, including measurements related to temperatures, surface composition, the presence of water molecules, and the existence of other conditions generally relevant to the question of life.

In late September 1971, astronomers who were keeping a watch on the planet saw a bright yellow cloud forming in the southern region known as Noachis. Dust storms had been seen on Mars before, but this one was of [177] special interest' as Mariner 9 was to experience it from nearby. When this storm was in its fifth week, it peaked-apparently worse than any that had ever been observed both in area and duration. By November 14, when the spacecraft was placed into orbit by firing its retro rocket, the worst of the storm had subsided, but only five distinct surface features could be identified. Conditions definitely were not those envisioned when the mission was planned, so plans had to be changed.

Fortunately, Mariner had the capability for reprogramming, a highly desirable feature for an exploratory mission, made possible by improvements in technology. Actually, observing the changes as conditions improved provided much new insight, and since Mariner 9 was viable for more than a year (the design goal was 3 months), a thorough study of this Martian dust storm was possible.

One of Mariner 9's revelations was a giant volcanic mountain, named Olympus Mons, and an almost unbelievable canyon system, far larger than Earth's Grand Canyon, named Valles Marineris in honor of its Mariner discoverer. Of course, the multiple-orbit imaging coverage provided by the long-term mission allowed cartographers to prepare detailed maps of Mars and provided scientists with several types of data for speculation about conditions on the surface.

Not long after the termination of the Voyager project, a new landing mission concept was born from the ashes. Advanced technology work had been continuing for several years on capsules designed to survive a hard landing; results were encouraging to those who hoped to obtain important data from the surface of Mars. In addition to the scientific stimulus, there had always been broad support for landing on Mars; this was, after all, a clear milestone in the space race that the Russians had been trying to achieve for a long time, if only for its propaganda value.

The new mission required a new name to give it a fresh start and to distinguish it from Voyager. Viking was the name chosen, and the first flights were proposed for the launch opportunity in 1973. The Viking program, proposed to be a bargain at only $364 million, was initially conceived to involve a Mariner-derived orbiter and a simple, hard-lander spacecraft. Congress approved the project in 1968, but it soon became apparent that funding and the scope of the mission did not mesh. After the grandiose studies and planning that had been done toward Voyager, we experienced difficulty in scaling down. Matters were made worse by a strong desire to make a quantitative advance beyond the 1971 orbiting missions, requiring [178] the landing spacecraft to have a greater capability than a hard-landing capsule. After a good bit of trauma and failure to match requirements and available funding, launch was postponed from 1973 to 1975, with a continuing program for development throughout the 2-year interval. Many of us were disappointed to pass the 1973 Mars opportunity, for the planetary orbit geometry of Mars and Earth at the time would have allowed the largest payload for a given sized launch vehicle for years to come. But as it turned out, the Viking missions in 1975-76 were probably better for the delay.

The postponement was considered to be a mixed blessing. The total project cost had to go up because people were kept at work longer; the delay provided opportunity for better planning and application of new technologies. Different management arrangements were worked out, borrowing from earlier project experiences and from the concept established for managing Voyager. At this time it was agreed that project management for Viking would reside in the field. Although I had accepted the plan for Voyager and had been named to direct the program from Headquarters, I never did think this was as sound in concept as making a field center responsible. The reason was simple: a manager needed a qualified staff at his fingertips to deal with management problems, and I did not think that we could ever assemble such a team at Headquarters. The Apollo program had been managed that way, but the Apollo Headquarters management team depended on Bellcomm, Inc., a complete systems organization under contract, which we could not have for Viking. Although Apollo was a successful program, I was never convinced that it could not have been managed successfully by a field center along the lines employed for lunar and planetary programs.

Viking missions were based on using the Titan 3C/Centaur launch vehicle instead of the Saturn, so direct involvement of Marshall Space Flight Center was no longer required. Lewis was responsible for Centaur development and for integration of the Titan 3C with the Centaur. Thus, Lewis was the obvious choice to manage the launch vehicle system. Either JPL or Langley might have been chosen for the project management center assignment, but three factors favored Langley: (1) Langley was truly a NASA center and not a "contractor" operation that, at the time, was somewhat out of sorts with NASA Headquarters, (2) Langley had successfully completed the Lunar Orbiter project and had a ready team with no other project assignment at the time, and (3) Langley had a strong research capability to back up development of a new landing vehicle. The landing craft was to be built by a [179] contractor, but the engineering and testing of landing systems, including entry aeroshell, parachute, and landing gear, nicely fitted the Langley background.

JPL was the obvious choice to manage the Viking orbiter system as well as the Deep Space Network. The Mariner-derived orbiter was to be an upgraded adaptation of a Mariner bus designed to transport the lander to Mars and provide injection into orbit. JPL would also play a vital role in space flight operations, as they were responsible for the Space Flight Operations Facility, where mission operations were conducted. However disappointed some JPL members may have been, there was no evidence of bitterness as they turned to the tasks assigned and performed them admirably.

By the time Viking began, there had been enough launches of satellites and interplanetary spacecraft that many people tended to think of Viking as just another, slightly more sophisticated mission, using existing technologies. Actually, this was not true, for the Viking project elements, including both hardware and software, were an order of magnitude more complex than anything that had gone before.

Perhaps the simplest way to explain this premise is to describe the Viking spacecraft just prior to launch. A launch vehicle manager might, out of habit, refer to it simply as the "payload" awaiting launch atop his rocket, but it was really a combination of four spacecraft, each with a different function and purpose. Completely separate yet tightly integrated entities were an interplanetary bus, an orbiter, an entry capsule, and a lander.

For transporting instruments and equipment to the vicinity of the planet there was the "bus," an interplanetary vehicle with attitude control, thermal control, power supply, communications link, midcourse correction capability, and all the systems required for a Mariner flyby mission. To perform the retro maneuver at the planet a relatively large rocket motor was required that could survive the long transit period of the transfer orbit and then be controlled precisely to inject the spacecraft into a preselected orbit.

After serving as an interplanetary spacecraft and injecting into Mars orbit, an additional duty of the bus, now an orbiter, was to serve as a launch platform for the entry capsule. This required precise attitude orientation, timing' and separation signals for ejecting the capsule so that it would enter the atmosphere and descend toward the surface. After this, the orbiter would observe Mars from orbit, in much the same way that an orbiting Earth resources satellite might observe our planet. One continuous and very [180] important supporting function for the orbiter throughout the mission was its role as a communications relay satellite for transmitting data from landers to Earth.

In the early days of missile development, learning to design and build vehicles capable of reentering Earth's atmosphere had been a significant technical challenge. To do the same thing for another planet with an atmosphere far less defined than that of Earth was the challenge faced by Viking entry capsule designers. Apollo and Viking had generally similar design requirements for an aerodynamically stabilizing shape, thermal protection from heating, and base structure and retro motor integration; however, the Viking entry capsule also had to deploy a landing spacecraft at the proper time without damaging its complex equipment and appendages.

A specially developed parachute system was carried to slow the descent for landing; after decelerating the entry capsule to about twice the speed of sound, further use of atmospheric drag was thus made. The parachute system demanded technological developments beyond those being used to return sounding rocket payloads to Earth because of the different atmosphere and approach conditions on Mars. The last official duty of the parachute system was to pull the aeroshell base structure away from the lander spacecraft so that it could extend its landing gear and prepare to land.

In addition to its engineering tasks, the entry spacecraft provided for scientific measurements during its passage through Mars' tenuous atmosphere. Data were collected and transmitted to Earth; thus, the Viking entry system also provided for in situ examination of the unknown Mars atmosphere. This alone was the equivalent of a sounding rocket mission into Earth's atmosphere.

Because attention was focused on the activities of the orbiter and lander spacecraft, the achievements of the entry spacecraft, its parachute, and complex systems were largely unheralded; a few years earlier, these would have been regarded as very significant. Of course, had any components of the entry systems failed to work, their importance to the success of the entire mission would have been painfully obvious.

Most people would recognize the lander spacecraft as a major design challenge, although by the time Viking was being designed the Surveyors had removed some of the doubts about the technical feasibility of developing such spacecraft. Nonetheless, designing landing spacecraft for Mars very nearly required starting from scratch. A major factor was Mars' atmosphere, for during the landing and touchdown phase, aerodynamics had to be considered for stability and control as well as rocket performance. This was not [181] a requirement for landing spacecraft on the airless Moon. Mars also has a significantly greater gravitational pull to overcome than does the Moon, so all-new requirements existed for engineering design. And, even though the attitude stabilization, Doppler radar, and retro rocket systems were generically similar to those of Surveyor, they were different enough to demand special detail in their design. An entire book could be written about the Viking landers and their almost unbelievable qualities; to discuss them in a few paragraphs is almost an injustice to those who should be credited for the design, development, and operation of these magnificent, self-sufficient, automatically controlled yet responsive machines.

The first duty of the spacecraft was to automatically land safely on the unknown surface of Mars without damaging any of the precious cargo of scientific instruments. To do this it had to determine how far above the surface it was, adjust descent and lateral velocities so as to touch down within prescribed limits, and then shut off the rocket motors at precisely the right speeds and altitudes. Because of the 20-minute lag in communications between Mars and Earth at the time of landing, the spacecraft's makers on Earth were absolutely no help in performing the real-time activities necessary to successful landings. I clearly recall discussions in the Space Flight Operations Facility at JPL during the period when we knew that either the landing had been done successfully or the lander had crashed, as we anxiously awaited data that would tell us what had happened. In some respects this was like watching a TV replay to learn the outcome of a sporting event that had already been decided.

After landing, the spacecraft became a science laboratory extraordinaire. It was at the same time a weather station, a geophysical observatory, a life sciences chemistry lab, a remote materials manipulator and processor, a data acquisition and processing station, and a data transmitter. It had its own power supply in the form of two radioisotope thermoelectric generators that used plutonium 238 to provide 70 watts of continuous power. It also contained a computer-centered "brain" called a guidance, control, and sequencing computer (GCSC), which could contain up to 60 days of instructions. Of course, the memory could be modified or updated from Earth when changes seemed necessary, but the spacecraft could easily take care of itself during the 12-hour periods when it was out of sight of Earth because of the rotation of Mars.

Because a major goal of Viking missions was the search for life, it was essential that Viking landers not take any form of life to Mars. Thus, the spacecraft had to be sterilized after they were built and tested. To achieve the [182] prescribed degree of sterilization necessary to satisfy internationally established planetary quarantine requirements, lander spacecraft were sealed in their bioshields and baked at temperatures above 113° C for about 24 hours. This had been shown to be adequate to ensure that the chances were less than 1 in 10 000 that a single organism would be transported to Mars from Earth.

Heating for the purpose of killing living organisms also produced risks to lander hardware, especially to electronic components. In order to plan for this last-minute treatment before launch, a great deal of component and materials testing had to be done before selections were made in the design process. Even so, this exposure to unusually high temperatures was made with a certain amount of concern about its effect on the lifetimes of critical components. Many lingering fears remained after the sad experiences with Ranger that were believed related to sterilization requirements.

The science instruments chosen for Viking lander spacecraft were selected very thoughtfully in accordance with major mission priorities and the state of the art in instrumentation technology. Not only was it critical that each instrument be capable of making contributions to knowledge on its own, but most instruments had to become components of a laboratory-like complex. Findings could be expected to be mutually reinforcing, such that the whole would be greater than the sum of the parts. In some cases, a component intended primarily for a scientific purpose also served a supporting function in another scientific investigation.

The choices for meaningful experiments were many; the final complement of Viking instruments was believed to address the highest-priority questions about Mars. There were cameras to see and observe as an inquisitive explorer would have done; meteorology sensors to measure and record the atmospheric conditions and report on the weather; "tools" for scratching the surface and for quantifying the physical properties of the soil; experiments to determine chemical constituents, mineral content, and composition of the soil and atmosphere; and, very importantly, there were three ways of measuring biologic activities that would answer burning questions about life on this neighboring world.

Of all the scientific instruments that have been carried into space, none are more appealing to most of us than cameras. Through our eyes we see things for ourselves; through the cameras onboard Viking our eyes were allowed to sense the mystery and beauty of this distant world as if we were there. The cameras used in the Lunar Orbiter were sometimes referred to as....

...."large Brownies," simply because they functioned much like the hand-held cameras seen around us every day. The other cameras commonly used in space were video or television cameras, somewhat less familiar at the time, but now commonly in use. Viking lander cameras differed from both of these [184] in several respects. They employed engineering principles that had been used before, but the method of implementation was different. They were called "facsimile" cameras because of the way they viewed and reconstructed scenes.

In principle, an image can be constructed by a light sensor that sees "elements" one at a time. If the elements are viewed in a row or line, and stored or transmitted electronically, they can be reconstructed as elements having the same intensity. Elements reconstructed and placed in contiguous lines combine to become whole images. Mariner 4's pictures of Mars were comprised of 200 TV lines with 200 picture elements per line (called pixels)-each represented by 64 shades of gray. Over 8 1/2 hours of transmission time was required to return the data for a single picture.

In talking about the process of data return and picture reconstruction, we jokingly discussed the possibility of putting up a large billboard with 40,000 nails on which to hang small square coupons. With a supply of coupons representing the 64 gradations, it would have been possible to hang the numbers in place as they were returned by telemetry so that a picture would be revealed, almost as if painting by number.

A process close to this actually materialized, as the numbers representing shades of gray were printed out sequentially on paper ticker tape. The columns of numbers representing a vertical strip 200 pixels long were then stapled side by side on a piece of beaver board, and colored crayons were used to color corresponding shades of gray. The result was a false-color image showing the edge of the planet in some detail, as well as the varying intensity sky above.

This historic picture was later framed and hung in the JPL Director's office area-a fitting memento of the first successful close-up imaging of the planet Mars. The display is now a museum piece, and destined to be of significant interest to future generations.

The Viking camera made use of light detector, lens, and mirror systems to perform a linescan. A nodding, rotating mirror allowed successive sweeps to reflect an image of the surroundings into the lens. Twelve detectors, three of which had color filters of red, blue, and green, allowed selective images to be recorded and reconstructed in color. By electronically recording the varying intensities of reflection, linescans were converted to digital signals that could be transmitted directly or stored in memory. With simple indexing of position and movement, contiguously placed reconstructions of each line became an image or "picture" fashioned from the composite bits of data. [185] For the facsimile principle to work, the features being imaged had to remain stationary long enough for the scanning process to be completed. Objects moving about as the facsimile scanning process occurred would not have been seen, except perhaps as small disturbances in line elements. Camera developers jokingly suggested that it was possible that a mobile Mars creature might have moved across the field of view of the Viking cameras without detection; however, since there were no signs that this had happened, even those hopeful of such discoveries had to be skeptical. Although the facsimile principle may seem rather simple, the Viking camera represented a significant advance in the state of the art at the time and was by no means a simple instrument.

Those who followed the Viking mission closely will recall that the first pictures released to the press showed Mars as a very red planet. Unfortunately, this portrayal was generally in keeping with scientific speculation, and 2 days later, when the image data had been thoroughly calibrated, red-faced NASA officials had to tell the press that there had been a slight misrepresentation. For several hours the ground reconstruction process was recalibrated and equipment adjusted; after this was done, some of the redness was reduced. In all fairness, the premature release was probably due to the terrific pressure produced by the desire to share findings with the public as quickly as possible, before completing the data processing checks known to be required.

In a recent discussion with Cal Broome, who had project responsibility for camera development, he indicated that his fondest memories from Viking were of the camera developments and the products they provided. He vividly recalled the experience of viewing the first picture and proudly took the position, "As far as I'm concerned, that's what Mars looked like that day."

He also recalled the trauma that resulted when the all-electronic scanning camera proposed by the Itek Corporation was being considered. A large amount of development effort by Aeronutronics, conducted during the Ranger project, had produced a successful facsimile camera that had been fairly well proven, involving both mechanical systems and electronics in its operation. While it appeared that the Ranger camera would have provided the necessary basic capabilities, it was neither as versatile nor as capable of electronic programming and selective applications as promised for the new concept. In reflecting on the situation, I believe this was simply an example of progress being made so rapidly in fast-moving technologies that excellent concepts became obsolete before they could be used. Regardless, it was the [186] judgment of those responsible for Viking that the advantages of the advanced system outweighed the risks involved in its development, and the Itek concept was selected. This belief was justified by the fact that the electronic scanning cameras worked well. When contact was lost with Viking Lander 1 in 1983, its cameras were still working after 7 years without a glitch.

In his excellent book The Martian Landscape, Tim Mutch, Team Leader for the Imaging Science Team, described in a clear and fascinating way the tradeoffs and other aspects of establishing camera design characteristics. When you and I choose a camera from the marketplace, we have no choice but to select from concepts generated by designers who decided what the public would buy. However, in the case of camera design for Viking, the team was able to establish requirements from scratch and iterate them against existing technologies. The most fundamental choice was resolution, the definition of image size for the smallest element to be seen. Selecting the smallest detail to be resolved also implied a maximum field of view, for such was the nature of the tradeoffs. According to Mutch, many of these tradeoffs had been studied for years by Fred Huck, an engineer at the Langley Research Center. With the collaboration of Glenn Taylor, also of Langley, the team was able to examine all the variables of camera performance, including those dictated by spacecraft constraints such as weight, power, and bit rate, and arrive at a balanced design for the hypothesized mission to Mars.

Superficially, the operation of the cameras seemed remarkably simple. The photosensor array and all the electronics that processed the points of incoming light were clustered in a small assembly only 3.4 centimeters (1.3 inches) across. Twelve photodiodes, each able to obtain image data, were mounted so that different focal lengths could be achieved. Some of the photodiodes were equipped with filters of red, blue, and green to permit recreation of color images.

A slot near the top of a small cylinder formed the "pinhole" window through which a small nodding mirror could peer. As the mirror nodded around a horizontal axis, it swept a vertical line, scanning reflected light from the objects in view, while electronic circuits recorded intensities. Five times a second the small cylinder was rotated so that the slot position allowed a new vertical line to be scanned. Indexing for these vertical lines and the timing for the nodding mirror had to be precisely controlled so that each pixel or picture element was contiguous. Actual positions had to be indexed to an accuracy of 0.01 millimeter-about one-tenth the diameter of a human hair--in order for the required resolution to be achieved.

[187] The capability of adjusting signal gains allowed images to be obtained and processed for various light levels, and the variety of photodiodes allowed a selection of amplification for either close-up or distant views. By simultaneously imaging the same scene with two cameras placed about 1 meter apart, stereoscopic views were obtained to permit three-dimensional viewing in the same way that our eyes perform. By any standards, these were remarkable cameras!

In a fashion typical of planetary missions, the launch vehicle used for Viking was specially integrated for this set of missions. The Titan 3 was a military vehicle originally developed by the Martin Marietta corporation for the Air Force. It included a two-stage core rocket system using liquid propellants, plus two large strap-on solid rockets. While not nearly as large as those used to help boost the Space Shuttle, these strap-on solids performed the same function of providing initial acceleration. They were 10 feet in diameter, and each produced about 1.2 million pounds of thrust for about 2 minutes. After burnout, they were jettisoned and dropped into the Atlantic Ocean.

The first stage of the Titan vehicle, also 10 feet in diameter, ignited just before the solids burned out for about 2 1/2 minutes. The second stage then separated and fired for 3 1/2 minutes. Both these core stages used a blend of hydrazine and unsymmetrical dimethyl hydrazine, with nitrogen tetroxide as an oxidizer.

The Centaur upper stage was basically the same General Dynamics-built liquid hydrogen, liquid oxygen rocket used for Surveyor. After separation from Titan, its two Pratt & Whitney RL-10 engines produced a total of about 30 000 pounds of thrust to send the spacecraft on its way. Its relight capability allowed the Vikings to be propelled into a 90-mile-high parking orbit until the right position around Earth was reached for injection into the transfer orbit The coast periods could vary from 6 to 30 minutes, depending on time of launch. After burnout, the final act of Centaur was to separate itself from the spacecraft and, by expelling its residual propellants, change its trajectory slightly so that it would have no chance of impacting and contaminating Mars. It then became a silent companion to Viking, slowly separating from the spacecraft as both objects coursed around the Sun in the general direction of Mars' orbit.

After the Voyager program was canceled, planning for Viking was begun in a very austere environment. The orbiter-bus was envisioned as a direct outgrowth of the Mariner '71 spacecraft, with a modest scale-up for the additional requirements of Viking. While actually resembling a Mariner and [188] benefiting greatly from its heritage, the Viking orbiter became an entirely different spacecraft. The propellant tanks, for instance, had to be roughly three times the size of those used to provide injection of Mariners 8 and 9 into orbit. The basic structure was enlarged to accommodate the lander aeroshell, and the solar panels were increased in size to provide more power. Over 15 square meters of solar cells supplied about 620 watts of electrical power at Mars, charging two 30-ampere hour nickel-cadmium batteries to be used when the cells were not in direct sunlight or when the spacecraft was oriented for pointing instruments or activating the capsule launch.

Another significant improvement included the addition of extra "brain power" to allow the orbiters to perform more complex functions. Viking orbiters possessed two 4096-word, general-purpose computers that could operate in parallel or tandem modes. These replaced the small special purpose computers contained in Mariners 8 and 9. The capability for more rapid picture taking allowed for better site surveys and special regional studies. This capability was augmented by tape recorder systems that could store 2.112 megabits per second, with a capacity of 55 TV pictures-over half a billion bits of information.

Viking orbiter communications systems used both S-band and X-band frequencies. A parabolic high-gain antenna, 57.9 inches in diameter, provided for the highly focused transmission and reception of radio energy to and from Earth. This antenna was backed up by a rod-shaped low-gain or omnidirectional antenna similar to the one on Mariner 4, so that no matter what the orientation of the high-gain antenna, communication at a low bit rate was possible. Orbiters also had relay antennas for receiving and transmitting signals to and from the Viking lander spacecraft; this allowed contact between Earth and the landers even when they were on the opposite side of Mars, provided they were in view of Earth.

Transmitter power for the orbiters was about 20 watts, allowing bit rates of 16 000 bits per second. While extremely small compared with the transmitter powers used by broadcast stations on Earth, this was about five times the power Mariner 4 used to provide a bit rate of 8 1/3 bits per second. Another significant factor was the development of very sensitive receivers and transmitters in the Deep Space Network, as highlighted by the huge 64-meter (210-foot)-diameter dishes.

While serving as the buses for transporting the landers to Mars, the orbiters had to serve as "hosts," providing the necessary power, thermal environment, engineering status, midcourse corrections, and attitude orientation [189] for capsule ejection. Of course, orbiters continued to serve the landers through the communications relay function after they reached the surface of Mars, but they really performed a major mission function on their own as scientific spacecraft. Their important scientific instruments included two television cameras for conducting site surveys and making maps and topographical studies, an atmospheric water detector, and a thermal mapper to allow studies of temperature variations and to look for hot spots. These alone were adequate justification for the orbiter missions, but in fact, these truly remarkable multipurpose spacecraft did the work of at least four special-purpose spacecraft.

As impressively self-sufficient as the Viking launch vehicles, landers, and orbiters were, three major systems that never left Earth were necessary to their success. These systems formed the connection between the people involved in the missions and the space machines. They were the launch facilities at Kennedy Space Center, the Deep Space Network (based at JPL but spread around the world), and the Space Flight Operations Facility at JPL, where mission operations were conducted.

Visions of the launch complex at Kennedy come to mind immediately; we have all seen television coverage of the gantrys and flame pits in action. Actually many more components-and even complete facilities-were just as vital to the launch operation. Although several were multipurpose, that is, they might also be used for other projects, most had to be especially adapted to Viking requirements. Orbiters were assembled in Building AO and mated with their propulsion systems in the Environmental Safe Facility. Landers were assembled in the Spacecraft Assembly and Encapsulation building, mated with the orbiters, and encapsulated in their heat shields before being moved to the launch pad, where they were mated with the Titan/Centaur launch vehicles. Many of these vital facilities that are sometimes taken for granted did not just happen to be in the right configuration at the right time, but the "heroes" who provided them will never be sufficiently recognized.

The Viking launch vehicle and spacecraft systems presented the most complex array of space hardware ever assembled for unmanned missions to another planet. The combined fleet of interplanetary, orbiting, entry, landing, and laboratory spacecraft that comprised the Viking expedition to Mars in 1975 and 1976 incorporated advanced technologies from almost every major discipline of science and engineering. Dedicated to a single set of goals, most of these automatons were programmed to function effectively with little human intervention; however, all were flexibly reprogrammable to [190] respond to requirements determined by human "masters on Earth. The Viking team, made up of humans and spacecraft, clearly proved that men and machines can work together in marvelous harmony, provided they are guided by common aims and a willingness to subjugate individual purposes to the greater good.

---