Figure 1: The damaged Apollo 13 service module, Odyssey, seen from the lunar module, Aquarius. Credit: NASA.

THE "SUCCESSFUL FAILURE" OF APOLLO 13, HALF A CENTURY LATER

The Apollo 13 mission was the fifth manned flight to the Moon and the third lunar landing attempt. After the first two successful test flights of Apollo 11 and Apollo 12 to the surface of the Moon, the Apollo 13 mission was more scientific oriented, with activities focused on geological and space physics experiments on the lunar surface, in addition to rock and soil samples retrieval. Moreover, mission objectives included the photography of candidate exploration sites, as well as to develop capabilities to work in the lunar environment. The landing site was selected to be in the Fra Mauro formation, near the Fra Mauro crater, for it was believed to contain impact material generated in the early history of the Moon.

For a number of reasons, in the popular culture the number 13 has been considered to be closely associated with luck; in some countries it represents bad luck, whereas in ohers, is a good omen. Superstition does not have any room in science and technology, so the number 13 for that particular Apollo flight just meant that it was the mission after Apollo 12. However, some people felt a bit uneasy with the designated mission number. Half a century after the historic flight of Apollo 13 everyone knows what actually happened. The phrase "Houston, we have a problem" became popular in modern culture, and the outcome of the mission has finally been understood as the consequence of a number of incidents, some of them even unnoticed, each one playing a part in the chain of events that ended up with three astronauts struggling to come back home alive. The happy ending of the mission was the result of the hard work of many people on the ground and the three astrounauts in their crippled spacecraft (Fig. 1), and not luck. Or wasn't it?

Apollo 13 lifted-off from lauch pad 39-A at the Kennedy Space Center in Florida on April 11th, 1970, at 14:13:00 EST, and its accident happened on April 13th. You can start counting the number of 13s in the previous sentence. From what it was publicly known at that time, we can say that the mission oddly started even many months before the launch. Orginally, the prime crew, composed by Jim Lovell (CDR), Ken Mattingly (CMP) and Fred Haise (LMP), was assigned to the Apollo 14 mission. The original Apollo 13 flight had the crew composed by Alan Shepard (first US astronaut, and CDR), Stu Roosa (CMP) and Ed Mitchell (Fig. 2). Shepard had been grounded for several years due to Ménière's deseas, and came back to flight status in 1968. Due to lack of proficiency and to allow him to get better prepared for the mission, in 1969 NASA decided to switch crews with that of Apollo 14. After months of training and only a week before the launch, the LMP of the backup crew, Charlie Duke, contracted rubella. Due to close proximity during training, all astronauts, from both prime (Fig. 3) and backup crews, were tested for rubella immunity. One of them, Ken Mattingly (Fig. 3), had never been exposed to the desease, therefore, he was not immune. Normally, under these circumstances, all three astronauts in the primary crew should have been replaced by those in the backup crew. However, in this particular case, Duke was part of the backup team which obligated to replace Mattingly by his backup counterpart, Jack Swigert (Fig. 4), only two days before the start of the mission, otherwise all astronauts would have been bumped to a later flight. This change in a crew with a such a short notice had never been done before.

Figure 2: Original prime crew of the Apollo 13 mission. From left to right: Alan B. Shepard, Jr. (CDR), Stuart A. Roosa (CMP), and Edgar D. Mitchell (LMP). To allow for a better preparation, considering the lack of proficiency of the commander due to his several years being grounded, NASA decided to move this crew to the next Apollo mission, thus switching crews. Credit: NASA

Figure 3: The new prime crew of the Apollo 13 mission. From left to right: James A. Lovell, Jr. (CDR), Thomas K. Mattingly II (CMP), and Fred W. Haise, Jr. (LMP). They had originally been selected for Apollo 14, but were reasigned to Apollo 13 to allow for more training time for Alan Shepard and his crew. Credit: NASA

Figure 4: The final crew of Apollo 13 that made the flight. From left to right: James Lovell, John L. Swigert Jr. (CMP), and Fred Haise. Jack Swigert was a last-minute replacement, taken from the back-up crew, to Ken Mattingly who had been exposed to rubella, contracted by back-up crew LMP Charles M. Duke Jr., and did not have immunity through prior exposure as the rest of the prime crew. Credit: NASA

The first two minutes of the mission were completely normal, without a glitch. However, during the second stage boost, the center engine of the S-II stage cut off about two minutes early. The engine shut down was instructed by the vehicle's guidance system due to a severe pogo oscillation detected. As a result, the remaining four S-II engines burned 34 second longer, and the S-IVB orbital incertion burn time was about 44 seconds longer as well to compensate. Orbital incertion and trans-lunar injection were successfully completed and the mission from there on proceded normaly, except for an anomaly with the reading of the quantity of liquid oxygen in tank No. 2 of the electrical power system (EPS) in the Service Module (SM) of the spacecraft (Fig. 5). At about 55 hours 55 minutes of ground elapsed time (GET; 10:08 PM EST) the crew reports a main bus B undervoltage condition following a large bang associated with it. The pressure in oxygene tank No. 2 was lost rapidly, and the current generated by fuel cells 1 and 3 droped to zero due to the loss of their oxygen supply. After some moments, a decision was made to abort the mission.

Postflight analyses revealed that the cause of the catastrophic failure of the SM was set up more than a year before the mission, without being noticed nor suspected. Was it indeed bad luck? A series of unfortunate events led to the unsuccessful lunar landing of Apollo 13, with the possible dead of the astronauts and the likely cancellation of the Apollo program. All these events relate to oxygen tank No. 2 of the EPS (Fig. 5). The first major one of them happened in 1968 during preparations for the Apollo 10 mission. The EPS is composed by three fuel cells, two liquid hydrogen tanks and two liquid oxygen tanks. Both oxygen tanks are mounted on a shelf that is installed inside Sector 4 of the SM. The oxygen assembly that was used for the Apollo 13 mission had been first installed in SM 106 for the flight of Apollo 10.

Due to electromagnetic interference problems with the vac-ion pumps on cryogenic tank domes in earlier Apollo spacecrafts, a modification to solve that was introduced. A decision was made to replace the complete oxygen shelf in SM 106 by one that already had that modification. In the deinstalling process, due to a bolt mistakenly left in place, the whole oxygen shelf was dropped about two inches back into its original position. This incident caused undue acceleration to the tanks, and photographs of the underside of the fuel cell shelf indicated that the closeout cap on the dome of oxygen tank No. 2 (Fig. 6) may have struck the underside of that shelf. After this, the oxygen shelf was removed without difficulty, visually inspected, and checked for self integrity. No problems were detectd, although a loosely fitting fill tube could have been displaced with the above incident. The tanks were modified to prevent the interference problems. The shelf passed the tests and was reinstalled in SM 109 (Apollo 13) in November 1968. The countdown demonstration test began in March 1970. During this test, difficulties to empty -or detanking- oxygen tank No. 2 were encountered. In the following days, detanking operations with tank No. 2 turned out to be difficult. It was considered as the source of the problem a possible leak in the path between the fill line (Fig. 6) and the quantity probe due to a loose fit. Perhaps the dropping of the oxygen shelf in 1968 contributed to the loose fit and the leak in the tank's fill line.

The detanking problems continued, and in order to empty tank No. 2 a decision was made to try to "boil off" the remaining oxygen by using the tank's heaters (Fig. 6). The tank got empty after about 8 hours of heater operation. And here there is a major issue that was unintendedly ignored. The heaters were energized with the 65 V DC power supply available on the ground. However, the original design of the tank's heating system considered operation with the 28 V DC spacecraft's power supply. Each heater was protected with a thermostatic switch (Fig. 6) which was intented to open the heater circuit when temperatures reached the 80 F. Test conducted after the mission showed that such switch would become defective and unable to open the circuit when working at the 65 V DC power. This mulfunction, unknown at the time, may have caused the temperature on the heater tube to rise to as much as 1,000 F during the "special" detanking procedure. This high temperature caused serius damage to the adjacent Teflon insulation, making wires inside the tank to become exposed. While NASA had ordered the manufacturer of the tanks to modify the switches to properly work at 65 V, this modification was never performed. More filling tests were conducted, and the detaking of tank No. 2 continued to be problematic. The solution again was to use the heaters to remove the excess oxygen. While extensive consideration was given to all possibilities of damage from a loose fill tube, very little attention was paid to the extended and repetitive operation of the heaters and fans (Fig. 6) during detanking sequences. The latter was simply not considered as potentially dangerous at the time. Since the filling of the oxygen tank No. 2 always proved to be successful, the leak of the fill line would not be a problem in flight. It was finally decided to leave the oxygen shelf and tank No. 2 in the SM and to proceed with preparations for launch of Apollo 13.

Figure 5: Diagram of the Apollo Spacecraft Service Module showing the location and configuration of the electrical power system (EPS) in Sector 4. EPS components include the 3 fuel cells, the 2 liquid hydrogen tanks, and the 2 liquid oxygen tanks. Oxygen tank No. 1 is located in the inner region of the sector, while oxygen tank No. 2 is located in the region further out. Credit: NASA

Figure 6: Diagram of the structure and different components of the oxygen tanks of the EPS. The heater and fan are energized simultanously, and the latter is used to stir-up the cryogenic oxygen in the tank to prevent its stratification and improve quantity level readings. The thermostat's function is that of preventing the temperature from rising to undesirable levels when using the heater. Credit: NASA

Now we encounter another element of luck, which may have been, at the end of the day, what saved the astronauts. Because of the cryogenic temperatures in the oxygen tanks, liquid oxygen had to be stirred to prevent stratification. The stratification of the liquid oxygen would provide an inaccurate reading of the quantity of oxygen remaining in the tank. This was the purpose of the fans inside the tanks (Fig. 6) which would operate at the same time of the heaters (Fig. 6). In previous Apollo missions the fans were turned on every 24 hours. During the first 46 hours of the mission the performance of oxygen tank No. 2 was normal. However, at about 46 hours 40 minutes GET, after a routine stir of the oxygen tanks, the quantity inside tank No. 2 went from the normal reading of about 82% full to an incorrect reading of over 100%. This issue may have been caused by a short circuit in the electrical wiring of the quantity gage. To better analyze the situation, Mission Control chose to perform two more stirings of tank No. 2, one at about 47 hours 55 minutes GET and another one at about 51 hours 08 minutes GET which showed off-scale high readings with no apparent adverse effects. A decision had been made to increase the frequency of the heater-cryo-stirs, going from the nomal once every 24 hours to once every 6 hours to help the reading of the gage. After veryfication procedures of the Lunar Module were completed, and before the next rest period of the crew, Mission Control requested a fifth stir of both oxygen tanks No. 1 and No. 2, following an indication of low pressure in tank No. 1, at about 55 hours 53 minuted GET. This last heater-cryo-stir was immediately followed by an explosion of tank No. 2 and the main bus B undervoltage condition reported by the crew. The rest is history.

The cause of the explosion was due to exposed wires inside tank No. 2 (Fig. 5 and Fig. 6) that, upon activation of the fan-heater system, created an arc that ignited the oxygen inside the tank. The explosion at least partially destroyed tank No. 2 and damaged tank No. 1 (Fig. 5) causing a leak in it. The excess pressure created inside Sector 4 of the SM due to leakage and additional burning caused by the loss of integrity of tank No. 2 blew off the corresponding outside panel of the SM (Fig. 1). Those exposed wires were the result of the extended and repetitive operation of the heaters on the ground with thermostats functioning at a higher voltage than that they had been manufactured for. And here is when the apparent bad luck of Apollo 13 may have been indeed a good one! Oxygen tank No. 2 explosion occured with the fifth cryo-stir. This means that, due to the unknown damage to the fan-heater system's wires, the explosion was going to happen anyways at the fifth use of that system. At a rate of one cryo-stir approximately every 24 hours, had not the quantity gage in tank No. 2 malfunctioned, the complete failure of the tank would have occured at about 120 hours GET. This means that the explosion in the SM would have occurred after the first EVA on the Moon, with Lovell and Haise on the lunar surface before a second EVA. They would have received an urgent call from either Mission Control or Jack Swigert in the CSM informing about the accident. Swigert's assessment of the situation and problem would have been very difficult without the help of his mission companions in the orbiting spacecraft. Moreover, the periodic interruption of telecommunications with Mission Control every time the CSM went behing the Moon would have added even more problems. Once the real magnitude and nature of the failure had become aparent, the EPS would no longer be producing oxygen, water and electrical power inside the CSM. The stay on the lunar surface would have been aborted, but with a crippled spacecraft and without the descent stage of the Lunar Module available as main propulsion system, all three astronauts would have died stranded in lunar orbit.

Now, let's consider another element of luck. The tank that exploded was oxygen tank No. 2, located further away from the inside of the SM Sector (Fig. 5). This meant that most of the energy of the explosion was released toward the outside of the spacecraft, thus blowing off the external panel (Fig. 1). Had tank No. 1 (Fig. 5) been the one being hit during removal from SM 106, the one with detanking issues (that required untested procedures), the one with the problem in the quantity probe, and ultimately that one that exploded, in that case most of the energy would have been released inside the SM, probably causing also tank No. 2 to explode and hence increasing the power of the explosion. Likely, the whole SM would have exploded perhaps with the Command Module as well, terminating with the mission and the life of the astronauts. Even more, if we consider the timing of the explosion itself, this may have been another element of luck, and a good one. Had the explosion occurred earlier than it did, before the fifth activation of the heater-fan system, something quite possible with exposed electrical wires powered in an environment full of pressurized oxygen, the very diminished supplies available in the spacecraft would have been totaly consumed before having completed the return trajectory back to Earth and around the Moon. The return trip would have been too long and the crew would have perished on the way back due to lack of oxygen. Now, one would think, why not immediately returning back to Earth, why having to go around the Moon? Although the possibility of an abort and immediate return to Earth was considered in every mission, this manoeuvre, specially that close to Earth, required the fully operational propulsion system of the CSM. Because of the explosion, the safest assumption was to consider that propulsion system as inoperative due to a potential major damage. Igniting the main engine would have been too risky. The lower power of the descent engine of the Lunar Module was insufficient to successfully accomplish such a drastic manoeuvre. Therefore, the only viable option was that of going around the Moon, a trip just long enough to keep the astronauts alive.

So in the end, the luck of Apollo 13 may have been a good one after all. In fact, the quantity probe malfunction, something that had nothing to do with electrical wires without insulation and which was not expected at all, may have led to the "successful failure" of Apollo 13. This problem prompted Mission Control to instruct the use of the heaters and fans in the oxygen tanks more often than normal which caused the explosion just at the right time in GET that enough residual power, water, electricity and oxygen would be available in the lunar lander to reach the Earth with the crew alive. Had not this malfunction presented itself, then the Apollo 13's story and that of the rest of the Apollo program may have been very different. That's luck!, don't you think?

References:

"Management Lessons of the Moon Program" by Andrew Chaikin, Goddard Space Flight Center Knowledge Management Workshop (July 2012)
Apollo 13, from Wikipedia, and references therein
Report of Apollo 13 Review Board, NASA (June 1970) (PDF version)
"50 Years Ago: Apollo 13 Passes Countdown Test" by Kelli Mars (Editor), NASA History (March 2020)
"13 Things that Saved Apollo 13" by Nancy Atkinson, Universe Today (April 2010)
"Why didn't Apollo 13 just turn around when the oxygen tank exploded?" by C. Stuart Hardwick, Quora (February 2019)

-April, 2020. (Last revised: July 2020)