1966

By 1966 Apollo had lost much of the emotional support of Congress and the public that had welled up five years earlier in the wake of the Soviet Vostoks. The drop was reasonable, since the successes of the Gemini and Saturn I programs had led many Americans to believe the space race with Russia had been won. Moreover, domestic and foreign commitments, made primarily in 1965, to President Johnson’s “Great Society” and to Southeast Asia had placed more demands on tax dollars than had been foreseen. For fiscal 1967, NASA submitted a budget request of $5.58 billion, the President cut it to $5.012, and Congress chopped it to $4.968. Apollo came through virtually unscathed; but its follow-on, Apollo Applications, felt the weight of the Budget Bureau’s ax.¹

Obtaining funds for space exploration might be becoming more difficult, but most NASA officials had no time to worry about future programs. Apollo boilerplate flight tests had ended, and production spacecraft would soon fly atop the Saturn IB. Manned Spacecraft Center Director Robert Gilruth told Chris Kraft, Director for Flight Operations in Houston, to get his people started on the job ahead.

By January 1966, Kraft’s group had drafted a preliminary “operations plan.” In February it distributed a more complete version that pinpointed the responsibilities and functions of everyone connected with flights, beginning with Director Gilruth. The plan listed 19 specific documents, ranging from the “mission directive” prepared by Joseph Shea’s Apollo office to the “postflight trajectory analysis” compiled by Kraft’s own directorate, that would be essential in conducting a mission. Kraft also named John Hodge as flight director for AS-202 and AS-203. Kraft, himself, would direct AS-204, the first manned mission in the program.^{* 2}

Qualifying Missions

Before starting Apollo-Saturn IB launches, however, the operations people had to clean up one outstanding matter in New Mexico. NASA had hoped to finish the Little Joe II abort qualification program by the end of 1965, but on 17 December the Flight Readiness Board refused to accept the booster and canceled a launch set for the next day. A month later, at 8:15 on the morning of 20 January 1966, the last Little Joe II headed toward an altitude of 24 kilometers and a downrange distance of 14 kilometers. Then, as designed, the launch vehicle started to tumble; the launch escape system sensed trouble and fired its abort rocket, carrying the command module away from impending disaster. All went well on Mission A-004-the launch, the test conditions, the telemetry, the spacecraft (Block I production model 002), and the postflight analysis. The spacecraft windows picked up too much soot from the tower jettison motor, but the structure remained intact. Little Joe II was honorably-retired, its basic purpose - making sure the launch escape and earth landing systems could protect the astronauts in either emergency or normal operations - accomplished.³

After the last Little Joe flight, the scene shifted to Florida, where a Saturn IB, the first of the uprated vehicles^** slated to boost manned flights into earth orbit, was ready. AS-201 did not get a lot of publicity, but Dale Myers and his North American crew considered its spacecraft CSM-009 their “teething” operation:

It . . . proved out our procedures, our checkout techniques, and proved that this equipment [fitted] together. . . . And we got lined up so we [were] able to handle operations both at the Cape and [in Downey]. Although spacecraft 009 had some problems in flight . . . we got what we were looking for from the primary objective, . . . real good data on our heatshield, which we just can’t get any testing on in any other way.⁴

The Saturn IB first stage, assembled by Chrysler and with its eight H-1 engines built by Rocketdyne, had been erected on Complex 34 at Cape Kennedy in August 1965. Command and service module 009 was hoisted atop the booster on 26 December. Between those dates, the new S-IVB stage built by Douglas, with its single Rocketdyne J-2 engine, had been mated to the first stage, checked out, and fitted with an 1,800-kilogram “instrument unit,” or guidance ring, made by IBM Federal Systems Division. The top third of the stack - the spacecraft-launch vehicle adapter, the cylindrical service module, the conical command module, and the pylon-shaped launch escape tower - had been North American’s responsibility. Once they were stacked together, NASA assumed control. It took two pages to list AS-201’s test objectives, but NASA’s main aims were to check the compatibility and structural integrity of the spacecraft and launch vehicle and to evaluate the spacecraft’s heatshield performance as the vehicle plunged through the atmosphere.⁵

Spacecraft 009 assembly began in October 1963 and continued throughout 1964, with the inner-shell aluminum-honeycomb pressure vessel taking shape concurrently with the stainless-steel-honeycomb outer shell and its ablative heatshield. By April 1965, 009 had reached the test division at Downey, where it spent the summer. After a review at the factory on 20 October, NASA’s Apollo engineers approved the spacecraft for shipment to Cape Kennedy. Three months of servicing and checkout followed before AS-201 was ready for its voyage.

On 20 February 1966, launch technicians at the Cape began a three-day countdown, fully expecting some of the spacecraft’s systems to delay the launch. But weather turned out to be the chief problem, causing two postponements. At 5:15 on the afternoon of the 25th, the countdown resumed. Three seconds before ignition - at 9:00 the next morning - a computer signaled that pressure in two helium spheres on the Saturn IB was below the danger line. The count was recycled to 15 minutes before launch and stopped. Discussions waxed hot between Huntsville and Cape engineers. Since no one could be sure how serious the problem really was, the mission was scrubbed at 10:45. Deciding that the drop in pressure was probably caused by either an excessive flow of oxygen in the checkout equipment or leakage in the flight system, Wernher von Braun’s Saturn team recommended advancing the ground pressure regulator to maintain a higher pressure in the spheres. Kurt Debus’ Cape crew agreed, and the launch was back on the track by 10:57.⁶

At 11:12 a.m. 26 February, AS-201’s first stage ignited and drove the combined vehicles up to 57 kilometers where, after separation, the S-IVB took over, propelling the payload up to 425 kilometers. The second stage then dropped off, and the spacecraft coasted in an arc, reaching a peak altitude of 488 kilometers. At the zenith, the service module engine fired for 184 seconds, hurtling the command module into a steep descent. After a 10-second cutoff, the rocket engine fired again, for 10 seconds, to prove it could restart. The two modules then separated. The command module, traveling at 8,300 meters per second, turned blunt end forward to meet the friction caused by the growing density of the atmosphere.⁷

Both booster and spacecraft performed adequately. From liftoff in Florida to touchdown in the South Atlantic, the mission lasted only 37 minutes. The spacecraft was recovered by the U.S.S. Boxer two and a half hours after splashdown. AS-201 proved that the spacecraft was structurally sound and, most important, that the heatshield could survive an atmospheric reentry.

There were several malfunctions, mostly minor. Three were serious. First, after the service propulsion system fired, it operated correctly for only 80 seconds. Then the pressure fell 30 percent because of helium ingestion into the oxidizer chamber. Second, a fault in the electrical power system caused a loss of steering control, resulting in a rolling reentry. And, third, flight measurements during reentry were distorted because of a short circuit. Although Mueller agreed that the mission objectives had been met, these three problems would have to be corrected.⁸

The service module engine received instant attention. North American’s Robert E. Field and Aerojet-General’s Dan David (the engine’s Apollo manager) ordered an analysis of what had gone wrong. The engine had operated well enough to finish the mission, but Field and David had to be sure that the Block II engine (undergoing ground testing) would not run into a similar situation during a lunar mission. They learned that a leak in an oxidizer line had permitted helium to mix with the oxidizer, causing the drop in temperature and pressure.

For all of Houston’s insistence on redundancy, this was one major system that had no backup. And it was a vital system. Because of the lunar-orbit rendezvous decision, it had a variety of jobs: midcourse corrections on the way to the moon, lunar-orbit insertion, and transearth injection (placing the spacecraft on the homeward path) on the return voyage. Weight penalties forbade a second propulsion system; the service module engine had to carry its own built-in reliability.⁹

To allow time for studying and solving propulsion system problems and to prevent program delays, NASA managers shuffled the launch sequence. Since AS-203 was not scheduled to carry a payload, it would be flown before AS-202. Billed as a launch vehicle development flight, the third Saturn IB was to place its S-IVB stage in orbit for study of liquid-hydrogen behavior in a weightless environment.^** On 5 July 1966, AS-203 was launched from Kennedy to insert the 26,500-kilogram second stage into orbit. Ground observers monitored the S-IVB by television during its first four circuits, watching the 8,600 kilograms of liquid hydrogen remaining in its tanks. Despite some turbulence, the S-IVB appeared capable of boosting the astronauts on a flight path to the moon.¹⁰

Mission AS-202 was twice as complicated as AS-201. It would last 90 minutes, reach an altitude of 100 kilometers, and travel two-thirds of the way around the world. Launched on 25 August, AS-202 had a host of objectives, but the focal interest was service module engine firings. With clockwork precision, the motor fired four times, for a total operating time of 200 seconds. After a steeper reentry than expected, the command module was plucked from the Pacific Ocean near Wake Island by the recovery forces ten hours after liftoff and placed aboard the U.S.S. Hornet. On the carrier, specialists found that the heatshield and capsule had come through reentry admirably.¹¹

Troubles and Troubleshooters

Saturn IB flights, for the most part, ran smoothly in 1966. Unfortunately, this was not true for all of Apollo. Early in the year, NASA Apollo Program Director Samuel Phillips and a cadre of analysts completed a survey of vehicles and management at North American, after several months of probing into activities at Downey, Seal Beach, and El Segundo. Phillips’ group noted that organizational and personnel weaknesses were hampering the contractor’s attempts to meet command and service module schedules, but the biggest problem was the S-II second stage of the launch vehicle, which threatened to block the chances of flying an all-up vehicle on the first Saturn V launch.

Despite two successful ground tests, on 29 December 1965 and 12 January 1966, the S-II was behind schedule and in trouble. North American realized this and hired a new manager, Robert E. Greer, a retired Air Force general, to get S-II development back on the track. By spring, Greer and his troops had gone to the Mississippi Test Facility, near the Pearl River north of New Orleans, to begin an intensive ground test program. For 15 seconds on 23 April, the five J-2 liquid-oxygen and liquid-hydrogen engines roared into action, producing the designed thrust of 4.5 million newtons (one million pounds).¹²

Three more firings were attempted - on 10, 11, and 16 May - but the engines were cut off too soon by faulty instrumentation. In two more tests, on the 17th and 20th, the engines fired for 150 and 350 seconds. The next scheduled 350-second test, on 25 May, met problems when fire broke out in two places on the S-II. Three days later, while the stage was being removed from the stand, a liquid-hydrogen tank exploded, injuring five persons and damaging the test stand.¹³

Although it was a gloomy day in Mississippi, 25 May 1966 was still a milestone for Saturn V. Two states away, in Florida, NASA ceremoniously rolled out its 2,700-metric-ton, diesel-powered, steel-link-tread crawler-transporter loaded with the 111-meter-tall, 196,000-kilogram^* Apollo-Saturn vehicle. Just before this impressive mass began moving at a snail’s pace away from the Vehicle Assembly Building, NASA Deputy Administrator Robert Seamans said: “I for one questioned whether a vehicle the size of Apollo Saturn could get out to the pad . . . or not.” It could.¹⁴

However well the rollout augured for Apollo’s eventual success, right then the S-II stage was in trouble. NASA Manned Space Flight Director George Mueller began sending weekly assessments of S-II progress to J. Leland Atwood, warning the president of North American that the S-II stood a good chance of replacing the lunar module as the pacing item in Apollo. But Atwood already knew it. That was why he had hired Greer - to bring the S-II more attention at a higher level of management.¹⁵

Mueller also told Atwood that Phillips, on his return from the West Coast, had pointed out problems with the spacecraft. Both earth-orbital (Block I) and lunar-orbital (Block II) versions of the command module were being plagued during manufacturing by late hardware deliveries from subcontractors and vendors. The most troublesome had been the environmental control unit being developed by AiResearch. Phillips had chided the subcontractor by letter for its poor performance. In October Atwood admitted to Mueller that this system was the most serious threat to meeting spacecraft schedules for the first manned Apollo flight.¹⁶

Phillips’ troubleshooting set a pattern for Apollo in 1966; many managers and subsystem managers found themselves dealing, often full-time, with the difficulties in getting qualified vehicles to the launch pad. One of the Houston managers who spent a lot of time trying to straighten out some subsystem that was in trouble was Rolf W, Lanzkron. Phillips had asked Shea to send Lanzkron to General Electric in late 1965 to help get the manufacturer of the ground checkout equipment onto the right path. While Lanzkron was there, GE’s general manager for the program, Roy H. Beaton, commented in a letter to Phillips:

As you might well guess he beat the living h--- out of us, . . . spurring us on to more effective utilization of our previously mammoth efforts. Despite the bruises, we feel that we are a far more effective organization now as a result of his leadership.¹⁷

And Lanzkron traveled elsewhere. On one occasion he went to Phoenix, where the Sperry Company was having a hard time with the guidance and navigation gyroscopes. For several years, Sperry had been using a commercial detergent, one that many housewives use for washing dishes, to remove grease from the gyro’s bearings. Suddenly something went wrong - the grease was not coming off. Baffled at first, Lanzkron and Sperry’s own troubleshooters finally discovered that Procter and Gamble had changed its product to include an additive that was supposed to make it better for dishwashing.

It may have helped the housewife, but the “improved” product certainly hindered the cleaning of the bearings.¹⁸ Solving the gyro problem was a minor achievement in getting systems ready for flight. Over in the state of New York, however, more complex technical, financial, and managerial problems would demand the attention of many, many troubleshooters.

Lunar Module

By 1966, the lunar module had achieved some degree of maturity. Grumman had brought the lander out of the design phase and was trying to move it in the production line. But there were indications that the contractor was going to have problems. Control of in-house costs was fairly efficient; the company’s chief difficulties lay in overruns by its subcontractors. R. Wayne Young, MSC’s lunar module project officer, estimated that by the end of June Grumman would spend $24 million more than its allotted funds. Moreover, since late 1965 Grumman’s scheduling position had been shaky, with delays indicated virtually across the board.¹⁹

In light of these severe overruns, Houston sent representatives to Bethpage to discuss cost-reduction measures. This conference produced a list of items to either be reduced or chopped from the major subcontractors. Meetings were then held with project manager at each of the subcontractor plants to ram through cutbacks in requirements and manpower. The reviews, lasting a month and a half, culminated in tightened test procedures and performance requirements. To make sure that cost-reduction measures were enforced, Grumman switched from quarterly to monthly meetings with its subcontractors, inviting the appropriate Houston subsystem manager to attend.²⁰

Despite these actions, lunar module costs had not leveled off by late spring. In-house cost control and forecasting had also begun to deteriorate, aggravating the problems already encountered. Against this backdrop, Gilruth met with Grumman’s new president, Llewellyn J. Evans, to discuss cost control and management of subcontractors. At Evans’ request, Gilruth sent a management analysis group to diagnose and recommend ways to remedy the company’s weaknesses. The NASA Management Review Team, headed by Wesley L. Hjornevik of Houston, was composed of members from both Houston and Washington.²¹

Hjornevik’s team assembled at Bethpage in June. After a ten-day review, the team reported its findings to company corporate officers and NASA officials. Looking upon the Hjornevik team as a “personal management analysis staff,” Evans promptly carried out most of its recommendations on program management, costs, subcontractor control, and ground support equipment. To make sure all orders were followed and all decisions were relayed speedily to operating organizations, Grumman installed Hugh McCullough at the head of a Program Control Office. George F. Titterton moved from his vice-presidential suite to the factory building that housed most of the spacecraft’s managerial and engineering staff, thus ensuring a high degree of corporate-level supervision.²²

To bring about the kind of cost forecasting and control that NASA wanted, Grumman adopted “work packages” - breaking the program down into manageable segments, with strict cost budgets, and assigning managers to ride herd on each package. By linking tasks to manpower, program managers could better judge and control work in progress. This approach was a real departure from the commodity-oriented approach used by Grumman until that time. Shea watched these operations closely and on 19 September expressed his belief to Evans that the work packages could control costs and might even effect some modest reductions. In the next two months, however, costs still exceeded budgets in some areas. Unless discipline were enforced, Shea warned Titterton on 18 November, the work packages could turn into so many worthless scraps of paper rather than effective management tools.²³

Hjornevik’s team also discovered that no one person had been assigned responsibility for overall subcontract supervision. As a result, this whole area suffered from splintered authority. Grumman appointed Brian Evans to the newly created position of Subcontract Manager, reporting directly to Program Director Joseph G. Gavin, Jr. Evans then assembled a staff of project managers and assigned each to a major subcontract, with jurisdiction over costs, schedules, and technical performance. The strengthened structure was a welcome tonic; hardware deliveries improved and subsystem qualification moved ahead. Titterton also instituted quarterly meetings with presidents of the major subcontractor firms, similar to those held by Mueller for NASA’s prime Apollo contractors.²⁴

The weaknesses in ground checkout equipment, which had been a millstone around the contractor’s neck since the early days of the program, had developed because Grumman leaders simply had not recognized the immensity of the task. In February 1966 Phillips had pointed out to Shea that this equipment had paced the start of propulsion system testing at White Sands, had hampered in-house activity at Bethpage, and threatened to delay operational readiness of checkout and launch facilities at Kennedy Space Center.^* Shea replied that Grumman had put checkout equipment engineering and manufacturing on a 56-hour work week and was adding manpower to do the job.²⁵

Despite Shea’s reassurances and Grumman’s attempts at remedial actions, the system failed to improve measurably. Grumman had made progress in engineering design, which was about 80 percent complete; the bottleneck was in fabrication. Phillips and Mueller became thoroughly alarmed. They suggested that Grumman purchase components for the system from General Electric and other vendors who were having more success in the field. Subsequently, Grumman did put a variety of ground support items up for competitive bid.²⁶

At Bethpage, the Hjornevik team’s difficulty in assessing the ground support equipment problem hinged on the fact that Grumman did not have a coordinated plan. The team suggested that Grumman devote more attention to specific areas such as deadlines for drawing releases, an intensified production effort, and a daily status review by program management. Llewellyn Evans named John Coursen to oversee ground-support-equipment manufacturing and set aside a separate building for the fabrication workers, whose numbers had grown considerably. Procurement was also strengthened, with Robert Brader heading a staff of a dozen purchasing people. And, finally, a “GSE command post” was established to track day-by-day progress.²⁷

Actions at Bethpage were complemented by moves in Houston. In mid-July, Wayne Young appointed a team to meet with Grumman every month to assess status and tackle problems. At the end of the summer, with the last Gemini flight mission scheduled before the end of the year, Charles Mathews and William Lee shipped some surplus Gemini checkout items to Bethpage.²⁸ Collectively, these measures brought a dramatic turnaround in Grumman’s checkout equipment progress. As Gavin later observed: “The tide was turned in midsummer. We were effectively on schedule in mid-october.”²⁹

Successfully overhauling management practices and fighting rising costs were commendable accomplishments, but the lunar module faced problems in other areas that were equally dangerous to Apollo. Downey and the command module had been the big technical worry during 1965, Shea said at a meeting in San Augustine, Texas. The lander, which had begun the program a year late, must not be allowed to stumble into the same pitfalls. Echoing Shea’s sentiments, William Lee commented that Apollo would be in deep trouble if the lunar module followed the pattern of Gemini and the command module.³⁰

A significant hurdle vaulted about mid-1966 was the final solution of the long-overdue radar-optical-tracker question, the last of the lander’s subsystems to be settled. Engineers in the Manned Spacecraft Center’s Apollo office and in Robert E. Duncan’s Guidance and Control Division had promoted an “olympics” - a contest that pitted the radar against the tracker - and performance trials took place in the spring of 1966. After tests and presentations by competing contractors RCA and Hughes Aircraft Company, a review board chose the RCA radar. Although both systems could be developed within the same time and cost ($14 million), the radar had more operational flexibility than the less versatile tracker. The radar was heavier, but the weight had little influence on the choice, because of Grumman’s weight-reduction program of the previous year.

Perhaps the decisive factor in the selection was the outspoken preference of the astronauts. When asked by Duncan to support the olympics, Donald Slayton stated forthrightly: “The question is not which system can be manufactured, packaged, and qualified as flight hardware at the earliest date; it is which design is most operationally suited to accomplishing the lunar mission.” In light of recent experience, Slayton and Russell L. Schweickart, the astronauts’ representative on the evaluation board, believed that mission planning should make maximum use of Gemini rendezvous procedures and orbital techniques. This should include, they said, “an independent, onboard source of range/range rate information . . . with accuracy on the order of that provided by the existing LEM rendezvous radar.” So Grumman, which had slowed down radar development, shifted RCA back into high gear.³¹

The lunar module engines, too, were still having technical troubles, troubles that seemed to defy solution, although none of them were grave enough to threaten eventual success. For the descent engine, these included rough burning; excessive eroding of the combustion chamber throat; burning of the throttle mechanism pintle tip, where fuel and oxidizer met and combustion began; and difficulty in getting presumably identical engines to operate alike.

Design engineers at the Thompson-Ramo-Wooldridge (TRW) Systems Group^** made several changes in the pintle tip, the most significant being a switch to columbium to improve thermal characteristics. Other revisions included removing a turbulence ring around the interior of the chamber and realigning the flow pattern of the fuel that cooled the sides of the chamber wall. Although qualification testing was delayed six months, the problems seemed to be solved.³²

Ascent engine technical problems were more fundamental. Bell was plagued by fabrication and welding difficulties and by severe gouging in the ablative lining of the thrust chamber. The injector, which had been fitted with baffles to combat combustion instability encountered during the shaped-charge bomb testing, was also a culprit. After an engineering review and resulting design revisions, including strengthening of the weld areas, Houston suggested that Bell begin work on a backup model. That would be expensive, but something had to be done. Subsequently, an improved injector demonstrated better burning characteristics. Late in 1966, however, another worry cropped up.

At a Manned Spacecraft Center senior staff meeting on 4 November, Max Faget reported two instances of unstable combustion: one, during a firing test at White Sands, with a flat-face injector; the second at Bell, during a bomb test for design verification of a supposedly improved, baffled model. In both tests, damages had been extensive. At this point in the program, with the first two flight vehicles already late for delivery, these failures were ominous.³³

Schedule difficulties for the lunar module were nothing new, of course. Grumman had been under the gun from the very beginning, when the mode selection made the lander a late starter in Apollo. But during the summer and autumn of 1966, schedules became crucial. In July, every vehicle on the production line through LM-4 was late. Moreover, because of tardy deliveries by vendors, a serious bottleneck was shaping up in the assembly of LM-1. By late November, however, the earlier remedial actions seemed to be having some good effect and this continual slippage appeared to have slowed. At a briefing for Olin Teague’s congressional Subcommittee on NASA Oversight in Houston on 6 October, Shea had said that he expected the first lunar module to be shipped early in 1967.³⁴

By the end of the year, LM-1 and LM-2 were in the test stands at Bethpage, and LM-3 through LM-7 were in various stages of fabrication and equipment installation. But the coming of the new year did not yield the progress Shea had looked for the previous October. Toward the end of January, it was revealed that LM-1 would not reach the Cape in February, as expected.³⁵ In short, the moon landing might be delayed because the lander was not ready. But the mission planners could not wait for the Apollo engineers to iron out all the problems. They had to plan for a landing in 1969 and hope that the hardware would catch up with them.

Plans and Progress in Space Flight

In mid-1966, Phillips asked Shea to set up a three-day symposium to review the status of Apollo. At this 25-27 June conference, Phillips requested that the 75 NASA and contractor experts consider carefully such subjects as command and service module maneuvers, lunar module descent and ascent, lunar landing sites, and the length of the visit to the lunar surface.

Shea opened the discussions by listing 23 steps, or rules, in design and operational philosophy (see accompanying list) that had evolved since the lunar-orbit rendezvous decision in 1962. Owen Maynard, deliberately simplifying the many complexities of a lunar mission, described nine plateaus, of which he said:

It is useful to think of the lunar landing mission as being planned in a series of steps (or decision points) separated by mission “plateaus.” . . . The decision to continue to the next plateau is made only after an assessment of the spacecraft’s present status and its ability to function properly on the next plateau. If, after such assessment, it is determined that the space craft will not be able to function properly, then the decision may be made to proceed with an alternative mission. Alternate missions, therefore, will be planned essentially for each plateau. Similarly, on certain of the plateaus, including lunar stay, the decision may be made to delay proceeding in the mission for a period of time. In this respect, the mission is open-ended and considerable flexibility exists.³⁶

These plateaus, representing the amount of energy expended in going from one step to the next, were widely used by the Apollo engineering team to map the pathway to the moon’s surface and back again. The plateaus were, logically, (1) prelaunch, (2) earth parking orbit, (3) translunar coast, (4) lunar orbit before lunar module descent, (5) lunar module descent, (6) lunar surface stay, (7) lunar module ascent, (8) lunar orbit after rendezvous, and (9) trans-earth coast. Breaking the journey into these segments, with identified stopping places, made the Apollo mission seem less complex and fearsome to the planners.

Near the close of the session, Shea commented that all stages of the Saturn V were at Kennedy, preparing for a flight test during 1967; that both the first Block II command and service modules and the lunar module should fly that same year; and that the time for the first lunar mission was rapidly closing in. Shea urged everyone at the meeting to review and comment on current plans and progress.³⁷

It was also time to get an active experiments program under way. Mueller reminded Gilruth that, because of the limitations of 1966-1967 funding, NASA should generate as many of the experiments as possible, instead of relying on contractors. On 14 February 1966, however, Robert O. Piland’s Experiments Program Office (established at MSC in the summer of 1965) was asked by Homer Newell, NASA’s Associate Administrator for Space Science and Applications, to contract for the development of an Apollo lunar surface experiments package (ALSEP). The following month, the Bendix Systems Division of Ann Arbor, Michigan, received a $17-million contract to produce four ALSEP units. Bendix was a good choice, having worked with the Jet Propulsion Laboratory on experiments for the unmanned lunar exploration program.³⁸

Getting started on what to take to the moon was fine; getting the facility ready to handle what was brought back from the moon was also important. Houston had to develop a new kind of facility, the Lunar Receiving Laboratory. Its two major jobs would be to protect against back contamination from the moon and to keep the lunar samples as isolated from earthly pollution as possible. Meeting these quarantine and control requirements resulted in greater construction costs than initially estimated, but the Space Science Board of the National Academy of Sciences had been adamant in its demands that no expense should be spared:

The introduction into Earth’s biosphere of destructive alien organisms could be a disaster of enormous significance to mankind. We can conceive of no more tragically ironic consequence of our search for extraterrestrial life.³⁹

A conference of experts, sponsored by the board in July 1964, had reaffirmed the potential hazards of back contamination and recommended preventive measures. The following year, planning sessions among NASA, the Public Health Service, the Department of Agriculture, and the Army Biological Laboratories mapped out a construction plan and set up precautionary procedures.

Thus, by February 1966, George Low of NASA and James L. Goddard of the Public Health Service had presented Congress with a case for the construction of a lunar sample and quarantine facility with six functions:

1. Microbiology tests of lunar samples to demonstrate to a reasonable degree of certainty the absence of harmful living organisms returned from the lunar surface;
2. Biologically isolated transport of the astronauts and persons required to have immediate contact with them between the recovery area and the quarantine facility;
3. Biological isolation of the astronauts, spacecraft, and other apparatus having a biologic contamination potential, as well as personnel required by mission operations to have immediate contact with these people and this equipment during the quarantine period;
4. Biological isolation during all operations on the samples that must be carried out during the quarantine period;
5. Biologically isolated processing of onboard camera film and data tape that had been exposed to a potentially contaminating environment;
6. Performance of time dependent scientific tests where valuable scientific data would be lost if the tests were delayed for the duration of the quarantine period.⁴⁰

Shortly after congressional approval of the laboratory, Headquarters reluctantly agreed that Houston should manage the design and development of the laboratory without the aid of the Corps of Engineers. Mueller wrote Gilruth on 13 May 1966 that the facility must be ready by November 1967 at a cost not to exceed $9.1 million. Gilruth and Low established a policy board, headed by Faget, and placed Joseph V. Piland in charge of construction. A contract was awarded, ground was broken, and building began in August.⁴¹

During 1966, planners of Apollo’s upcoming operational phase studied the results of other programs for information that might be useful. Perhaps the two they scrutinized most carefully were Gemini VIII, which proved that one vehicle could find another in space and safely dock with it, and Surveyor I, which showed that a craft could land softly on the moon without sinking into the soil - at least in the area of Oceanus Procellarum.

Neil Armstrong and David Scott rode Gemini VIII into orbit on 16 March to chase an Agena target vehicle already in flight. An onboard radar acquired the target when the two vehicles were 332 kilometers apart, and the crew members saw the Agena when they were 140 kilometers away. Six hours into the flight, Armstrong and Scott, after inspecting the Agena closely, nudged the nose of their spacecraft into the docking cone, recording the first docking of two vehicles in orbit. Twenty-seven minutes later, Scott’s instruments told him that the spacecraft was not in the planned attitude. The docked vehicles then began to gyrate. Armstrong steadied the two craft with the thrusters, and Scott hit the undocking button. Almost immediately, the spacecraft started spinning at the rate of one revolution per second. Armstrong had to use the reentry control system^* to straighten out his vehicle. With the help of the flight controllers in Houston and along the Manned Space Flight Network, the crew made a safe emergency landing in the Pacific Ocean - rather than in the Atlantic, as planned.⁴²

Even before Gemini had chalked up the world’s first docking, the successful rendezvous of Gemini VI-A with VII the previous December had affected the thinking of Apollo mission designers. The inability of the Saturn IB to toss the command and service modules and the lunar module into orbit together had forced planners to consider “LM-alone” flights. Gemini’s successful dual missions suggested that it might be possible to launch a crew aboard a command module to hunt down a lunar module launched by a different Saturn IB. Two of the crewmen would then transfer to the lander and carry out an earth-orbital operation previously planned for a Saturn V flight.

Although the dual flight for Gemini had been greeted with enthusiasm, the proposal for an Apollo tête-à-tête met with resistance. John D. Hodge, Kraft’s chief lieutenant in the mission control trenches, said there would be problems in simultaneously tracking four booster stages and in operating two mission control rooms. Planning continued, anyway, and Howard Tindall started working up flight rules - such as which launch vehicle would go first, the one with the command and service modules (AS-207) or the one with the lunar module (AS-208). A spate of “Tindallgrams” ensued. By May, Tindall agreed with Hodge about the complexity of the proposed mission.⁴³

While planning proceeded on mission AS-207/208, which seemed to be gaining favor in Washington, the Soviet Union announced on 4 April that Luna 10 was in lunar orbit - a space first. As the Russian spacecraft sent back information on its voyage around the moon, the United States made its own unmanned lunar exploration spacecraft ready for flight. Surveyor I, launched by an Atlas-Centaur from Cape Kennedy on 30 May for a 63-hour trip, was programmed to land softly on the moon to test bearing strength, temperatures, and radar reflectivity and to send television pictures back to the earth. With only slight midcourse corrections, Surveyor I flew straight to its target. On 2 June, the vehicle fired its braking rockets, slowing its speed from 9,650 kilometers per hour to 640. Four meters above the surface of the crater Flamstead, it was moving at a mere 5.6 kilometers per hour. The three footpads touched safely down within 19 milliseconds of each other.

During the next two weeks, more than 10,000 detailed pictures were transmitted to the Goldstone antenna and processed at the Jet Propulsion Laboratory. They showed rubble scattered over the surface in the Ocean of Storms region. The Surveyor craft scanned the horizon and sky better than had been anticipated; its pictures of the stars Sirius and Canopus gave triangulations for its exact location; and its solar cells, radars, computers, and test gear all worked well. The craft did not encounter either hard or porous rock; nor did it find a moon covered by a thick layer of dust. It landed, instead, on a surface composed of finely granulated material with particles that adhered to each other and not to the spacecraft. After all the doubts and waiting, Surveyor I demonstrated that a lunar module could land safely on the moon and that its pilots could get out and walk on the surface.⁴⁴

The Astronauts and the Gemini Experience

Because of the heavy workload in Gemini and the upcoming missions in Apollo, Robert Gilruth had convinced George Mueller the previous year that he needed more astronauts. On 4 April 1966, NASA announced that 19 new flight candidates had been selected, bringing the roster up to 50.^* Donald Slayton presided over the corps, selecting and training the crews that were flying Gemini missions almost bimonthly.

Preparations for Gemini IX, the second mission scheduled for 1966, began the year in tragedy when its prime crew, Elliot See and Charles Bassett, crashed their aircraft into the building at McDonnell Aircraft Corporation that housed the mission spacecraft. Both were killed. Thomas Stafford and Eugene Cernan took over their duties. On 17 May, an Atlas booster attempted to put an Agena target vehicle into orbit for Gemini and failed. NASA launched a substitute vehicle, called the augmented target docking adapter, on 1 June. Stafford and Cernan were ready to follow, but problems with their guidance system and computer forced them to wait two days before Gemini IX-A was launched to start the chase. Once they caught up, they found that the launch shroud had stuck to the substitute target, making it look, as Stafford said, “like an angry alligator.” Although hopes for a second docking in space were dashed, Stafford and Cernan carried out rendezvous maneuvers in a variety of ways and Cernan spent two strenuous hours outside of the spacecraft, trying in vain to ride an astronaut maneuvering unit. Apollo mission planners examined these flight results closely, looking for better operations and training procedures, especially for extravehicular activity.⁴⁵

Six weeks after the Stafford-Cernan flight, on 18 July, John Young and Michael Collins pushed off aboard Gemini X to rendezvous with a pair of Agenas, one launched for their own mission and the other left in orbit by Gemini VIII. They had trouble making the initial rendezvous and used too much fuel; but, once hooked up to their Agena, they found both high-altitude flight, to 763 kilometers, and a meeting with the second Agena fairly simple. Using a hand gun, Collins had such a successful period outside the spacecraft that some NASA officials believed most of the extravehicular problems had been overcome.⁴⁶

But on 12 September, with Charles Conrad at the helm of Gemini XI, Richard Gordon found that moving about in space was as difficult as Cernan had said. Gordon became totally exhausted trying to hook a line between the spacecraft and target vehicle so the two craft could separate, spin, and produce a small amount of artificial gravity. He managed to finish the job, but at great physical cost. Nevertheless, Gemini XI expanded manned space exploration to a distance of nearly 1,400 kilometers above the earth to demonstrate that Apollo spacecraft could travel safely through the trapped radiation zones on their way to the moon. More importantly, perhaps, the crew carried out a first-orbit rendezvous, to simulate the lunar module lifting off the moon to meet the command module in lunar orbit, and made the first computer-controlled reentry. Conrad checked his onboard data with mission control, cut in his computer, and flew in on what amounted to an automatic pilot - much as Apollo crews would have to do to hit the narrow reentry corridor on their return to earth.⁴⁷

In the Gemini finale, NASA was intent on eliminating some of the mystery of why man’s work outside his spacecraft was so difficult. In preparation for this, the astronauts began underwater training, which simulated extravehicular activity more closely than the few seconds of weightlessness that could be obtained during Keplerian trajectories in aircraft. The pilot-controlled maneuvering unit was canceled after Gordon’s difficulties, so the Gemini XII crew could concentrate on the “fundamentals” of extravehicular movements. When James Lovell and Edwin Aldrin left the ground on 11 November, this was really the chief objective of their mission. By this time, crew systems personnel had attached enough rails and handholds here and there about the spacecraft to give Aldrin a relatively easy five hours of work outside the spacecraft.⁴⁸

Gemini made major contributions to Apollo and to the astronauts. Flight control and tracking network personnel learned to conduct complex missions with a variety of problems, and mission planners understood more about what it would take to land men on the moon. Rendezvous was demonstrated in so many ways that few engineers remembered they had ever thought it might be difficult. Perhaps the biggest gain for the astronauts was that 16 of the 50 had flown, operated controls, and performed experiments in the weightlessness of space.

Apollo astronauts, however, would rely more on simulators than on Gemini experience. There were, or soon would be, three sets of these trainers - two at Cape Kennedy and one in Houston - modeled after the command module and the lunar module. The simulators, constantly being changed to match the cabin of each individual spacecraft, were engineered to provide their riders with all the sights, sounds, and movements they would encounter in actual flight. Slayton had told George Mueller that the crews would need 180 training hours in the command module simulator and the flight commander and lunar module pilot an additional 140 hours in the lunar module trainer - about 80 percent more training time than the pilots of the early Gemini flights had required.⁴⁹

Preparations for the First Manned Apollo Mission

For a time, the mission called AS-204 had two flight plans. AS-204A, manned by Gus Grissom, Edward White, and Roger Chaffee,^* was “to verify spacecraft crew operations and CSM subsystems performance for an earth-orbit mission of up to 14 days’ duration and to verify the launch vehicle subsystems performance in preparation for subsequent operational Saturn IB missions.” The flight would be in the last quarter of 1966 from Launch Complex 34 at Cape Kennedy. AS-204B, on the other hand, would be an unmanned mission with the same objectives (except for crew operations), to be flown only if spacecraft and launch vehicle had not qualified for manned flights. And there were doubts. Gas ingestion in the service module propulsion system in AS-201 and the resulting erratic firing had caused some misgivings, although these had been somewhat allayed by AS-202.⁵⁰

As in early Mercury and Gemini manned flights, stress was laid on engineering and operational qualification rather than on experiments - whether medical or scientific. In December 1966, with only 9 experiments assigned to AS-204, 30 operational functions had a higher priority. And even then Slayton complained that the crew was not getting enough time in the new simulation and checkout facilities because of the experiments. Despite his arguments, the second Apollo crew (Walter Schirra, Donn Eisele, and Walter Cunningham, with Frank Borman, Stafford, and Collins as backups), announced on 29 September, was scheduled for a heavier workload of experiments.⁵¹ As technical troubles came to the fore, however, emphasis on experiments shifted.

North American should have shipped spacecraft 012 from Downey to Kennedy in early August, but “eleventh hour problems associated with the Command Module Environmental Control Unit water glycol pump failure resulted in a NAA NASA decision to replace the ECU with the unit from SC 014.” The Customer Acceptance Review revealed some environmental control items that still needed to be corrected, but NASA allowed North American to ship 012 to Florida on 25 August anyway. Once it arrived, John G. Shinkle, Apollo Program Manager at Kennedy, complained about the amount of engineering work that still had to be done. More than half of it, he said, should have been finished before the spacecraft left the factory.⁵²

While flight-preparation crews were having problems, Grissom, White, and Chaffee were finding bottlenecks in training activities. The chief problem was keeping the Apollo mission simulator current with changes being made in spacecraft 012. At the Cape, Riley D. McCafferty said, there were more than 100 modifications outstanding at one time. Grissom, McCafferty later recalled, would “tear my heart out” because the simulator was not keeping up with the spacecraft. Eventually, the first Apollo commander hung a lemon on the trainer.⁵³

Getting the spacecraft to the Cape did not really improve conditions. The environmental control unit needed to be replaced again, which held up testing in the vacuum chamber. AiResearch shipped the new unit from its West Coast plant to Kennedy on 2 November. Within two weeks, it was installed and testing was begun. It was then returned to California for further work. By mid-December, the component was back in Florida and in the spacecraft. Meanwhile, the service module had been waiting in the vacuum chamber for the command module. While it was sitting there, a light shattered, and falling debris damaged several of the maneuvering thrusters.⁵⁴ But this was not the only cause for worry about the service module.

On 25 October at the North American factory, the service module for spacecraft 017 was undergoing routine pressure tests of the propulsion system’s propellant tanks when the tanks suddenly exploded. No one was injured, but North American and NASA engineers were baffled as to the cause for the next few weeks. The tanks had not been overpressurized, test procedures had not been relaxed, and no design deficiencies were apparent; yet the fuel storage tank had failed with a bang. Since the service module for spacecraft 012 had been through identical tests, Shea was vitally concerned with unraveling this riddle before Grissom and his group flew.

William M. Bland and Joseph N. Kotanchik were sent from the Manned Spacecraft Center to Downey to help North American hunt for the trouble, and Houston set up a parallel test to verify the results. They learned that the methanol (methyl alcohol) employed as a test pressurant fluid caused stress corrosion (or cracking) of the titanium alloy used for the propellant tanks. Replacing the methanol with a fluid that was compatible with titanium would eliminate this problem. In the meantime, the tanks were removed from service module 012 and found to be free of any dangerous corrosion.⁵⁵

In September, Mueller reminded Gilruth of the upcoming Design Certification Review. Board membership would, he said, include himself, Gilruth, von Braun, and Debus. The group met on 7 October and agreed that the space vehicle conformed to design requirements and was flightworthy, provided several deficiencies were corrected. Phillips sent the list to Lee B. James at Marshall, Shinkle at Kennedy, and Shea at the Manned Spacecraft Center, urging speedy clearance. Shinkle had already registered his complaints about spacecraft 012; now he added that Houston should insist on better spacecraft being shipped to the Cape. He pointed out the major problems that had been found: a leak in the service propulsion system, problems with the reaction control system, troubles in the environmental control unit, and even design deficiencies in the crew couches that required North American engineers to travel from Downey to the Cape to correct them.⁵⁶

In early December, NASA reluctantly surrendered its plans for launching the first manned Apollo flight before the end of 1966. Mueller and Seamans then reshuffled the flight schedule, delaying AS-204 until February 1967 and scrubbing the scheduled second mission. Experimenters who had planned to place their wares aboard Schirra’s spacecraft were brushed aside. Following AS-204, NASA planned to fly the lunar module alone and then a manned Block II command and service module, No. 101, in August 1967 to rendezvous with unmanned LM-2, the LM being lofted into orbit by a Saturn IB in a mission dubbed AS-205/208.

If everything went well, NASA hoped to get two crews besides Grissom’s spaceborne before the end of 1967, with at least one riding a Saturn V. Replacing the Schirra team as the second Apollo flight crew were James McDivitt, David Scott, and Russell Schweickart (backed by Thomas Stafford, John Young, and Eugene Cernan) for a workout of the command module and lander in earth orbit. To fly the Saturn V mission, AS-503, NASA picked Frank Borman, Michael Collins, and William Anders (with Charles Conrad, Richard Gordon, and Clifton Williams as backups); they would ride the spacecraft into orbit and out as far as 6,400 kilometers above the earth.⁵⁷

After all this flight shuffling, the Apollo program seemed to be in fair shape at the end of 1966. North American had finished the last of the manufacturing work on the earth-orbital version of the command and service modules on 16 September and could now concentrate on improving the lunar-orbital spacecraft.⁵⁸ The lunar module still had problems, but Grumman was making headway in resolving them. The pathway to the moon appeared to be clearing, as NASA stood on the threshold of Apollo manned space flight operations.