Challenger Launch Explosion


Photographs from the first seconds of the take-off of space shuttle Challenger on January 28th, 1986 show grey smoke coming from the right Solid Rocket Booster, and a stream of vapourised material indicating that the joint was not completely sealed. The area from which the smoke came faced towards the External Tank. The puffs of smoke were rhythmical (indicating some flexing in the joint) and increasingly black, indicating that the source was likely to be a combination of grease and the insulation and O-rings in the joint seal, and the cause the hot propellant gasses.

From 37 to 64 seconds after launch, wind shear was encountered, forcing course corrections and more activity on the part of the steering system than in any previous flight.

About 59 seconds after lift-off the first signs of flame were seen from the area of the seal, almost simultaneously with a difference in pressure between the two boosters, providing more evidence of a leak.

The flame played onto the External Tank and supporting structures. At about 65 seconds after launch the flame changed colour as it ignited hydrogen leaking from the breached external tank, and the pressure loss from the External Tank backed up the theory that the tank had been breached.

72 seconds into the launch the supporting structures in the path of the flame severed, allowing the right booster to twist.

At about 73 seconds into the launch, massive amounts of hydrogen were released from the External Tank, producing a sudden forward thrust pushing the hydrogen tank into the intertank structure, and at roughly the same time the right booster impacted upon the same structures and the liquid oxygen tank.

A massive explosion resulted almost immediately, destroying the Challenger and killing all seven onboard.

The Commissions findings blamed the failure on the failure of the seals in the joint between the lower segments of the right Solid Rocket Booster

The temperature at launch was 36 Farenheit, 15 lower than any previous launch, and the area of the joint that failed was facing away from the sun decreasing the temperature further. The O-rings were known to be more responsive to decompression at higher temperatures, returning to their original shape faster. Application of motor pressure to actuate the O-ring and seal was ideally meant to be applied to ensure that the O-ring was properly sealing the joint, but this may not have occurred in time.

Despite its frequent occurrence, neither Thiokol nor NASA developed a solution to the O-ring erosion and blow-by, and the dependency between the occurrences and low temperatures was not noticed.

The commission concluded if those making the decision were aware of the O-ring problems (four of 21 launches at 61 Farenheit or above showed signs of O-ring thermal distress) it would have been unlikely that they would have approved the launch. One example of a waived requirement of the operational envelope that the launch should not proceed when below 53 degrees Farenheit, but this waiver was not required to be approved at all levels of management, and pressure had brought about management decisions that ignored engineering advice. No safety representative was present for various decisions, including that to proceed with the launch.

Early design of the joint seal did not expect that exhaust gasses would contact the seal, but early testing showed burns. The resultant action was to lift the acceptable level of damage to cover the level occurring.

Fix?

An incident investigation may have noticed the link between low temperatures and launch problems.

Fix?

A greater level of independence was required in the decisions relating to the launch.