About six minutes had passed since the accident.
Over the loop, Kranz asked Liebergot for recommendations. Kranz had once said that the hardest part of being a flight controller was being “the last man in the decision chain.” That occurred, Kranz said, when a problem was passed to a flight controller at the last minute and he had to solve it all by himself; if he made a mistake, he did it in front of the whole world, and possibly jeopardized the mission. This was Liebergot’s situation now. On his telemetry screen, he could produce far more information about the spacecraft’s electrical system than the astronauts themselves had; the trouble was that he had no idea where to start looking. Half of the lights before him were amber, and he recalled that the only other time this had happened was shortly after lift-off during the Apollo 12 mission. He also recalled that on that occasion the spacecraft had been hit by lightning—a recollection that did him no good whatever, since there was no lightning two hundred and five thousand miles out in space. He now set about trying to piece together what he knew. The main question he had to answer was why there had been an undervolt in Main Bus B. Spread out before him on the console was a diagram of the spacecraft’s electrical system. It made a sort of chain, leading from the hydrogen and oxygen tanks to the fuel cells and from those to the buses, and ending with the equipment in the spacecraft that was receiving—or was supposed to be receiving—power. At the moment, they were connected as shown in the following diagram (the hydrogen tanks are not included).
Fuel Cells 1 and 2 were supplying electricity to Main Bus A, which was still working. Main Bus B, however, was drawing its electricity from Fuel Cell 3 alone, and it didn’t take Liebergot long to find out that Fuel Cell 3 had stopped generating power. Then he learned that Fuel Cell 1 was generating no power, either. It looked as if the spacecraft had lost two fuel cells—an unprecedented situation. The spacecraft was getting along solely on the electricity that Fuel Cell 2 was supplying to Main Bus A—and, inexplicably, the power put out by the one remaining good fuel cell was beginning to drop, too, though not yet enough for Liebergot to worry about.
The two apparently dead cells, Liebergot knew, drew their oxygen from the same two tanks, but since the good cell drew its oxygen from them as well, he didn’t give the tanks much thought. There was little reason to, for although the two tanks were not, strictly speaking, redundant—they shared the system of pipes leading to the fuel cells—there were so many safety valves that they might as well have been. They were separated by valves that insured that the oxygen would flow only out of them, and, as a further safeguard, each fuel cell could be cut off from the oxygen by a valve of its own, called reactant valves. The protection afforded by all these valves was critical, because in addition to fuelling the electrical system and producing water, the two tanks provided all the command module’s oxygen for breathing. (In the cabin, there was a small emergency supply in a tank called the surge tank, and in three one-pound bottles, but this had to be conserved for breathing when the command module plunged alone through the earth’s atmosphere.)
The implication of two dead fuel cells was so staggering in itself that Liebergot couldn’t bring himself to believe that such a state of affairs was possible. At NASA, backups don’t fail. Liebergot was encouraged in his disbelief by the flight controllers’ operating procedure, which required them to make presumptions against such failures—partly because of the admittedly imperfect quality of telemetry. According to the standard procedure, before Liebergot could think about such a thing as the oxygen he had to make sure that the fuel cells really were dead. Perhaps there had been an instrumentation failure, or perhaps there was some other simple explanation. Until he was certain, Liebergot didn’t even report the loss of the cells. “You can’t alarm the crew unnecessarily—you’ll look like a big ass unless you’re sure,” he later said.
One simple explanation that occurred to Liebergot was that the jolt or the bang, whatever it was, at eight minutes past nine had disconnected the two fuel cells from the two buses. He therefore suggested to Kranz that Swigert check on whether the cells were in fact hooked up to the buses. Any connecting or disconnecting that the astronauts did was by means of switches at their consoles; the switches for the cells and buses were in front of Haise but within reach of Swigert, and Swigert now flicked them down and back. (Flight controllers call such flicking “cycling the switch.”) Swigert reported no change in the electricity level of the buses. That meant that the trouble did not lie in anything as simple as broken connections. Liebergot couldn’t make any sense out of it. He wished he could be almost anywhere else. He couldn’t, of course, because, among other things, Lousma kept asking Kranz if there were any more recommendations he could pass on to the astronauts, and Kranz kept asking Liebergot. Liebergot felt rather cornered. He was the one on the spot, and all the electrical engineers in the world couldn’t help him. At length, because it was possible to switch circuits among the fuel cells and buses, he suggested that the astronauts switch the lines from the two dead cells so that each fed the other bus. That way, Liebergot figured, something might develop to give him a clearer picture of what was going on. Also, fuel cells, like flashlight batteries, sometimes worked better if they were changed around. Kranz, however, refused to go along with the suggestion. He had to be cautious, because nobody knew what was wrong and he didn’t want to do anything that might make matters worse. At the moment, all the power in the spacecraft was coming from Main Bus A, and he didn’t want to risk disturbing it. For the time being, Kranz planned to be very deliberate and very methodical about authorizing any changes. Less than seven minutes had passed since the bang.
The only other idea that Liebergot could come up with just then was that in order to keep up the power in the good bus—which was continuing to drop—the astronauts might augment it by feeding into it electricity from one of the storage batteries that were supposed to supply power to the spacecraft during its reëntry through the earth’s atmosphere. There was, it seemed, very little choice; in fact, as Liebergot made the suggestion, he could see on his telemetry screen that the astronauts were already hooking the battery to the bus.
Although the brunt of the difficulties fell on the EECOM, all the other flight controllers were having trouble, too. Their voices over the loop were almost unnaturally calm, but one of them said later that he could tell from their tones that “a lot of stomachs were turning over.” At the moment of the bang, the spacecraft began pitching and yawing about like a depth-charged submarine. Two identical balls set in the dashboard, one in front of Lovell and the other in front of Swigert, appeared to spin erratically. They were the flight-director attitude indicators, or F.D.A.I.s—sort of three-dimensional compasses that showed which way the spacecraft was pointing. Appearances to the contrary, the balls were actually still, as the compass card in a ship’s binnacle is; it was the spacecraft that was doing the turning. The guidance computer in the spacecraft, which normally held it steady by automatically firing sixteen small thruster rockets outside the service module whenever necessary to correct its attitude, had been unable to stop the wobbling. Now Lovell was trying to do so by firing the thrusters manually, using a pistol-grip hand control at the end of the armrest on his couch. He wasn’t having much luck, for the spacecraft kept buffeting and yawing as if something were venting from it and imparting an unwanted thrust. Part of Lovell’s difficulties stemmed from the fact that the sixteen thrusters operated electrically; each of the main buses supplied current to fire half of them, and the eight thrusters dependent on Main Bus B weren’t working. The good bus couldn’t accommodate all sixteen, so Buck Willoughby, the Guidance and Navigation Control Officer,—a tall ex-Marine flier from Colorado who sat on Liebergot’s right—had to figure out which were the best thrusters to keep operating; he had to make sure that one thruster was working in each direction for the three motions of the spacecraft—up and down, left and right, and roll.
The GNC had a special interest in getting Lovell to steady the spacecraft,
for once it had been brought to the right attitude Lovell could set up the gentle roll—the passive thermal-control roll—that kept it turning once every twenty minutes, so that the sun would heat it evenly on all sides. The delicate electronic instruments for guidance and navigation, which were the GNC’s responsibility, and some of the spacecraft’s propulsion systems were especially sensitive to extremes of temperature, and without the regular thermal roll the part of the spacecraft left facing the sun could get as hot as two hundred and fifty degrees, and the part left facing away could get as cold as absolute zero. However, even after Lovell had plugged the thrusters into the buses in the way the GNC thought best, he still couldn’t control the spacecraft’s attitude.
The wobbling was causing other problems as well, for if the spacecraft should happen to roll into certain attitudes the guidance system would lock. The heart of the guidance system was the inertial-measurement unit, a spherical structure in the lower equipment bay, at the foot of the center couch, containing the guidance platform. This was a small metal block that swung freely on three gimbals (like those that keep a ship’s compass level), so that a set of gyroscopes could maintain it in the same position in relation to the stars regardless of the attitude of the spacecraft. Its attitude was electrically relayed to the ship’s guidance computer. The platform had been aligned with certain stars before launch, and it was still aligned with them, for whenever the spacecraft rolled, pitched, or yawed the spinning gyroscopes adjusted the gimbals to keep the platform true. The trouble was that if the three gimbals lined up in certain ways, they would lock and the spacecraft would suddenly be without any reference point in space. In effect, the astronauts would be without a compass. Two or three times, the GUIDO broke in on the loop to tell Kranz that the spacecraft was wobbling toward what he called “gimbal lock,” and at the warning Lovell would point the spacecraft in another direction as hastily as a helmsman would steer a ship away from a reef.
A spacecraft is such a welter of interdependent elements that any one problem can set off a whole series of other problems. The erratic spinning threatened radio communications, because it was next to impossible to keep the antennas aimed at the earth. After the accident, the INCO, the radio controller, who sat at Kranz’s left, had advised shifting from the high-gain antenna—the stick that had been jarred at the time of the bang—to the omnidirectional antenna system, which didn’t have to be pointed so precisely. There were four omnidirectional antennas—big scimitars—spotted around the spacecraft, and normally as it rolled the INCO would keep switching on the one that happened to be on the side nearest the earth. However, with the wobbling, there were times when neither the INCO nor the astronauts knew which antenna was facing the earth. Sometimes communications stopped altogether.
The astronauts were so busy avoiding gimbal lock, checking antennas, and transferring thrusters and other equipment from one bus to the other that they didn’t have a chance to worry much about exactly what sort of danger they were in, and most of the flight controllers didn’t, either. However, one person who had a little time on his hands for worrying was the TELMU, Robert Heselmeyer; because he was a lunar-module man, he was somewhat removed from the situation. Although the LM was powered down, it was still using a little electricity drawn from the command module to warm some of its equipment. Heselmeyer sat in silence as he watched the current being fed to the LM go down and down. When the current stopped altogether, he reported the fact to Kranz. Kranz asked Heselmeyer to get back to him later, because he had enough on his hands at the moment. Heselmeyer continued to worry, for it had crossed his mind that if anything serious happened to the command module the astronauts might have to use the lunar module as a lifeboat. He rummaged around on his console for instructions on such lifeboat procedures.
Liebergot, who was still trying to come up with some ideas for reviving the two apparently dead fuel cells, suggested to Kranz that they both be unhooked from the buses, in the hope of separating bad equipment from good. Kranz, who was still being cautious and didn’t want to disturb too many things at once, agreed to disconnect only Fuel Cell 1, which was attached to the good Bus A. Part of his thinking was that by isolating one section of the system and then another it might be possible to pinpoint the trouble spot.
In the meantime, Liebergot had requested Lovell to read to the ground all the gauges having to do with the electrical system, to see if they bore out the information that the ground was receiving. Lovell at last got as far as the pressure gauges for the oxygen tanks. “Our Oxygen No. 2 Tank is reading zero. Did you get that?” he said. Now he floated up out of his seat and pressed his face against the window so that he could look backward toward the service module. He saw a thin sheet of vapor, like a cirrus cloud. “It looks to me that we are venting something,” he reported. “We are venting something out into space.” Now he understood why he had been unable to steady the spacecraft: venting imparted motion as surely as firing a rocket. Thirteen minutes had passed since the bang. To Lousma, the CAPCOM, this was the most chilling moment of the flight. Although Lovell’s voice was calm, what he was saying was as alarming as if a ship’s captain had reported seawater rushing in through the hull—only the astronauts wouldn’t be able to jump into anything as hospitable as even an Arctic sea.
Lovell had no doubt that what he had seen venting was a gas, for in space liquids form hard nuggets, not thin sheets of cloud. And he had a pretty good idea that the gas was oxygen, for he now noticed that not only was the pressure in Oxygen Tank No. 2 at zero but the pressure in Tank No. 1 was dropping as well; it was oxygen from this tank that was leaking into space at the moment. Lovell felt that it was just a matter of time before the command module itself would go dead.
The other flight controllers, unlike the CAPCOM, so far had no such forebodings. As the disaster unfolded step by step, they continually seemed to be left incredulous, one step behind. One reason for their incredulity was that they were missing a key fact: they did not know that the original tank failure had been a violent one. Liebergot was thinking in terms of a gentle leak, and he did not suspect that the rupture of Tank No. 2 had been explosive; what had actually happened was that it had ripped out pipes and valves between the two tanks, in an area called the manifold, where several pipes joined, also causing the oxygen in Tank No. 1 to slowly dissipate. And there was something else the flight controllers didn’t know: the jolt had shut the reactant valves on Fuel Cells 1 and 3—the immediate cause of the power failure, as the valves cut off the flow of oxygen to the two cells. The command and service modules’ electricity would last only as long as oxygen from Tank No. 1 continued to reach Fuel Cell 2. As the oxygen in the service module was what the astronauts were breathing, it had already crossed Liebergot’s mind that the astronauts could be what he called “belly up” in a matter of hours. He couldn’t quite believe it, though, because of his trust in the soundness of Tank No. 1. Kranz gave a short talk to the flight controllers over the loop, urging them to keep cool; guessing would just make matters worse, he said. Optimistically, Liebergot checked to make sure that there was no instrumentation problem.
The astronauts, however, by now had no illusions. Haise said later, “The ground may not have believed what it was seeing, but we did. It’s like blowing a fuse in a house—the loss is a lot more real if you’re in it. Things turn off. We believed that the oxygen situation was disastrous, because we could see it venting. The ground may have been hoping there was an instrumentation problem, but on our gauges we could see that the pressure was gone in one tank and going down in the other, and it doesn’t take you long to figure out what happened.”
The chief reason the flight controllers didn’t tumble to the seriousness of the oxygen problem was that the correct answer was also the unthinkable one. In a number of respects, the situation was like the sinking of the Titanic, another craft that was admired as a nation’s greatest technical achievement. The ship had reputedly been unsinkable, because its hull was divided into redundant watertight com
partments, but the collision with an iceberg sliced open too many of them, and it sank. There were obvious differences between the two incidents—Apollo 13, for example, carried its own iceberg within itself. A remark made later by a NASA engineer was strongly reminiscent of the worldwide reaction to the earlier accident: “Nobody thought the spacecraft would lose two fuel cells and two oxygen tanks. It couldn’t happen.” Swigert himself wrote afterward, “If somebody had thrown that at us in the simulator, we’d have said, ‘Come on, you’re not being realistic.’”
For some time, the flight controllers, in their pursuit of solutions, attempted to go in two directions at once: they tried to save the moon-landing mission while simultaneously preparing for the worst. About four minutes after the astronauts reported the venting, Liebergot suggested to Kranz that the crew start powering down the command module, to put less strain on the surviving bus, which was continuing to lose power. Liebergot told Kranz he wanted the astronauts to work their way through the first half of page 5 of what was called the Emergency Power-Down Checklist, which told how to turn off equipment in the proper sequence so that an instrument wasn’t switched off before another one that depended on it. For the moment, the procedure should ease the power crisis. In the spacecraft, the astronauts, to find the proper checklist, had to riffle through twenty pounds of instruction sheets before they got the right ones. Kranz still had no intention of giving up the ship; he made sure no equipment was turned off that would preclude landing on the moon—a possibility he had by no means abandoned. Like Kranz, Liebergot was hoping there was still some way of saving the mission, and as he went about selecting more equipment for the astronauts to turn off he was thinking, in another part of his mind, of possible ways to get more oxygen out of Tank No. 1, so that they could power up again.
The power shortage in the command module was now well beyond the help of the reëntry battery, so Liebergot ordered the astronauts to disconnect it. They were reluctant to do so as the battery was providing a good deal of the current they were able to draw from the good bus, but Liebergot insisted. He knew that as the power-down continued, there would be less need for the battery anyway. He had suddenly begun to fear that nothing could be done—that it was just a matter of time before the command module lost all its power. In that case, the astronauts would have to use the LM as a lifeboat to bring them back to earth. Then, just before they hit the atmosphere, they would have to abandon the LM and find some way to fly the dead command module, which was the only part of the Apollo spacecraft with a heat shield capable of withstanding the high temperature of reëntry. If they were to do that, they would need every ampere of electricity in the reëntry batteries. For the same reason, Liebergot urged that the astronauts immediately isolate the supply of oxygen in the surge tank, which was normally connected with the service-module supply. Kranz, who was still thinking in terms of conserving the oxygen in Tank No. 1, wanted to know why Liebergot was ordering the change, and the EECOM replied that he was now more worried about conserving the reëntry oxygen. Kranz saw what he was driving at.
Thirteen: The Apollo Flight That Failed Page 3