Apollo: The Race to the Moon (14 page)

On top, stretching into the sky alongside the Saturn V, were the umbilical tower on one side and the mobile service structure on the other. The mobile service structure, as tall as the launcher, had five work platforms that gave access to the launch vehicle and also contained several enclosed rooms in its base. It was moved up to the vehicle after the launcher was in place, then moved back again to its parking area more than a mile away from the pad eleven hours before launch.

The launcher itself was the platform on which the Saturn sat, a steel box 25 feet deep, 160 feet long, and 135 feet wide. The engines of the Saturn were positioned over a 45-foot-square hole in the middle of it. Beneath the whole structure was the wedge-shaped flame deflector, which sat in a trench between embankments on either side. So from ground level on up, the addition went like this: 48 feet for the flame trench, 22 feet between that and the bottom of the launcher (to make room for the transporter), and 25 feet for the launcher, which finally put Buchanan, staring at his drawings, at the Zero Deck—95 feet above the ground, roughly the height of a seven-storey building.

Then above the launcher was the 380-foot umbilical tower, structurally integral with the launcher. The tower had two high-speed elevators to carry men and equipment to the nine swing arms that carried electric, propellant, and pneumatic lines to the Saturn. The swing arms themselves were massive, wide enough to drive a jeep across and averaging about 24 tons in weight—but also movable, designed to disconnect from the Saturn and swing away from it in the first few seconds after the engines ignited. Atop the umbilical tower was a 19-foot-high, 25-ton hammerhead crane, and atop that was a 48-foot retractable lighting mast. Thus, the total height from the base of the flame deflector to the top of the lighting mast was 542 feet.

As they began to work on the designs for the launch pad in more detail, Buchanan began marking up the side of their office building, a converted cotton mill in Huntsville, putting yellow tape where the different levels of the structure would be. It got his people “thinking in the right proportions,” Buchanan said. Then, to make sure that they really understood what they were up against, Buchanan got a large photograph of the tallest building in Huntsville, the Times Building, 107 feet high. Buchanan superimposed on the photograph a sketch of Launch Complex 39. The top of the Times Building was only 12 feet higher than Zero Deck of the launcher. Buchanan would show the picture to his Huntsville friends, and they reacted more or less as he had when Petrone first gave him the job. “Well, you’re probably crazy to even build that thing,” they would say to Buchanan, “but then we know you’re crazy when you say you’re going to move it.”

In the beginning, Debus and Petrone hadn’t been that worried about how to move the Saturn to the pad. They knew they wouldn’t be able to move it horizontally, but they had always assumed that some sort of barge system would be the solution. Barges were a good way to carry great weight. And so from the earliest planning through the summer and fall of 1961, the plans called for a canal to be built from the V.A.B. out to the launch pad. A huge motor-powered barge would carry the assembled Saturn and its umbilical tower.

But after Buchanan was put to work on the problem in October 1961, he found that the more they worked on the barge design the more problems they encountered. The onboard propulsion system they had planned consisted of a set of large outboard motors. But tank tests revealed that they couldn’t use such a system in a canal of the planned width using a barge of the required dimensions. There would be so much backwash that more motors would be needed than they had places to put them, and they couldn’t widen the canal enough to get rid of the backwash effects. The next version of the plan scrapped the outboard motors for a rail system running alongside the canal—a twentieth-century version of the mules that used to pull barges along canals a century earlier. It would work, but it was also making the system much more complicated and expensive.

Then they discovered that it was going to be impossible to steer the barge: The Saturn and the umbilical tower would act like a huge sail, forcing the barge to the banks in a cross breeze. The rail system would have to be modified so that it not only pulled the barge, but kept it clear of the canal banks. And there were other little things. If they had a three-and-a-half-mile canal cutting through the middle of the Merritt Island facility, they would have to build a system of drawbridges, which would be complicated and expensive. A canal going into the V.A.B. would create high humidity in a place where they wanted as little humidity as possible. Contending with that would also be complicated and expensive. And after figuring out solutions to all these problems, Buchanan had to decide what to do when the barge and its stack reached the launch pad. No matter how Buchanan fiddled with different ideas, it came down to the same choice. Either he had to jack the whole launcher up a hundred feet so he would have room to put a flame deflector under it, or he had to evacuate the water out to a depth of a hundred feet, like a big bathtub. “Neither option was all that attractive,” Buchanan said dryly. As he looked long and hard at the barge option, he began to think that too much had been assumed too quickly.

By the end of January 1962, Debus and Petrone were still committed to the mobile launch concept for reasons that by now had nothing to do with the speed of launch. They wanted the flexibility that a mobile system gave them, the protection from the elements, the room for growth. But the underlying technical rationale for the system had originally been that it would be the most efficient way to launch large numbers of vehicles, and the break-even point had generally been calculated as twelve. By this time, realism had prevailed. There was no more talk of a hundred launches per year or of $10 million per-launch costs. On the contrary, headquarters was now confident that the lunar program would require considerably fewer than twelve large Saturn launches per year.

Two studies by NASA contractors had already come out recommending a fixed system. Debus knew further that a systems analyst at headquarters had put a three-man team to work reassessing the launch system and that their thinking was moving in the direction of a fixed system. If the mobile system was going to survive, they must pin its pieces in place, which meant reaching a firm decision on the mode of transport. At a meeting at Huntsville on Tuesday, January 30, 1962, Debus made it clear that he wanted a decision, he wanted it soon, and that unless they had something better, they would have to go with the barge.

On Friday of the same week, one of the men who had been in the meeting, O. K. Duren, got a call from a man named Barry Schlenk. Schlenk was at Huntsville that day talking to some people about an overhead crane for the Titan system. Schlenk had overheard someone talking about Marshall’s problem with moving the Saturn down at the Cape. Did Duren know that Schlenk’s company, the Bucyrus-Erie Company, had a rig that was being used for mining in the Kentucky coalfields? It crawled along the surface of the ground on tank treads, stripping the overburden from the mining area. It was huge and it was self-leveling. Would Duren be interested in hearing about it? Duren would, and the two spent the afternoon looking at pictures of the machine. Duren was enthusiastic and called his boss, Albert Zeiler.

Zeiler told Debus about the shovel, adding that it was the dumbest thing he’d ever heard of. But Debus told him to go take look at those shovels and come back and tell him why they shouldn’t consider them. Reluctantly, Zeiler took Theodor Poppel, head of the Launch Facilities and Support Equipment Office, Duren, and Buchanan up to Paradise, Kentucky—a most inaptly named place, in Buchanan’s opinion—to see this machine. The story that was passed down in Cape lore in later years was that they went out to the coalfield and climbed onto the vehicle and examined it, and then Zeiler and the others sat down, grabbed onto the nearest handholds, and said, “We’re ready! You can start!” And their puzzled hosts said, “What do you mean? We’ve been moving for a couple of minutes now.”

The shovel was too slow, only twenty feet per minute, and the leveling technology wasn’t up to aerospace standards. But the fact remained: Here was this great big machine transporting an extremely heavy weight with no vibration. Buchanan began to look into it more carefully. Studies were conducted, technical comparisons were made, and only four months after the NASA engineers had so skeptically gone to Paradise, Debus approved Buchanan’s recommendation to discard the plans for a barge and to proceed instead with a gigantic transporter-crawler.

Because of its much larger scale, the final result bore only a passing resemblance to the big stripping shovel in the Kentucky coalfields. But Buchanan used its basic design—a platform with a truck at each of its four corners. Each truck looked rather like a tank, with a double set of belted links, or shoes (known to the engineers familiar with the budget as “them golden slippers”). Each link, or shoe, was nearly eight feet long and weighed a ton. Each belt—eight in all—consisted of fifty-seven shoes.

Describing the scale of his creation, Buchanan used to suggest that his listeners imagine the infield of a major-league baseball diamond being cut out and made into a platform. Then they should imagine the platform, elevated on its monster treads, moving along a track for three miles. Now all Buchanan’s listeners had to do was imagine the elevated baseball infield moving along a track for three miles carrying two-thirds of the Washington Monument on top, coming to a five-degree slope, and then climbing it. Last of all, they must imagine that as the assembly climbs the slope, the baseball infield is hydraulically jacked so that it remains level throughout the climb. That was the crawler.

It was a marvel for which Buchanan was honored with some of the most prestigious awards in the engineering fraternity, but the ways in which it was a marvel are for the most part too technical to be appreciated by anyone except other engineers. For the lay reader, perhaps this is enough: During its voyage from the V.A.B. to the launch pad, including its climb up that five-degree slope, the tip of the Saturn, 363 feet above the platform, never moved outside the vertical by more than the dimensions of a basketball.

All this Buchanan promised Debus that spring of 1962, and he got his chance to make good on it. Debus and Petrone were not to be stopped: They overcame the objections of the Office of Manned Space Flight and Congress alike and received permission to proceed with the mobile launch system.

“We often looked at the launch site and those things on the ground that don’t fly as Stage Zero [of the booster],” Rocco Petrone said. “You had to have all the intricacies of a stage, things like swing arms, hold-down arms, feeding the gases in, all the propellants. When you’ve released, your Stage One is flying, but if you haven’t done all these things on Stage Zero, Stage One would never get a chance to fly.” The Cape’s task was to design all this even as the Apollo-Saturn stack itself was being designed. “That was a hell of a challenge,” Petrone continued, “and, I think, a challenge not very well understood. Here you have the launch vehicle stages going down the road with the spacecraft, and they develop needs for more juice, more wire, more propellants. You’ve got to be in the right rhythm, because when they come together you must all be at the crossroads at the same time.”

Chapter 7. “We had more harebrained schemes than you could shake a stick at”

In the months following Kennedy’s speech, Bob Gilruth would occasionally remark to the others in the Space Task Group that everything Kennedy had said in that speech was fine except for that one little word “safely,” as in “… and returning him safely to earth.” “That’s not simple,” Gilruth would say to the others. “We don’t know how to do that.” That was why he had been aghast when he heard Kennedy’s speech. Gilruth was confident that NASA could get a man to the moon safely—eventually—or that they could get a man to the moon within the decade if they were willing to take a pretty high chance of killing him. But as of 1961 they still didn’t know how to do it both safely and within the decade.

It was hard enough just deciding what “safely” meant. Caldwell Johnson and Bob Piland, who wrote the work statement for the Apollo spacecraft, decided they had to have a number stating how reliable the spacecraft must be. That one demand would make a huge difference in the cost of the program because, as Johnson pointed out, “if you can afford to lose half of the spacecraft and half the men, you can build them a damn sight cheaper.” On the other hand, Johnson continued, “in this country you just don’t go around killing people. So we had to pick a number. Piland, he didn’t dare pick it. And God knows I wasn’t going to pick it. Nobody wanted to pick the number. So one day we walked down to see Gilruth and we said the time had come to bite the bullet.”

Max Faget was there. Faget didn’t have much respect for this reliability analysis—it was okay in theory, he said, except that it depended upon data that didn’t exist. In this objection lay a controversy that would fester for the next two years.

The theory for computing reliability is simplicity itself. If a machine has a part that fails twenty out of a hundred times, its reliability—the probability that it will work—is .80. If the machine has a second part that will fail ten times out of a hundred, its reliability is .90. The probability that both parts will work is the multiple of the two individual reliabilities, .80 times .90, or .72. The complete machine (if it consists of just the two parts) will work seventy-two times out of a hundred.

In a spacecraft, there would be tens of thousands of parts that all had to work in order to complete the mission successfully. To compute the overall probability of mission success, it was necessary only to know the reliability of each of those individual parts, and then calculate their combined product, which could be done in a comparatively short time on an ordinary calculator. Some smaller number of parts all had to work just to get the crew back safely (a less demanding task than completing the mission successfully). For the probability of getting the crew back, just multiply the reliability of all those parts.