A/V Space History – The Apollo 13 Accident

posted in: Project Apollo | 2

Also known as figuring things out by dropping stuff down a big tube (and filming it)

When we begin to create a collection for a new DVD set, the first thing we do is sit down and list what we consider to be “must haves” for that set. Some are easy to determine (and find), and some items are added to the list as production proceeds.

For the Apollo 13 set, I knew about one set of footage that wasn’t widely distributed, but simply had to be included on the set – testing that was done at Lewis Research Center and Langley Research Center that lends dramatic support to the findings of the accident review board.

The Accident

To understand the footage, let’s review what happened to the Apollo 13 oxygen tank (#2).

Design of Apollo 13 oxygen tank. NASA.
Design of Apollo 13 oxygen tank. NASA.

The original 1962 specification given to Beech Aircraft (manufacturers of the service module oxygen tanks) specified the use of 28 Vdc for the tank and heater assembly, which is the voltage used in the spacecraft. In 1965, these specifications were revised, and the heaters were to be changed for compatibility with the 65 Vdc used in ground support equipment. Beech did not change their parts specifications to meet the 65 Vdc requirement, and it was not caught by other oversight.

Oxygen tank number 2 began manufacture in 1966, this was the eighth block II tank built. Several minor flaws were discovered during manufacture. The upper fan motor was noisy and drew excessive current. The tank was disassembled and the heater assembly, fans and heaters were replaced with new parts. The tank was reassembled and sealed, and the space between the inner and outer shells was pumped down over a 28-day period to create the necessary vacuum.

The tank was leak tested at 500 psi and proof tested at 1335 psi with helium. The tank was then tested with liquid oxygen and no difficulties were recorded in the test or in the detanking after completion. However, the acceptance test indicated that the rate of heat leak into the tank was higher than permitted by specifications. The tank was reworked, but never met these specs. The tank was accepted with a waiver of this condition. The tank was accepted and shipped on May 3, 1967.

Arrangement of components in SM bay 4. NASA.
Arrangement of components in SM bay 4. NASA.

The tank was assembled on an oxygen shelf with another tank and installed in service module 106 on June 4, 1967,  for flight in the Apollo 10 spacecraft.

Due to electromagnetic interference problems with pumps on cryogenic tank domes in earlier Apollo spacecraft, a modification was introduced and a decision was made to replace the complete oxygen shelf in SM 106. On October 21, 1968, the shelf was removed.

After the lines and bolts were disconnected, a fixture suspended from a crane was placed under the shelf and used to lift the shelf and extract it from bay 4. One shelf bolt was mistakenly left in place and, after the front of the shelf was raised about 2 inches, the fixture broke allowing the shelf to drop back into place. Indications noted that the closeout cap on the dome of oxygen tank #2 may have struck the underside of the shelf during this incident.

The remaining bolt was removed, the incident was reported, and the shelf was taken out. Ultimately, the review board found that the possibility of tank damage due to this incident was rather low. It is possible, however, that a loosely fitting fill tube could have been displaced by this event.

The shelf passed testing and was installed in SM 109 on November 27, 1968. In June of 1969 the spacecraft was shipped to Kennedy Space Center.

Up to the time of the countdown demonstration test for Apollo 13, nothing unusual had occurred with oxygen tank #2 in testing. During the CDDT on March 16, 1970, during the time when the oxygen tanks are normally partially emptied to about 50 percent of capacity, oxygen tank #1 behaved normally, but #2 only went down to 92 percent of capacity. It was decided to proceed with the CDDT and look at the detanking problem in detail after the test.

On Friday, March 27, 1970, detanking was resumed. Tank number 2 was about 83 percent full, and was vented through its fill line, bringing it down to 65 percent. Discussions between KSC, MSC, North American and Beech concluded the problem might be due to a leak in the path between the fill line and quantity probe due to loose fit in the sleeves and tube. At this point a discrepancy report against the spacecraft was written.

A normal detanking procedure was then conducted on both tanks, #1 emptied normally, tank #2 did not. A decision was made to “boil off” the remaining oxygen by using the tank heaters. The heaters were energized with the 65 Vdc ground support power supply and after 6 hours the quantity was down to 35 percent. The fans and heaters were finally turned off after about 8 hours of operation.

At this point a decision was made that if the tank could be filled properly, no other problems would be experienced in flight, since the problem with the tank seemed to be the detanking process only. The tank was tested on March 30, twelve days prior to the launch. Replacing the shelf would have been difficult and would have taken at least 45 hours.

As the launch date approached, the detanking problems were considered by the Apollo organization. At this point the “shelf drop” incident was not considered, and the discussions focused on the loose fill tube and not the effect on the tank from the 8 hours of heater operation. In fact, many of those in the discussions were not aware of the extended heater operations.

Later investigation showed that the thermostatic switch, mounted on the heater tube, which is intended to open upon reaching a temperature of 80 degrees F, failed to open when powered from the 65 Vdc ground supply. The switches used were rated for the 28 Vdc spacecraft power supply, and could not open properly at 65 Vdc. Tests showed the temperature on the heater tube may have reached as high as 1000 degrees F during the detanking, which can cause serious damage to the teflon insulation. Such damage almost certainly occurred. This was not known at the time.

The decision was made to leave the oxygen tank in place and proceed with the launch of Apollo 13.

At 55 minutes and 53 seconds into the flight, oxygen tank #2’s fans were turned on. Here’s what happened:

The evidence points strongly to an electrical short circuit with arcing as the initiating event. About 2.7 seconds after the fans were turned on, a current spike and voltage drop were recorded in the spacecraft electrical system. A pressure rise began in the tank 13 seconds later. Such a time lag is possible with low-level combustion at that time. Tests showed that the energy from the arc is more than sufficient to ignite teflon of the type contained within the tank.

The pressure rise lasted more than 69 seconds, consistent with teflon igniting in the tank (as opposed to other materials). The relief valve was fully open at 1008 psi, thereafter the tank lost integrity. Telemetry from the spacecraft was lost for a period of 1.8 seconds. The panel of bay 4 was ejected from the spacecraft at approximately the time of the communication loss.

The release of the oxygen pressurized the shelf space of bay 4 in the service module. If the hole in the tank were large enough, the escaping oxygen would have been sufficient to blow off the panel. However, it is also possible that the escape of oxygen was accompanied by combustion of mylar and kapton which would augment the pressure. The ejected panel struck the service module high-gain antenna, which accounted for the loss in communication.

The explosion also caused a slow leak in the tank 1 system. Odyssey would very soon be out of oxygen and power.

The shock of the explosion closed the oxygen feed valves to fuel cells number 1 and 3. After trying to stem the loss of oxygen from tank 1, an effort that failed, the LM became a lifeboat.

The Footage

Lewis Research Center Zero-G Research Facility. NASA.
Lewis Research Center Zero-G Research Facility. NASA.

Lewis concentrated on analyzing what happened inside the tank, primarily teflon flame propagation. To examine the problem properly, the test required a zero-g environment. These tests were conducted at Lewis Research Center’s 5-second zero gravity facility.

The Zero Gravity Research Facility was designated a national historic landmark in 1985. It consists of a concrete lined shaft, 28 feet in diameter, that extends 510 feet below ground level. An aluminum vacuum chamber, 20 feet in diameter and 470 feet high is contained within the concrete shaft. It is the largest microgravity facility in the world.

Combustion chamber for oxygen tank tests. NASA.
Combustion chamber for oxygen tank tests. NASA.

A combustion chamber was designed and constructed with a sapphire window to permit high speed photography. The footage shows the results of the tests, at 400 frames per second. During this test the chamber is falling within the zero-g research facility, in a vacuum chamber.

At Langley, tests focused on panel separation. A one-half scale dynamic model was constructed, and panels were attached through replica-scaled joints to a test fixture that simulated SM geometry and volume. A high-pressure gas system was used to rapidly build up pressure behind the cover panel. The tests were conducted both in atmosphere and in a vacuum.

Complete separation of the panels in vacuum was demonstrated. Separation was achieved by rapid pressurization of the oxygen shelf space.

One of the more chilling of the findings from these tests was that the pressure in the oxygen shelf space, required to blow off the panel, was about 25 psi. If the panel had not blown off, and the center tunnel of the service module had pressurized, only 10 psi was required to immediately sever the tension ties securing the CM to the SM. In short, the command module could have been immediately separated from the service module as a result of the explosion, which almost certainly would have led to a loss of the crew.


Parting note: It took about 6 months to track down this footage, as it wasn’t routinely available in the normal NASA channels. In fact, it was finally located, after noticing a notation on some accession papers, in a particular box at the National Archives in College Park, MD. The accession had not yet been processed into the center, but because the box number was available we were able to have it pulled and make a transfer of the footage. Such is the way these things work. I’m pleased we were able to supply it as part of our Apollo 13 collection. 

This description used material from the Apollo 13 Review Board Report.

2 Responses

  1. Stephen M. Zumbo
    | Reply

    Thank you for this description and video. I heard Gene Cernan say that in another documentary video that we came closer to losing the crew of Apollo 13 than most people realize, and that bringing them back safely was perhaps NASA’s greatest success, even greater than the first lunar landing. I’ve always felt bad that Jim Lovell lost his chance to walk on the Moon, but I’m glad he lived to speak and write about his experiences.

  2. Raymond Ramirez
    | Reply

    I was a college student when the flight of Apollo 13 happened, and I followed the news. I even bought a copy of the pictures that Air & Space Magazine published of the stricken Service Module. Later I saw the movie “Apollo 13” and I see many errors, starting with the paint scheme of the Saturn V. I hope that the Spacecraft Films documentary of the real Apollo 13 drama will reach out to clear the misinformation that the movie has left.

Leave a Reply