Challenger: the O-Ring Chart and the Launch-Eve Teleconference

Challenger lifted off on 28 January 1986 at 11:38 Eastern, with the air temperature near the pad at 36 degrees, 15 degrees colder than any launch before it. The flight lasted 73 seconds, and all seven crew members died. The night before, the engineers who built the booster had recommended against launching in the cold. About half an hour later, the answer changed.

A joint nobody expected to be touched

A shuttle's solid rocket booster is built in sections. Where two sections meet, the joint is sealed by two rubber rings, one backing up the other to keep hot gas from leaking through. The Rogers Commission wrote that neither the contractor nor NASA expected the rubber O-rings to be reached by hot motor gas, much less partially burned.

Twenty-four shuttle flights had gone before Challenger, all successful. At liftoff, puffs of smoke came from the right booster's aft field joint between about two-thirds of a second and two and a half seconds after ignition, roughly three puffs per second. In the commission's own tests, with a tight gap like the one on these joints, sealing without gas blow-by did not occur consistently until the temperature was raised to 55 degrees. On launch day, the commission found, a ring at that cold temperature would be very slow to return to shape, so it was probable it would not seal the gap in time.

At about 58 seconds the leak was clearly visible as a flame at the joint. At about 73 seconds the vehicle broke up. The commission concluded the cause was failure of the pressure seal in the aft field joint of the right booster, due to a faulty design, unacceptably sensitive to temperature, and several other factors.

The teleconference

The night before launch, the contractor's engineers in Utah and NASA at Kennedy and Marshall held a teleconference about the cold. Its second phase began at about 8:45 p.m. Eastern. The contractor's engineering team presented the history of O-ring erosion and blow-by and their test results. Their conclusion was a chart: do not launch with O-ring temperatures below 53 degrees, the lowest of any previous flight.

The NASA side was not pleased. One manager was reported to be appalled; another testified he may have said words like, "When do you want me to launch? Next April?" The engineers were asked to quantify their concern. One said he had no data to quantify it, but that it was away from goodness in the current database.

According to an engineer's testimony, a NASA manager at Marshall said he would not launch against the contractor's recommendation, and a contractor manager, asked if he would launch, said no because the engineering recommendation was not to.

The caucus and the hat

The contractor's managers asked for a five-minute offline caucus. It lasted about 30 minutes, and the call recessed at about 10:30 p.m. In the room, two engineers kept objecting. One sketched the joint on a pad in front of the managers and stopped when he realized he was not getting through. The other tried once more with photographs and also stopped.

A senior manager asked the head of engineering to take off his engineering hat and put on his management hat. The new reasoning was that the primary ring could erode to three times the previous worst case, and even if it did not seal, the secondary was in position and would. The head of engineering told the commission: "We had to prove to them that we weren't ready, and we couldn't prove absolutely that the motor wouldn't work." The commission's view was that NASA appeared to be requiring a contractor to prove it was not safe to launch, rather than proving it was safe.

The people who approved the launch never heard most of this. The commission found they were unaware of the recent history of O-ring problems and of the contractor's first written recommendation against launching below 53 degrees. The contractor's certification of flight readiness, signed on 9 January, never mentioned the O-ring problems; the review on the 15th identified no problems with the booster; and the final mission management meeting on the 25th reported no technical issues. The commission found that ice on the pad was not a cause, and it did not conclude that NASA's decision specifically overrode a no-launch recommendation from the contractor, whose own management had by then reversed its position.

A problem rooted in history