Skip to content

Challenger

Lessons From the Challenger Tragedy, Retro Report on PBS, downloaded from PBS LearningMedia. Asset copyright © 2019 Retro Report, Inc. All rights reserved. Producer and editor: Bret Sigler. Used under the PBS LearningMedia license (educational, non-commercial, no derivatives).

What happened

On the morning of 28 January 1986, Space Shuttle Challenger lifted off from Kennedy Space Center on mission STS 51-L at 11:38 a.m. Eastern time, carrying seven crew members including Christa McAuliffe, the first participant in NASA's Teacher in Space program. It was the coldest launch in the program's history: the overnight temperature had fallen well below freezing, ice coated the launch tower, and the ambient temperature at liftoff was about 36 °F, some fifteen degrees colder than any previous Shuttle launch.

Less than a second after ignition, launch-pad cameras recorded puffs of black smoke escaping from the aft field joint of the right-hand Solid Rocket Booster, the joint between two of the steel casing segments that hold the solid propellant. The smoke stopped after a few seconds as burned residue temporarily plugged the gap. At 37 seconds the vehicle flew into the strongest wind shear any Shuttle had experienced, and at 58 seconds a flame appeared at the same spot on the joint. Within two seconds it had grown into a blowtorch playing on the external tank. At 72 seconds the booster's lower attachment strut failed and the booster pivoted into the tank; at 73 seconds the tank's hydrogen and oxygen tanks ruptured, and the vehicle broke apart at about 48,000 feet under aerodynamic loads it was never designed to survive. All seven crew members were killed.

The night before the launch, engineers at Morton Thiokol, the booster's manufacturer, had recommended against launching. Their data showed that the rubber O-rings sealing the field joints stiffened in the cold and had allowed hot gas past them on earlier cold-weather flights, and they argued that no Shuttle should fly with a seal temperature below 53 °F, the coldest they had flown before. Managers at NASA's Marshall Space Flight Center pushed back hard; one asked whether Thiokol wanted him to wait "until next April" to launch. After a recess in which Thiokol's senior vice president told the engineering manager to "take off his engineering hat and put on his management hat," Thiokol management reversed the recommendation, and the concern was never passed up to the officials who made the final launch decision.

The engineering failure

  • A joint that did not work the way it was designed to. The field joint relied on two O-rings seated in a tang-and-clevis groove. Under the pressure of ignition the joint rotated open slightly, widening the gap the O-rings had to fill. A warm O-ring could expand quickly enough to follow the gap; a cold one could not. Both the Rogers Commission and the House Committee on Science and Technology concluded that the design was faulty, that neither NASA nor Thiokol fully understood how the joint behaved, and that the test and certification programs were inadequate.

  • Warnings that were normalized. O-ring erosion and "blow-by" had been observed on flight after flight since 1981 and had been briefed to every level of Shuttle management, including NASA Headquarters. Because the damage had never yet led to disaster, each new occurrence was treated as acceptable, and flights continued under waivers while a leisurely 27-month redesign program went forward. The House Committee found that the "fundamental problem was poor technical decision-making over a period of several years by top NASA and contractor personnel, who failed to act decisively to solve the increasingly serious anomalies in the Solid Rocket Booster joints."

  • Schedule pressure. NASA was trying to reach a launch rate of 24 flights a year and to present the Shuttle as an operational, routine system. The House Committee found that this drive "created pressure throughout the agency that directly contributed to unsafe launch operations," and NASA's own 1988 lessons-learned study concluded that "managers were pressuring the work force to break management rules in an attempt to maintain flight schedules."

  • A launch decision made on the wrong question. The Thiokol engineers were asked to prove it was unsafe to launch rather than to show it was safe, an inversion of every previous flight-readiness practice. NASA's lessons-learned report puts it plainly: questions "were constituted in the form which elicited answers fitting a why we should not launch rather than why we should launch philosophy."

  • A safety organization that had withered. Safety, reliability, and quality-assurance staffing had been cut as the Shuttle was declared operational, safety reviews had come to rely on paperwork rather than physical checks, and the problem-reporting system never carried the joint anomaly above the third tier of management. The Rogers Commission called the safety program "silent."

  • Reliability estimates that were fantasy. Working engineers put the odds of losing a vehicle at roughly one in a hundred; NASA management quoted figures as low as one in a hundred thousand. Commission member Richard Feynman found that management "exaggerates the reliability of its product, to the point of fantasy," and that the reasoning used to accept eroded seals amounted to "a kind of Russian roulette."

Lessons

The two investigations disagreed about emphasis, the Rogers Commission stressing flawed communication and the House Committee stressing flawed technical judgment, but they agreed on what had to change. Safety oversight must be independent of the programs it oversees and staffed for the life of the program, not wound down when a system is declared "operational." Go/no-go decisions must rest on data and analysis rather than schedule convenience, with the burden of proof on those who want to proceed. Anomalies are evidence that a system is not behaving as designed, and the fact that they have not yet caused a catastrophe is no evidence that they will not. Rules exist to be followed, and when they must be waived, everyone with a stake in the outcome needs to know. And a flight schedule that cannot be met safely is not a schedule; it is a hazard.

Feynman and the Rogers Commission

Richard Feynman, the Caltech physicist and Nobel laureate, served on the Presidential Commission chaired by former Secretary of State William Rogers. During a televised hearing he famously dropped a piece of O-ring material clamped in a small C-clamp into a glass of ice water and showed that it had lost its resilience, cutting through weeks of testimony in a few seconds. He then went beyond the O-rings to investigate the main engines and avionics, and wrote his own account of what he found. Chairman Rogers, worried the criticism would damage public confidence in NASA, asked him to tone it down; Feynman refused and threatened to take his name off the report. The compromise was to publish his observations, unaltered, as Appendix F of the Commission's report.

The CNN coverage below, from June 1986, reports on that dispute as the Commission delivered its report to President Reagan, and includes a live interview in which Feynman explains why he thinks of NASA's attitude to its repeated warnings as that of a child who keeps running into the road because "nothing happened" the last time.

Feynman's appendix is short, plainly written, and worth reading in full. It ends with a sentence that has become the epigraph of every engineering-ethics course since: "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled."

Read Appendix F: Personal Observations on the Reliability of the Shuttle

Sources

Local copies of the two government reports are in this directory, and the Rogers Commission's main report and Feynman's appendix are reproduced as pages of this site.

  • U.S. House of Representatives, Committee on Science and Technology, Investigation of the Challenger Accident, House Report 99-1016, 99th Congress, 2nd Session, 29 October 1986. PDF
  • NASA Office of Safety, Reliability, Maintainability and Quality Assurance, Lessons Learned From Challenger, February 1988. PDF
  • Report of the Presidential Commission on the Space Shuttle Challenger Accident (the Rogers Commission), 6 June 1986. Full text of Volume I (local copy); Appendix F, Feynman's personal observations (local copy); NASA History Office online edition with all five volumes.