by Neil Sheehan
The Black Saturday briefing, held in the Arbor Vitae control room, became the major event of the month at headquarters. Getz was its maestro. Prior to arriving in California, his nickname had been “Bill,” from his middle name of William. Soon, behind his back, he was being called “Cecil B. De Getz,” after Cecil B. DeMille, the famous Hollywood producer and director of spectacular epics like The Ten Commandments and Samson and Delilah. Getz’s productions began at 8:45 in the morning and extended until well after 5:00 in the afternoon, with ten-minute personal relief breaks at midmorning and mid-afternoon and an hour off between noon and 1:00 for lunch. He ran the sessions with ruthless efficiency, because he knew this was what Schriever wanted in order to pack as much as possible into a single day. A missile team would be allotted fifty minutes to make its presentation. If a briefer was starting to run over on his time, Getz would flash on a screen behind him a sketch of a hook pulling a man off a stage. The trick never failed to elicit laughter from the briefing room, but if the briefer ignored the message, Getz would abruptly inform him, even in the middle of a sentence, “Time’s up!”
Schriever’s subordinates had aptly dubbed the monthly briefing Black Saturday because “the Boss” was not interested in good news. At a time when the upper echelons of the American military were beginning to suffer from the disease of professional arrogance and lack of imagination, and the fare of a normal briefing was “Progress,” Schriever took an opposite approach. He wanted to know the problems, on the not irrational assumption that if they were solved, success would take care of itself. “I don’t like to be surprised,” he would say. “Give me the bad news. I can take it. I will not fire you for giving me the bad news. I will fire you if you don’t give me the bad news.” The briefings thus tended to be succinct recitals of woe, with frequently tentative ideas on how a team was going to solve a problem that continued to baffle. The discussions that followed sometimes helped, sometimes not. The paperwork burden and the time consumed in team meetings to reach agreement on the status of every aspect of every project made Getz the most unpopular man in the command. Yet no one dared ignore him because they knew that Schriever wanted what he was demanding. And there was no revolt because all understood that the system enforced discipline and teamwork. Everyone was made aware of what everyone else was doing and thus could pitch in to assist. Most important of all, the focus stayed where it mattered—the gradual elimination of impediments to getting the job done.
64.
THE TRIALS OF ATLAS AND A CHRISTMAS SURPRISE
The launching of Atlas at Cape Canaveral began on June 11, 1957, with an ostensible failure, a left punch for Schriever to absorb after taking a right, as the disappointment came just twenty-two days after Thiel and Mettler incinerated Thor 103 on its pad by overpressurization of the LOX tank. In Atlas, an ICBM designed to hurl its warhead 6,330 miles, Bennie was dealing with a missile larger and more complicated than the intermediate-range Thor and thus considerably more prone to trouble. Atlas was approximately seventy-five feet long, ten feet in diameter, and had in excess of 40,000 parts. In firing position, it was as high as a seven-story office building. When fully loaded with fuel and a simulated hydrogen bomb in its reentry vehicle warhead, it weighed 243,000 pounds, in contrast to Thor’s 110,000.
Where the 1,725-mile Thor could loft itself with one of Hall’s booster engines, improved to 150,000 pounds of thrust, Atlas needed 360,000 pounds. To obtain it, three engines were lined up at the base of the rocket, two 150,000-pound boosters, one on each side, with a 60,000-pound sustainer engine set between them. Atlas was what is referred to in the guided missile business as a stage and a half rocket. All three engines at the base fired simultaneously for liftoff, but two minutes into the flight, after the rocket had long cleared the dense air of the atmosphere and was on its way to maximum speed, the two booster engines were cut off by a radio control signal from the ground during testing and by the onboard inertial guidance system after it had been deployed. Another signal fired a release mechanism on the framework to which the big boosters were mounted and they fell away back to earth. The 60,000-pound sustainer engine in the middle, separately attached to the bulkhead at the base of the fuel tank fuselage, was kept burning for close to another three minutes to bring the Atlas within a fraction of the 16,000 miles per hour necessary to hurl the warhead the full 6,330 miles. With a last signal, tiny “retro” rockets at the front of the fuselage blasted into life. They snapped the warhead free from the fuselage, sending it off on its curved journey upward through space, reaching a point more than 800 miles above the earth at the apogee, before arching down to its target. This, in any case, was how the Atlas was supposed to work. Getting it to do so was another matter now to be undertaken.
Because they were moving through uncharted terrain, everyone involved—Convair, the Ramo-Wooldridge group, and Bennie’s project officers—had decided to test the Atlas in four stages. The Series A missiles would check out the functioning and air-worthiness of the fuel tank fuselage and the propulsion system. These missiles would be the lightest in the series at 181,000 pounds, as they would be equipped with only the two main booster engines, not the sustainer. They would also be flown the shortest distance, a mere 530 miles. The completeness of the missiles and the length of the flights would then gradually increase through Series B and C, until, in Series D, missiles identical to those that were to be deployed would be tested at the full range of 6,330 miles.
Atlas 4A, the first readied for flight (missile numbers often did not correspond to launch sequence because flaws would be discovered in preflight tests and another missile substituted), reached Cape Canaveral in late March 1957. It came, as all of its relatives would, by trailer truck on a 2,622-mile journey across the continent from the Convair plant at San Diego. Even with the nose cone removed and shipped separately to shorten it, the missile was still too long and bulky to fly to the Cape in a C-124 Globemaster. And so a special sixty-four-foot-long trailer was fashioned, a steel cradle on wheels, and the missile, wrapped in a canvas shroud, was loaded into it and the trailer hooked to a truck. The trip took nine days because, for safety reasons, driving was restricted to daylight. There were armed guards on the trailer truck and in accompanying vehicles.
At dawn on June 11, 1957, Atlas 4A stood on the launch pad, the stainless steel of its fuel tank fuselage section gleaming in the first washes of the sun rising over the Atlantic. The day of a missile launching at the Cape could no longer be kept hidden. There were too many leaks and giveaway signs of preparation, and so on this day thousands of spectators lined Cocoa Beach five miles to the south to watch America’s first intercontinental ballistic missile soar in the inauguration of a new epoch. The countdown in the blockhouse had started earlier, at 5:00 A.M., half an hour before sunrise. The Atlas had passed months and months of preflight checks of the components in California and then of the assembled missile at the Cape, including a short, static firing of the engines on the launch pad. It was rigged out with telemetry sensors to monitor its performance during the flight. The care of the preparations was evident in the monotonous but confident manner in which the countdown unfolded for three hours and twenty minutes. There were only two hitches that paused it, both toward the end of the sequence. A joint in a line feeding LOX from a storage tank to the missile sprang a leak and had to be replaced. Then an electrical circuit breaker tripped, cutting off the connection between a control console in the blockhouse and the missile. A Convair technician walked out of the blockhouse to the electrical transfer room near the launch pad and the now fully fueled missile and reset the breaker. The blockhouse door was closed again and the last steps of the count completed. Little green lights flashed across the control panel of the Convair test conductor, who had led the countdown, as he pressed the button to start the ignition sequence.
To the relief and joy of those on the Cape so intimately involved and the bystanders on Cocoa Beach, Atlas 4A rose and began a magnificent flight, for twenty-four seconds. Then, all of
a sudden, the engines lost thrust. The flare of rockets no longer lit the sky. Only orange smoke billowed from the engine nozzles. The Atlas flipped wildly through a loop-the-loop and fell back into its trail of fire. The voice of the range safety officer at Central Control came up on the blockhouse intercom: “Destruct.” This time justifiably, he punched the button flashing a radio signal to the packet of explosives on the missile and scattered the Atlas in pieces of flaming debris. “That was a total waste,” someone in the despondent blockhouse said. “Hell, no,” replied Edward Doll, one of the Ramo-Wooldridge engineers, pointing out that what looked bad was actually good news. They had just watched the Atlas gyrate through a series of extreme contortions in the sky and the missile had not broken up from the stress. Karel Bossart’s radical weight-saving design of the Atlas fuselage that doubled as its fuel tank had been a worry for everyone. John Medaris and Wernher von Braun had just been shown to be self-serving Cassandras in predicting that the “balloon,” as they had scornfully referred to the Atlas, would crumple under the strains of liftoff and flight. “We proved it could stand three G’s,” Doll said, engineer’s shorthand for three times the force of gravity. And Bossart was present at the Cape that day to witness the vindication of his idea. Bennie could thus console himself with a partial success, but he knew that the Pentagon and the White House, like the spectators on Cocoa Beach, would see the launch as another of Schriever’s missiles gone down in flames or burned up on the pad.
Partial success was certainly not enough after the second Atlas launched, 6A, which took almost three and a half months to ready, put in a virtually identical performance on September 25, 1957. The rocket flew for thirty-two seconds before the failure of a LOX regulator, as the telemetry would reveal, led to another loss of thrust and destruction. Jacobson’s promising accomplishments with Thor provided some diversionary comfort, but this vanished that October 4 with the shock of Sputnik. The pressure on Schriever ratcheted up enormously. The third Atlas, 12A, had to fly as promised and everyone involved, from the launch crew in the blockhouse at Canaveral to those waiting at the other end of the direct Teletype line at the Ballistic Missile Division in Los Angeles, shared the unnerving suspense. The rocket’s booster engines burned faultlessly for the full two minutes after liftoff on December 17, 1957, the fifty-fourth anniversary of the Wright brothers’ flight, sending the missile the 530 miles down the Caribbean range for which the flight had been programmed. Eisenhower was in Paris, where he could pass on the encouraging word to the other Allied leaders at a NATO meeting he was attending. Even Democratic Senate Majority Leader Lyndon Johnson, who was cranking up his Preparedness Investigating Subcommittee to give the administration a thrashing, had a compliment. “That is mighty good news,” he said.
It was clear by this time that there was a major flaw in the 150,000-pound-thrust booster engine, a flaw that had been responsible for the loss of thrust on the original Atlas launch of June 11, 1957, and for some of the failures in the Thor and Jupiter launches. The flaw was the worst kind an engineer could face, because it appeared randomly, sometimes twice in a row but on average about every five or six launches. The rest of the time the engines functioned fine. And to make the problem still more intractable, the engineers disagreed on where the flaw lay. Thiel’s former German colleagues on the von Braun team told him right away, and as it turned out correctly, that the defect was in the turbo-pump, which mixed the RP-1 and LOX together at extremely high speed as they were fed into the burn chamber of the engine. They were convinced that the force of liftoff caused the bearings within the pump to shift. The bearings were then seizing up in flight, stopping the pump and the flow of fuel to the engine and, if the pump overheated enough, causing it to blow up and take the missile with it. The answer, they said, was to put a restraining mechanism on the bearings to hold them in place. The Ramo-Wooldridge rocket engine specialist disagreed. He said the failures were being caused by a misalignment of the outlet from the fuel tank to the pump.
Ed Hall, who had been assigned by Bennie to develop a revolutionary ICBM with a solid-fueled rather than a liquid-fueled engine, was off on his own brainstorming and did not get involved. The argument endured for months with others injecting their guesses and no solution in view. Bennie, who had come to have a particular trust in Mettler, asked for a memo advising him what to do. Mettler urged him to keep firing missiles until they could sort out an answer because the failures were random and they were learning so much with each launch. Most Air Force generals, their careers at stake as Schriever’s was, would have halted, focused on fixing the engine no matter how much time was lost, and then resumed testing. In his race with the Soviets, time was a commodity with which Bennie Schriever was unwilling to part. He took Mettler’s advice and pressed on. The turbopump was not fixed until well into the fall of 1958.
After that first successful Atlas flight on the fifty-fourth anniversary of the opening of the aerial age, there were days of triumph and months of heartbreak, but the ultimate goal of an operational intercontinental missile force as a deterrent to a Soviet surprise attack was always in sight. On August 2, 1958, the second Atlas in the B Series, 4B, gave, on signal, a perfect rendition of the five prescribed steps of flight. The booster engines shut down after two minutes, the release mechanism jettisoned them, the sustainer continued to burn for nearly another three minutes until it too was cut off, the two diminutive vernier engines made the final corrections in speed and angle, and the miniature retro rockets then came to life and freed the warhead to take flight through space. On November 28, another missile in the B Series, Atlas 12B, became the first to fly the entire 6,330-mile course.
Then came a Christmas surprise thought up by Mettler and several Convair engineers. On December 28, 1958, Atlas 10B, fitted with a special aerodynamic nose cone, blasted off a pad at Cape Canaveral. Inside the nose cone were two battery-powered tape recorders fitted to two radio transmitters, a pair of each in case one of the devices should fail. Instead of being sent on the high looping course of an ICBM, Atlas 10B was directed along a lower course parallel to the earth, and instead of cutting off the booster and sustainer engines, they were kept burning until the entire missile reached the speed of 17,300 miles per hour, escaped the gravity pull of earth, and flew into orbit. Every 101 minutes, it completed a circle of the globe. On the thirteenth pass over the United States, another radio signal from Canaveral turned on the tape recorders and transmitters and the voice of Eisenhower broadcast Christmas greetings to the peoples of the earth. “Through the marvels of scientific advance, my voice is coming to you from a satellite circling in outer space,” the president said. “My message is a simple one. Through this unique means I convey to you and to all mankind America’s wish for peace on earth and good will toward men everywhere.” Project Score, as the project had been code-named, lacked the shock of Sputnik, but it was still quite an accomplishment. Spent of its fuel, the Atlas had become a satellite weighing 8,800 pounds. It continued circling the earth for thirty-three days, traveling 12.5 million miles before falling back into the atmosphere near Midway Island in the Pacific on January 21, 1959, burning up in a fiery climax.
The next six months were the most heartbreaking time. Testing of the C and then the D Series, the model that was to be deployed initially, began with a string of failures. Only two of the eight missiles of both series launched during the first half of 1959 were truly successful. The other six did not just blow up on the pad, but neither did they meet most of their test objectives. The mid-1958 “Ph.D. type” operational capability that Trevor Gardner had dreamed of back in 1954 had proven impossible to meet, but so did the June 1959 deployment date that Schriever set. The testing did improve the accuracy of the missile during the first half of 1959, eventually attaining a consistent accuracy of 2.3 miles by substituting an aerodynamic ablative reentry vehicle for the blunt-nose Mark 2 heat shield type first designed for Atlas. As had happened with Thor, occasional random misses were occurring in the launches aimed at the circl
e of hydrophones off Ascension Island in the South Atlantic. And again as with Thor, analysis convinced the engineers that the high winds in the upper atmosphere were from time to time catching the heat shield of the Mark 2 and pushing it off trajectory. While Bennie had been willing to tolerate the flaw in Thor because it was an intermediate-range missile and he was in a rush to finish the project, he was unwilling to ignore it in America’s first ICBM.
To obtain an ablative RV, they could not simply copy the one von Braun had pioneered for Jupiter, because the Atlas’s nose cone would be reentering the atmosphere at the much higher speed of 16,000 miles per hour, thus generating a lot more heat. As the ablative type worked by coating the nose cone with a compound of plastic and other material that burned off on reentry—deflecting heat from the nose cone itself and the hydrogen bomb inside—the question was exactly how much and what composition of coating was required. At the suggestion of Simon Ramo, Schriever had commissioned Lockheed to create a three-stage rocket called the X-17, or Athena, for the precise purpose of mimicking the conditions under which an ICBM warhead reentered the atmosphere. A scaled-down model of the RV was mounted on the rocket and the first two stages launched it into space. The third stage was then ignited and fired the nose cone back down into the atmosphere. The X-17 turned out to be an example of the technologist outsmarting himself with gimmickry a bit too fancy for the moment. The X-17 declined to go fast enough on the downward leg to replicate the reentry heat of an ICBM warhead.
