As Skylab progressed from blueprint to hardware, the program office at Houston focused attention on flight operations. Skylab operations would differ significantly from Apollo missions, in which a series of time-critical events had bound the operation to a rigid schedule. Spacecraft failures were anticipated by contingency plans that left little to chance. In Skylab, extensive earth and solar observations dictated a more flexible schedule. The operations teams in Houston and Huntsville also needed greater staying power; while Apollo missions had lasted no more than two weeks, Skylab's would run for months. Data management was another concern. Unlike Apollo missions, the workshop would be out of contact with ground stations much of the time. Data would have to be stored on board until Skylab passed over a ground station, when the telemetry would be "dumped" into a ground receiver. Skylab operations would also force the Houston center into new relationships with Huntsville and the scientific community. Marshall and the principal investigators would exert considerable influence on Skylab. Crew training would have to be expanded to meet the scientific objectives. The dual requirement for training in both science and spacecraft operations laid a considerable burden on Houston's training office and also touched off a lengthy dispute over crew selection.1
During Skylab missions, Houston and Huntsville would achieve a remarkable degree of teamwork, quite unlike the disharmony that characterized early Apollo Applications planning. That disagreement had originated in George Mueller's determination to get AAP under way using the lunar module and a low-cost wet workshop. While Houston had no confidence in this concept, Huntsville was willing-even anxious-to develop the hardware. Relations were further exacerbated by Huntsville's desire to have more say in flight operations. As development center for the workshop, MSFC would certainly play an active role; working  out the details of this new relationship, however, required lengthy negotiations.
The two centers began preparing for AAP operations in late 1966, focusing initially on communications and the role of Huntsville's Operations Support Center. From time to time, Huntsville officials suspected that their counterparts in Houston were using obstructionist tactics to prevent MSFC's participation in planning and executing AAP flights. Martin Marietta's integration contract, which made no reference to Huntsville's support role, was particularly galling. Nevertheless, it appeared likely at Huntsville that Houston would eventually give ground. Huntsville's involvement with the workshop "made it technically very difficult to exclude [Marshall] from operations support."2
In 1967 Huntsville pressed for more responsibility in flight operations. At the very least, the center wanted a supporting role on AAP flights; ideally, the workshop would help Marshall become a leader in spacecraft design and operations. MSC gave ground grudgingly. In June, Director of Flight Operations Christopher Kraft agreed to use Marshall engineers for AAP flight operations, provided they were integrated into his organization. This was unacceptable to Huntsville, which wanted the group to remain separate with the lead engineer reporting to MSC for requirements. By November 1967 the two centers had agreed that Huntsville would staff a Systems and Experiments Section within MSC's Flight Control Division.3
By early 1969 it seemed that the two centers were near a modus vivendi. In February, MSC's program manager Robert Thompson assured Belew, his opposite at Marshall, that Huntsville would be kept aware of all developments "by the necessary coordination of our two offices and by MSFC review and concurrence with the evolved operating procedures." Houston would initiate change proposals through Marshall's program office to preclude any appearance of meddling with that center's contractors. Supplemental contracts, added to Marshall's basic contracts, would formalize Houston's relations with McDonnell Douglas and other firms.4
Huntsville officials were pleased with these concessions, but still wanted a formal agreement spelling out the "total operations interface." Such an agreement, while recognizing Houston's direction of mission operations, would also honor Marshall's "cradle-to-grave" responsibility for hardware. Huntsville exercised this responsibility for its launch vehicles, analyzing their performance in flight and establishing operational procedures and limits. KSC's launch team, offspring of the von Braun organization, considered Marshall's involvement on Saturn tests a natural extension of its design responsibility; Huntsville hoped to gain similar recognition from MSC for Skylab operations. At an April planning meeting, Marshall's staff approved a recommendation that the  center seek "an active voice in real-time operations decisions affecting either individual MSFC hardware modules or the integrated cluster." 5
Two months after the dry-workshop decision, center representatives met in Houston to discuss operations. MSC reviewed its specific requirements for flight planning, Huntsville outlined the functions of its support center, and a debate on management philosophy ensued. Before adjourning, the two sides reached agreement on several points: (1) The mission requirements document, prepared jointly by the two program offices, would serve as the basic instrument for mission planning, and both offices would use it as their formal communications link to MSC's operations team; (2) Houston would prepare the operational data book, using information from Huntsville and contractors; and (3) Marshall would provide Houston access to its configuration control boards.6
Other aspects of Skylab operations, however, continued to divide the two centers. While Houston sought to upgrade the workshop, Huntsville clung to the no-change dictum. Matters reached a low point at the telescope-mount design review in May 1970 when, as an MSC official recalls, the two sides "slugged it out to a standstill." Thereafter, relations improved markedly, and by year's end the two centers had agreed on the basic framework for Skylab operations. A flight management team, comprising program managers and MSC's operations managers, would set policy. Although Houston had a majority on the team, Huntsville and KSC were assured a voice in all matters. Daily operations remained in the hands of MSC's flight control teams. If problems involved hardware, the flight director could seek assistance from a Marshall liaison team stationed nearby in the Flight Operations Management Room. The liaison team could, in turn, call for help from a much larger group of engineers at Huntsville's operations center. An elaborate communications system tied the two centers together, providing Huntsville with detailed information on Skylab's condition.7
Attitudes in Houston changed appreciably after Apollo 11 and the dry-workshop decision (pp. 109-10). Until July 1969, most MSC officials viewed Skylab as an unwelcome diversion; after the lunar landing it became the next major program.
The Flight Control Division began preparing an operations plan in August 1969. Division Director Eugene Kranz hoped to retain man Apollo features in the Skylab operation, but certain changes were dictate by the longer missions and the larger number of flight systems. Houston could not afford to keep a full complement in its Mission Control Center throughout the Skylab missions, as it was doing for Apollo. Besides, a earth-orbital mission required less support. During the astronaut working hours, a "high-level" shift would run operations; at night MSC  would maintain a skeleton crew with additional engineers on call. A second concern involved staffing of the mission control room, the heart of Houston's operations complex. Two flight controllers divided the responsibility for spacecraft systems on Apollo. Since Skylab had five distinct units (including the telescope mount), a similar division, plus other requirements, would bring the staff to nearly 30 engineers. Kranz feared that such a large group might hamper the flight director: "You would end up caucusing instead of making decisions. " His preliminary plan allowed one systems expert for each spacecraft; the plan also consolidated some other duties. With these changes, Kranz expected to have no more than 20 flight controllers in the control room during periods of peak activity. Normal operations would require only 11.8
Kranz renewed Skylab planning in November as part of a larger review conducted by the flight operations directorate. Manning requirements were a major topic, but a number of other issues were also discussed: the impact of the new 50° orbital inclination on operations, Houston s relations with principal investigators, and the requirements for unmanned operations. Kranz listed 11 aspects of the Skylab operation that had no precedent in Apollo missions and asked for a thorough review of these key mission issues."9
During the next 30 months, the flight-control organization was restructured. Several instructors were retrained as systems engineers. Men assigned to the experiments required extensive training; several took lengthy courses in solar physics. In October 1972, one flight-control team was assigned full time to Skylab. One of its first tasks was to develop procedures for data processing; another was to conduct several mission simulations with the flight crews. After Apollo 17's splashdown, the rest of the division turned its full attention to Skylab.10
When the missions began, the division's preparations proved sound in most areas. One exception proved to be the transmission of data. Signals transmitted from the spacecraft were picked up by 1 of 13 stations in the tracking and data network and forwarded through Goddard Space Flight Center to Houston. About a quarter of the time, Skylab would be close enough to a station to transmit data as it was acquired. Most of the time, however, data were recorded to be "dumped" when the workshop reached the next station. Skylab's telemetry system required only five minutes to transmit data that had taken two hours to gather.11
The system was a major change from Apollo, and Houston's flight-control teams had trouble adjusting. On lunar missions, flight controllers had seen only 10% of the data, but they had been able to call up specific information when needed. Increased telemetry from the workshop and the long periods between transmissions ruled out immediate access to data during Skylab. Instead, using a process called "redundancy removal, only changes to data reached Mission Control. The new equipment  was installed late, and some flight controllers failed to master it. The shortcoming became rather painful during the crisis that followed launch of the workshop. According to Kranz: "Because of the lack of proficiency in the data retrieval task, the flight controllers were generally inefficient in accomplishing contingency analyses." After the first manned mission, 12 persons were trained specifically for data retrieval. 12
While Kranz's division prepared for operations, the Flight Crew Operations Directorate began work on a Skylab flight plan. Eventually, a plan would provide a detailed schedule for each crew's activities in space. The initial drafts, however, served different purposes. They were, first of all, training vehicles for flight planners who found their Apollo background of limited value. The drafts also served to point up crucial issues, define crew-training requirements, and uncover problems with experiment priorities. Much of the necessary information came from the mission requirements document: objectives, experiment requirements, extravehicular activity, recovery zones, information on television and photography. General guidelines for scheduling crew activities were set within the Flight Crew Operations Directorate. Initially, these guidelines were fairly rigid (e.g., all crewmen would eat together), but as scheduling complexities increased some flexibility was allowed. Although computers were used, the actual scheduling was done by hand. The goal was to meet all the objectives of the mission requirements document. When this proved impossible, the program offices revised the document, usually reducing the number of times certain experiments were repeated. The books, checklists, cue cards, maps, and charts used in planning each mission totaled more than 10 000 pages.13
Huntsville began preparing for mission support in mid-1970 by identifying 17 major tasks and appointing a manager to handle each requirement. The Mission Operations Office coordinated planning principally through monthly meetings of task managers, prime contractors, and representatives of Marshall's major divisions. Much of 1971 was spent preparing documents; in the end 19 plans for mission support were written. Marshall engineers met frequently with the Houston operations team, a particularly important series of meetings in mid-1972 reviewed hardware characteristics and operating procedures. In October Huntsville tested two years of work with a mission simulation, a prelude to participation in Houston's dress rehearsals.14
Marshall consolidated its support in the Huntsville Operations Support Center, an organization that had proved itself during Apollo. Skylab requirements would be handled by 10 mission support groups, each staffed to service a major system, e.g., attitude control. Initial manpower  projections for the support groups totaled more than 400 engineers, some to be drawn from the program office, others from MSFC laboratories and contractor teams. Saturn engineers would monitor launch vehicle operations during checkout and early stages of flight. Other personnel managed a complex communications network of voice, television, and high-speed digital-data lines connecting Huntsville with Houston and the Cape. The Mission Operations Planning System, an asset unavailable during earlier manned missions, allowed support personnel to draw on Houston's computers for immediate printouts of current flight and experiment data.15
MSFC officials divided the Skylab mission into five phases. Prelaunch support began in October 1972. During this phase both launch vehicle and workshop engineers would be at KSC's call. The second phase, workshop launch and deployment, lasted only a few crucial hours but produced peak activity at the support center. Launch of a crew represented a third phase; the first part of each launch would also require peak operations. Manned operations were the fourth phase. The support center's coordinating staff would serve at full strength while the astronauts were at work and at partial strength the rest of the time. Members of the mission support groups would handle Skylab problems during the normal work week. Nights and weekends, they would remain on call for emergencies. The last phase, unmanned operations, required MSFC monitoring, because several workshop systems continued to operate, as did the solar telescopes.16
The choice of Skylab crewmen was bound to cause hard feelings among Houston's astronauts. The group had expanded rapidly in the mid-1960s, and as NASA's fortunes declined it was clear that some of them were not going to fly-at least not until the 1980s. The problem was aggravated by Houston's selection policy. As director of flight crew operations, Deke Slayton determined who would fly. His recommendations went through Gilruth to Headquarters, but Slayton's choices were usually approved. He placed a premium on experience; consequently astronauts moved in a natural progression from Gemini flights through service on Apollo backup crews to an Apollo flight. His policy favored those pilots who had entered the program by 1963 and those test pilots in the 1966 group who received an early assignment. At a disadvantage were the scientist-astronauts brought into the program in 1965 and 1967. By the time these men had finished the required year of Air Force flight training, they were last in line.
Dissatisfaction among the scientist-astronauts surfaced after the first lunar landing. Despite speculation that subsequent missions would....
....stress science, Slayton chose only test pilots for the next three Apollo flights. In October 1969, scientist-pilots complained to Headquarters about selection criteria that emphasized operations at the expense of science. Slayton's rebuttal stressed the hazards of a lunar mission-no one would benefit from a dead geologist on the moon-and downplayed the importance of scientific competence in lunar exploration.17
During 1970 opportunities for the scientist-astronauts declined further. In January NASA canceled one Apollo flight; in September, two more. It seemed likely that no scientist would explore the moon. Late that year the Space Science Board sought assurances from NASA that two scientists would fly on each Skylab mission. The board's action coincided with a resurgence of dissatisfaction among the scientist-astronauts in Houston. Homer Newell, NASA's top-ranking scientist, went to Houston in January 1971 to hear their complaints and see what could be done about them. One by one the scientist-astronauts told Newell that they could not get a fair shake as long as a test pilot (Slayton) picked the crews. As they saw it, his choices were determined by flying time, special skills, and personal relations. Science was not a consideration; in fact, those who showed more interest in science could be at a disadvantage. Several astronauts recommended that Headquarters establish criteria for crew membership, preferably with some appreciation for science. The group felt strongly that each Skylab mission should include two scientists. One of them noted, "flight operations take only a small fraction of the time required for science and other objectives.'' 18
Newell incorporated much of what was said in his recommendations to the administrator. On the sensitive issue of crew selection, he urged that Harrison Schmitt (the only astronaut with a Ph.D. in geology) be assigned to a lunar landing as early as possible and that two scientists  be considered for each Skylab flight. He also proposed a review of NASA's crew-selection process and suggested restructuring the scientist-astronaut program to allow a greater commitment to a scientific career Since he had heard only one side of the issue, Newell labeled his recommendations "tentative." 19
The recommendations touched off several months of debate concerning the makeup of Skylab crews. Slayton and Gilruth argued against more than one scientist per flight, reasoning that hardware problems would demand a high level of systems expertise, an area in which test pilots were thought to excel. Gilruth informed Dale Myers in June that reliability studies indicated "a high probability of systems problems . . . during the mission." Since the workshop's systems could not be modified after launch, Houston was directing most of its training to "systems management and malfunction procedures." He also pointed out that Skylab missions had been planned around a concept of maximum cross-training, which would give each crewman roughly the same degree of proficiency on all major experiments. Consequently, an astronaut's specific academic background was relatively unimportant.20
Myers wanted to accommodate the scientists by including a second scientist on at least one mission, but Gilruth's arguments were persuasive, and Myers remained undecided. When three Soviet cosmonauts died on 29 June during reentry, however, he agreed that NASA should give operational considerations top priority. On 6 July Myers recommended approval of Houston's plan for Skylab crews; two pilot-astronauts would go on each mission with one scientist-astronaut. On the first flight, the scientist would be a physician. Myers left selection of specific crew members to Houston. Newell expressed some misgivings, but the plan was adopted.21
Crew selections were made late in the year and formally announced on 19 January 1972. Charles "Pete" Conrad, the ranking Skylab astronaut, headed the first crew. Conrad had flown three previous missions commanding Apollo 12's flight to the moon. Two astronauts new to spaceflight made up the rest of an all-Navy crew. Joe Kerwin had earned his M.D. at Northwestern University before joining NASA in 1965, Paul Weitz had entered the program a year later. Alan Bean, commander of the second mission, was the only other veteran selected for Skylab; he had gone to the moon with Conrad in November 1969. Owen Garriott, an electrical engineer with a Stanford Ph.D., filled the scientist's slot and Jack Lousma, a Marine major, received the pilot's assignment. Another Marine test pilot, Gerald Carr, headed the third crew, which included Edward Gibson, a Caltech Ph.D., and Air Force Lt. Col. William Pogue.i The selections represented a compromise among NASA  interests: less experience-only two veterans-than Slayton wanted and fewer professional scientists than Newell wanted.22
In retrospect, the importance of crew makeup was overstated. On all three missions, test pilots performed experiment work creditably while scientist-astronauts proved adept at repairing spacecraft systems. Success depended more on teamwork and individual attitudes than on academic training. Although the medical directorate had fought hard to send a physician on the second mission, their fears about a 56-day flight proved groundless. Apollo telescope mount experimenters were well served by Garriott and Gibson. Ideally the second or third crew should have included an earth-resources specialist, but the earth-resources experiments had been added late in the program and none of NASA's scientist-astronauts was particularly qualified with the hardware. Furthermore, given the experimental nature of those instruments, expertise might have been wasted. Slayton's contention that the flight plan would allow little time for independent research proved largely correct.23
Crew training began in October 1970, largely because of prodding from the Apollo telescope mount investigators. The Naval Research Laboratory's Richard Tousey had first approached Houston about a solar physics course for astronauts in 1967. He renewed his request in February 1970 in a strong letter to the program office. Recounting his earlier suggestions, Tousey noted "that little has been done as yet to arrange for scientific training of the crew." He acknowledged that astronauts could operate the telescope mount without an understanding of solar physics, but the data thus obtained would be inferior. For that reason NASA had promised that its crewmen would have appropriate scientific training. Tousey feared that Houston's procrastination would necessitate a cram course a few months before launch, "when systems operational training will be paramount." Ideally, training should begin 24 months before liftoff. With a July 1972 launch date (according to early 1970 schedules), there was little time to waste.24
Houston was not particularly eager to begin crew training, for the astronauts were heavily involved in design reviews and training chief John Von Bockel had his hands full with Apollo. By June, however, MSC had taken steps to satisfy the telescope-mount investigators. At a meeting in Denver, it was agreed that Skylab astronauts would begin a 10-week, 60-hour course in solar physics that fall. Principal investigators would take an active part. All crewmen would be given the same level of training, regardless of their background.25
Principal investigators were generally pleased with the course outline prepared by Dr. Frank Orrall, University of Hawaii physicist. Tousey....
The one-g trainer at Manned Spacecraft Center. Above, exterior of the workshop and Apollo command module. ML71-7650. Right above, upper deck (forward compartment) of the workshop. The square port with the coiled metallic hose hanging on it, left of center, is the scientific airlock. The double ring of storage lockers and water tanks would be easily accessible in zero-g. S-72-51657. Right below, the lower deck with compartments labeled. ML72-5059. See following pages for remaining modules of the trainer.
 ....suggested several changes, including observations of the sun during the course, rather than afterward. He also proposed to augment Orrall's presentation with several lectures on the role of solar physics within the larger framework of science. He hoped this would stimulate the astronauts' interest by pointing up the applications of solar physics "outside the study of our sun as merely a thing in itself."26 When the course got under way in late October, most of the astronauts found the instruction quite a challenge. One admitted, "I was right up to my eyeballs in trouble the whole time, trying to keep up and understand what was going on." Most of them had trouble communicating with the investigators- professionals in an esoteric specialty. For Jerry Carr, the course went much better after he gave up trying to be a solar physicist and instead looked for ways to become a competent observer.27
While the astronauts were learning solar physics, MSC's training office began work on a much larger program encompassing all Skylab training. Robert Kohler took the lead in preparing the syllabus, assisted by a team from Martin Marietta. Kohler laid out a 2200-hour program stretching over 18 months. The schedule was based on a 28-hour training week; previous programs indicated that astronauts would spend another 20 to 25 hours in travel, physical exercise, flying, and reviews. Kohler's program included 450 hours of briefings and reviews, 450 hours of experiment work, and nearly 700 hours of simulator training. It was a demanding schedule compared to Apollo missions, which had averaged 1200 hours of training.28
Briefings constituted a large part of training in 1971. Experiment...
....briefings were handled in two phases. Principal investigators lectured the astronauts on the theory, objectives, and judgment involved in gathering data; later, Martin Marietta instructors provided a nuts-and-bolts presentation on operational procedures, maintenance, safety, and support equipment. North American Rockwell conducted a lengthy block of instruction on the Apollo spacecraft-130 hours of briefings and nearly twice as much time in the simulator. Although Skylab crews would spend relatively little time in the Apollo spacecraft, those few hours would encompass a number of events where an error could prove fatal. The largest block of instructional time was devoted to the workshop, with Martin Marietta covering the telescope mount and McDonnell Douglas the remaining systems.29
Through most of 1971, the training office worked its schedule around spacecraft testing. Traditionally, astronauts had played an active role in testing flight hardware. The Skylab syllabus provided 100 hours for this purpose; the crews would eventually spend twice that much time. The scheduled hours, moreover, reflected only part of the time actually invested. Most tests were conducted at contractor plants or other NASA centers. Frequently, crews would travel to Huntsville or St. Louis only to have a test postponed. Schedule slips at Huntington Beach were the biggest headache; workshop delays cost the training office hundreds of man-days. After the missions were completed, Von Bockel would recommend against astronaut participation in future spacecraft testing.30
A number of other training requirements kept astronauts on the go Crews reviewed navigational stars and received instruction on the stellar experiments at the Morehead Planetarium in Chapel Hill, North Caro...
....lina. Work with the astronaut maneuvering units took them to Denver and to Langley Research Center in Virginia. Apart from spacecraft testing, extravehicular training in Huntsville's neutral buoyancy trainer (p. 170) required the most travel. Beginning in February 1972, one crew or another used the tank nearly every month.31
Training moved from theory to practice in early 1972 when crewmen occupied the Skylab simulators. A computer system in the workshop mockup displayed images similar to those the astronauts would see in flight. The telescope mount console was its most prominent feature; crewmen spent as much as 200 hours studying solar activity on its video screens. The computer could also display normal and abnormal conditions on a half-dozen other control panels. Frequently, while one crew trained in the workshop, a second worked in the command-module simulator, practicing flights to and from Skylab. Two other Apollo simulators provided special training for launch aborts and rendezvous procedures Astronauts could operate the simulators independently or in conjunction with Mission Control. When complex display systems were not required, crews worked in one-g mockups, training models that duplicated the Skylab configuration.32
Houston's basic principle was that all crewmen should become proficient with the major experiments; at the same time, however, the variety of systems required a degree of specialization. The commander was given responsibility for the Apollo spacecraft; the scientist took charge of extravehicular activities, the solar telescope, and medical experiments; workshop systems and the earth-resources equipment fell to the pilot. This division of labor was apparent in the training performed  by the crews. Conrad, despite his considerable flight experience in the command module, spent 400 hours in the Apollo simulators, 55 hours more than Paul Weitz. Weitz, in turn, spent nearly twice as much time on earth resources as either of his crewmates. Kerwin's preparation for the medical experiments, 181 hours, considerably exceeded that of either of his partners. The pattern generally held true for the other crews. The syllabus was a guide rather than a rigid yardstick. Schedules could be changed by the crew commander and the mission's training coordinator. Commanders exerted a great deal of authority; for example, Conrad insisted that 20 hours was not enough training for workshop activation, and his crew eventually spent 125 hours mastering the task. Instructors evaluated progress by operational competence demonstrated, rather than hours of exposure.33
The start of "mini-aims" in September 1972 marked the transition from individual to team training. These sessions in the workshop simulator kicked off at 6:00 a.m., reveille on a mission day, and ran until bedtime at 10:00 p.m. The crew received instructions from a teleprinter as it would in flight. Voice contact with the ground was limited to times when the simulated flight brought the workshop over a ground station,....
.....but instructors could answer specific questions at any time. Mini-sims were an excellent investment of time; crews benefited from the integrated training, and flight planners uncovered a number of scheduling constraints.34
Pressures mounted in the last months before launch as training schedules were disrupted by simulator breakdowns, reviews, and last-minute demands on the astronauts. By January 1973 the first crew had fallen behind schedule and work weeks stretched to 60 hours. Late that month Bill Schneider moved the workshop launch from 30 April to 14 May because of delays at the Cape. The extra two weeks gave the training office a little breathing room, but the crews continued to work at a hectic pace.35
After the missions were over, Von Bockel was reasonably satisfied with the training program, though he would have made some changes. He had sought unsuccessfully to train only one backup crew, considering the 5000 man-hours invested in the second as unnecessary. Slayton, however, needed two; since one prime crew included a doctor and the other two a physical scientist, he had to be prepared to replace both. Von Bockel acknowledged that his instructors did not always stay ahead of the students. The astronauts were eager to learn, and program engineers seldom ignored their questions. "If the crew wanted to know something," he recalled, "people seemed to come out of the woodwork." Instructors, on the other hand, frequently had trouble getting information. Von Bockel recommended that in future programs, training materials should be prepared well in advance of instruction.
Skylab's biggest training problem, as indicated by the flights, was the long interval between instruction and performance of certain critical tasks. The last crew's deactivation and reentry came 13 weeks after training, and they made a procedural error-quickly rectified-that could be attributed to unfamiliarity with procedures. Von Bockel recommended that future missions allow time for refresher training during the flight. 36
i APP. E contains biographies of the Skylab astronauts