by Gene Kranz
At last the Agena target performed as advertised, rising from Pad 14 and then five minutes later separating from the Atlas and reigxniting its engine to maneuver into a 180-mile circular orbit. At cutoff, Brooks was all smiles when the Agena responded to his commands. Hodge’s attention now turned to the astronauts, Armstrong and Scott, and the Gemini on Pad 19.
Since I was spending most of my time planning for Gemini 9, I had little experience with this crew. Hodge had debated the mission rules, run the simulations, and briefed the pilots. My association with the crew was limited to the handful of reentry simulations to get my team pulled together for the mission.
For Gemini 8 we had an experienced Gemini team, a novice Agena team that had never seen a spacecraft in orbit, a crew that would be docking with an Agena for the first time, and MCC and remote site computer systems running brand-new software. Adding to the level of concern was the fact that the Agena had failed on its previous mission—many at MCC considered it the potentially fatal weak link.
The Gemini 8 launch was textbook. Hodge’s team smoothly guided the crew through the rendezvous maneuvers to the handover point where the crew took it in on their own. The mission was progressing smoothly, and I arrived for my twelve-hour shift just as the crew and the MCC were comparing data for final rendezvous maneuvers. The Agena was performing flawlessly. During the docking it would be positioned perpendicular to the direction of orbital travel.
To maneuver the unmanned Agena from the ground we had to send time-tagged commands known as stored program commands (SPC) to its onboard computer. The commands were prepared at the MCC and sent to the remote sites. When the sites transmitted the commands to the Agena there was a complicated error-checking routine on board the Agena to make sure the commands were correctly inserted into the Agena computer’s memory, which the ground controllers could read using the site computer and could automatically compare with the commands transmitted from the site. This was a new computer program, used for the first time on Gemini 8. As a backup the site recorded the commands as they left the antenna and checked these automatically with the intended command load. We called it an echo check. The favored technique was always to compare the commands actually in the Agena’s memory. If you couldn’t automatically compare the data, the Agena controller could perform a manual data comparison using the recorded telemetry of the Agena memory. This process, however, often took several hours. To avoid glitching the mission time-line, the controllers would normally give their Go based on the echo check if the auto memory comparison failed.
At MCC in the second row of consoles, Brooks was struggling to complete the SPC load to configure the Agena for the after-docking maneuver. This maneuver used the Agena control system to align the docked spacecraft with the direction of orbital travel. Brooks, after manually assembling the command sequence and adding to the time tags for each event, made a final eyeball check of the load and sent it to the Rose Knot Victor where it awaited the arrival of the Gemini 8 spacecraft.
At acquisition Keith Kundel at Rose Knot Victor uplinked the command load to the Agena while Armstrong and Scott were in the braking maneuver. The load was accepted by the Agena, but the Agena memory dump and automatic data comparison could not be performed. Perplexed, Chuck Gruby, the Agena systems engineer, reran the comparison routine to no avail, then began the laborious process of printing out the memory data and manually checking every one of the thousands of data bits. Gruby called Brooks and advised him that although they did not get the auto compare, all of the commands were uplinked and, from the echo check, he was certain that the Agena command load was okay. It would take Gruby several hours to do the manual data comparison.
Over the next half orbit, Armstrong slowly closed on the Agena, finally maintaining a a station-keeping position a few feet from the docking collar. The Rose Knot CapCom, Keith Kundel, looked both spacecraft over, got the nod from his systems controllers, and gave Armstrong the “Go for docking!” Neil had been patiently waiting, standing off a few feet, the nose of the Gemini aligned with the Agena docking adapter. With the Go, he closed on the target, moving a few inches closer each second.
Six hours and thirty-four minutes after liftoff, another American record entered the books as Armstrong reported to the Rose Knot: “We’re docked, no noticeable oscillations, very smooth.” In Houston, a brief cheer rose from the team, then we all settled down to listen to the remainder of the pass.
Kundel and his team were busy, rapidly assessing the status of both spacecraft. As the docked Gemini-Agena approached LOS, Kundel gave Armstrong a “looking good. . . . The planned maneuver load has been uplinked.” Chuck Gruby, the Agena controller, remained at the console reviewing the pass record and continued the bit-by-bit comparison of the command load. He advised Kundel once again that he was sure that the load got in properly. On board Gemini 8, the crew began their post-docking checks. Brooks advised Hodge that the Rose Knot Victor did not get a “compare” on the load but that he was sure there was no problem in the load or the Agena. Hodge’s CapCom, Jim Lovell, noted the load compare problem as something that Armstrong and Scott should be aware of.
After a military coup in Zanzibar in January 1964, President Johnson had ordered that the site be closed, and much of the air-ground communications equipment was relocated to Tananarive, at the center of the island of Madagascar in the Indian Ocean. With the new communications installation, the MCC could communicate remotely through the site transmitters to the Gemini. (We did not get telemetry from this site; only voice communication.) The Tananarive pass was very short, the spacecraft passing low on the horizon at the very fringes of radio coverage. Toward the end of the pass Lovell advised Armstrong and Scott of the SPC maneuver load status and concluded, “If you run into trouble and the Agena attitude control system goes wild, just send the command to turn it off and take control of the spacecraft.”
The station pass concluded with Lovell’s request for a propellant quantity readout. Approaching the Coastal Sentry, Scott, in the Gemini spacecraft, commanded the Agena to execute the maneuver.
The command sequence began, the Agena jets firing to maneuver the docked spacecraft 90 degrees in the direction of orbit travel. Dave Scott clocked the turn and was pressing on through the flight plan checklist when he looked up. The docked spacecraft had rolled 30 degrees off the horizon, according to his attitude indicator. A check by Armstrong also showed a roll. Using the Gemini thrusters, Neil maneuvered back to the correct attitude. When he released the hand controller, the docked spacecraft resumed its motion. The astronauts conferred briefly, both believing the problem was a stuck-on Agena thruster. Dave disabled the Agena attitude control.
(Unknown to Armstrong and Scott, inside the Gemini spacecraft an electrical short was triggering a twenty-three-pound roll thruster to fire. None of us knew at the time there was a problem. At first it was intermittent, and then it came on continuously. With no ground station in view, the crew was on their own.)
For a few moments the problem seemed corrected, then a roll rate developed, causing the spacecraft to spin erratically like a small centrifuge. Armstrong took control, fighting the gyrations with the Gemini’s thrusters. As their orbital fuel levels plummeted, Scott cycled the Agena controls once again. Nothing seemed to be working. To isolate the problem they quickly decided to separate from the Agena. Prior to jettison, Scott, with a test pilot’s remarkable presence of mind, reenabled ground command control of the Agena. This would give the ground team the ability to troubleshoot the Agena later in the mission.
When Scott hit the emergency release Armstrong fired the thrusters in one long burst, pulling the Gemini away from the Agena. At undocking the Gemini spacecraft shed almost half its mass. Now, as a much lighter spacecraft, the effect of the continuously firing Gemini thrusters virtually doubled.
The Gemini was now rolling and tumbling. Using every test pilot skill, Armstrong and Scott fought for survival as the spacecraft completed a turn every second. As the orbit propellant drop
ped below 25 percent, there could be no doubt that the problem was in one of the Gemini thrusters.
The Coastal Sentry control team, stationed east of Okinawa, heard the docking report over the air from the Rose Knot and had received the Teletype message that all was Go at loss of signal. At the planned acquisition time the Coastal Sentry ship’s technicians reported difficulty in locking up the Gemini telemetry. They had the impression that the Gemini was spinning. Finally, with the telemetry lockup, Jim Fucci, the horrified Coastal Sentry CapCom, reported to the MCC, “The crew has undocked . . . they are rolling 360 degrees per second . . . orbital fuel is down to 20 percent.”
All we could do in the MCC was hold on and pray that Coastal Sentry’s team could somehow give the crew some help. Scott reported, “We have serious problems, we’re tumbling and have separated.” Then Armstrong continued, “We’re rolling up, we can’t turn anything off.” Armstrong then activated the reentry jets, and killed power to the orbit thrusters.
On the verge of losing consciousness from the accelerating spin, the crew regained control, using the fuel normally reserved for the reentry phase. Coastal Sentry CapCom Jim Fucci related: “They’ve activated their entry fuel system, they are firing the entry jets, the spinning is slowing, rates are coming down, it looks like they are starting to get control.”
Fucci continued reporting: “They’ve just about used up one entry fuel system, they’re getting it just about stable now!” With the Gemini craft stabilized, Armstrong started flicking the switches in the overhead panel to determine which Gemini jet had stuck and nearly killed them.
As the Gemini left the Coastal Sentry the mission was a mess. Gemini 8 was now entering the period where the only site coverage was provided by the two ships. Within three hours, we would see the crew only once every ninety minutes, which was not good. Although the docking objective had been satisfied, the rest of the mission was a bust. We were out of orbit fuel and half of the fuel in one of the two reentry fuel systems had been used up. The reentry fuel was of course intended for reentry. It was time to come home. No mission rules covered the exact predicament we were in. To get to the next day’s landing area, Hodge would have to limp through sixteen hours with no fuel and limited tracking station coverage. His alternative was to call it a day and bring the mission down in the West Pacific prior to entering the period of limited site coverage. Hodge moved quickly to terminate the mission and bring the spacecraft down in the last suitable landing area covered by the recovery forces. He turned and told Kraft he was bringing them home.
My team had come into the MCC for the docking and we were critically aware of the mission status. Hodge, drained and groggy after eleven hours on the console, told me, “Your team is trained for entry. I want you to bring Armstrong and Scott home.” I stood up, put on my white vest, and over the voice loop told my controllers to get on the console and start the handover.
Within minutes, my team had messages circling the globe, activating the recovery forces. The West Pacific recovery forces were alerted that the planned target point was 500 miles east of Okinawa. Within minutes, four para-rescue aircraft with life rafts and parachutists, as well as an amphibian (seaplane), began the race to the landing point. The destroyer USS Mason, 160 miles from the landing point, turned east and put all four boilers on line.
The White Team handover in the midst of the crisis was seamless. Our job was simply to get the crew home in the West Pacific landing area. My assets were the three remaining site passes and the MCC team. We had a single shot to do it right. A wave-off would send the crew off the tracking network critically low on fuel, shooting the gap for another thirteen hours before reaching the next suitable landing area. It was like changing quarterbacks in the middle of a handoff. In simulations we had practiced this kind of switchover, just as a football team rehearses plays.
The Trench, the EECOM, and the GNC pulled out the checklists and smoothly prepared the briefings at the next two ship sites. The Trench read the targeting orders to the computing complex, then relayed the resultant landing data to Recovery and then prepared the detailed deorbit data for the crew. Forty minutes after Armstrong’s undocking, my team had completed the deorbit planning and had passed the reentry data to the two ships.
Fucci, on the Coastal Sentry, in crisply measured tones read the long string of the deorbit data, recovery call signs, pickup times, and last minute reminders to the crew. This was our last go-for-broke pass, and the remote site performance was superb. We would hear no more until the recovery forces made contact. Gemini would be on its own, deorbiting out of station contact over Central Africa for the West Pacific landing site.
A rescue aircraft was overhead as Gemini 8 splashed down, and fifteen minutes later, a para-jumper landed with rafts and flotation gear within swimming distance of the spacecraft. The USS Mason arrived after the crew had spent three hours bobbing in the heavy seas in their spacecraft.
Gemini 8’s flight lasted ten hours and forty-one minutes, but the flight of the Agena, thanks to Dave Scott, continued under a carefully crafted plan for the next two days. Brooks performed ten maneuvers with the Agena, expending all maneuver fuel and electrical power and vindicating the record of the Agena in the process. Over 5,000 commands later, the Agena died, a valiant first effort for Brooks’s team. They were ready.
My debriefing was short and pointed: “We got our crew home safely and the control teams did a damn fine job under real-time pressure. I know this is going to sound like Monday morning quarterbacking, but the lessons from this mission are how we screwed up in planning and training.
“The crew reacted as they were trained, and they reacted wrong because we trained them wrong. We failed to realize that when two spacecraft are docked they must be considered as one spacecraft, one integrated power system, one integrated control system, and a single structure.”
I continued: “The next thing is that many of us did not trust the Agena. Only Brooks’s team thought it was a great piece of hardware. If we don’t trust a spacecraft, we have to fix it. We were lucky, too damned lucky, and we must never forget this mission’s lessons.”
Treating docked spacecraft as a single system was one of the most important lessons to come from Gemini. It had a profound effect on our future success as flight controllers. The lesson learned on Gemini 8 would be invaluable on Apollo 13.
9
THE ANGRY ALLIGATOR
Spring 1966
John Hodge departed the ranks of Gemini flight directors to prepare for the unmanned Apollo Saturn rocket flight testing. Glynn Lunney and Cliff Charlesworth joined me as flight directors for Gemini 9. Both men had grown up in the trajectory world, but there the similarity ended.
Glynn was an early entrant to the Space Task Group. He was smart as a whip, boyish and trim, the youngest of the flight directors, freewheelng, with a tendency to get ahead of his team in moving to a solution. Cliff was the oldest, a civilian who had previously worked in the Navy and in Army ordnance labs. He had a laid-back personal style that earned him the nickname “The Mississippi Gambler.” I regarded Glynn Lunney as my friendly but intense competitor; he ran the Trench, and I ran the Flight Control Branch. Our branches were two of the seven in Hodge’s Flight Control Division that staffed the Mission Control Center. Glynn and I both wore two hats, as flight directors in the MCC and as branch chiefs in Hodge’s organization.
I had the good fortune to have grown up under several outstanding leaders who had given me a lot of hands-on experience with people and technically complex missions. I didn’t have Glynn’s innate talent, so I surrounded myself with smart people and relied on them to work with me as a team to get the job done. My credo: always hire people who are smarter and better than you are and learn with them.
My team respected me because I did the dirty work and never pulled rank. I assigned the work responsibilities and once I set up the plan I stuck to it. I took the risks to let people stretch and grow and I took the heat, and deserved it, especially when I deci
ded there would be no holidays, no vacations. Which meant that I got to answer the calls from agitated wives. I was blessed that Marta was not among them. She understood the pressures at NASA and, in turn, I tried not to bring them home. In truth, there was no room for them. Our sixth child, and fifth daughter, Jean Marie, was born on April 16, 1966, between Gemini 8 and 9. The house we had moved into just a year before was crowded again.
With the increasing frequency and complexity of the flights, we now had to prepare for several missions simultaneously. To integrate the Gemini mission design, training, and planning, Hodge, in his role as division chief, assigned a lead flight director to coordinate the mission strategy of the flight directors, astronauts, and program office. During the mission, the lead provided a strategy overview, resolving any conflicts among the flight directors.