Failure Is Not an Option

by Gene Kranz

  The roll of the dice put Glynn Lunney, Cliff Charlesworth, and me on the schedule for the first lunar mission. This was a reunion of sorts, the first time the three of us had worked together since Gemini 12. Since the Apollo 1 fire, Hodge had rotated the lead assignments among Glynn, Cliff, and me. When I became Flight Control division chief, I saw no reason to shuffle the lead responsibilities, so Charlesworth retained the lead flight director role for Apollo 11.

  One of Cliff’s initial tasks was assigning the flight directors for the mission phases. There were eight major phases of a lunar mission; five of the eight would have been demonstrated on either Apollo 8 or 10. The three new mission phases were the lunar landing, surface EVA, and lunar ascent.

  Our experience made it a toss-up for the phase assignments. Cliff had launched the most Saturn rockets; Glynn would have been to the Moon twice with Apollo 8 and 10, and I would have the most lunar module experience. We had all worked manned command modules. Cliff and Glynn were considered by the controllers as trajectory biased, favoring work on the trajectory aspects of the mission, while I was considered as systems-biased, favoring work with the CSM and LM control teams. I hoped that when Charlesworth put the pieces together, he would give me the assignment for the landing phase of the mission.

  The meeting where that actually happened was almost anticlimactic. Cliff walked into my office, stood at the window, stared out for a moment, and then turned with a smile on his face. He knew that I wanted to get down to business, but he just toyed with me for a few moments, passing the time of day, then abruptly said, “I think it’s time to decide on the Apollo 11 phase assignments.”

  With no further preliminaries, he continued, “I think I should launch Apollo 11 and do the EVA. Milt [Windler] will take the entry. This leaves Glynn for the lunar ascent . . . and you with the landing. Is that okay with you?” I nodded, and the meeting was over. The entire session had taken less than sixty seconds.

  I had drawn the flight director assignment to put the first man on the Moon.

  I had to tell someone and the list would start with Marta. Seldom would I call Marta bearing glad tidings; it was usually, “I’m going to be late,” or “I’m not going to be home to help one of our kids with a math test preparation.” This time I couldn’t wait to tell her. “Marta, guess what? Cliff gave me the lunar landing assignment!” When I called the staff in, they could tell from the way I was ricocheting around the office and from my ecstatic expression that I had gotten the glittering prize—the Moon landing. They were as happy as I was.

  My landing team formed up quickly. Each team was constructed on a mission-by-mission basis. When the flight directors received their mission phase assignment, the branch chiefs carefully matched the personalities and strengths of controllers to those of the individual flight directors and their capabilities to handle the mission events.

  Bob Carlton, nicknamed “The Silver Fox” for his prematurely gray hair, had the responsibilities for the LM navigation, control, and propulsion systems. His call sign was CONTROL. During the last seconds of landing, his slow, deliberate Alabama drawl would be the only voice on the intercom, calling out the seconds of fuel remaining.

  Don Puddy, the tall, intense Oklahoman who never wasted a word, was the self-appointed leader of the lunar module team, responsible for communications, power, and life support systems, call sign TELMU.

  Steve Bales, filling the GUIDO (guidance) position, was one of the first of the computer whiz kids. Steve in many ways was more like the systems controllers. He had not yet developed the arrogance so characteristic of the typical Trench inhabitant. He spoke rapidly, running his words together occasionally, then letting them stumble out. You could tell how he felt by his voice inflection. His large, round black-rimmed glasses set him apart from most of the controllers.

  FIDO was Jay Greene, the pipe-smoking New Yorker, a rabble-rouser who did not like it when things got quiet. He liked to coach the flight director along a decision path. He was elite in the ranks of the FIDOs, cocky and crisp with his calls. Bostick, my branch chief with responsibility for the Trench, knew that I was the weakest flight director when it came to the trajectory, so he gave me an experienced FIDO who would teach me the new stuff I needed to know for the landing.

  To Greene’s left sat the RETRO, Chuck Deiterich. Like Greene, he had a classic disdain for any controller who did not immediately surrender to the wisdom coming from the Trench. Chuck would either bury you with more data than you needed or cut you at the knees . . . once telling me during a debriefing, “You don’t want to hear about that, Flight . . . it’s too technical!”

  The combination of Chuck’s Texas and Greene’s New York accents during the rapid-fire exchanges on the voice loops made for interesting listening during time-critical operations. Their voices were unmistakable.

  The final member of the Trench was Gran Paules, who would work with Bales. Gran was a tall, blond, taciturn controller who had the habit of turning to look at you when you called. His nasal inflection reminded you of someone constantly suffering from an allergy. Like Steve, he was typical of the next generation in the MCC.

  Slayton selected Charlie Duke from the astronaut class of 1966 as my CapCom. Duke was well experienced in the operation, having worked on Glynn’s team for the previous mission. For flight directors and CapComs, the principal tools used during the mission were the MCC intercom and crew voice loops. Our common job was to listen, integrate, communicate, and act.

  During mission preparation, Duke provided the communications conduit. Aware of both crew and MCC concerns during meetings, he brokered and summarized the resulting actions with the crew. Watching him operate, I knew why the Apollo 10 and 11 astronauts wanted him in the CapCom’s slot. Of all the astronauts, he would have made a hell of a good flight director. I had a great feeling about the easy confidence Duke showed during planning sessions. He contributed to making my mission preparation successful, helping to bring the controllers to the highest pitch of readiness in the three months before the lunar landing mission. The SimSup and his team came from the Flight Control Division’s Mission Simulation Branch. SimSup was my other partner in team building. He worked with the flight directors and the branch chiefs in carefully monitoring controller performance during training and certifying them suitable for mission support. There was no way one flight director could do this job by himself.

  My lunar module team, the four controllers in the Trench, and Charlie Duke were the core controllers for landing. Our job was to get the LM, an odd-looking contraption—like a praying mantis, Mike Collins said—close enough to the surface to let the crew take over and attempt landing. “Close enough” was subjective. Only the crew in the LM would know whether to land or abort in the last few hundred feet. It was our job to get them to their decision point.

  There were three members of the command module team in the control room, looking over Mike Collins’s shoulders in the spacecraft. Ed Fendell was making the transition to a consolidated (CSM and LM) communications position. Ed was intuitive in responding to problems and he developed great young controllers. He was also noisy, poked fun at every controller, and could be disruptive. You had to earn his respect and keep him on a short leash. I worked with Ed many times during Gemini and liked his spirit, commitment, and willingness to step up to responsibility. Above all, I liked him because he never left the flight director hanging and never catered to anyone. His independence irked many of the controllers, but they respected him. Glynn and Cliff considered him a pain in the ass. The rest of the team consisted of Buck Willoughby and John Aaron. During the landing phase their job was to monitor Mike Collins’s solo operations in the command module. During the descent, if needed, we would use the Command and Service Module as a communications relay point and possibly an orbital rescue vehicle.

  With the assignments completed, I called the first meeting of the White Team to finish working out the detailed landing strategy. Personal and team readiness would emerge from our study and t
he team working sessions on the trajectory, flight plan, and the mission rules. Then the simulation training would integrate the ground team with the astronauts and test our mission planning. The White Team had a total of eleven days of simulation to get ready for the landing. Only seven of these were with the crew. Four were with math models and a simulated astronaut.

  Bill Tindall had started weekly meetings on the descent phase in April and had released a barrage of “Tindallgrams” and assorted notes. “Tindallgram” was the name given to Bill’s comic and highly treasured memos of the techniques meetings he conducted from 1966 to 1970 to document key engineering and operational decisions. In May 1996 the memos were bound in a single volume and distributed to “Bill’s many friends.” Tindallgrams were converted into new procedures, flight plan entries, and the jargon used by the controllers in their Go NoGo.

  One of the Tindallgrams really grabbed our attention and also gave us a few laughs. It began, “There is another thing about powered descent crew procedures that has really bugged me. Maybe I’m an ‘Aunt Emma’—certainly some smart people may laugh at my concern, but I just feel that the crew should not be diddling with the computer keyboard during powered descent unless it is absolutely necessary. They will never hit the wrong button, of course, but if they do, the results can be rather lousy.” The next day we started a review of every crew computer keystroke and its effect throughout the descent phase.

  Another “Aunt Emma” note challenged the terms used by my flight controllers after landing, “Once we get to the Moon does Go mean ‘stay’ on the surface, and does NoGo mean abort from the surface? I think the Go NoGo decision should be changed to Stay NoStay or something like that. Just call me ‘Aunt Emma.’ ”

  We changed the procedures for the entire after-landing process into a series of Stay NoStay decisions. Tindallgrams, spiced with humor, idiosyncratic grammar, and personal “revelations,” got the job done. After Apollo 11, at a post-mission beer party, Flight Control made Tindall an honorary flight director, with the team color Gray. His color is retired, like that of many flight directors, and now hangs in the third floor of the Mission Control Center.

  I asked him to sit next to me at the console for the lunar landing. Tindall, ever modest, declined, but I persisted in my request and he finally agreed. He was one of the great pioneers of manned spaceflight.

  During the mission rules sessions, Buzz Aldrin was the crewman usually involved, demonstrating his knowledge of a variety of subjects, and generally dominating the crew side of the conversations. Neil Armstrong seemed more the observer than the participant, but when you looked at his eyes, you knew that he was the commander and had all the pieces assembled in his mind. I don’t think he ever raised his voice. He just saved his energy for when it was needed. He would listen to our discussions and, if there were any controversy, he and Aldrin would try out our ideas in the simulators and then give feedback through Charlie Duke to the individual controllers. Buzz and Neil seldom took a strong position during the meetings, but they were good listeners. They knew enough about us to trust us, to give us the benefit of the doubt. Mike Collins used a different tactic. He worked directly with the Trench and systems guys. By the time we got to the rules sessions, all the problems were ironed out.

  We published our first complete set of rules for the Apollo 11 mission on May 16, two months prior to launch. With no landing simulation experience, this first set of rules represented the sum total of our knowledge from our meetings. With the simulations starting in early June, the learning curve would be steep, resulting in planned rules updates weekly until launch.

  Simulation training is broken into two parts, nominal and contingency. The nominal training occurs early in the simulation period. It lasts only two to three days and is used to establish crew-controller action timing, locate the Go NoGo decision points, and exercise the procedures for the planned mission. The contingency training tests the crew-controller decision process in a mission environment while solving complex trajectory and systems problems. Training scripts are developed by SimSup’s team, and problems are programmed into the simulators without the crews’ or controller’s knowledge. The training environment becomes as close to the real thing as possible, with the training team testing the flight teams’ strategy, knowledge, and coordination while probing into the psyche of the crew and controllers. Nothing is sacred; no quarter is given and none asked. Training for a lunar mission was a daunting task. Training to cover every conceivable aspect of the first lunar landing bordered on the impossible.

  To get a handshake on the unwritten rules for the landing, I had a final strategy session before simulation start-up with Neil, Buzz, Mike, and Charlie Duke. It was in this session that I outlined the landing strategy. We had only two consecutive orbits to try to land on the Moon. If we had problems on the first orbit we would delay to the second. If we still had problems, we would start the lunar descent to buy five additional minutes to solve the problem. If we couldn’t come up with answers, we would abort the landing and start a rendezvous to recover the LM, then jettison it and head back home. If problems surfaced beyond five minutes, we would try to land and then lift off from the surface after a brief stay. We would try for the landing even if we could only touch down and then lift off two hours later when the CSM passed overhead in lunar orbit with the proper conditions for rendezvous. (The LM had limited electrical power. The lunar liftoff time had to be precisely established to allow rendezvous and docking before the LM batteries failed.)

  I knew Armstrong never said much, but I expected him to be vocal on the mission rule strategy. He wasn’t. It took time to get used to his silence. As we went through the rules, Neil would generally smile or nod. I believed that he had set his own rules for the landing. I just wanted to know what they were. My gut feeling said he would press on, accepting any risk as long as there was even a remote chance to land. I believed we were well in sync, since I had a similar set of rules. I would let the crew continue as long as there was a chance.

  The SimSup presided over some of the most complex technology of the space program. The only real things in the simulation training were the crew cockpit and the consoles in Mission Control. Everything else was make-believe, created by a bunch of computers, lots of formulas, and a skilled bunch of technicians. The business was more art than science because we were training for something that had never been done before.

  Every time an astronaut threw a switch or a controller sent a command, a series of equations was triggered in the simulation computer. This pulse started a chain reaction: the data cascaded through sixteen other computers, sending impulses to the cockpit or the controllers’ displays. SimSup needed a thick skin. When the simulation hardware and computers misbehaved, he had an unhappy crew and control team while we waited hours to restart the training. If they balked too often, Slayton, Kraft, and the flight directors would get involved. The squeaky wheel invariably got the grease, priorities were rearranged, and someone downstream, another flight director and group of astronauts, would be stuck with less and would start screaming about being cheated out of their training. Simulator training was routinely conducted sixteen hours each day. Both primary and backup crews were being trained for the subsequent two or three missions on two sets of training hardware. For the coming mission in the sequence SimSup had a prime and backup crew and four teams of controllers to get ready for a mission in ninety days, and he always had the priority.

  The simulation team, like the flight director, worked to the launch day deadline. When the controllers clocked twelve-hour days, the training team worked fourteen. When he was not training, SimSup was studying the controllers, crews, and mission strategy, looking for the holes and developing new training runs to exploit the perceived holes.

  SimSup for the descent phase was Dick Koos, who struck me as a quiet young academic. In fact, he was a discharged sergeant from the Army Missile Command at Fort Bliss, Texas. Dick was an early hire into the Space Task Group, assigned to train the
crews and control teams. His background was in computer guidance for the early ground-to-air missiles. Simulation was an entirely new field, but with his computer experience he took to his work with a passion.

  Koos went up through the ranks rapidly and developed into an excellent SimSup. Dick was a thin guy, wore wire-rimmed glasses, expressed himself in incomplete sentences, and seemed unsure of what he was trying to say. His external demeanor set you up for his training sessions, which were like a rapier, cutting so cleanly that you did not know you were bleeding until long after the thrust. Koos was a worthy adversary and an excellent choice for training my White Team for the Apollo 11 landing.

  SimSup had months to design the training sessions; we had but seconds to minutes to solve problems he posed. The early sessions were rugged. SimSup’s team attacked every aspect of our knowledge, even the relationships between space and ground teams. They pounded away at the strategy and timing. When we were through, we were in a place beyond exhaustion.

  Landing training kicked off at launch minus eight weeks. My team, fresh from Apollo 9, had a “hot hand,” and from the beginning it seemed we could do no wrong. The first two training days were used to nail down the timeline for the landing preparation, establish the Go NoGo points, give the controllers a walk-through of the landing sequence, and become familiar with the three-second communications delay when working at lunar distances. Initially the three-second delay didn’t seem like much, but if things start to go wrong in the final seconds before landing you could quickly find yourself in the corner of a box with no more options. The mission rules and procedures were refined during the initial two days of training and I felt well prepared to start the landing abort simulations.