Home Table of Contents What's New Image Index Copyright ScienceViews Search


On Mars: Exploration of the Red Planet. 1958-1978

 
 
SCIENTISTS, INSTRUMENTS, AND SUBCONTRACTORS
 
 
 
[220] The Viking project stretchout affected management of the scientific experiments for the Mars mission. Originally, the Viking Project Office had planned to negotiate contracts with the scientists and select instrument subcontractors during the first weeks of 1970, and most of the science teams met in early January to review their plans. With the switch to a 1975 mission, that schedule had to be reevaluated and those activities reprogrammed. On 13 January the science teams, except those working on biology instrument and the lander imaging system, were told to terminate their Viking activities. 36
 
Jerry Soffen advised all of the scientists in late January that the Viking Project Office's main goal was to make the transition to a revised schedule as smooth as possible, while protecting against any unnecessary cost increases or further schedule delays. "During this transition period," Soffen hoped that the scientists would "not lose sight of the Viking objectives," and he reminded them that "scientific research has never been an easy way of life. We expect to find favorable aspects of this Viking deferment in the form of improvements in the investigations and the better use of Mariner 71 results." 37 The Viking Project Office worked out a procedure for keeping the science team leaders in the instrument definition process during the transition without having to include them in formal contract negotiations. After selection of a subcontractor to negotiate to build a science instrument and before negotiations began, a technical review would be held. Martin Marietta, the Viking Project Office, the science team, and the subcontractor (or "vendor") would thoroughly review the procurement drawing, especially where changes in specifications were required. The science team [221] leader could participate in discussions leading to prenegotiation specification. Then, during negotiations, any additional changes would be coordinated with the team leader through the Viking office. 38
 
For the scientists as a group, the next big gathering scheduled was the Viking science review in mid-April 1970. By that time, Martin Marietta had chosen Itek Corporation's Optical Systems Division to develop and build the lander camera system and was evaluating biology instrument proposals from Bendix Aerospace Systems Division and TRW Defense and Space Systems Group. JPL was in the process of evaluating a breadboard model of the gas chromatograph-mass spectrometer, and Martin Marietta's planning for the construction of the upper-atmospheric mass spectrometer breadboard was under way. 39
 
For three days, 13-15 April, 42 scientists (about two-thirds of the total team membership) met with representatives from the project office and lander contractor. After receiving reports from the Viking managers the first morning, each team leader presented a 10- to 20-minute summary report on the status of his experiment that afternoon. On the 14th, a series of concurrent team meetings gave the scientists time to talk with their teammates and discuss matters of common interest with other teams. Later that day, a number of special science meetings took up investigative considerations affecting more than one team, such as site alteration, organic contamination, landing site characteristics, atmosphere. The final day of the gathering was given over to a session of the Science Steering Group. The scientists found all the meetings educational but agreed that the smaller "think" groups they had participated in the second day were particularly stimulating. Viking's schedule may have been stretched out, but nearly everyone agreed that much work would still have to be done by all to meet the 1975 launch date. 40
 
The pace of work was moderately slow at first because of the limited money available, but in retrospect that may have been fortunate, because many technological problems lay ahead. Three scientific instruments-the ones given first priority for the dollars available-were particular problems: the gas chromatograph-mass spectrometer, the biology instrument, and the lander imaging system. * While the story of these instruments is a tale of amazing accomplishment, the facts also indicate that if Viking had flown in 1973 it probably would have been launched without the gas chromatograph-mass spectrometer and the biology instrument. Without those experiments, Viking would have been a vastly different mission. Those instruments were ready to fly in 1975, and the story of their design and fabrication deserves to be told. For the men and women who worked the extra hours, sweated out the successive problems, and reveled in personal [222] satisfaction when the experiments actually worked on the surface of Mars, it was "their" lander, "their" experiments, and "their" triumph.
 
 
Gas Chromatograph-Mass Spectrometer (GCMS)
 
 
Development of a GCMS prototype had initially been assigned to the Jet Propulsion Laboratory by Langley in August 1968. This responsibility remained with JPL when the Viking project was officially established. Before selecting a contractor to build the flight hardware, the California lab had the task of developing, fabricating, and testing a lightweight portable breadboard of the GCMS that could be used to carry out surface organic analysis by pyrolysis. Gas chromatography and mass spectrometry in the laboratory were one thing; shrinking the equipment to a size that could be placed on a spacecraft was another. 41 Requirements for such an instrument were not easy to meet for a laboratory model; restrictions put on the design to qualify it for spaceflight made it extremely difficult.
 
Pulverized Martian soil would be placed in the instrument and heated to temperatures up to 500°C. The gases given off would be carried into a gas column, a long tube packed with coated glass beads that would selectively delay the passage of gases according to their adsorptive qualities. The column would then be heated progressively to 200°C at a rate of 8.3°C per minute. Each level of temperature would release different organic molecules, separated into narrow family groupings. A palladium separator unit, porous only to hydrogen, would filter out that gas, leaving only the vaporized organic compounds, which would be drawn into the mass spectrometer to be ionized. The stream of ions would be focused in the electrostatic and magnetic sectors of the device. When the stream of focused ions struck the electron multiplier tube, generating electrical impulses, that activity would be amplified and recorded, producing a profile of each compound. Finally, the profiles would be converted into digital signals that could be transmitted to Earth. 42
 
Although the GCMS was a complex piece of equipment, no one predicted the difficulties that JPL encountered in its development. At first, dollars and failure to agree on priority for the instrument's development were causes for delay. But by the summer of 1970, serious engineering and managerial problems were plaguing GCMS development. 43
 
In September 1970. Cal Broome told Jim Martin that the GCMS, nominally under the purview of Henry Norris's Viking Orbiter Office, was a stepchild not getting proper supervision because of the decentralized management structure at the lab. 44 A five-day GCMS engineering model review, held 25-30 January 1971, was a disaster. Jack Van Ness told Langley Director Cortright that between 200 and 300 "request for action" forms resulted from the review; he anticipated that 100 to 150 of those items would be assigned to JPL for its attention. "It is expected that the major output of the review will be a critical reassessment of the requirements imposed upon [223] the instrument and its subsystems, with an eye towards reductions in instruments complexity." 45 Two weeks later, Van Ness reported that JPL had take steps to strengthen its managerial control. John J. Paulson, head of the GCMS project office, would henceforth report directly to Robert Parks, assistant laboratory director for flight projects. This shift put the GCMS on the same management plane as Mariner Mars 71 and Viking Orbiter. The Viking Project Office hoped this visibility would help solve some of the stepchild's troubles. 46
 
Jim Martin was not pleased. At a Science Steering Group meeting 2-3 March 1971, he indicated that funding increases, technical problems, and schedule slips had caused him and his colleagues considerable concern about the future of the GCMS. Although the recent management change at JPL was encouraging, the instrument's progress would be watched closely during the next few months. If progress was not satisfactory, Martin would have to consider an alternate or less ambitious design. 47 The project manager's attitude toward the GCMS difficulties was not enhanced by his unhappiness over the science subsystem preliminary design review at Martin Marietta on 1-2 March. The part of the PDR covering the science experiments integration laboratory (SElL), to be built in Denver, was particularly unsatisfactory. Martin told the lander contractor that the SElL PDR would be repeated and that no funds would be spent on equipment for that instrument until a satisfactory review had been held. 48 (The SElL was canceled in July 1971; instruments tests would be performed on the system test bed lander at Martin Marietta.)
 
On 18 March, the GCMS engineering breadboard was operated for the first time ass completely automated soil-organic-analysis instrument. Several problems of the kind usually associated with first tries were encountered, but everyone in the Viking Project Office interested in the development of the GCMS considered its major step forward. 49 Meanwhile, an ad hoc GCMS requirements review panel, established by Martin after the unsuccessful engineering model review in January, met to discuss possible ways of simplifying the design. ** Preliminary results of the ad hoc panel's study were presented at the June 1971 Science Steering Group meeting. Martin noted several discouraging facts at this session: by this date the start of GCMS science testing had slipped by six months (from early 1971 to October 1971); after four years of work the breadboard was just ready; and the GCMS was now getting too heavy. Originally projected to weigh about 9.5 kilograms, the GCMS was weighing in at about 14.5 kilograms. The ad hoc panel presented five GCMS design variants with weight projections between 11 and 14 kilograms, but they requested and were given more time to study the science impact of these alternatives. 50
 

[224] The development model, top left, of the gas chromatograph-mass spectrometer was the first step toward spacecraft hardware. After a breadboard model, completed in October 1971 to perfect functioning of the instrument, designers worked on weight, size, and modifications to integrate it into the lander. The mockup, top right, is 35 centimers wide. Finally, the flight GCMS is tested and prepared for its long journey through space to investigate Mars.

 
 
As the reconsideration of the GCMS continued, the Viking Project Office sponsored the first "Viking science symposium," structured to provide extended discussions of the chemical and biological premises on which two of the project's major investigations-biology and the molecular analysis experiment-were based. While much of the material presented was old information to seasoned Mars hands, for many of the attendees it was the first time they had been exposed to these scientific assumptions underlying the Martian search for life. In addition, several new interpretations of old phenomena or refined Mars data were presented for discussion. Alan Binder [225] of the Illinois Institute of Technology's Research Institute suggested an alternative explanation for the so-called "wave darkening." The most common reason given for this phenomenon had been an increase in atmospheric humidity as water sublimed from one polar cap and moved toward the other. New observations indicated that the wave, which progressed at a speed of 30 kilometers a day, might actually be a wave of brightening. Earth-based photometric measurements had compared dark areas to bright areas on the assumption that it was the bright areas that were unchanging. If the bright areas were getting brighter, then water or vegetation were not needed to explain the change. Instead, the explanation might be some simple mechanism, a dust storm, for example. Some microbe hunters who saw this as one more strike against the possibility of Martian water might not have been pleased, but the reasoning was more consistent with other investigations that indicated limited water on the Red planet. 51
 
Toby Owen of the State University of New York at Stony Brook and Michael McElroy of Harvard reported that Mariner 6 and 7 had provided new clues about the composition of the planet's atmosphere. It was 95 percent carbon dioxide. Nitrogen probably existed in quantities less than 4 percent, and perhaps as little as 0.5 percent. Traces of carbon monoxide, molecular oxygen, ozone, and water vapor were likely. While these were not very encouraging comments for those who wanted to find life on Mars, Carl Sagan repeated his oft-given summary that the only way to make such a determination was to go there and check out the planet. Such an examination might not end all speculation, but it would certainly give them better data. To make that trip worth the effort, the GCMS and the biology instrument would have to work.
 
The problems encountered with the gas chromatograph-mass spectrometer were not made any better by renewed money problems. A special meeting held 19 September to discuss the budget led to some very bitter reactions by several scientists. Martin told those investigators that they would have to reduce their projected costs by a further $17 million to $22 million. Before the next discussion of the science budget reduction in early October, Jerry Soffen received some amazing letters in response to his comments about scientific priorities. There was a decided lander-versus-orbiter outlook among the scientists, and a dichotomy between the build-the-experiment-hardware-yourself group and the more theoretically oriented investigators.
 
Harold P. Klein, biology team leader, was among the first to write. He concluded that it was more important to get results from the lander than from the orbiter. "I say this for a number of reasons: by 1975, we will have had several missions to the planets-with flybys and orbiters, but no lander mission; we have learned a great deal about Mars from the Mariner series and so there is no doubt that these have shaped our views of the planet, and that Mariner 9 should add immeasurably to this store of information.'' But there had never been a direct measurement made from or of the surface of Mars.
 
[226] "What I am emphasizing is something which science recognizes as first order science - i.e., it is generally easier to refine your techniques, and repeat your experiments with more sophisticated equipment than to start investigating in unknown territory." But Klein noted that "it is much more exciting to try something completely new and different - to do something first." He would be willing to sacrifice the orbiter imaging system rather than subtract anything from the landed group of experiments.
 
On the lander, we are proposing a number of investigations-and while these will all be "first time" investigations. and therefore of great potential interest, it is obvious that some are concerned with answering really colossal questions and others are not. It is no surprise-at least to me- that there is a direct relationship between the magnitude of the scientific question being asked, and the complexity, uncertainty and therefore, the expense involved in the equipment concerned with each investigation.
 
Klein would prune the orbiter science to only that needed to support the lander. While dropping the large imaging payload, he would maintain the atmospheric water detector and the infrared thermal mapping device. He hoped that no lander experiments would have to be eliminated, but if deletions were necessary the big experiments-the GCMS, the biology instrument, and lander imaging must be preserved. 52
 
Don L. Anderson, seismology team leader, was equally strong in his opinions. "First of all, I feel that Viking was poorly conceived from the beginning, and this, of course, was headquarters' fault." With that shot across the NASA bow, he continued:
 
The way science was selected was ill-conceived, and headquarters was repeatedly warned that one does not decide what needs to be done and then try to find someone to do it. In the past, the scientists designed the experiments and, by and large, the instrument. The Viking scientists have little experimental experience and virtually no equipment experience. They were chosen because they expressed an interest in so area-not because of any demonstrated wisdom on the important problems of Mars or of the solar system. As a group they cannot provide you guidance in scientific policy matters of priority. As individuals they are ineffective, because of the system, in riding herd on their own experiments, particularly the costs.
 
Translated, the exobiologists might be asking the "colossal" questions, but it was Anderson and his colleagues who were doing experiments with which they had first-hand experience. They could create hardware and deliver it as a reasonable cost and on time. Anderson accepted, to a degree, that "one can argue that the first mission to Mars should have biological emphasis," but the realities were "that the biological and organic experiments were not ready when the payload was selected, are not ready now, and probably will not be ready in 1975." Anderson admitted that physical [227] measurements, such as seismology, were relatively easy, but that complex experiments like the GCMS and the biology investigation were more difficult than anything NASA had ever flown. One could argue parenthetically that the molecular and biological investigations were closer to real laboratory science than anything ever done before in space. These experiments required more than data gathering; they demanded elaborate manipulations of sample materials in miniature laboratories. As he noted, such biological investigations as these were "not even routine measurements on the Earth." They were "not ready to fly a biological mission to Mars. Even if the instruments arc ready the chances are high that they will not work on Mars, and if they do, will give ambiguous results." This team leader represented one camp of scientists who wanted to make "straightforward" measurements; Klein and his associates preferred to pioneer a new "first order" science in space. There were strong arguments for both points of view, which did not make Soffen's or Martin's tasks any easier. The Viking Project Office managers had their hands full-with complicated and troublesome hardware, independent and troublesome scientists. A firm discipline would have to be applied to both. 53
 
The issues raised in the September-October 1971 Science Steering Group meetings would not be resolved immediately. But the discussions led to several changes, as the minutes recorded:
 
l. Reduction of science team support-By deleting certain efforts of the scientists, holding fewer meetings, and supplying less assistance....This will save 33 M[illion].
 
2. Reduction of the Molecular Investigation-Current technical problems with the GCMS have resulted in substantial cost increase over the original estimate. Most team leaders agree to the importance of the investigation but feel that there should be a cost ceiling. By reducing the requirements and simplifying the instrument, it should be possible to assure technical feasibility and to bring the costs down to a level consistent with the present project plans ($35 M). This involves a reduction of the number of samples analyzed, deletion of direct [mass spectrometer] analysis and [deletion of a detector portion of the gas chromatograph]. The cost saving is $3.0 M.
 
3. Relaxation of the Biology Instrument Requirements major requirements involving temperature control and waste management, and several minor ones, can be relaxed at considerable savings....The total cost reductions of $2.0 M has been agreed upon.
 
4. Limitation of Viking Orbiter Science Mission Planning....The saving is $1.0 M.
 
5. Reduction of Meteorology Investigation....to result in a "weather station" type experiment....The saving should be $1.6 M.
 
6. Limitation of the Physical Properties Investigation to Current Baseline... [The saving would be $0.15 M.]
 
7. [228] Use of fixed masts for the Viking Lander Cameras....The cost saving is $0.3 M.
 
8. End Mission B at the beginning of conjunction. . . . The savings are essentially in operations: $0.5 M. 54
 
These changes totaled up to a possible saving of $11.5 million. Decisions that were postponed at that meeting included eliminating photometric calibration of the orbiter camera ($1.6 million) and deleting the X-band radio ($1.1 million), the image-motion-compensation device for the orbiter camera ($0.4 million), the retarding-potential analyzer from the entry science experiment ($2.3 million), and deleting either the infrared thermal-mapper ($3.3 million) or one of the biology experiments ($1.9 million). (Deletion of the orbiter imaging system was also seriously considered at this time. That proposal is described in chapter 9.)
 
Between October 1971 and March 1972, there were numerous conversations among Viking Project Office personnel members, JPL authorities, and the contractor, Litton Industries, about the fate of the GCMS. Jim Martin was not very happy with JPL's management of this activity, So he told the lab on several occasions that he wanted JPL to monitor the contract the way Martin Marietta was monitoring its science subcontracts. He did not want JPL trying to build the GCMS; that was Litton's responsibility. As early as October 1971, Martin was considering finding another organization to handle the GCMS contract, and the project office awarded Bendix Aerospace a contract to study the feasibility of using an organic analysis mass spectrometer (OAMS) in place of the GCMS. Similar in the information that it produced, the OAMS did not use a gas chromatograph. To demonstrate his concern, Jim Martin added the GCMS to the "Top Ten Problems'' list on 26 October. "Specifically the problem is the systems design and program redefinition of a simplified GCMS." Shortly thereafter, Klaus Biemann and his colleagues of the molecular analysis science team requested that Alfred O. C. Nier, the entry science team leader, be added to their group because of his background in mass spectrometry. 55
 
The addition of Nier to the GCMS activity was another blow to JPL. He had written to Jerry Soffen in September 1971: "While I regard a properly devised and managed GCMS experiment as one of the most important things we could do on Mars, the history of this endeavor leaves so much to be desired I really wonder whether it has not disqualified itself already." Nier thought that JPL's record in this area was "dismal." Nier also shared Don Anderson's complaint about the GCMS scientists' lack of experience in inventing and building instruments. He believed that it was "most unfortunate that in NASA's selection of the team some regard was not given to this factor in view of JPL's weakness in this very difficult area." By these statements, Nier did not mean to detract from the caliber of the individuals on the GCMS team, but he felt that it was necessary to underscore the nature of the problem facing the project managers. 56
 
[229] Continued troubles with development scheduling for the gas chromatograph-mass spectrometer and the lack of confidence among the scientists in JPL's ability to manage the instrument's development and fabrication led Martin to transfer the management of the GCMS instrument contract from JPL to his Viking Project Office at Langley. As a preparatory measure, he announced that effective 29 February 1972 Cal Broome, lander science instruments manager, would report directly to the Viking project manager. This shift was one more step to tighten control over the lander science payload and give those experiments the visibility that they seemed to require. Further-as a consequence of Klaus Biemann's presentation on the GCMS and the OAMS made at the February Science Steering Group meeting, in which Biemann had noted that each instrument had advantages and disadvantages that could not be directly compared-Martin decided in favor of continuing the development of a simplified version of the GCMS. His action was prompted primarily by the cost projections, which indicated that it would be cheaper, by about $7.5 million, to retain the GCMS and transfer management of the instrument to Langley. NASA Headquarters approved this recommendation on l0 March, and Martin appointed Joseph C. Moorman as the GCMS manager and J. B. Lovell as the Viking Project Office resident engineer at Litton Industries, Although the development and fabrication of the instrument was still far from ensured, at least a more responsive management-contractor structure had been established to deal with the problems that would emerge later. 57
 
 
Viking Biology Instrument
 
 
Nearly everyone associated with the Viking project realized the Viking biology instrument was going to be a technical and scientific challenge, but no one was able to predict just how much time, energy, and dollars would be required by this complex scientific package. Devising a biology instrument that held three experiments ins container less than 0.027 cubic meter in volume and weighing about 15.5 kilograms was more of a chore than even the most pessimistic persons had believed. Certainly the TRW Systems Group personnel who won the Viking biology instrument subcontract in competition with Bendix Aerospace Systems Division did not expect its original estimated cost of the completed flight instruments and test articles to soar from $13.7 million to more than $59 million. 58 A box about the size of a gallon milk carton, the instrument contained some 40,000 parts, half of them transistors. In addition to tiny ovens to heat the samples were ampules containing nutrients, which were to be broken on command; bottled radioactive gases; geiger counters; some 50 valves; and a xenon lamp to duplicate the light of the sun. It was a complicated and sophisticated miniature laboratory.
 
The Viking biology instrument was originally conceived as essentially the integration of four individual life-detection schemes. According to....
 

 
[230] Gas chromatograph-mass spectrometer cost projections.
 

 
....Loyal G. Goff, Viking Program Office, NASA Headquarters, "the transition from these early hardware models to an integrated, automated, and miniaturized flight unit capable of surviving all of the environmental conditions of sterilization, launch, cruise, and landing was a horrendous undertaking." These environmental requirements, with the performance specifications, demanded considerable examination and testing of the materials used in the biology instrument. The initial design concepts for the experiment were developed by Ball Brothers Research Corporation, Boulder, Colorado, and the Applied Technology Division of TRW Defense and Space Systems Group, Redondo Beach, California, under contracts managed by NASA's Ames Research Center. 59
 
On 3 September 1970, when the TRW team was given the go-ahead by Martin Marietta, four direct biological tests had been selected for the [231] instruments that could examine the Martian soil for traces of living organisms through the measurement of some aspect of the metabolic process. Three of the procedures could in principle detect "resting" metabolism, although all would be more reliable if growing organisms were present. The first experiment, originally called carbon assimilation but later known as pyrolytic release, would be performed with a minimum addition of external substances (that is, only radioactive carbon dioxide [14CO2], radioactive carbon monoxide [14CO], and water vapor) to the samples. Experiment two, originally known as Gulliver and subsequently called labeled release, was to add extremely diluted solutions of labeled (carbon 14) organic matter to the Martian soil samples under conditions that barely moistened the samples. Experiment three, called the gas-exchange experiment, provided for adding greater amounts of organic materials and water to the samples. Because it was rich in nutrients, Jerry Soffen and others referred to this as the "chicken soup" experiment. The fourth experiment (subsequently eliminated) was the light-scattering experiment, or Wolf Trap as it was better known. Requiring the growth of organisms in the sample, this investigation provided the least Marslike environment because it would suspend the sample in an aqueous solution. But if microorganisms did grow, they would turn the liquid cloudy, and the light sensor would detect the change. Together, the four experiments represented a range from very dry to saturated solutions, and experimenters hoped they would provide a check on each other while giving Martian microbes a choice of environments in which to grow. 60
 
The first year of work leading up to the preliminary design review was spent making a breadboard model for each of the experiments. The PDR, originally scheduled for July 1971, was postponed three months so that a number of changes could be made in the biology instrument design. In October, TRW submitted new "estimated cost at completion" figures to Martin Marietta; the cost had risen to $20.2 million. TRW had greatly underestimated the complexity of the task, which accounted for about half of the $6.5-million jump. The rest was due to modifications in the experiment definition.
 
The 4-6 October preliminary design review in Redondo Beach, California, disclosed a number of problem areas in the design and management of the Viking biology instrument. Rodney A. Mills, Walter Jakobowski's deputy, feared that Martin Marietta and TRW could both be blamed for poor management. 61 Of particular concern were the complexity of the waste management system, which would store the water and organic materials after they had beets tested; the complicated nature of the sampling system; the increasing instrument weight, which would lead to higher costs; and the numerous elements that, should they fail, would render the whole instrument useless. On I July 1971, Jim Martin issued Viking project directive no. 6: "It is project policy that no single malfunction shall cause the loss of data return from more than one scientific investigation ." 62 Each [232] of the biology experiments was considered to be one scientific investigation under this philosophy, and there were numerous "single point failures" that could terminate the data return from the instrument. At the October PDR, no single experiment stood out as a particular problem, but Martin, Broome, and their colleges were worried about the overall complexity of the TRW design. 63
 
During November and December 1971, TRW and Ames Research Center personnel under Harold Klein worked to simplify the biology Instrument. Deleted from the design were the Martian gas pump, the onboard carbon dioxide gas system, one control chamber each for the gas-exchange and light-scattering experiments, and related valves, plumbing, and wiring. But it became apparent at a biology instrument review in late December that more drastic changes would have to be made. During the final days of January 1972, Martin concluded that one of the experiments would have to be eliminated to reduce the volume, weight, complexity, and cost of the package. Walt Jakobowski and Richard Young from NASA Headquarters met with representatives from the Viking Project Office, Martin Marietta, and TRW on 24-25 January to discuss ways to remedy the problems, especially cost, which had escalated to $33 million. 64
 
That meeting was not a satisfactory one from Jakobowski and Martin's point of view. TRW was not able to suggest any acceptable engineering cost reduction items without removing two or more experiments. Additionally, all of TRW's cost reduction proposals had high-risk factors for scheduling, testing, or both. Martin Marietta personnel who had reviewed TRW's schedule and manpower figures were also unable to offer any alternatives. To find solutions to their problems, Martin formed an ad hoc panel for the examination of imposed and derived requirements on the Viking biology Instrument under the chairmanship of Howard B. Edwards of Langley's Instrument Research Division. While that panel met to determine which, if any, of the scientific and engineering requirements could be relaxed or eliminated to reduce cost, weight, size, and complexity of the overall instrument, Klein, Joshua Lederberg, and Alex Rich, biology team members who were not affiliated with any particular experiment, met to discuss priorities for deleting one of the experiments.
 
Dropping an experiment was a painful experience for the men who made the recommendation and those who implemented it. By 13 March, NASA Headquarters had decided that the light-scattering experiment, the Investigation based on the least Marslike premise, should be terminated. The men in Washington cited possible difficulties in interpreting results and a potential for further cost growth as reasons for their action. It was John Naugle's unhappy responsibility to tell Wolf Vishniac that his Wolf Trap would not he included in the Martian biology instrument. Noting that "this was one of the more difficult decisions" that he had had to make since joining NASA, Naugle told Vishniac that they had to "simplify the biology experiment-its history of growth in cost and complexity had [233] forced this position.'' In deciding how to reduce costs, the managers at NASA had tried to consider both scientific and engineering factors:
 
On the science side, we are assured that the deletion of the light scattering experiment, while undesirable, is the least damaging in terms of data lost. I won't go into detail here since you have talked at length with Drs. Lederberg and Rich on this subject. On the engineering side, it seems that the light scattering experiment might be considered one of the least complex in terms of number of parts and detail of design, but is one of the more difficult to actually build into a problem free device.
 
Following advice from all members of the biology team, Naugle stressed the desire that Vishniac continue to participate as a member of that group. 65
 
Although the biology team seldom acted as a cohesive group, the decision to eliminate the light-scattering experiment did draw members together temporarily. As a group, they aired their dissatisfaction with the decision, the manner in which it was made, and the limited likelihood that it would reduce significantly the cost of the biology instrument. At a biology team meeting in March, Dick Young and Jerry Soffen were on the hot seat as they once again explained the need for cost reductions in an era of tight budgets. Klein, the team leader, wrote to Naugle on behalf of the whole group:
 
Naturally, the Team is not very happy that the scope of the biological experiments was reduced...This science reduction is all the more difficult to accept because it is not at all clear just what factors dictated this decision. Recent discussions with TRW....leave little doubt that no savings in weight or in volume will follow from the elimination of the light scattering experiment....Whether, at this late date, any cost savings wilt accrue from the deletion is also problematical.
 
While stopping short of mutiny-and still promising to work hard-Klein said that the team wanted a better explanation of why Wolf Trap was dropped. 66
 
Understandably, Wolf Vishniac was not happy with the decision, He criticized Lederberg and Rich for not being familiar with the development status of his experiment: "I am shocked to find that a judgment on the value of an experiment was based upon such complete ignorance on the present state of the instrument..." Much of the discussion regarding Wolf Trap concerned "matters which have long ago been settled and solved," Some of the data the NASA managers had used in their decision-making process had been gathered by the Ames Research Center, Vishniac was told by persons at Ames that they had sent headquarters "some old reports which we had lying around," When the scientist asked why "old" material was used, he was given some surprising news: "It doesn't really matter, we have long ago decided that light scattering is to be eliminated," The more Vishniac investigated the elimination of his experiment, the more he was displeased. [234] He believed that there had been some anticipatory preparation for dropping Wolf Trap. And according to Vishniac, Lederberg and Rich were not really suited or capable of making an informed decision. "Their aloofness from the team, their ignorance of the mechanical details and the apparent predisposition of Ames to leave out the light scattering experiment makes me question the value of their recommendation." 67
 
In a compassionate review of the decision and the process by which it had been made, Naugle tried to allay Vishniac's frustration and anger. The associate administrator pointed out that something had to give, as the budget could not be increased. They had been forced to review and revise all of the Viking experiments on the orbiter and lander. If Lederberg and Rich had not participated in the examination of the biology instrument, someone entirely unfamiliar with the instrument and the search for life on Mars would have.
 
We recognized that we were asking them to undertake a very difficult and personally distasteful job of reviewing four experiments which had originally been very carefully selected and had just recently been certified as complementary and an excellent payload for Viking, and recommending which of the four could be removed with the least impact on the overall biology experiment. They reluctantly agreed.
 
In the guidelines we gave them we said the decision should be primarily made on the basis of the scientific merits of the experiments since there was no substantial engineering factor to use to select the experiments to be deleted. . . . .
 
Dr. Lederberg and Dr. Rich's recommendations were clear-that all four experiments should fly, but if one must he dropped, it should he the light scattering experiment. They also make it clear that although the experiment should be dropped, the experimenter (Dr. Vishniac) should not!
 
Naugle thought that the deletion would "contribute" in a very real way to the solution of their Viking payload problem. "I am assured that we will save at least two or three pounds [0. 9- 1.4 kilograms] by this action. This will be applied directly to the weight deficit already incurred by the biology package." Additionally, space would be saved for other biology requirements, at a saving of at least $2.3 million. 68 In the short run, the projected cost of the biology instrument did drop, but by the fall of 1973 the cost estimates would escalate wildly, leading to another major review of the biology package.
 
Wolf Vishniac faced other disappointments in the loss of his Mars experiment. While he continued to participate constructively in the biology team's work, he no longer had any NASA funds to support his research projects and personnel. Vishniac soon discovered that he would have to pay a high price for having gambled on spaceflight experiments. He had been the first person to receive exobiological research support from the agency, [235] but now that the money was gone he discovered a hostility on the part of many scientists directed toward those who had accepted "space dollars." In spring 1973, Vishniac wrote to Soffen telling him that he could not attend a particular meeting. "I will do whatever is essential in the Viking program but I simply must place my priorities on my university work. The consequences of my change in status in the Viking Team have been far-reaching as you know, not to say disastrous." He was finding it difficult to obtain support for laboratory research because of his work with the space agency. The National Institutes of Health had refused a grant application; "I was told unofficially that it received low priority because I was 'NASAing' around." The National Science Foundation had decided not to renew a grant for Vishniac, partly because of his association with NASA. The exobiologist told Soffen that "it is essential that I recapture some sort of standing in the academic world and I must therefore limit my participation in Viking essentials only." 69
 
In 1973, Vishniac was still pursuing his research into the origins of life and the possibility of life on other worlds when he fell 150 meters to his death in Antarctica's Asgard Mountains. Searching for life in the dry valleys of that bitter cold and windswept region, Vishniac was attempting to prove that life forms could adapt to extremely hostile environments. Early in 1972, he had found microorganisms growing in what had previously been thought to be sterile dry valleys. This discovery by Vishniac and his graduate student assistant Stanley E. Mainzer, using a version of the Wolf Trap light-scattering instrument, was a bit of good news for the believers in life on other planets but a contradiction of the findings of Norman Horowitz and his colleagues Roy E. Cameron and Jerry S. Hubbard, who in five years of research had yet to detect any life forms in that barren land.
 
The dry valleys of South Victoria Land, Antarctica, with a few other ice-free areas on the perimeter of that continent, formed what was generally agreed to be the most extreme cold-desert region on Earth. The area was also the closest terrestrial analogy to the Martian environment. These valleys, which covered several thousand square kilometers, were cut off from the flow of glaciers out of the interior of the continent by the Transantarctic Mountains. Although the valleys were ice-free, their mean annual temperature was -20°C to -25°C, with atmospheric temperatures rising to just the 0ÉC mark at the height of the summer season. Liquid precipitation and water vapor were almost nonexistent, and the limited snowfall usually sublimed to the vapor phase without ever turning to liquid. It was in this region that Horowitz's colleagues discovered what was believed to be the only truly sterile soil on the face of Earth. From their research in the dry valleys, Horowitz and his associates concluded:
 
These results have important implications for the Mars biological program. First, it is evident that the fear that terrestrial microorganisms carried to Mars could multiply and contaminate the planet is unfounded.
 
 

 
[236] Scientists attented a Viking planning meeting at Langley Research Center in 1973. Lfet to right are Dr. William H. Michael, Jr., leader of the radio science team; Dr. Wolf Vishniac, assistant biology team leader; and Dr. Richard S. Young, Viking program scientist from NASA Headquarters in Washington.
 
The Antarctic desert is far more hospitable to terrestrial life than is Mars, particularly in regard to the abundance of water. In other respects, too-such as she ultraviolet flux at the surface-Mars is decidedly more hostile than the Antarctic.
 
Second, Martian life, if any, must have evolved special means for obtaining and retaining water. . . .This has been known for some time. What is new in these findings is that even under severe selective pressure microbial life in the Antarctic has been unable to discover a comparable mechanism. To some this may suggest that life on Mars is an impossibility. In view of the very different histories of Mars and the dry valleys---we believe that such a conclusion is not justified.
 
Finally, the Antarctic has provided us with a natural environment as much like Mars as we are likely to find on Earth. In this environment, the capacity of life as we know it to adapt and survive is pushed to the limit. The concentration of living things around the sources of water in the dry valleys and their rapid drying out in the most arid locales may be useful as a model of the distribution of the life we may, if we are lucky, find on Mars. 70
 
 
But in 1972, Vishniac detected microorganisms with Wolf Trap in exactly those regions that Horowitz had declared sterile, Life had found ways to survive in the inhospitable, Marslike dry valleys. In December 1973, Vishniac went back to Antarctica to learn more about these hardy microbes. He wanted to know where they obtained their life-sustaining water and nourishment. Alone on a steep slope in the dry valleys, Vishniac slipped, fell, and died. 71 Vishniac and his Wolf Trap life detector had been successful [237] on Earth, but he would not see Viking go to Mars, and his instrument would not be applied to searching for elusive Martian microbes. A man who had done much to give exobiology legitimacy as a field of research was gone. The loss of Vishniac from the biology team was repeatedly felt in later years. He had been an arbiter and a man of good cheer. As the biology instrument continued to increase in cost and to raise more and more technological hurdles to be overcome, a man with his talents and humor was sorely needed.
 
During the first half of 1973, work progressed on the design development units for the biology instrument and the gas chromatograph-mass spectrometer. Science tests for the biology instrument had begun in mid December 1972, with biology team members participating in the trials of their experiments. GCMS testing began in early May. After the first round of testing, the Viking managers held a critical design review on 23-25 May for the biology instrument, and even though they discovered no major problems with the package, Martin Marietta and the Viking Project Office were less than pleased with the review. The GCMS critical design review in mid-July disclosed only three major concerns, which was encouraging news considering the problems that piece of hardware had caused earlier.

Unhappily, new trouble with the management of the biology instrument surfaced in mid-July. At a meeting held at TRW, Jim Martin learned that completion of the design development unit had slipped by three weeks and the projected delivery date of the proof test capsule unit was behind by five weeks. The problem, Martin found, was failure to plan ahead; TRW lacked the skilled manpower to assemble and checkout these crucial units. As the July session went on, the discussion of the biology instrument came "unglued," according to Martin; he feared that the work at TRW was "out of control" with no credible schedule or cost plan. 72 By that autumn, the situation was even bleaker. On 15 October, Ed Cortright wrote to George Solomon, vice president and general manager at TRW. Cortright had been monitoring TRW's handling of the biology instrument problems with the intent of reporting to Hans Mark, director of Ames Research Center. His report was to give the center better data for judging prospective contractors-of which TRW was one-of experiment hardware for the Pioneer spacecraft scheduled to visit Venus. Cortright's report to Ames would not be favorable. He thought that TRW, Martin Marietta, and NASA had underestimated the complexity of the biology instrument task: "The original TRW proposed cost was grossly underestimated with the result that the current estimate at completion is $30.9 million, which is $18.4 million or 147 percent over the original estimate." Of that amount, $12.4 million was TRW's overrun; $6 million had been spent on redefining the experiments.

 
Cortright told Solomon that the TRW management had placed too much emphasis on the company's previous performance and had been reluctant to face the fact that the biology instrument was getting into serious trouble. "You are currently beset with a rash of technical problems....
 
 

[238] The development model of the Viking biology instrument's mechanical subsystem, top left, conveys some of the external complexity of the experiment. The mockup, top right, minus the essential electrical and plumbing connections exposes the hardware to view. At lower left, a diagram shows the biology instrument after deletion of the light-scattering experiment. At lower right, in final stages in 1975, the automated equivalent of a well-equipped biological laboratory makes up a package of less than 0.03 cubic meter to land on Mars.

 
....which further threaten schedule and cost. It is clear that if the job were on schedule, there would be more time to adequately cope with the necessary fixes." Impressed with the steps Solomon had taken to strengthen management of the biology package, Cortright nevertheless believed that "heroic action" would be necessary to ensure "a successful experiment on the surface of Mars." 73 Two weeks later, after the schedule had slipped even further and the biology instruments had been put on Martin's Top Ten [239] Problems list, Cortright again wrote the general manager about the "potentially catastrophic" situation and sent a similar letter to Richard DeLauer, TRW Space Systems executive vice president. To DeLauer, Cortright bluntly said. "It is imperative that you bring to bear on these problems the most talented individuals you can find within your Company, and elsewhere, and quickly weld them into a problem solving team to get this job done. I know you have taken steps in this direction and I cannot fault individuals who are currently working the problems. However, I must believe that you have not yet applied your maximum effort, for which there is no longer any substitute." 74
 
The problems at TRW were twofold. The engineering tasks imposed by the experiments were very difficult, and TRW's management of the project was poor. At very low temperatures, valves and seals failed, and other hardware difficulties surfaced as the initial pieces of equipment were tested. But most serious was the absence of a strong, driving manager at the California firm overseeing the work. In November 1973, production of the flight units was essentially stopped while the biology instrument was redesigned. But design quality and workmanship problems persisted, causing test failures and schedule difficulties. To meet the launch date, TRW was required to conduct design-development concurrently with qualification testing and fabrication of the flight units. By the first of February 1974, several independent analyses of the situation at TRW pointed to the possibility that the final flight units of the biology instrument would not be ready until July 1975. That would be very close to the scheduled launch dates (August and September) and too late for adequate preflight science testing.
 
Cal Broome, who had been appointed NASA biology instrument manager in December 1973, in a private note to Jim Martin on 7 February 1974, stated that his own view of the situation at the subcontractor's was that the "engineering organization, and, to a lesser extent, the manufacturing organization [at TRW], are running out of control." Furthermore, "The TRW engineering 'culture' simply cannot accept scheduling and discipline in connection with engineering problems." Broome was also worried that others would nut share his opinion of TRW's failings and simply view his pessimistic outlook as a case of Broome having panicked again; but Hatch Wroton, the Martin Marietta resident engineer at TRW, and Dave Rogers, the JPL resident at TRW, had independently assessed the biology instrument's status and agreed with Broome's bleak prognosis. 75
 
During the remaining months before the Viking launches, time lost in the schedule would be made up, only to be lost again when some new difficulty appeared. In July 1974, Martin had Walter O. Lowrie, lander manager at Martin Marietta, and Henry Norris, orbiter manager at JPL, study contingency plans for flights without the biology instrument and single flights of the Viking spacecraft in 1975, 1977, and 1979. Days later, progress on the instrument at TRW looked more promising, but by the end [240] of the year, when the performance verification tests of the completed instruments were being conducted in Redondo Beach, new doubts about meeting the schedule plagued the Viking managers. 76
 
The seesaw between failure and progress finally stopped in the early spring of 1975. On 7 March, Martin wrote to the three men who had seen the biology instrument through some of its most difficult moments- Eugene M. Noneman, TRW; Hatch Wroton, Martin Marietta; and Roy J. Duckett, Viking Project Office: "I was pleased today to be advised that Viking biology instrument S/N 106 is in its shipping box ready for delivery. I believe that you and your team members have achieved a very significant and important milestone. While there is still much work ahead of us, having a flightworthy biology instrument ready to ship to the Cape is a gratifying accomplishment." Martin extended his personal congratulations to every member of the team. 77 On 28 May, Cal Broome could at last recommend to Jim Martin that the GCMS and the biology instrument be removed from the Top Ten Problems list. Those had been the final items on the list of troubles. The hardware units were finally ready for shipment.
 


Table 42

Viking Biology Instrument Schedule, 1971-1975

Milestone

Original Contract

Actual

Delay

Delivery Date

Delivery Date

in Months

Preliminary design review

July 1971

Oct. 1971

3

Critical design review

Aug. 1971

May 1973 a

9

Design-development testing complete (S/N 001)

July 1971

Dec. 1973 b

17

Qualification unit delivery/ qualification testing complete (S/N 102)

Sept. 1973

Mar. 1975 c

18

Proof-test capsule unit delivery (S/N 103)

June 1973

Nov. 1974

17

Flight unit - l delivery S/N 105 on Viking lander capsule #1

Jan. 1974

Mar. 1975

14

Flight unit - 2 delivery S/N 106 on Viking lander capsule #2

Apr. 1974

Mar. 1975

11

Flight unit-3 delivery (S/N 104)

July 1974

Apr. 1975

9

Spare flight unit

Added Dec. 1973 d

Deleted Oct. 1974 e

-

a Martin Marietta contended that a realistic CDR was not completed until Mar. 1974

b Design development testing was completed on a nondeliverable unit: one of the deliverable units was canceled: the other deliverable unit's mechanical subaseembly was simulated in system test bed testing.

c Qualilication testing was different from original plans and not as comprehensive.

d This unit not included in the original contract was added in Dec. 1973

e Unit deleted Oct. 1974 when requirement for spare lander was eliminated.


 
Biology Instrument Cost Projection.
 

 
[241] The cost in individual time and effort on these two items had been high; the dollar costs were equally great. By launch, the GCMS bill read $41 million, and the biology instrument had cost $59 million. 78
 
There was, of course, more to the Viking lander science package than the gas chromatograph-msss spectrometer and the biology instrument. Each of the other instruments went through a similar series of problems met and problems solved. The GCMS and the biology instrument were unique because of the magnitude of the difficulties and the expense. With time, all problems with the instruments were resolved, and interaction among the scientists improved. Still, each team remained a collection of individuals, [242] and among the teams only a loose confederation existed. Before the missions were flown, a stronger discipline would have to be forged. Operation of the orbiters and landers would be a complex task, and each sequential operation would have to be carefully planned and precisely executed. Jerry Soffen, Jim Martin, and Tom Young had many difficult tasks ahead of them, and one was establishing tighter control over the Viking scientists without stifling their inquisitiveness-exercising discipline so as to get maximum science return, but not in such a manner as to eliminate flexibility when scientific targets of opportunity appeared.
 
As the Viking science teams and their instruments matured, Jim Martin faced other technical problems with the lander, each of which had to be solved before the spacecraft could fly. Complexity and technological challenges abounded. Building Martian landing craft was genuinely hard work.
 

* Thomas A. Mutch has described the hisetory of the lander cameras in The Martian Landscape, NASA SP-425 (Washington, 1978), pp. 3-31.
 
** Panel membres included Chairman H. B. Edwards, K. Biemann, T. Owen, R. S. Young, J. J. Paulson, and G. C. Broome.