Chapter 13

Agena on Trial

Manned space flight and NASA faced the new year of 1966 in an ambiguous position. High achievement had marked 1963, capped by the exciting and important “76” mission at the very end of the year. But the key to more sophisticated missions, the Agena, was in serious technical trouble. Only with Agena could Gemini hope to realize a range of still-to-be-attained goals - docking, re-rendezvous, rendezvous with two separate targets during a single mission, and high-altitude flight goals that would be indispensable to Apollo, the program to land men on the Moon. But many doubted that Agena could be ready in time to meet Gemini’s tight launch schedules. Year’s end saw Agena’s career in manned space flight once again called into question - and this time a substitute target had already been approved for development.

Agena, though most critical, was not the only problem. Extravehicular activity (canceled in the three previous missions) was supposedly ready to enter a more advanced stage. Unexpected development troubles demanded a last-minute effort (reminiscent of Gemini IV) to qualify equipment. Edward White had succeeded in his “space walk,” but NASA faced a tougher EVA task - testing the Air Force’s Astronaut Maneuvering Unit (AMU), a far more complex personal propulsion system than White had used. Step-by-step progress having been skipped, the EVA set for Gemini VIII in mid-March had to bridge the gap.

At the beginning of 1966, then, the Gemini program had met with success in seven straight missions, five with crews aboard. Not all its goals had been attained, but many had. Now the Apollo program neared its operational stage. Might NASA halt Gemini to concentrate on Apollo? Administrator James Webb had used similar reasoning to conclude Project Mercury earlier than many desired. Although George Mueller, Associate Administrator for Manned Space Flight, knew of no move afoot to close down Gemini, he foresaw that many engineers in Houston might worry that they were nearly out of jobs. To assuage their misgivings, in December 1965 he made a case for flying all 12 Gemini missions. Even a cursory glance at the program’s aims, Mueller said, showed healthy returns for nearly every item. While medical fears had been erased by the outcome of 14 days in space, NASA still needed to perfect techniques for rendezvous and extravehicular activities. Then, too, an experienced cadre of flight crews was essential, not only for flying missions but for astronaut and flight control training as well. LeRoy Day, Mueller’s Deputy Director for Gemini, passed this reassurance on to Gemini Program Manager Charles Mathews in Houston.¹ That potential morale threat allayed, the engineers could focus on such technical problems as making Agena work.

Sickness to Health

Agena’s woes were by now chronic. The Gemini Agena target vehicle (GATV) was pacing the program by mid-1965, prompting GPO to consider removing the first production model, GATV 5001, from its job as a test vehicle so it could be used in Gemini VIII.² All such plans went up in smoke with the explosion of GATV 5002, which ignited the most demanding piece of engineering detective work in the entire Gemini program. Efforts to cure Agena’s ailments spanned more than four months, much of it on a three-shifts-a-day, seven-days-a-week schedule.³

An hour after the Agena failure of 25 October 1965, Mission Director William Schneider had left Houston for Florida, where Colonel John B. Hudson, SSD Deputy Commander for Launch Vehicles, had called a meeting of a subpanel of the Agena Flight Safety Review Board for the 26th.* The subpanel members had learned enough from telemetry data to list the tasks to be done: find out why the Agena had failed and what the fixes would entail for design, performance, and schedule; decide if it would be possible to use GATV 5001 and how long it would take to get it ready for launch; and begin cutting red tape that might slow the work.⁴

Lawrence A. Smith, Gemini Manager for Lockheed, had already sent the taped record of telemetry signals to the plant in Sunnyvale, California, where W.R. Abbott took charge of the failure-search team. Most likely causes of the disaster were a “hard-start” backfire or an electrical short; Abbott’s group soon narrowed its search to the engine as the more probable source of trouble. After reporting to Major General Ben Funk’s full Agena Flight Safety Review Board, Hudson took his subpanel to Sunnyvale on 1 November; they agreed with Abbott’s analysis that a hard start (similar to an automobile engine backfire) had been the cause and that it had resulted from fuel being injected into the firing chamber before oxidizer.⁵

The problem was rooted in NASA’s original specification for a Gemini target vehicle able to start and stop its main engine five times during a mission, in contrast to the Standard Agena’s two-start engine. This 150 percent increase in demands on the engine at once raised the problem of fuel and oxidizer economy. In the two-start engine, the oxidizer began flowing first, while a pressure switch restricted fuel flow until a given amount of oxidizer had reached the firing chamber. This was known to enhance the engine’s starting characteristics, but it was also wasteful. Oxidizer leaked through before engine firing, and some continued to flow after shutdown; the oxidizer would be gone long before the fuel ran out. So Lockheed accepted a proposal by the engine subcontractor, Bell Aerosystems Company, to remove the pressure switch and thus allow fuel to enter the chamber first.⁶

Abbott concluded that in space the presence of fuel in the thrust chamber (perhaps in considerable quantity) had caused the engine to backfire when the oxidizer reached the chamber, causing an explosion. When Funk’s review board met in Los Angeles on 3 November to make tentative plans for an engine requalification program, Abbott presented his findings, which were discussed the next day.⁷

But Abbott’s and Hudson’s groups were not the only ones working on the problem. At NASA Headquarters, Associate Administrator Robert Seamans told George Mueller to form a NASA review board to look into all aspects of the failure, both technical and managerial. Mueller appointed MSC Director Robert Gilruth co-chairman of a Gemini Agena Target Vehicle Review Board and asked Air Force Major General Osmond Ritland to serve with Gilruth.** ⁸

And down at Cape Kennedy, Lockheed’s Wulfgang C. Noeggerath was working with MSC engineer Horace E. Whitacre to pinpoint the cause of the failure. Unsure that the two of them could explore the matter in the depth needed, Whitacre suggested that Lockheed sponsor a symposium of rocket experts from around the nation. Noeggerath convinced his superiors that it was a good idea.⁹

The two-day symposium began on 12 November, with 19 scientists and engineers in attendance*** ¹⁰ Noeggerath and Whitacre told the visiting experts that the most likely cause of the Agena explosion had been a premature engine shutdown. Engine firing had produced severe oscillations and mechanical damage. Temperature decreases had indicated fuel spillage. When electrical circuitry failed, the engine stopped, but a valve that controlled tank pressure as fuel was being used remained open. As fuel stopped flowing, pressure built up in the tanks, which ruptured and destroyed the vehicle - a planned flight safety precaution. Whitacre and Noegerrath also reported that the engine had not been tested at simulated altitudes higher than 34,000 meters, since no one believed that the environment above that level made any difference for engine firings.

Although Abbott’s backfire theory accounted for the oscillations that had triggered the explosion, not everyone agreed that a single cause was enough to explain what happened. But the symposium could come up with nothing better. On 15 November, it recommended to SSD that engines should be modified so oxidizer entered the chamber first and should be tested at simulated altitudes closer to where Agena would be working - above 76,000 meters.¹¹

Funk now formed a “super tiger” team of three senior engineers**** to review everything that had been found about the explosion and to suggest some answers to the NASA review board. The three agreed with oxidizer starting and with firings at simulated altitudes above 76,000 meters. They also wanted Bell Aerosystems to conduct ground ignition tests for data on engine-firing characteristics. The super tigers presented these recommendations at a meeting in Houston on 20 November,# then to the Gilruth-Ritland review board, which approved them. Lockheed announced the formation of a Project Surefire Engine Development Task force to carry out the program. This did not end the analysis of the trouble. Reports and recommendations from other NASA centers committed to come to Gilruth until 9 March 1966, one week before the Gemini VIII flight.¹²

An Alternative Target

While Agena’s sponsors labored to nurse it back to health in time for the Gemini VIII mission, McDonnell engineers had been thinking of other ways to achieve rendezvous and docking. During launch preparations for the Gemini VII/VI-A mission, McDonnell’s Gemini Program Technical Director, John Yardley, invited several NASA officials to his motel room in Cocoa Beach, Florida. He outlined a plan for making a poor man’s target by bolting a target docking adapter to the rendezvous and recovery section of a spacecraft and fitting it to the Atlas booster. An enthusiastic Mueller told Mathews to prepare a defense of the concept for Seamans, who would have to approve it. To avoid any hint that a new development program was in the offing, they decided to call it simply an “ATDA,” for augmented target docking adapter, which accurately reflected its status as a rearrangement of already developed and qualified hardware.¹³

An immediate question was whether the Atlas launch vehicle could handle the proposed ATDA; it was much lighter in weight than the Agena but lacked an engine to boost it to orbit. A call to General Dynamics in San Diego posed the weight question (without disclosing the as yet unapproved plan) and received an encouraging response.¹⁴

By 5 December 1965, Mathews had the case for the alternate vehicle ready. While Day filled in Seamans’ staff, Mueller described the plans to the Associate Administrator himself, who approved it. Four days later a statement of work for the ATDA was ready and McDonnell began building the substitute Gemini target.¹⁵ This target adapter became something of a sword of Damocles over Lockheed, a weapon that GPO was willing to use at more than one level. Jerome Hammack spurred Lockheed’s efforts by sending Smith a picture of the alternate vehicle (often called the “glob"), and Mathews asked Flight Crew Operations for an alternative flight plan, eliminating all Agena maneuvers from Gemini VIII.¹⁶

Project Surefire was already running into trouble. The crucial simulated high-altitude tests of the modified engine could only be run at the Air Force’s Arnold Engineering Development Center in Tennessee, but it was booked solid. Time was running out for Gemini VIII, scheduled for mid-March 1966. John Hudson flew to Vandenberg Air Force Base, where he persuaded General Bernard Schriever, Air Force Systems Commander, to sign a letter moving Agena to the head of the line at Arnold.¹⁷

The Agena test program also got a priority from NASA, when Mueller decided that Apollo lunar module engine tests at Arnold could be slipped. On 17 December, after Bell had completed the Project Surefire modifications, the Air Force accepted the main engine for GATV-5003. Bell had already begun the series of 48 sea-level firings that the super tiger team had recommended.¹⁸

Setting up the test program, however, was only part of the problem. Another source of delay loomed in Mueller’s demand that GATV-5003, which had arrived at the Cape on 18 January, undergo a static firing before it was committed to Gemini VIII. A worried Agena team, fearful that this demand would so delay their efforts that the outcome of the Arnold tests would be meaningless, met late into the night of 4 February 1966, the eve of another meeting of the Flight Safety Review Board, to discuss means of getting Mueller to postpone his decision.

The next day, Lockheed’s Smith and Aerospace’s Hohmann told Mueller that a static firing at this point was just not worth the delay it would entail. Mueller quizzed them closely and asked for a written report on the pros and cons of static firing as well as on everything that had been done to ensure that Agena would not fail again. The pressure was still on, but Mueller had at least not closed the door on Agena - it still had a chance to compete with the ATDA for Gemini VIII.¹⁹ February 14 became the deadline for making the choice, while GPO kept working on both the ATDA and Agena.

Late in January, GPO engineers went to St. Louis to conduct a design review of the ATDA, and Gemini procurement received word to put through the final papers for its purchase.²⁰ ATDA development was quick because its parts had already qualified for space flight and good luck held its cost down. A spacecraft rendezvous and recovery system fished from the sea for postflight examination after the Gemini VI-A mission could be used in building the ATDA. McDonnell put it together by 1 February, and NASA conducted the acceptance review the next day.²¹ The stand-in was ready to assume the starring role.

Agena was clearly trailing its rival, but its sponsors hoped to regain lost ground when the second act of the test program began at Arnold on 7 February.²² Meanwhile, Hohmann and Smith had sent their written reports on static firing to Mueller. In Hohmann’s view, static firing was mainly useful for training launch crews, not for proving rockets. He pointed out that Mercury-Atlas 1 had failed at launch, even though it had been static fired, and that static firing would not have disclosed GATV 5002’s problems. Smith stressed the penalties in money and time.²³ A quick poll of opinion from NASA Headquarters and the manned space flight centers supported the Hohmann-Smith viewpoint,²⁴ and Mueller dropped the notion of static-firing the Agena.

That was a plus for Agena’s prospects, but the test program at Arnold produced less happy results. After the first six tests, problems with mismatched hardware had already compelled GPO to direct McDonnell to speed up its ATDA testing.²⁵ The seventh test, on 12 February 1966, was nearly fatal. Fuel lines contaminated by alcohol and water caused a hard start that badly damaged the engine. Fortunately, Bell had just about finished its series of sea-level tests and could send that engine to replace the damaged one. As Agena’s time seemed to be running out, its proponents literally worked around the clock, juggling, cajoling, scheming, begging, and snarling when necessary, to reach what had begun to seem an impossible goal. More than once, Day and Mathews pleaded with Mueller to keep the Agena. Finally he gave them one week to return the vehicle to good health in time for the review board meetings to be held in Washington 6 and 7 March. Day later recalled his feeling, all during Project Surefire, that Mueller was just putting pressure on MSC and Lockheed and never really intended to cancel the Agena. Mueller did object to the cost of the modifications. He was not willing to brook what appeared to be a $15 million overrun and so was studying proposals to cut Agena out of the program, using the ATDA or two spacecraft for rendezvous.²⁶

On 1 March, the new test series began at Arnold. By the end of the fourth day, 22 firings at simulated altitudes of from 83,800 to 114,300 meters had proved the success of the modifications. Meeting as planned, the Design Certification and Air Force Flight Safety Review Boards approved the modified Agena for flight.²⁷ The Agena had been requalified just in time to fly on Gemini VIII, for its rival, too, was ready, having also completed its test program on 4 March. Now the ATDA went into storage at Cape Kennedy, to be called out if the Agena again faltered.²⁸

Extravehicular Activity Preparations

While one group in the manned space flight program struggled to restore the Agena, another faced qualification problems, on a smaller scale, with extravehicular flight equipment. White’s spectacular venture into space during Gemini IV was backed by a comparatively simple technology. That first step into space had merely required an astronaut to leave the cabin and see what he could do. White did it in grand style, making up his flight plan as he went along. His successor would not be so free as he attempted such specific tasks as retrieving experiment packages. So easy a first venture misled planners into thinking that EVA would present few, if any, problems. No one really worried when space walks were deleted from the missions between Gemini IV and VIII, even though the Astronaut Maneuvering Unit was still scheduled for Gemini IX.

The Air Force, however, was disconcerted. Colonel Daniel McKee, head of the Air Force Systems Command Field Office at MSC, complained about being kept in the dark about the plans for White’s exercise. His office should have been involved in the orderly planning that was to lead to the use of the AMU, since at about $12 million that was the single most expensive Defense experiment to be carried on Gemini. The AMU was designed to make an extravehicular astronaut independent of spacecraft systems. A boxlike backpack with sidearm controllers, it consisted mainly of three beams and supporting shelves on which such parts as tanks to store the hydrogen-peroxide propulsion system and the life-support oxygen supply were mounted. Because the spacecraft was so small, the AMU was housed back in the adapter section. The astronaut would go out the hatch tied to a tether, make his way to the rear of the spacecraft, and strap himself into the AMU. This special propulsion system weighed about 76 kilograms (168 pounds), which was no burden at all in the weightlessness of space.²⁹

On 20 September 1965, NASA had named the crew for Gemini VIII. Neil Armstrong, a civilian test pilot with long experience in the X-15 rocket research aircraft program, was the command pilot, as he had been in the backup crew for Gemini V. His fellow crewman, David Scott, was new to the Gemini program. For the backup crew for Gemini VIII, Navy Lieutenant Commanders Charles Conrad and Richard F. Gordon, Jr., the pattern was similar. Conrad had been pilot on Gemini V, and Gordon was newly assigned to Gemini.³⁰

Some three weeks before the crew was announced, McDonnell held a briefing on the extravehicular gear Gemini VIII’s could carry. It comprised two major units: an Extravehicular Life Support System, called the ELSS by the engineers who worked on it, and an Extravehicular Support Package, known as the ESP or, more commonly, the backpack. The life support system was a chestpack designed to do just that - feed vital oxygen to the spaceborne astronaut from the spacecraft’s supply, from a primary source in the backpack, and from its own emergency supply. The backpack did more. Designed like the AMU to fit into the spacecraft adapter section, it carried, aside from its own oxygen supply, a radio and 8 kilograms (18 pounds) of propellant for a zip-gun maneuvering unit. The backpack was connected to the spacecraft systems by an 8-meter oxygen-hose tether. Once the astronaut had switched over from the spacecraft to the backpack oxygen supply, he could add a lightweight 23-meter tether to the shorter hose and, theoretically at least, maneuver as far as 30 meters from the spacecraft.³¹ Armstrong attended the St. Louis meeting and asked for help in the training program. The crews would need a realistic spacecraft adapter with which to practice donning the backpack. Armstrong also wanted the pilot to leave the spacecraft in the altitude chamber and test the combined backpack-chestpack.³²

As soon as he was assigned to the mission, Scott began concentrating on the extravehicular exercise, eventually going through over 300 airplane zero g parabolas and more than 20 hours on an air-bearing table. The astronauts practiced EVA maneuvers, supported by an air cushion of 0.0254 millimeter (0.001 inch), on a table roughly 6 by 7 meters. They used a zip gun to move from one place to another, which gave them some idea of what it would be like to start and stop in space. This strenuous training raised some questions.³³

Scott’s zip gun had about 15 times more propellant than White’s and used Freon instead of oxygen as fuel, further multiplying the gun’s total impulse, since Freon has a density about three times greater than oxygen. How oxygen acted in vacuum was fairly well known, hut Scott worried about how Freon would behave. One problem soon showed up: at low temperatures, the Freon caused the zip gun’s poppet valve to stick open when triggered, and the escaping gas threatened to tumble the astronaut in space. New seals solved the problem and two new shutoff valves added a safety factor.³⁴

By December, Scott and Armstrong were voicing a number of doubts about the equipment, ranging from nitpicks to serious complaints. One that fell into the latter class was the threat of an oxygen ejector in the chestpack freezing and blocking the flow of oxygen from both the spacecraft and the emergency supply in the chestpack. The life support system had been icing up during tests. Although test conditions were more severe than Scott would meet in space, he could hardly be expected to ignore the warning. The designers obliged by installing 20-watt heaters near the ejector.³⁵

Another problem was the tangle of umbilicals, tethers, and jumper cables that made donning the chestpack inside the spacecraft so difficult. During early tests, Scott found his movements restricted and his vision nearly blocked by his pressure suit while he was trying to connect everything. Late in December 1965, however, Scott satisfied himself that he could strap on the unit, hook it up, and open the hatch in the McDonnell altitude chamber at a simulated 46,000 meters. Scott went through a full-dress rehearsal in the last few weeks before flight, in the MSC 6-meter vacuum chamber, putting on the chestpack inside the spacecraft, going outside, and then donning the backpack housed in the adapter.³⁶

Mission Planning Comes of Age

Technical problems in qualifying the Agena and the extravehicular equipment commanded center stage in Gemini VIII preparations, adding to the already heavy burdens of planning the mission. Project Gemini was entering a more advanced phase, as both spacecraft and target faced missions of growing complexity that would test their capabilities to the limit. Program leaders had to balance their concern for reaching the program goals against the dangers of trying too much too soon. A persistent problem such as the Agena presented could not but raise doubts and cause second thoughts about going forward with some as yet untried operation. Even under the best of circumstances, trying to foresee and counter everything that might go wrong with four major dynamic systems - spacecraft, booster, Atlas, and Agena - made mission planning an arduous task. With major technical difficulties further clouding the issue for Gemini VIII, plans changed quickly and often.

In the summer of 1965, MSC’s Mission Planning and Analysis Division had started tailoring a plan for Gemini VIII, the first results of which were discussed on 26 and 27 August. Among the alternative modes of rendezvous being considered was a rendezvous sooner than the fourth revolution of the spacecraft - the “standard” rendezvous that had been scheduled for Gemini VI. Despite doubts that the flight control team could support any rendezvous earlier than that, the scheme called “M equals 2” (rendezvous in the second revolution) being studied by the mission planners was worth thinking about. Another subject was a proposed phantom rendezvous with an imaginary target, requiring a thrust from the Agena’s main engine of at least 150 meters (500 feet) per second, to take place shortly after the first sleep period. The pilot would then exit the spacecraft for more than two hours of extravehicular activity - that is, floating freely around the world! After that, the spacecraft would undock and withdraw from the Agena, to return later for a second rendezvous. Finally, Gemini VIII’s Agena would be left in orbit as a passive target for Gemini IX.

No sooner had the Gemini VIII plan been committed to paper than caution flags were raised. One issue was an old one that the earlier crews had fought - sleeping alternately. Lockheed recommended that one astronaut remain awake whenever the spacecraft and Agena were docked. Mathews consulted with Whitacre, then denied the request on the ground that sleep at this time (after launch, rendezvous, and docking, and before EVA) was necessary for both men. Besides, Whitacre’s analysis showed that the tracking network could cope with almost anything that might go wrong. Another question was time. Fuel cell development problems had imposed a limit of two days on rendezvous flights. Could so elaborate a plan be carried out in such a short time? Maybe the time could be expanded. Since the fuel cell’s troubles seemed headed toward resolution, McDonnell was asked to see if later rendezvous missions could be extended to three days.³⁷

Meetings continued throughout the fall of 1965, as spokesmen for NASA Headquarters, McDonnell, and MSC began to stress re-rendezvous, which they thought might be good training for Apollo. Discussions on firing the two Agena propulsion systems remained inconclusive, as did talk about flying for three days. When a McDonnell study indicated that the fuel cells could support a 72-hour flight, if all supplies were carefully husbanded, that question appeared to be settled.³⁸ Firing the Agena main engine while docked with the spacecraft, however, was finally rejected for the same reason that it had been on Gemini IV - it was not yet deemed safe enough. That meant the phantom rendezvous was out.³⁹

Toward the end of February 1966, with problems seemingly well in hand, a “final” version of the flight plan appeared. Like the Gemini VII plan, this was more an outline than a precise schedule of events. Crew and flight controllers had a range of options to deal flexibly with circumstances as they arose in the course of the mission.⁴⁰

Operations planning was being paralleled by experiments planning. By November 1965, the Manned Space Flight Experiments Board at NASA Headquarters had approved eight tasks for Gemini VIII. Eventually, ten experiments were approved for the mission,* three of them requiring extravehicular activity. Of these, two were scientifically oriented - S-9, Nuclear Emulsion, to expose an experimental package to radioactivity in space (especially in the South Atlantic Anomaly), and S-10, Agena Micrometeorite Collection, to study the micrometeorite content of the upper atmosphere. In the third, proposed by the Department of Defense, Scott would use a power wrench for weightless work. He would go to the adapter area, pull out a box containing a torqueless motor-driven wrench, use the tool to take five nuts off a special plate, and then rebolt the plate to the box. This simple task - with and without knee tethers - would be compared with doing the same thing on the ground to show the differences in working in one gravity and in weightlessness. Scott and George C. Franklin of the Flight Crew Support Division decided to augment this experiment by adapting a cheap standard socket wrench to fit the nuts and the pressurized glove. They believed that comparing the muscle-powered and the electrically-operated tools would say something useful about energy usage in space.⁴¹

Mission plans and flight schedules were inseparable, and Apollo again began to intrude. Apollo mission 201 was planned for February 1966; if there were any delays, it would slip into March. The problem was not with launch pads nor, in most cases, with people. A tracking ship, the Rose Knot Victor, was the source of conflict. For Apollo 201, a suborbital flight, the ship would be sailing the Atlantic Ocean. But its station for Gemini VIII was in the Pacific. Mueller ruled that the Gemini flight had priority; but Apollo 201 flew as scheduled on 26 February, giving the slow-moving Rose Knot time enough to keep its date with Gemini VIII.⁴²

Flight Control also shifted for Gemini VIII. Christopher Kraft, who had directed flights for Mercury and all Gemini missions through VII/VI-A, had to leave Gemini to begin planning for lunar landing missions as Apollo neared operational status, although he expected to keep an eye out for Gemini lessons that might be of use to Apollo. Kraft’s move left Gemini Mission Control short of experienced flight directors. His successor, John Hodge, who headed the Flight Control Division, divided flight direction into 12-hour shifts with Eugene Kranz, Chief of the Flight Control Operations Branch. Clifford E. Charlesworth, flight dynamics officer on past Gemini missions, began training as a flight director.** ⁴³

In the two weeks before the scheduled launch, equipment problems remained a threat. The extravehicular gear, in particular, was still in trouble, with lines icing and valves cracking. Then, at Cape Kennedy, the spacecraft environmental control system began acting up; and, over on pad 14, Atlas fueling ran into some difficulties. These last two problems did cause a day’s delay, from 15 to 16 March. Then everything was ready to go.⁴⁴

The Whirligig

On 16 March 1966, five months after Walter Schirra and Thomas Stafford had been left at the starting gate in NASA’s first attempt to launch two vehicles toward rendezvous on the same day, NASA tried again. This time nothing marred the countdown of the Atlas-Agena or the Gemini space vehicle.⁴⁵

The target launch vehicle lifted from pad 14 at 10 o’clock in the morning. Its trajectory was at first low and to the right (south) of the intended flightpath. The sustainer engine rammed the target back on track. In a little more than five minutes, the Atlas had done its job. Now it was Agena’s turn. After a short coast, its secondary propulsion system burst into life. The crucial test for the Agena came with the firing of its main engine, and the engineers crossed their fingers and held their breaths. But it worked. The engine ignited and carried the target into a 298-kilometer circular orbit.⁴⁶ Planners had wondered if the Agena could so position itself that astronauts could catch it. The answer was yes!

With one up and one to go, attention turned to 19. Fourteen minutes before the Atlas-Agena lifted, Armstrong and Scott slid through the spacecraft hatches into their couches. As the flight-preparation crew helped harness Scott to his parachute, they found one of its catches full of glue. Backup command pilot Conrad and McDonnell pad leader Guenter Wendt began digging it out. Just a little thing like that, Scott thought, “might have cost us a launch,” but he could not help smiling as he watched Conrad sweat over the job. The catch came unglued and Gordon, the backup pilot, tried the fitting a few times to prove to Scott that it was working. Learning of the Agena’s nearly perfect orbit, Armstrong said, “Beautiful, we will take that one.”⁴⁷

Given the Agena’s orbital parameters, the Gemini launch vehicle should lift off at 10:40:59 am. The powerful engines of Titan II throbbed into life exactly on time, and Armstrong and Scott felt the hold-down bolts shear for breakaway. GLV-8 started off a little low, as had Atlas, but soon straightened to boost the 3,788-kilogram (8,351-pound) spacecraft into an elliptical orbit 160 by 272 kilometers.⁴⁸

After the first hurdle had been vaulted, the next challenge was catching the target. Procedures were much the same as those for Gemini VI-A, although this time there was no friendly target to point its attached transponders toward the spacecraft’s radar. Armstrong and Scott began the chase 1963 kilometers behind the Agena.

Thirty-four minutes into the flight, the Sun set and, in the engulfing darkness, the crew could see brilliant fires streaming from their spacecraft’s thrusters. As the radiator in the adapter expelled water, the thrusters fired to compensate for a sideward turn. The Carnarvon, Australia, tracking station told them the radiator was not much of a problem and passed to them the Flight Director’s “go” for a day’s flight.⁴⁹

Over the Pacific, the two astronauts had some time to sightsee. Molokai, Maui, and Hawaii hove clearly into view. Armstrong tried to see Karnai and Oahu, but cloud banks obscured them. Minutes later, Scott said to his partner, “We’re going over Baja California now. Can you been it?” But Armstrong had his eyes on the Los Angeles ship basin in the other direction, and his response was, “Oh, look at all those ships!” Armstrong then spotted the Rogers Dry Lake bed. He looked for, but was not certain he found, Edwards Air Force Base, where he had spent seven years piloting experimental airplanes. Over Texas, both men wanted to see if they could spot their homes, but work preempted this scenic interlude. At the low point of their first circuit of Earth, Armstrong aligned the inertial platform for a height adjustment maneuver. At 1:34 hours elapsed time, he touched off a five-second burst of the thrusters for a small retrograde change in velocity, to lower the apogee slightly. Armstrong noticed a problem in cutting off residual thrust. This resulted in varying computer readings and made it difficult to tell the exact deceleration obtained.⁵⁰

On their mission, Schirra and Stafford had been so preoccupied that they had not taken time to eat, which left them hungry, as well as tired, when they caught up to Borman and Lovell. Scott and Armstrong knew they would be very busy all three days of their mission, so each grabbed a package of food and started preparing a meal, which seemed to take longer than they thought it would. When they had to stop and align the platform for a maneuver to raise the perigee, they placed the food packages against the spacecraft ceiling. Weightlessness was handy.⁵¹

Nearing second apogee (2:18:25 hours), Armstrong fired the thrusters to add 15 meters per second to their speed. Again, tail-off residuals made it hard to get a computer reading.⁵² After this maneuver, Armstrong and Scott pulled their food from the ceiling. Although Armstrong’s chicken and gravy casserole had been mixed with water for half an hour, it was still dry in spots and not much like home cooking. But he finished it and washed it down with fruit juice to keep from dehydrating. Then he tried a package of brownies, which were stuck together and crumbly. They were hard to eat without scattering weightless scraps all over the cabin.⁵³

The next maneuver was designed to push the spacecraft into the target’s orbital plane. Armstrong yawed Gemini VIII’s nose 90 degrees south of the flight path. Over the Pacific Ocean, 25 minutes before completing the second revolution (2:45:50 hours), the command pilot punched the aft thrusters to produce a horizontal velocity change of 8 meters per second. He waited for the ground controllers to tell him if any adjustment was needed. Hearing nothing, he assumed his thrusting had been correct. Over the Guaymas, Mexico, tracking station, Lovell, the Houston CapCom, suddenly cut in on the remote site line to order him to add 0.6 meter per second to his speed. With only a minute to get ready, there was little time to turn the spacecraft and no time to align the platform. “It was a pretty quick loose burn . . . without much preparation,” Scott said.⁵⁴

Armstrong and Scott then began the rendezvous radar test. They did not expect to get radar contact as quickly as Schirra and Stafford had, but the Westinghouse development team had promised target acquisition at 343 kilometers. The radar locked on solidly at 332 kilometers, which was good enough.⁵⁵

Over the Tananarive tracking station, 3:48:10 hours after launch. Armstrong nosed the spacecraft down 20 degrees and applied the aft thrusters for an in-plane (with the target) velocity change of 18 meters per second. This gave them a nearly circular orbit close to 28 kilometers below that of the target. The spacecraft was now in position to start the terminal phase of rendezvous.⁵⁶

The crew sighted a shining object 140 kilometers ahead, which must be the Agena. After closing to a range of 102 kilometers, all doubts were erased - the target gleamed in the sunlight. Scott switched the computer from the catchup to the rendezvous mode and watched the distance dwindle on the slide, automatically. Just before sunset, the Agena suddenly disappeared, but at twilight its acquisition lights blinked into view.⁵⁷

When the Agena was at the proper angle (10 degrees) above them, Armstrong aligned the inertial platform for the translation maneuver. Then he pitched Gemini VIII’s nose up 31.3 degrees and canted the vehicle 16.8 degrees to the left. At 5:14:56 hours, ground elapsed time, the command pilot fired his aft thrusters, later making two small corrections. High over the Coastal Sentry Quebec tracking ship, stationed near Antigua Island, at 5:43:09 hours, he braked the spacecraft. Since he could see the Agena, Armstrong judged his braking action by eye as Scott called out radar range and range rate. At a distance of 46 meters, relative velocity between the two vehicles had been canceled. The second rendezvous in the Gemini program had been achieved.⁵⁸

For 36 minutes after rendezvous, Armstrong’s delicate maneuvering kept his spacecraft on station with the target vehicle. As the command pilot drove, Scott inspected the Agena - checking antennas, docking lights, and the like. Finding it hard to see all of the target’s instrument panel displays near the docking cone, he used the telescopic sight of a hand-held sextant. But a really good look would have to wait until they were docked, when these instruments would become a second dashboard. Meanwhile, Armstrong studied the general appearance of the Agena. It seemed stable, and he nudged the spacecraft to within a meter (about three feet) of the target. Then, at 6:32:42, Keith K. Kundel, CapCom on the Rose Knot Victor, radioed, “Go ahead and dock.”⁵⁹

Armstrong eased Gemini VIII toward the target at a barely perceptible rate of 8 centimeters (3 inches) per second. “About two feet [60 centimeters] out,” he told the Rose Knot Victor. In a matter of seconds, Armstrong gleefully reported, “Flight, we are docked! It’s . . . really a smoothie - no noticeable oscillations at all.” For a moment, the flight controllers in Houston could not realize that they had really done It. Then pandemonium broke loose, with back slaps, hand shakes, cheers, and tremendous grins.⁶⁰

Because there had been some difficulty in verifying the Agena’s uplinked* stored program commands for the planned docked yaw maneuver and in loading the target’s velocity meter, the flight controllers suspected that Agena’s attitude control system might be misbehaving. In fact, Lovell, on the remote link through Tanarive just before the spacecraft passed out of communications range, told the crew, “If you run into trouble and the attitude control system in the Agena goes wild, just . . . turn it off and take control with the spacecraft.” With this warning raging in their ears, Armstrong and Scott began their docked chores.⁶¹

The Agena was designed to obey orders from the spacecraft, as well as from ground control. Scott commanded the target’s attitude control system to turn the vehicle combination 90 degrees to the right. It took five seconds less than the full minute expected. Scott next dialed an order to start the Agena’s tape recorder and looked over toward Armstrong. As he did, his gaze skimmed the control panel in the spacecraft. Something had to be wrong - Gemini VIII should be in level flight, but the “ball” indicator showed a 30-degree roll. He knew there was no use checking the horizon out the window, as they were passing through Earth’s shadow. There would be no help from the ground tracking stations either; they were still out of communications range.⁶²

"Neil, we’re in a bank,” Scott said. He thought perhaps his spacecraft attitude ball had tumbled, but Armstrong’s indicator showed an identical mark. The command pilot managed, with bursts from the OAMS, to stop the motion temporarily, but it soon started again. Their immediate reaction was to blame the Agena. As soon as the vehicles were steady enough, Scott commanded the target to turn off its attitude control system, as the communicator had instructed. For four minutes, the two craft steadied and straightened up; the trouble seemed to be over. Armstrong started maneuvering to get the docked vehicles into the correct horizontal position; suddenly they began to roll again, faster and faster. “What’s the problem now?” the pilots wondered. They were supposed to do a small test to find out what stress and strain the linkage between the two vehicles could tolerate. That issue was now academic; the immediate question was whether it could stand up under these wild gyrations.

While Armstrong struggled with the controls, Scott photographed the interaction between the two vehicles out of his spacecraft window. The command pilot soon reported that the OAMS propellant had dropped to 30 percent, a strong clue that a spacecraft thruster might be causing the trouble. While Armstrong fought the controls, Scott cycled the target vehicle switches off and on and off again. Then Armstrong jiggled the spacecraft switches as well, to see if they could isolate the problem. Nothing they did seemed to have any effect.⁶³

The crew realized that they would have to break away from the Agena to analyze the situation. Past simulation training gave them no clues to what was happening or how to handle it. Scott transferred control of the Agena to the ground stations (which had been locked out to prevent spurious signals), and Armstrong labored to steady the vehicles enough to divorce them. “Go,” Armstrong said, and Scott hit the undocking button. Armstrong gave the thrusters a long hard burst, and the spacecraft pulled straight back.⁶⁴

Almost immediately, suspicion about a spacecraft control problem became an established fact as the spacecraft rolled even faster. “And then we really took off,” Armstrong and Scott later reported. Gemini VIII soon came into acquisition range of the Coastal Sentry Quebec. James R. Fucci, CapCom aboard the ship, was concerned and perplexed. He could not get a solid electronic lock-on, but a blinking light signal indicated that the craft had undocked. Unaware that the spacecraft was rolling, so the antennas could not remain in position, he put in a call to the crew to try to find out about these strange signs he saw on his console.

Fucci: Gemini VIII, CSQ CapCom. Com check. How do you read?

Scott: We have serious problems here . . . we’re tumbling end over end up here. We’re disengaged from the Agena.

Fucci: Okay. We got your SPACECRAFT FREE indication here. . . . What seems to be the problem?

Armstrong: We’re rolling up and we can’t turn anything off. Continuously increasing in a left roll.

Fucci: Roger. [37 seconds later] Gemini VIII. CSQ.

Armstrong: Stand by.

Scott: We have a violent left roll here at the present time and we can’t turn the RCS’s off, and we can’t fire it, and we certainly have a roll . . . stuck hand control.⁶⁵

After backing away from the Agena, the spacecraft had started to whirl at a dizzying rate of one revolution per second. Armstrong suspected that the maneuvering thrusters were about finished. He and Scott were also having trouble seeing the overhead panel dials; their physiological limits seemed near. They were dizzy, and their vision was blurred. Something had to be done. “All that we’ve got left is the reentry control system,” Armstrong said. “Press on,” Scott responded. The two men began to throw switches to cut out the OAMS and cut in the reentry control system. Armstrong tried his hand controller - nothing. Scott tried his - still nothing. They started switching circuitry again - maybe something had been set in the wrong position.

The hand controllers responded!

Armstrong steadied the motion and then turned off one ring of the reentry control system to conserve fuel. He then carefully reactivated the maneuver thrusters; now they were able to tell that No. 8 had “failed on” - that is, it had stuck open!⁶⁶

Using the reentry control thrusters meant that the Gemini VIII mission would have to come to an end as soon as possible. That was a mission rule. True, the spacecraft was operating in a backup mode - but it was the prime mode for reentry. If these thrusters developed leaks, the crew would have absolutely no means of getting the spacecraft into position for the critical retrofire that would return them to Earth. Attitude control before and after reentry was essential to reenter the atmosphere safely. Here was a case where the fail-safe maneuvers that Headquarters had insisted on early in the program were impossible - there was virtually no maneuverability left in the orbital thrusters. Armstrong and Scott also remembered, wistfully, that Kraft, the flight controllers, and engineers had nursed other missions to completion. Could the same be done for them now? This was but a fleeting hope, as the Hawaiian tracking station communicator told them to get their spacecraft into position for reentry.⁶⁷

Gemini VIII’s problems were certainly the most frustrating of any Gemini had yet encountered. The flight control team’s ability to respond to real problems on previous missions, keeping spacecraft flying to wring all useful data from failures as well as successes, had bolstered confidence in the program and promoted “real-time” planning. But Gemini VIII’s failure had forced the astronauts to resort to a last-ditch mode for attitude control before the ground crews had a chance to provide the options that might have allowed the flight to go on.

John Hodge, in his first trial as chief flight director, now had only one choice left - which contingency recovery landing area would be best? If he waited much longer, it would take a full day (or 15 revolutions) for the crew to reach a splashdown point from which they could be quickly recovered. Since the orbital track had precessed westward, landing during the sixth or seventh orbits would have to take place in the Pacific Ocean. When the Landing and Recovery Division recommended a touchdown in the seventh circuit, Hodge agreed.

Kranz had dropped by to listen to the spacecraft and target docking. Since Hodge had been at the flight director’s console for 11 hours, he and Kranz decided that the second shift should report for duty immediately, catch up on all information, and direct the final phases of the mission. Had the flight continued for three days, reentry would have taken place on Kranz’ shift, anyway, and he and his men had more practice in recovery procedures than Hodge and his group.⁶⁸

The engineers who had worked so hard on the Agena’s problems found their situation just as exasperating as that of the flight controllers. After the docking, Smith, Harold W. Nolan, and others from Lockheed had retired to nearby motel rooms to celebrate the momentous event. Very shortly, Smith called Nolan, saying, “We’ve got trouble!” Nolan switched on his television, only to hear newscasters reporting that the Agena was at fault. Smith’s motel room became the initial Lockheed failure-analysis command post, the first guess being that the target’s attitude control system had failed.

Many other engineers and program officials also heard about the spinning spacecraft while out of touch with minute-by-minute developments. Mueller, for instance, had remained at Cape Kennedy only through launch and the early phases of the mission. Then he took off for Washington to attend the annual Robert H. Goddard Memorial Dinner, sponsored by the National Space Club. The pilot of the NASA aircraft heard what was happening over the plane’s radio and informed Mueller. They returned to Florida, where Merritt Preston met Mueller’s party with a motorcycle escort, the prelude to a hair-raising ride to the old Mercury Control Center in time for spacecraft retrofire.⁶⁹

Most of NASA’s leaders at Headquarters had, in fact, already headed for the Goddard dinner - the prestigious social event of the year for the space community. At the opening reception, Deputy Administrator Seamans** was called to the telephone to learn of Gemini VIII’s plight. He immediately phoned Houston Flight Control and learned that the spacecraft spinning had been stopped. When he told the chairman of the dinner about the trouble, Seamans was asked to make a brief announcement: he said the flight would have to be aborted, but the crew seemed in no immediate danger. Vice President Hubert H. Humphrey, the principal speaker, asked to be told as soon as the crew had been successfully recovered. Before he had finished his address, Humphrey was able to inform his listeners that Armstrong and Scott had landed safely. Seamans vowed that never again would he be caught in a public position during the critical phase of any succeeding flight. He needed privacy and better communications with the Control Center.⁷⁰

As a rule, McDonnell (the spacecraft contractor) sent several of its experts from the Cape to Houston after launch and first orbit to be available as troubleshooters. On 16 March 1966, a NASA Gulfstream left Florida for Texas with about 14 passengers, among them several high-ranking McDonnell engineers. Over New Orleans, the pilot cut in a commercial radio broadcast to the cabin. The announcer was talking about an imminent recovery in the Pacific. This was all the startled passengers heard, since the news announcement ended there. Something had obviously gone wrong, but what was it? There was nothing to do but wait until they got to Houston.

Raymond Hill, McDonnell’s Gemini manager at the Cape, recalled that his company’s policy changed radically after “I . . . was caught with my pants down.” In the future, senior McDonnell officials - Hill, Walter Burke, John Yardley, and Robert Lindley - would not be in transit at the same time during a flight. Hill stayed at the Cape, Burke went to Houston for the first day of the flight and then back to St. Louis, and Yardley and Lindley went to Houston and stayed until the mission was over. McDonnell specialists, who had previously remained in St. Louis to handle test set-ups and to answer questions by telephone and teletype, were shifted, along with their subcontractor counterparts, to Houston to work directly with GPO systems engineers during the mission.⁷¹

Meanwhile, Navy recovery forces in the Pacific were swinging into action. A destroyer, the U.S.S. Leonard F. Mason, steamed at flank speed toward the expected landing point 800 kilometers east of Okinawa and 1,000 kilometers south of Yokosuka, Japan.⁷²

With Gemini VIII now flying over the southern latitudes, Kranz had only three tracking stations in position to keep in touch with the crew*** - Coastal Sentry Quebec, Rose Knot Victor, and Hawaii.⁷³ The spacecraft was in darkness over the Congo when Kranz’s Houston flight controllers began the final countdown for retrofire. Through the remote stations, Scott reported, “Props OFF,” and Armstrong said, “Hang in there.” Seconds later, Scott said, “Okay. Four retros bed in AUTO RETROFIRE. . . .”

Armstrong was worried that he and Scott might land in some remote wilderness where they would be hard to find. He later said he had been thinking of the steamship Andrea Doria, which had gone down in the Atlantic on 26 July 1956. Although the liner’s radios were operating, it had taken the rescue vessels a day and a half to find the sinking ship. He wanted Scott to doublecheck his every move - “I keep thinking there’s something we’ve forgotten about,” he said, “but I don’t know what it is.” Scott answered reassuringly, “We’ve done everything, as far as I know.” Over China, Gemini VIII slipped down in the fringes of the atmosphere.⁷⁴

Everything clicked off properly during descent. As they neared a landing, Armstrong asked his partner, “Do you [see] water out there?” Looking into the first faint light of dawn, Scott replied, “All I see is haze.” Then his voice quickened, “Oh, yes, there’s water! It’s water!” Less than two minutes later, Scott yelled, “LANDING - SAFE.” The flight had lasted 10 hours 41 minutes 26 seconds.⁷⁵

The crew went quickly through the postlanding checklist, putting switches and valves in their correct positions. Then antennas were extended so they could communicate with the recovery forces. “Naha RESCUE 1, Naha SEARCH 1,” Scott called, but no answer came. They were not very worried, however, as Houston Flight Control had told them the rescue planes would get to them shortly and the Mason should reach them in three hours. This meant their landing had been very close to the contingency touchdown point.⁷⁶

Several aircraft, including two HC-54 Rescuemasters - one from Naha Air Base, Okinawa, and the other from Tachikawa Air Base, Japan - had raced to fetch the crew. The HC-54 from Naha got there first. Suddenly the pilot shouted, “I got it!” He had seen the spacecraft, with its main parachute in full bloom, drifting to the ocean’s surface. Three pararescuemen were equipped and ready to jump. Armstrong and Scott saw one of the three as he parachuted down. Because of the waves, the frogmen had trouble hooking the flotation collar to the spacecraft. The rough sea also made them queasy, a feeling shared by the astronauts. But the swimmers persisted and secured the collar within 45 minutes of spacecraft landing. Demonstrating excellent cooperation with NASA and careful planning, the Department of Defense recovery forces had reacted to the emergency landing as though it were normal. Armstrong and Scott had few complaints about recovery in this remote area.⁷⁷

Three hours later, as promised, the Mason pulled alongside and fastened a line to the spacecraft. Climbing the Jacob’s ladder in sea swells of 4 to 5 meters was hard, but they made it. On deck, the tired astronauts managed smiles and greetings for the welcoming sailors. Still feeling nauseated, the Gemini VIII crew headed immediately for sick bay. Medical personnel helped them strip off their pressure suits. Their undergarments were soaked with sweat. They were thirsty, but clinical examination showed minimal dehydration. The Mason reached Okinawa the next day, and the two astronauts flew on to Hawaii, then home.⁷⁸

Once the manned phase of the Gemini VIII mission was over. Hodge and Kranz turned back to the target vehicle. Because Scott had the foresight to pass the control of the target back to the ground, there was a chance to put the Agena through its paces and see how it reacted to commands. There was still hope that the Agena for Gemini VIII might be used as a passive target for Gemini IX or X.

Agena’s Solo

After the undocking, the Agena had stabilized quickly. In the 15th revolution over the Carnarvon station more than 21 hours after launch, Flight Control commanded the main engine to fire twice, to place the Agena in a circle 407 kilometers above Earth. The first burn produced its half of the goal, but the second did not. Instead Flight Control found the parameters were now 407 by 626 kilometers.⁷⁹

Melvin F. Brooks, the Agena systems monitor in Flight Control, immediately began conferring with the Lockheed engineers to figure out what had happened. They suspected that the vehicle’s center of gravity had been miscalculated. How could they command the vehicle to offset this? On the next main engine burn, the center-of-gravity compensation attempt failed. Brooks and Lockheed engineers huddled again. What could be wrong now? They finally agreed that there also seemed to be trouble in the yaw hydraulics, allowing the engine to gimbal more than it should. The target’s orbit now measured 211 by 476 kilometers.⁸⁰

If this Agena were to become Gemini IX’s or X’s passive target, there were two major problems to contend with, and Flight Control had to decide what to do about them. There was definitely too much fuel aboard* and the orbit was still too high. Hodge and his controllers decided not to try any more plane-change maneuvers; they would simply try to get the vehicle to the altitude they wanted. The next firing, a retrograde maneuver, convinced them that they had the hang of operating the vehicle. So Flight Control concentrated on reducing the fuel supply in both the primary and secondary tanks.⁸¹

In all, ten maneuvers were made using the two propulsion systems, sometimes with both firing at once. This was considerably more than the five starts required by the contract. The Agena’s command and communications system had accepted a total of 5,439 commands (45 from the Gemini VIII spacecraft). Lockheed’s contract had only called for 1,000.⁸²

Just before the Gemini VIII-Agena docking, Scott had commented that he “bet those Lockheed guys are just jumping up and down.” And so they had been. Jubilation died quickly when the news came that the spacecraft was in trouble. Agena’s solo maneuvers wiped away any suspicions of wrongdoing on its part. Somebody else must shoulder the blame for Gemini VIII’s early landing. Why had thruster No. 8 failed in the open position?

From its landing spot in the Pacific Ocean, the spacecraft had been hauled back to its place of birth - the McDonnell plant in St. Louis - so the engineers could analyze its problems. Set up in a controlled laboratory where the investigations could proceed unmolested, the spacecraft was checked over completely for more than a month. Only the most probable cause of the trouble could be identified. Scott Simpkinson’s evaluation team decided that:

The valves on thruster 8 opening unintentionally was probably caused by an electrical short, . . . there were several locations in the spacecraft at which the fault could have occurred.

To prevent a recurrence of the thruster problem, McDonnell changed the attitude control circuit switch so that when it was in the “off” position no power could go to the thrusters. Formerly, turning the power to the electronics packages did not stop power going to the thrusters. They could still fire.⁸³

Thus, the Gemini VIII mission ended on a dissonant chord - high success (the first space docking), undeniable failure (abbreviation of the mission), and much relief (safe recovery of the astronauts from a dangerous situation). The timing of the failure was especially frustrating. Being out of communications left the flight controllers and engineers helpless. Time after time in later interviews they repeated: if that spacecraft had just been over a ground station, telemetry would have told them that thruster No. 8 was firing continuously; they could have told the crew what to do before the reentry control system was activated and it was too late. Although the Gemini team was chagrined that the crew had been forced to land early, the knowledge that docking could be achieved with relative ease somewhat assuaged their anguish. Moreover, the Agena solo had demonstrated the target vehicle could help fly more elegant missions. There would be no pause in the program.⁸⁴

Press on to Gemini IX!