Apollo: The Race to the Moon (26 page)

Time would not soften Joe Shea’s judgment about the situation he found when he went to ASPO in Houston two years after North American began work. “I do not have a high opinion of North American and their motives in the early days,” he said many years later. “I think they were more interested in the financial aspects of the program than in the technical content of the program. I think Storms was a very bad general manager, I think Atwood [North American’s president] had dollar signs in his eyes. Their first program manager was a first-class jerk. There were spots of good guys, but it was just an ineffective organization. They had no discipline, no concept of change control. If anything, they were interested in pumping the program up rather than in what the program really was.” Shea was not alone in his opinion. During the first years of the C.S.M. contract, North American acquired a very bad reputation around NASA.

It wasn’t all North American’s fault. The Apollo spacecraft was a completely new machine and more complex than any aircraft ever built. The Mercury capsule had been much simpler—yet McDonnell Aircraft Company had run into the same kinds of problems that North American was stumbling over now. For its first year, many at NASA had thought McDonnell was a money-grubbing, incompetent contractor that was shipping half-finished, misbegotten capsules to the Cape where they had to be rebuilt from scratch. By the time McDonnell was making Gemini capsules, it had gotten so good at its job that the Geminis routinely arrived at the Cape in launch-ready condition, and McDonnell had in NASA’s eyes become a paragon of corporate virtue.

Nor should it be forgotten that many of the NASA people who were managing the hardware contracts had come out of Langley. When Sam Phillips came to the Apollo Program at the end of 1963, at the same time Shea went to Houston, he was startled to find such a complete contrast between “a tremendous technical capability [and] a very, very thin capability to manage a large-scale program activity,” and this was nowhere truer than at Houston. No one had learned how to manage contracts at Langley, because there weren’t any. The men of the old Space Task Group were engineers in bone and spirit, and suddenly they had become managers by decree. Some of them became good managers, but it took them a while. Many never were comfortable in that role. Was there ever a less plausible manager than Caldwell Johnson—without a shred of diplomacy, a loner, a natural designer who by his own admission lost interest in his designs once they were “pretty well patted into shape”? Yet when Shea arrived to take over ASPO, there was Johnson, managing NASA’s C.S.M. contract with North American, poring over work schedules and time charges and management systems. (Shea shifted him back to a design job forthwith.)

Lee Atwood thought that much of the criticism of North American was really part of an inevitable charade to which contractors must reconcile themselves. “The contractor is always the underdog when something comes up,” he said later. “He has to absorb the criticism.” NASA’s technical people love to think they invent things: “That’s kind of a handicap for the contractor’s technical people, because they are being second-guessed by a large gathering of technical people in NASA. If the work goes wrong, the contractor is wrong. If it goes right, the NASA technical man gets his medal or whatever.”

But however the blame is parceled out, the fact remains that the first years of North American’s effort to build the command and service module were rife with wasted effort, lack of coordination, and intermittent warfare between contractor and customer.

To get control of the program, Shea called in his new assistant, Tom Markley, and said to him that somehow they had to keep this thing on track. “Go out and fix me a management tool,” he told Markley.

Markley sent his men out to look at the major aerospace companies and their sophisticated management control centers. Then Markley himself went to visit what everyone said was the best of them all, at the Martin Company. It was beautiful, Markley thought, sort of a management war room. He figured he would set something like that up for Joe—which would be a big relief to Markley because, until he did, Shea was expecting him to know all the answers. Whenever Markley didn’t know something, he had to live with Joe riding the living hell out of him until he found out—“so it was best if I got a good system going.”

Shea turned the Martin model down flat. Markley submitted a revised plan, scaled down a little. Shea turned that down flat too. This is ridiculous, Markley thought, so finally he walked in and asked Shea what kind of management system he wanted. Shea told him. “What he wanted,” Markley said, “was to get a notebook every Thursday night, and in it he wanted the entire program structured, meaning all the interfaces with Marshall, everything going on with the command module, everything going on with the service module, everything going on with the SLA, everything going on with guidance and navigation, everything going on with the LEM, everything going on with the launch pad, the ground support equipment, with General Electric, everything Boeing was doing, everything everyone was doing.”* All this was to be in one notebook, and should include not only what had been done in the last seven days but also comparisons against the programmed schedule and costs.

[* “LEM” refers to the lunar excursion module, the lunar lander. The word “excursion” was insisted upon by Owen Maynard and others at Langley who wanted to emphasize its limitations. Eventually headquarters shortened the official title to lunar module, and the official abbreviation was shortened to LM. Everyone continued to call it the “lem,” so we have compromised with purity by spelling it “LEM” throughout the book. The “SLA” was the spacecraft lunar adapter, the slanted section just below the C.S.M. that contained the lunar module during the launch. SLA is pronounced “slaw,” as in cole.]

And so Markley went away to produce the first edition of the loose-leaf notebook that he and his staff would prepare for Shea every week for the next 165 weeks. Each notebook ran to more than a hundred pages and was on Shea’s desk by close of business Thursday. Shea would get up at 4 A.M. Friday and start to annotate it, usually with technical comments and instructions, occasionally with remarks such as “You’ve gotta be kidding.” He would work on the notebook intermittently from Friday morning through the weekend and return it to Markley on Monday morning. By Monday afternoon, everyone in the ASPO network would have gotten his response from Shea.

With the annotated notebook in his hands, Markley became the action officer. “I’d take this notebook, and then I’d fan it back out very quickly and say, ‘Hey, we’ve got these questions and I want the answers in the notebook by this Thursday.’ We’d get the answers back and then we’d take Joe’s marked-up notebook and start off in the front, anything he had marks on. We’d have the answers there. And then we’d have the brand-new section with the newest status too.” The scrawled notations to and fro were a way of keeping up “a running communication,” in Shea’s words, without having to type up long memos. “Even if I didn’t see [the project officers] directly, I’d at least been in touch with them, and they’d been back in touch with me.”

“I’d never seen it done before and I haven’t seen it done since,” recalled the project manager for the LEM. “Shea came into the organization and he said, ‘I want only those things that you want me to read and that you want some kind of answer on. Just don’t tell me things are going along great, but if you want some decision, do it through your weekly activities report.’ So we’d send them in on Thursday and on Monday morning you’d have your copy back with his comments on it.” He could never figure out where Shea got the time.

By now, in 1964, the program had exploded in size. Faget’s engineering division alone employed 1,400 men, serving (among other duties) as ASPO’s technical resource. At North American’s Downey plant near Los Angeles, 4,000 engineers were working on the command and service module. Grumman was going full speed on the lunar module. Other contractors were making the space suits, the guidance and navigation systems, and a dozen other major systems, not to mention the hundreds of smaller contracts and subcontracts. And this was just Joe Shea’s shop, having nothing to do with the one being run out of Marshall to build the Saturn V and a third one at the Cape to build the V.A.B. and Complex 39. All of these systems were being designed, developed, and assembled simultaneously. The most minute change in one system was likely to require a cascade of changes in others. The complexity of the program and the coordination required to keep all the pieces growing in harmony increased exponentially.

The coordinating power for the spacecraft was in Shea’s hands, and he exercised it without a blink. As huge as the program was becoming, with its growing paraphernalia of collective decision making, any decision about the C.S.M. or the lunar module ultimately came to Shea, and Shea didn’t worry about getting a consensus. In the fall of 1964, for example, he established a Change Board that would systematize and limit the changes being made to the spacecraft. But when it came to the final decision, the one man whose opinion really counted was Joe Shea. An engineer who had to deal with Shea remembered what it was like. If you had a change you wanted to propose in the spacecraft, and if it got by the Change Board, the last step in the process was to go into Shea’s office on the seventh floor over in Building 2 at M.S.C. You’d walk in, and there Joe would be, by himself, behind his desk, and you had to explain to him what it was you wanted and why it was absolutely essential that this change be made in the spacecraft. And then Joe would begin to grill you. There was no point in trying to sidestep or fudge an answer. It made no difference if Shea wasn’t a specialist in your area. “If you understand it, you can make me understand it,” Shea kept saying, and you had better understand it absolutely cold or Shea would dissect you.

Also, you had to keep ahead of him, which could be difficult. “He’d stop you in the middle of a presentation and say, ‘You’re going to have to include that you need four batteries in the circuit,’” one engineer recalled. “And you’d say, ‘I’m going to get around to that.’ And he’d say, ‘Well, tell me right now why the hell we need four batteries. And goddamn it, if you can’t tell me why we need four, then you ought to come back and see me next week. You’re wasting my time.’”

At the end of the meeting, Shea would say yes or no. And that was it. It took people a while to get used to that, and sometimes during his first months at Houston, an item that Shea had decided against would be brought up at another meeting. “I want to tell you that an adverse decision is not a decision delayed,” Shea icily informed an engineer one day, and after that the practice stopped. “I never ran the Change Board as a democratic process,” Shea acknowledged.

“The better is the enemy of the good,” Shea told them again and again.* The biggest problem with a new product in its developmental phase, Shea thought, was that a good engineer could always think of ways to make it better. This was fine, except that they couldn’t keep changing the spacecraft forever. Sooner or later they had to lock it into one configuration, so that they could make that configuration work. Keep the changes down.

[*There are variants on the phrase, which has been popular among engineers for many years. Bill Donovan, head of the O.S.S. during World War II, liked to say “the perfect is the enemy of the good.” Shea’s version enjoyed a vogue during the Apollo Program. Jim Webb and Robert McNamara both used it, always referring to it as an old Russian proverb. One historian of the Apollo Program, Robert Sherrod, was able to track a variation back to Voltaire (“Le mieux est l’ennemi du bien”). Voltaire in turn seemed to think it was an old Italian adage.]

Keep it simple—that was the other half of the Shea doctrine. To Shea, it seemed as if everyone was saying, “Apollo is the most complex job in the world; therefore, every part has to be as complicated as possible.” And that was exactly backwards. What they should be trying to do was figure out how to do things as simply as possible, with as few ways as possible to go wrong.

Fuel gauges, for example, in the tanks in the reaction and control system (R.C.S.) on the spacecraft. Okay, Shea agreed, it’s tough to measure how much fuel you’ve got in a tank when you’re in zero gravity. The usual devices don’t work. And furthermore, it was vital that they keep enough fuel in those tanks right up to the end of the flight, because the R.C.S. was what put the spacecraft in the right attitude to enter the atmosphere. But the people who wrote the specifications for the fuel gauges had reasoned that, because it was so important, they needed a highly accurate fuel gauge system. The way they would do it was to build a Geiger-counter arrangement that would infer the bulk of fuel left in the tank by the attenuation of radiation sources that would be transmitted through the fuel. It was a challenging bit of technology, no doubt about it—so challenging that the gauge they had developed so far was accurate only to about 10 percent and it kept failing quality tests for reliability.

Shea let this process run for a couple of months and then stepped in. “This thing doesn’t make any sense to me,” he said. “We don’t need anything that complicated.” Shea had a Karmann Ghia that didn’t even need a fuel gauge, because it had a little reserve fuel tank that he could switch to. And that’s what they ended up doing for the R.C.S. fuel tanks. They had a big tank that they used during the mission, and a little tank, not to be touched unless the big tank ran out, but that could get the crew home safely. No more nuclear gas gauge.

There was the problem of the heat shield. The specification the North American engineers were trying to meet required, logically enough, that the heat shield maintain its integrity under the extreme shifts from cold to heat that it would experience on its way to the moon and back, depending on whether it was in sun or shadow. But when they began testing the heat shield material under entry conditions, they found to their consternation that after exposure to severe cold it began to crack and craze and flake. They were going to have to create a new heat shield material, which would take untold quantities of money and time. Shea asked how long it took for the heat shield to cool down to the point where the problems began. The answer was about thirteen hours. So why did the spacecraft have to stay in the same attitude for that long? Why couldn’t it rotate, so the heat shield would remain nice and warm all the time? And that was the origin of what came to be known as the barbecue mode, or passive thermal control, in which the spacecraft rotated once an hour all the way out to the moon and back. Keep it simple.