SLOW RECOVERY

Within days of the publication of Floyd Thompson’s damning report into the Apollo 1 fire, the first efforts were implemented to fulfil its recommendations. Of paramount importance was the redesign of the hatch, which would change from a complex two – piece device into a ‘unified’ single section. Although it was heavier than the hatch which had prevented Gus Grissom, Ed White and Roger Chaffee from escaping the inferno of Spacecraft 012, it could be opened in as little as five seconds and had a manual release for either internal or external operation. At the same time, fire and safety precautions were upgraded at Cape Kennedy and a slidewire was added to Pad 34’s service structure to allow crews to rapidly descend to ground level.

By the beginning of May 1967, a sense pervaded NASA and North American that the first steps to recover from the fire were underway; so much so that George Mueller proposed an unmanned test flight of the gigantic Saturn V lunar rocket as soon as possible. A crewless demonstration of the improved Apollo system was definitely needed and, utilising a command and service module combo known as ‘Spacecraft 017’, was pencilled-in for the early autumn of that year. By that time, four manned missions had also been timetabled, one featuring the command and service module on its own, the other three inclusive of the lunar module, after which an attempt to actually touch down on the Moon might go ahead. Certainly, Time magazine told its readers on 19 May that unmanned Apollos were scheduled for September, October and December, followed by an inaugural manned mission in March 1968. NASA Headquarters were even more optimistic. Some managers suggested that a lunar landing could occur on the fourth manned Apollo flight, but their counterparts in Houston expressed more caution. Chris Kraft, for one, had warned George Low, who replaced Joe Shea to head the Apollo Spacecraft Program Office, that a lunar landing should not be attempted ‘‘on the first flight which leaves the Earth’s gravitational field’’.

Others, including Mueller, wanted to skip the flight of a manned command and service module in Earth orbit entirely and press on with a complete ‘all-up’ test of the entire Apollo combination, including the lunar module. ‘‘Bob Gilruth got in the way of this one,’’ wrote Deke Slayton. ‘‘For one thing, the Apollo CSM was a sufficiently complex piece of machinery that it needed a shakedown flight of its own. Why try to test two manned vehicles for the first time at the same time? We thought a CSM-only flight was the way to go before the fire and nothing we were going to learn was likely to change that.” Moreover, the lunar module itself was running months behind schedule and a manned flight was not anticipated until at least the end of 1968. Mueller was finally persuaded to accept a command and service module flight in Earth orbit for the first manned Apollo mission.

Despite the increased optimism, concerns remained. The schedule for the first unmanned Apollo test atop the Saturn V – designated ‘Apollo 4’ or ‘Apollo-Saturn 501’ (AS-501) – was extremely tight. In particular, the Saturn’s S-II second stage had undergone a difficult year of testing in 1966. Nonetheless, at the stroke of 7:00 am on

9 November 1967, the entire Cape Kennedy area received a jolt when the five F-1 engines of the Saturn V ignited with what Brooks, Grimwood and Swenson later described as ‘‘a man-made earthquake and shockwave… the question was not whether the Saturn V had risen, but whether Florida had sunk!’’ Deke Slayton, who had come to the Cape to watch the behemoth fly, later recounted that he had ‘‘seen a lot of launches… but nothing was ever as impressive as that first Saturn V. It just rose with naked power, lots of noise and light’’. Fellow astronaut Tom Stafford, also there, commented that Walter Cronkite’s CBS News trailer almost shook itself to pieces. ‘‘Suddenly,’’ added Mike Collins, ‘‘you realise the meaning of 7.5 million pounds of thrust – it can make the Cape Kennedy sand vibrate under your feet at a distance of four miles… ’’

The merest mention of the name ‘Saturn V’ implies power. From a height, weight and payload-to-orbit standpoint, it remains the largest and most powerful rocket ever brought to operational status, although the Soviet Union’s short-lived Energia had slightly more thrust at liftoff. It evolved from a series of rockets, originally dubbed the Saturn ‘C-1’ through ‘C-5’, of which NASA announced its intent to build the latter in January 1962. It would be, the agency revealed, a three-stage launcher with five F-1 engines on its first stage, five Rocketdyne-built J-2 engines on its second stage and a single J-2 on its third stage. These engines, when tested, had shattered the windows of nearby houses. It would be capable of delivering up to 118,000 kg into low-Earth orbit or up to 41,000 kg into lunar orbit. Early in 1963, the C-5 received a new name: Saturn V.

When a mockup of the rocket was rolled out to Pad 39A at Cape Kennedy on 25 May 1966, it amply demonstrated its colossal proportions. It stood 110.6 m tall and

10 m wide, only a few centimetres shorter than St Paul’s Cathedral in London. It comprised an S-IC first stage, an S-II second stage and was topped by the S-IVB which would be restarted in space to boost the Apollo spacecraft towards the Moon on a so-called ‘translunar injection’ (TLI) burn. All three stages used liquid oxygen as an oxidiser. Fuel for the first stage was the RP-1 form of refined kerosene, while the S-II and S-IVB utilised liquid hydrogen. Eighty-nine truckloads of liquid oxygen and 28 of liquid hydrogen, together with 27 railcars filled with RP-1, were needed to fuel the Saturn V.

The S-IC first stage, built by Boeing, was 42 m tall and its five F-1 engines, arranged in a cross pattern, produced over 3.4 million kg of thrust to lift the Saturn to an altitude of 61 km. The four ‘outboard’ engines could be gimballed for steering during flight, whilst the centre one was fixed. The S-II, built by North American, was

Spectacular panoramic view of the Cape Kennedy landscape as ‘Moon-fever’ gripped NASA in mid-1966. Clearly visible are a Saturn Y test vehicle, the gigantic Vehicle Assembly Building (VAB) and the Launch Control Center (LCC).

25 m tall and would make history as the largest cryogenic-fuelled rocket stage ever built. Finally, the Douglas Aircraft Company’s 17.85 m-tall S-IVB would be used to place the Apollo spacecraft into Earth orbit, then restart a couple of hours later for a six-and-a-half-minute-long TLI burn. It also provided a ‘garage’ to house the lunar module.

The Apollo 4 spacecraft was an old Block 1 with many features of the upgraded Block 2 design, including an improved heat shield and the new unified hatch. The aim of its mission was to evaluate its structural integrity, its compatibility with the Saturn V and its ability to enter an elliptical orbit and re-enter the atmosphere to land in the Pacific. The mission ran perfectly: the Saturn V boosted the spacecraft into a 185 km parking orbit and, after two circuits of the globe, for the first time, its S-IVB third stage restarted to propel Apollo 4 to an apogee of more than 17,000 km. Next, the service module’s SPS engine ignited, sending the spacecraft out to 18,000 km for a four-and-a-half-hour-long ‘soak’ in the little-known radiation and temperature environment of deep space. In doing so, Apollo 4 dipped its toe into the conditions that astronauts would one day experience as they traversed the 370,000 km translunar gulf.

Finally, with the command module’s nose pointed Earthward, the SPS fired a second time to bring it home. The service module separated and the command module hit the upper atmosphere, just as it would on a lunar return, at 40,000 km/h. Nine hours after its launch, Apollo 4 hit the waves of the Pacific, near Hawaii, just 16 km from the primary recovery ship Bennington. As successful as the mission had been, a long road remained before an actual lunar landing could be accomplished. Certainly, an additional uncrewed flight was highly desirable to many within NASA, providing further confirmatory data that the enormous rocket was capable of delivering men safely to the Moon. One crucial vehicle which still needed an ‘all-up’ performance test was Grumman’s lunar module, the first flight-ready version of which – designated ‘LM-1’ – was delivered to Cape Kennedy, three months late, at the end of June 1967.

By a strange twist, Apollo 5, which would consist solely of the lunar module, with no command and service module aboard, was assigned the Saturn 1B originally meant to carry Gus Grissom’s crew into orbit. In the immediate aftermath of the fire, it had been destacked from Pad 34, checked for corrosion or damage and finally restacked on Pad 37 on 12 April 1967. With the lunar module installed in its nose, the 55 m rocket looked unusual, ‘stubby’ even, since it lacked the command and service modules and an escape tower. The LM-1, encased in the final stage of the Saturn, had an incomplete environmental control system and was not fitted with landing gear, since it was destined to burn up during re-entry into the atmosphere.

Loading propellants aboard the rocket proved troublesome, mainly due to procedural difficulties and minor irritations such as clogged filters and ground support equipment glitches, but a simulated launch demonstration ended success­fully on 19 January 1968. Three days later, at 5:48 pm, Apollo 5 set off and was inserted perfectly into orbit. Forty-five minutes into the flight, LM-1’s attitude control thrusters pushed it away from the S-IVB and a lengthy checkout of its systems began. Two orbits later, its TRW-built descent engine – the world’s first-ever

The legless Apollo 5 lunar module is prepared for flight.

throttleable rocket, capable of slowing it down for landing on the Moon – was fired for 38 seconds, but was ended abruptly by the lunar module’s guidance system when it sensed the vehicle had not accelerated fast enough. In response to the cutoff, flight controllers moved to an alternate plan: firing the descent engine on two further occasions, then igniting the ascent engine. With all primary tests done, LM-1 re­entered the atmosphere to destruction and its remains plunged into the Pacific, several hundred kilometres south-west of Guam, on 12 February. So successful, in fact, was Apollo 5 that a further unmanned test of the lunar module was considered unnecessary. Its next flight, atop the Saturn V, would be carried out with a crew aboard.

However, the lander still had many problems of its own. The instability of its Bell – built ascent engine, in particular, caused concern throughout 1967 and for much of 1968. Although both George Mueller and Sam Phillips felt that Bell had a good chance of solving the engine’s fuel-injector problems, the agency nevertheless hired Rocketdyne to develop an alternate device. Despite difficulties in both cases,

Rocketdyne was ultimately chosen to outfit the lunar module’s fuel injector. Other problems with the bug-like lander included windows blown out and fractured during high-temperature tests, broken wiring and stress corrosion cracks in its aluminium structural members; the latter led to the formation of a team to identify the cause and implement corrective actions. Grumman analysed more than 1,400 components and heavier alloys were employed for newer sections of the lunar module. Weight, too, posed an issue. In 1965, more than 1,100 kg had been shaved from the lunar module and NASA even offered incentives to Grumman to remove yet more unwanted bulk. The LM-1 flight had been good enough for NASA to cancel an unmanned LM-2 test, but LM-3 – the first mission to fly manned – would not be ready until at least the end of 1968.

Meanwhile, the performance of the Saturn V on the Apollo 4 mission fired up hopes that it could soon be entrusted with a human crew. Nonetheless, another test flight, that of Apollo 6, was still required … and rightly so, for the rocket’s second mission, AS-502, almost ended in a disaster. On 13 March 1967, the S-fC first stage arrived at Cape Kennedy and, inside the cavernous interior of the Vehicle Assembly Building, was mated to its S-ff second stage in May. By February of the following year, topped by the S-fVB third stage and the Apollo 6 command and service module, it was rolled into wind-driven rain towards its destination: Pad 39A, today revered as one of the most famous and historic launch platforms in the world. Despite communications difficulties, which forced a two-hour halt, the stack arrived at the pad at 6:00 pm.

Aside from being a second unmanned test of the Saturn V, the Apollo 6 mission would put Spacecraft 020 through its paces on the final flight of the command and service modules before a human crew headed aloft on Apollo 7. Originally scheduled for launch in the first quarter of 1968, the flight was postponed several times. First, the tank ‘skirt’ on another service module split during structural tests, prompting an inspection and restrengthening of Apollo 6 to prevent a similar problem. Next, after rollout to the pad, water seepage was detected in the Saturn V’s S-ff second stage and some parts had to be replaced. Eventually, at 7:00 am on 4 April, the rocket thundered into the heavens, seemingly with perfection. . . and then, things began to go wrong.

Throughout the first two minutes of its climb, the five F-1 engines burned steadily and normally, then experienced thrust fluctuations which caused the entire rocket to oscillate longitudinally like a pogo stick for around 30 seconds. Low-frequency modulations were recorded in the Apollo 6 command module, exceeding design criteria, but otherwise the first stage completed its work. However, the time soon came for the S-ff second stage to exhibit problems: two of its five J-2 engines suddenly stopped, four minutes into a six-minute firing, requiring the others to burn for 59 seconds longer than planned to compensate for the abrupt power loss. The rocket did not tumble and explode, however, because the failed J-2s were adjacent to one another and the Saturn survived by gimballing its remaining ‘good’ engines. Still, the second stage did not achieve its desired velocity and ended up at a higher altitude than it should before its fuel was exhausted.

This meant that the S-fVB had to burn for correspondingly longer. ft ‘‘was

confusing to the computer guiding the S-IVB,” wrote Deke Slayton, “which realised it was higher than it should be… and slower. So while it added 29 seconds to the burn, it actually pointed itself down toward the centre of the Earth.” At length, after a difficult ascent in which the S-IVB pitched itself back upwards and entered orbit firing backwards, Apollo 6 was inserted into a wild 178-367 km elliptical orbit, instead of a 160 km circular path. The Saturn’s troubles, though, were still not over. An attempt to restart the S-IVB – just as it would be required to do in order to boost Apollo crews toward the Moon – failed when the third stage refused to ignite. “If this had been a manned flight,’’ wrote Deke Slayton, “the escape tower on the Apollo would have been commanded to fire, pulling the spacecraft away from the Saturn for a parachute landing in the Atlantic.’’

An ‘alternate’ mission was now inevitable and the command and service module were duly separated from the S-IVB and the SPS engine burned for seven minutes, simulating a TLI manoeuvre and pushing the apogee of Apollo 6’s looping elliptical orbit to 22,200 km. This gave it enough altitude to mimic a lunar-type return, but not enough velocity, and it splashed down in the Pacific, missing its impact point by 80 km. Ten hours after launch, the command module was hauled aboard the amphibious assault ship Okinawa. Despite a NASA press release which declared that preliminary data indicated the spacecraft had done its job well, many felt that, overall, the mission had not been a success. The Saturn V might need a third unmanned test before it could be flown with astronauts aboard.

In fact, pogo effects had been observed, to a lesser extent, during the Apollo 4 launch and its apparent cause was traced to a partial vacuum created in the fuel and oxidiser suction lines by the rocket engines. The condition, wrote Brooks, Grimwood and Swenson, produced a hydraulic resonance; in effect, the engine ‘skipped’ when bubbles caused by the partial vacuum reached the firing chamber. Engineers later determined that two of the Saturn V’s engines had been inadvertently tuned to the same frequency, which probably made the problem worse. In future, all clustered engines were tuned to different frequencies to prevent any two or more of them from pulling the rocket off-balance and changing its trajectory.

As part of efforts to rectify the issue, Rocketdyne began retesting the F-1 engine in late May, injecting helium into the liquid oxygen feed lines to interrupt the resonating frequencies which had caused the unacceptable vibration levels. In four of the six tests, the ‘cure’ proved worse than the ‘disease’, by making the oscillations more pronounced. Attempts at NASA’s Marshall Space Flight Center in Huntsville, Alabama, used the same technique, but produced quite different results; no oscillations were observed. Elsewhere, the cause of the J-2 failures proved more of a mystery. During tests, engineers discovered that frost forming on propellant lines when the engines fired at ground temperatures served as an extra protection against the fuel lines rupturing. However, frosting did not take place in the vacuum of space, pointing at a possible cause of the failure. The chances of American bootprints on the Moon before the end of 1969, it seemed, was still very much touch-and-go.