Failure Is Not an Option
Page 19
The crew for Gemini 5 was one of my favorites. The commander was Gordo Cooper, my welcome wagon driver from day one at the Cape, the friendliest of the original astronauts, and one who had shown great poise on Mercury 9 when his electrical system went to hell. Pete Conrad, his partner, had the most engaging personality in the second group of astronauts. Notwithstanding his flap with Hunter at Carnarvon, I liked Conrad’s toothy grin, intensity, and frank and total openness. You had no doubt that he would do what he said.
August 21, 1965, Gemini-Titan 5
The Kraft, Kranz, and Hodge mission team cycle on Gemini 5 was virtually the same as on Gemini 4. At the first site after reaching orbit we became concerned about an unexplained drop in the fuel cell oxygen pressure. Since this was the first experience with fuel cells and cryogenics in orbit, there was no frame of reference for Kraft’s Red Team EECOM.
During the first pass over Carnarvon, Cooper reported that the oxygen pressure was dropping rapidly. Troubleshooting, the crew found that the circuit breaker for the heater that warmed up the spacecraft’s liquid oxygen supply for the fuel cells had tripped. EECOM, listening to the crew’s report, recommended resetting the circuit breaker and cycling the heater switch. The crew concurred and acted on his recommendations as they coasted across the Pacific.
By the time the spacecraft arrived over Hawaii, the pressure was well below the point at which the fuel cells were expected to keep running—and continuing to drop. The crew advised Hawaii that they had abandoned the rendezvous test and had powered down most of the spacecraft equipment. The operation of the fuel cells required a very precise pressure balance between the oxygen and the hydrogen and the pressure of the water produced in the cell. We had no data to guide us when the cells were operating at the current level. Additional equipment was turned off as the spacecraft continued on a path across the United States.
It was customary for the second shift of controllers to be present for launch. This gave the launch team some extra controller resources if problems occurred. It also gave the second shift a good start at working and understanding the problems they would be given at handover. John Aaron from my shift had been sitting next to Kraft’s Red Team EECOM since launch. He walked over to Kraft’s console with the plot of the pressure data, his face reflecting the grim news. The rate of pressure loss, however, was starting to decrease. Faced with a possible fuel cell failure, but needing to buy more time to evaluate the situation, Kraft elected to continue for a few more orbits to see what would happen. The only good reentry options would occur for the mid-Pacific landing areas on orbits four, five, and six; then we would have to shoot the gap of poor coverage, a no-man’s-land. Kraft advised recovery to deploy aircraft to the orbit six landing area. Within the hour, six aircraft, a destroyer, and an oil tanker had received orders to proceed to the landing area southeast of Hawaii.
As Kraft continued to weigh the options, my control team was reporting to its consoles wondering whether we would get a shot at the problem. Meanwhile, Kraft was masterfully playing the options and assessing the alternatives. When the fluctuating oxygen pressure finally stabilized, and the fuel cells were still operating, I knew Kraft was going to go for a full day. He queried Cooper: “Gordo, I think the oxygen has bottomed out. We’ve got thirteen hours on the batteries. I think we should go for it.” Cooper’s response was immediate. “I was hoping you would say that. Let’s give it a go.”
I was mentally going through my handover questions for Kraft when he abruptly stood up and started to put his headset away. Surprised, I glanced at his log. At shift handover the controller coming on shift would check the log entries of the controller he was relieving in order to see if there were any outstanding items or problems. The log was a summary of the status of the mission. There was no set plan to follow—and I suspected he was ready to leave for the press conference. So I asked him, “Chris, what do you want to do?” His reply was crisp and curt. “You’re the flight director, it’s your shift. Make up your own mind.” Kraft had given me the job to shoot the gap!
The spacecraft flies sixteen orbits per day. During a mission, Go NoGo assessments are made of the spacecraft and crew to determine whether is safe to continue to the next point. These decisions are made by factoring in the spacecraft mission status and the locations of the naval recovery forces. The decisions are for orbits one, three, six, and sixteen, and multiples thereafter. The major daily Go NoGo is the one that shoots the gap of poor ground station coverage between orbits six and sixteen.
This event in my life as a controller stands out as the moment I finally came of age in Mission Control. Chris was right. I was the flight director, it was my shift, my decision, and I had better get going. Like my first solo with Mr. Coleman, it was time to spread my wings and give it a shot. I was damn happy to have Aaron on my shift. He was born to be a systems engineer; he could tell me in plain language what the status was of any element of the spacecraft.
In Mission Control, there is no such thing as a first team. Every team must have the leadership and technical knowledge to sustain the effort during a crisis. The White Team controllers stepped up to the plate and took their turn at bat.
The Trench worked with the recovery forces to select the best landing areas as we shot the gap. Aldrin, in between calls to the crew, worked with the flight planners to develop alternatives to the rendezvous. The systems team led by Aaron refined the power-down procedures in case we had to use the batteries. The team was on top of the job, and its response to the problem was electrifying. Within a few hours the engineers at McDonnell were running a laboratory test of a fuel cell in St. Louis to replicate the problem we were having in flight.
At John Aaron’s urging, I decided to run a load test on the fuel cell. As the team talked, we developed a plan to perform a power-up followed by an oxygen purge to the cell. We selected a time for the test so that if the cell failed, we would have enough battery power to get back the next morning to the landing area patrolled by an aircraft carrier. We developed the procedure, talked to the crew through the power-up, and held our breath. All eyes were on the data from the Hawaii site as the power-up and purge progressed. A purge shoots a stream of pure oxygen into the fuel cell to flush out any moisture and impurities in the cell. The spacecraft was again powered down while we evaluated the data.
The fuel cell sustained the load and with a few more data points we were convinced the pressure would gradually increase. We handed our data to Hodge at shift change and recommended a plan to power up incrementally over the next three days.
The press conference was Aaron’s show. When I got the press question, “What would it look like if we could open up an oxygen tank?” I handed the briefing over to him. I was proud of my team and Aaron in particular. Sitting next to me was a fresh college graduate teaching the world all about fuel cells.
We limped through the flight of Gemini 5 a day at a time. By day four, we had full power even though the oxygen tank pressure had only risen to 140 psi rather than the design operating pressure of 875-900 psi.
Before liftoff a clock was set in the MCC to read 119:06. At launch the clock started counting down. At the beginning of Kraft’s sixth shift, as the clock approached zero, the retrofire officer (RETRO) counted down the seconds over Kraft’s voice loop. At zero, Kraft lit up a cigar and proudly announced, “America has just set a new space record!” The Gemini 5 crew had eclipsed the Russian manned flight duration, and it was suggested to Conrad that he should perform a victory roll. Conrad wisely demurred, “I ain’t got the fuel, sorry!”
The next two days became really limited. The seesaw effect of powering up and down took a toll, and then the Gemini thrusters started to fail, further limiting the flight plan. With the crew in drifting flight and many experiments canceled, McDivitt, now a CapCom, decided it was time to pipe some Al Hirt jazz up to the crew for wake-up. Another MCC tradition was born.
During my final shift, I cut the mission short by one orbit. Tropical Storm Betsy had turned
into a hurricane, and the recovery forces were moved off the storm track, north and west to the vicinity of Grand Turk Island in the Atlantic. The final shift allowed me to pull a “gotcha” on my friend and judo partner, Llewellyn. The Public Affairs Office prepared a news update for the in-flight crew and this one included a report that Scott Carpenter, now also an aquanaut, was descending to the Sealab off the California coast for a thirty-day undersea mission.
Llewellyn had never forgotten Carpenter’s remarks implying that Llewellyn did not know where Scotty landed on Mercury 7. John is intense, emotional, and trusting, perfect characteristics for an easy set-up. The defining moment that brings every mission to an end is John’s stentorian second-by-second countdown to retrofire. This is an event looked forward to by the entire team, and one John relishes like a fine cigar. I called FIDO, RETRO, and CapCom over to my console. I told them that we wanted to get Llewellyn to believe that Carpenter would perform the countdown to the Gemini 5 retrofire from the Sealab. The team readily agreed to give it a shot and see if we could pull a gotcha on John.
At shift change time, Llewellyn sauntered into the room, exchanging banter with the various controllers until he got to FIDO. When FIDO did not respond, Llewellyn tried to pump him and RETRO to find out what was wrong. RETRO reported, “It’s that damn test with Sealab. It doesn’t make sense.” Llewellyn, concerned, responded, “What test?” RETRO advised him to talk to me about this mysterious test.
It was tough to keep a straight face as a troubled and deeply serious Llewellyn approached the console. He said, “RETRO said to talk to you about the test.” I told him to sit down, then I solemnly said, “John, I know how you feel about Carpenter, and you’re not gonna like what we have got to do.” His frown deepened as I continued. “We have orders to rig communications to Sealab, so that Carpenter can make the retrofire countdown. I don’t like it and have argued all night with Kraft, but he says, ‘Do it!’ ”
Llewellyn looked as if he had been poleaxed, first puzzled but then furious. At that moment, rookie astronaut Dave Scott walked by and asked, “John, what’s the matter?” Scott now listened to Llewellyn’s explanation of the Sealab command control test. Scott became even more furious than Llewellyn, loudly stating this was some kind of a “half-assed” decision and then complaining that he had doubts about whether he wanted to be an astronaut if such crazy decisions were going to be made in the future.
The team let Llewellyn and Scott dangle on the hook for a few minutes more before telling them it was all a joke. The gotchas in Mission Control were usually irreverent and often silly, a way of sticking a pin in the bubble of someone’s ego. They relieved the tension, poked fun, or just let one of the chiefs know that you will take their orders and respect their rank, but you won’t run scared. But I knew that I was in for a tough bout the next time I stepped on the judo mat with Llewellyn.
With the successful Gemini 5 mission behind us, it was time to go for a rendezvous and docking. The Russians, after their fast start, had been unable to rendezvous two spacecraft. We suspected they were hampered by inadequate computer and guidance capabilities. On two previous dual spacecraft missions, in 1962 and 1963, their spacecraft came within three miles of each other. Close does not count. A rendezvous means achieving a stable position that allows docking. It was now time for America to try for rendezvous.
The rendezvous target was an Air Force Agena upper-stage rocket. In early 1965, I made a fortunate series of personnel selections and formed a new systems branch combining Mel Brooks and Jim Hannigan, giving them responsibility for the Agena target and the lunar landing module. Brooks’s innovativeness and can-do attitude were perfectly balanced by Hannigan’s conservative do-it-by-the-numbers approach. Together, they built a great systems team for Flight Control.
The Agena was normally used to place super-secret military satellites in orbit. It was operated by an onboard programmer that issued coded instructions to operate the systems. The Agena was modified by NASA to provide a restartable engine, a docking adapter, a status light panel, and electrical connectors that provided a limited display and control capability for the crew. Many of my Agena controllers had cut their teeth at Lockheed and were rock-solid confident in their spacecraft. Other members of the MCC team were less impressed, disparagingly referring to the Agena as a restartable cigar.
I had a hell of a scare with the Agena rendezvous target rocket during a command test. Five days before launch, I was supporting the Agena readiness demonstration. In the middle of the evening, with everything quiet at the MCC, I heard the Atlas Agena test conductor call out on the loop, “Who the hell is transmitting engine start commands?” He continued, “Houston, have you been sending commands?” Just as I was about to respond, “Negative,” I heard the voice of my command technician state calmly, “Flight, the command system has failed, we have been in continuous command transmission. We have belched out every command in our inventory!” We were lucky that day. If we had been in orbit and our command system failed, the commands would have been transmitted to the Agena, possibly starting the rocket engine, changing control system modes, and turning the telemetry system off. The unplanned commands would have wreaked havoc on the mission. We amended our procedures, but we were still uneasy about the next step. Loading new software into new computers and using it for the first time was like playing Russian roulette. It demanded and got a lot of respect.
October 25, 1965, Gemini-Titan 6
Brooks’s team was sharp as a scalpel as the test conductor pressed the button and the Atlas/Agena rose majestically from launch pad 14. To the north, at nearby launch complex 19, astronauts Wally Schirra and Tom Stafford sat atop the Titan II rocket in their couches and listened as the Gemini test conductor updated them on the Atlas/Agena launch progress. To set up the conditions for a rendezvous, we first launched the Atlas/Agena. When the Agena passed over the Cape at the end of its first orbit, we would launch the Gemini spacecraft from another pad into a slightly lower orbit to begin the catch-up phase of the rendezvous. Launch of the target spacecraft was normal. The Agena separated from the Atlas booster and the Agena engine ignited. Brooks’s data flickered briefly, then stopped updating. FIDO reported tracking was lost. Then came a report that Range Safety was tracking multiple pieces of debris falling into the ocean.
Meanwhile, the Gemini countdown had continued on Pad 19, but the reports going to the crew became progressively worse. The reports from Canary Islands and Carnarvon were negative. At launch minus fifty minutes, we scrubbed the Gemini countdown and gave the crew the bad news. The Agena was destroyed, pieces scattered in the Atlantic. They had no target for their planned rendezvous.
I stayed with Brooks and his Agena team until there was no hope. Brooks was defeated, utterly spent. Many thought that the Agena had lived up to its less than sterling reputation. I was sad for my team. It was time for a few beers at the Singing Wheel before we started to regroup.
This watering hole was a two-story, barnlike building, the place we went when we needed some R&R. It was located a mile west of the center on state Highway 3 and hosted the Gemini-Apollo generation of flight controllers. The floor in the barroom tilted toward the wall to a degree that made it impossible to lean back in a chair without falling over. The Singing Wheel sold Lone Star beer by the pitcher. Nelson Bland, the owner, knew all the controllers and ran remarkably accurate tabs for all of them. A wall-to-wall mirror stretched behind the ponderous and scarred bartop. The tables in the back room were covered with checkered oilcloth. John Llewellyn’s wife, Olga, occasionally tended bar. After judo sessions, Llewellyn, Dutch, and I, wearing our sweaty gi outfits (white cotton trousers and a heavy, kimonolike jacket), would drive over for a beer or two after calling it a night.
It was, for all of us, a place of refuge where we could celebrate on the good days—and lick our wounds on the bad ones. Today was one of the bad ones.
8
THE SPIRIT OF 76
November 1965
The rendezvous
in space continued to elude us. The Russians had tried twice and failed—but we were impressed by their dual launch capability. We couldn’t even get our target rocket into orbit. Accomplishing this became the highest priority for Gemini and the American space program. Until now, a rendezvous in space was something only mathematicians really believed was possible. They worked out elegant equations and said, “If you launch it at this time, and go this fast, in this direction, you eventually are going to catch up to a target. If you perform the maneuvers properly, the two spacecraft will end up side by side.”
Proving that this theory would work became not only a goal, but an obsession. If we accomplished a rendezvous, we would validate the software that controlled the Gemini spacecraft as well as the crew’s fallback manual backups. There was no time to waste; we needed to dramatically improve our learning curve in order to be ready for the far more complex and sophisticated rendezvous and docking procedures necessary for a lunar landing. Within hours of the launch failure of our Agena target and the consequent scrubbing of Schirra and Stafford’s Gemini flight, we were discussing an alternative mission. A proposal from McDonnell’s senior management seemed to offer the most promising option. Walter Burke, the McDonnell vice-president and general manager for space and missiles, and his deputy, John Yardley, suggested we take a page from the Russians’ script by launching two Gemini spacecraft in rapid succession from the same launch pad. He proposed using the Gemini 7 spacecraft, flying the subsequent long duration mission, as the rendezvous target for Gemini 6.