Challenger STS-51-L: The Seals, the Cold and 73 Seconds

All seven crew members died when the space shuttle Challenger broke apart 73 seconds after liftoff at 11:38 on 28 January 1986. Among them was Christa McAuliffe, a New Hampshire teacher chosen to be the first teacher in space. The mission, 51-L, was Challenger's tenth flight. The warning had been written down six months earlier by a Thiokol engineer, Roger Boisjoly, who told his managers that a failure of the booster seals would be a "catastrophe of the highest order, loss of human life."

The crew and the mission

The crew was Francis R. Scobee, Michael J. Smith, Judith A. Resnik, Ellison S. Onizuka, Ronald E. McNair, Gregory B. Jarvis and Christa McAuliffe. Four had already flown in space. Scobee, the commander, had piloted Challenger in April 1984 on a mission to capture and repair a satellite. Smith, the pilot, had been selected by NASA in 1980 and was still waiting for his first flight. Resnik had flown on Discovery's first voyage in 1984, and Onizuka had flown a classified military mission in January 1985. McNair had played his saxophone in orbit in 1984 and planned to record a solo for the French composer Jean-Michel Jarre.

Jarvis was a Hughes Aircraft satellite engineer whose seat had twice been given to someone else, first to a senator and then to a congressman. Challenger was his third assignment. More than 11,000 teachers had applied for the teacher-in-space role that McAuliffe won. The crew was to release a satellite and study Halley's Comet, and McAuliffe planned to teach two lessons from orbit.

The launch delays

The launch, planned for 22 January, slipped repeatedly. On 27 January the crew was strapped in, but a problem with a hatch handle and strong winds at the emergency landing runway ended the attempt. That night the temperature began to fall.

The night before

NASA's launch managers expected clear skies and temperatures in the low 20s Fahrenheit for about 11 hours. They discussed water drains, safety showers and heaters, but no one raised the booster seals.

Each booster was built in four segments. Each joint was sealed by two rubber rings, barely a quarter of an inch thick, which had to spring back after the joint flexed at ignition. Cold rubber is stiff and slow to spring back. Every flight with seals estimated at 63° or colder had come back with signs of heat damage. In January 1985, Discovery flew with seals estimated at 53°, and hot gas got past a primary ring for the first time.

At about a quarter to nine that evening, Thiokol's vice president of engineering, Bob Lund, recommended against launching unless the seals were at least 53 degrees. NASA's Lawrence Mulloy challenged the data and the new temperature rule, saying that with it "they might not launch until April." George Hardy said he was appalled but would not launch against the contractor's advice. A requested five-minute break lasted half an hour. Thiokol's senior manager, Jerry Mason, asked Lund to take off his engineering hat and put on his management hat. Four Thiokol managers made the final recommendation, and the engineers were not asked.

At about 11:00 Thiokol came back on the line, called the data inconclusive and recommended launch. No one on the line disagreed, and a signed recommendation was faxed to Florida just before midnight. Allan McDonald, Thiokol's representative at the Cape, kept arguing for delay, but none of it reached the officials at the top of the launch decision.

Morning of the launch

Overnight water left trickling in the pad's lines froze, and at half past 1:00 an ice team found icicles across more than 100 feet of the tower. Their handheld instrument later read about 25° on the surface of the left booster and about 8° near the base of the right. There was no rule for booster temperatures, and the readings were not reported. Rockwell said it could not assure that it was safe to fly, but NASA judged that the ice would fall clear and held the launch to let it melt. By late morning the air at the pad was 36°, still 15° colder than any previous shuttle launch. Parts of the booster joints were later calculated to have been as cold as 28°.

The flight