Apollo 6

• 1. Introduction

  ›

  Apollo 6 launched from Kennedy Space Center on the 4th of April 1968, at precisely 7:00:01 in the morning Eastern time. That same day, Martin Luther King Jr. was assassinated in Memphis, Tennessee. The nation's attention was pulled so completely elsewhere that one of the most consequential rocket tests in history went almost unnoticed by the press.

  The Saturn V that rose from Launch Complex 39A that morning was only the second of its kind to fly. Its job was to prove the machine could carry human beings to the Moon. What followed instead was a cascade of engine failures that, by any ordinary measure, should have grounded the program. The question the rest of this documentary will answer is how a mission that its own director would later call "a failure" still cleared the Saturn V for crewed flight.

• 2. A Rocket Built in Pieces

  ›

  The S-IC first stage of the Apollo 6 launch vehicle arrived at Kennedy Space Center by barge on the 13th of March 1967, and was standing upright in the Vehicle Assembly Building just four days later. The S-IVB third stage and the Instrument Unit computer arrived on the same day: March 17.

  The S-II second stage was not ready, so engineers substituted a dumbbell-shaped spacer that matched the S-II in height and mass and carried all its electrical connections. When the S-II finally arrived on the 24th of May, it was stacked and mated into the rocket on July 7.

  Apollo 6 was also the first mission to use High Bay 3 of the Vehicle Assembly Building, and the new facility immediately showed a problem: its air conditioning was not up to the job. Portable high-capacity cooling units had to be brought in for equipment and workers alike.

  Progress was slow throughout 1967, partly because so much personnel and equipment was tied up preparing the Saturn V intended for Apollo 4, the very first Saturn V launch. After Apollo 4 flew on the 9th of November 1967, work on Apollo 6 accelerated. The command and service module was mounted atop the launch vehicle on the 11th of December 1967, and the entire stack rolled out to Launch Complex 39A on the 6th of February 1968. The rollout took all day, was conducted largely in heavy rain, and was interrupted for two hours when communications failed. The vehicle did not reach the pad until after dark.

  Launch was first set for the 28th of March 1968. Problems with guidance equipment and fueling pushed it to April 1, then April 3, and finally to April 4. Every remaining issue was resolved during built-in countdown holds, so the mission launched without any further delay.

• 3. What the Mission Was Supposed to Do

  ›

  Apollo 6 carried a command and service module designated CSM-020, paired with a Lunar Test Article called LTA-2R. LTA-2R was a simulated lunar module, built not to fly but to be shaken. Its ascent stage was ballasted aluminum studded with sensors to record vibration, acoustics, and structural loads; its descent stage tanks held a water-glycol mixture and freon rather than real propellant.

  The plan called for the Saturn V's third stage, the S-IVB, to push the spacecraft onto a trans-lunar trajectory. That trajectory would carry it past the Moon's orbit without encountering the Moon. Once the burn was complete, the service module's main engine would slow the craft, dropping its apogee to 11,989 nautical miles and pulling it back toward Earth. On the return leg, the engine would fire again to boost reentry speed to 36,500 feet per second at a reentry angle of negative 6.5 degrees, simulating the conditions an Apollo crew would face coming home from the Moon. The entire mission was designed to last about ten hours.

  CSM-020 arrived at Kennedy Space Center from North American Aviation with only 23 mostly routine problems, a sharp contrast with the Apollo 1 command module, which had arrived with hundreds of unresolved issues. Kenneth S. Kleinknecht, the Command and Service Module manager at the Manned Spaceflight Center in Houston, was pleased with the spacecraft's condition, though he objected to it arriving wrapped in flammable mylar. One of the modifications to CSM-020 was a new crew hatch, designed to replace the one condemned by the Apollo 1 investigation board as too difficult to open in an emergency. That flaw had contributed to the deaths of three astronauts in the Apollo 1 fire of the 27th of January 1967.

• 4. The Pogo Problem

  ›

  For the first two minutes after liftoff, the Saturn V performed normally. Then, as the S-IC first stage continued to burn, pogo oscillations began shaking the vehicle. The thrust variations produced a g-force of plus or minus 0.6 g on the structure. The Saturn V had been designed for a maximum of 0.25 g.

  NASA Associate Administrator for Manned Space Flight George Mueller explained the mechanics to the House Committee on Government Operations. Combustion inside any engine is never perfectly uniform, he said, producing thrust fluctuations that travel through the fuel lines like sound through an organ pipe. The pipe has its own resonant frequency, the vehicle structure responds like a tuning fork, and under the right conditions the whole system begins to oscillate longitudinally.

  The Saturn V had been "detuned" before launch, meaning NASA believed it had been adjusted to avoid vibrating at its natural frequencies. The pogo episode on Apollo 6 showed that belief was wrong. Nevertheless, the vehicle survived with only one piece of damage: a panel of the Spacecraft-Lunar Module Adapter broke loose during the oscillations.

  After the mission, roughly 1,000 government and industry engineers worked on solving the pogo problem. The solution involved filling cavities in the valves leading to the F-1 and J-2 engines with helium gas just before takeoff, using it as a shock absorber to damp pressure oscillations before they could build into a resonance cycle.

• 5. Engine Failures in Sequence

  ›

  After the S-IC first stage separated, the S-II second stage ignited its five J-2 engines, and engine number two began showing performance problems at 225 seconds after liftoff. At T+319 seconds the problem worsened sharply. At T+412 seconds the Instrument Unit shut the engine down entirely.

  Two seconds later, engine number three also shut down, even though it had been running normally. A wiring error was the cause. The shutdown command sent to engine two was cross-connected to engine three as well. The Instrument Unit compensated: the three surviving engines burned for 58 seconds longer than planned. The S-IVB third stage also fired for 29 seconds longer than planned, though it experienced a slight performance loss.

  The root cause traced to a component called the augmented spark igniter, a miniature rocket engine mounted at the top center of each J-2. Its job was to produce a flame large enough to ignite the full fuel and oxidizer flow when the engine started. The ASI ran on cryogenic propellants fed through lines with metal bellows sections, included to accommodate thermal expansion.

  On the ground, frost and liquid air forming on those cold lines would dampen vibrations. In vacuum, there was no such protection. On engine two, the liquid hydrogen line to the ASI failed. The liquid oxygen line kept feeding, and the resulting excess oxygen in a hydrogen-rich chamber produced extreme heat, burning through the engine dome. Pressure dropped, the Instrument Unit sent a shutdown signal, and the cross-wiring took engine three down with it.

  A similar failure in the ASI propellant lines of the S-IVB's single J-2 engine almost certainly prevented it from restarting to send the spacecraft on its trans-lunar path. For all future missions, the metal bellows were replaced with rigid bends and the lines were strengthened.

• 6. Improvising a New Mission

  ›

  With the S-IVB unable to restart, flight director Clifford E. Charlesworth and his team in Mission Control chose a pre-planned alternate profile. The service module's Service Propulsion System engine fired for 442 seconds, lifting the spacecraft to an apogee of 11,989 nautical miles with a low enough perigee to guarantee reentry, replicating what Apollo 4 had done.

  The improvisation had a cost. There was not enough propellant left for the second SPS burn that was supposed to accelerate the spacecraft to a speed mimicking a lunar return. The command module entered the atmosphere at 33,000 feet per second instead of the planned 37,000 feet per second. This meant the mission fell short of one of its primary objectives: fully simulating the thermal and deceleration loads of a real Moon-return reentry.

  While the spacecraft was at high altitudes, the command module returned data on how well the spacecraft's skin shielded future astronauts from the Van Allen Belts. Ten hours after launch, the command module landed 43 nautical miles from the planned touchdown point, in the North Pacific Ocean north of Hawaii, and was recovered.

  The S-IVB's orbit gradually decayed, and it reentered the atmosphere on the 26th of April 1968. The service module was jettisoned just before the command module reached the atmosphere and burned up on reentry.

• 7. Enough to Fly

  ›

  At a post-launch press conference, Apollo Program Director Samuel C. Phillips acknowledged "there's no question that it's less than a perfect mission," but called the launch vehicle's ability to reach orbit after losing two engines "a major unplanned accomplishment." George Mueller praised the mission as "a good job all around" and "a successful mission" in which NASA had "learned a great deal," though he later stated flatly that Apollo 6 "will have to be defined as a failure."

  NASA's rapid analysis and corrective response satisfied the Senate Committee on Aeronautical and Space Sciences, which reported in late April that the agency had diagnosed the abnormalities and taken corrective action. Engineers at the Marshall Space Flight Center in Alabama concluded that a third uncrewed Saturn V test flight was unnecessary. The next Saturn V would carry people.

  There was one remaining debate within NASA after the mission: whether to configure the spacecraft's emergency detection system to abort automatically if pogo exceeded certain limits. Director of Flight Crew Operations Deke Slayton opposed automatic abort, and work shifted instead toward a "pogo abort sensor" that would let the crew decide. By August 1968, it was clear that the pogo fix itself made such a sensor redundant, and work on it was abandoned.

  Command module CM-020 was transferred to the Smithsonian Institution after the mission and is now on display at the Fernbank Science Center in Atlanta, Georgia, where visitors can see the hatch that replaced the one that cost three astronauts their lives on the 27th of January 1967.