Columbia STS-107: How a Foam Strike Led to the Shuttle’s Loss

Seven astronauts died when Space Shuttle Columbia broke apart on its return to Earth on 1 February 2003. The loss traces back to a piece of insulation foam, about 6 in by 10 in by 16 in, that came off the external tank 81.9 seconds after liftoff on 16 January and struck the left wing.

A mission years in the making

STS-107 was a 16-day scientific mission to study the effects of microgravity on living organisms. It had been planned for 2001 and was postponed 13 times. Rick Husband, 45, commanded it on his second spaceflight. Before launch he called the parents of each crew member. Laurel Clark, a flight surgeon, had been selected in 1996 after an earlier application in 1994. Willie McCool was the pilot.

Columbia was the first orbiter, flown first on 12 April 1981, and STS-107 was its 28th mission. Its thermal protection system used 24,300 silica tiles under the belly, rated for 1,000°, and reinforced carbon on the nose cap and leading edges, rated for 1,600°. The documentary calls this system the shuttle's major weakness.

The foam strike

The orange foam on the external tank kept the fuel cold. Pieces tended to rip off under the aerodynamic forces of takeoff. Of the 112 flights before STS-107, 80% had some foam shedding, and tile damage from it was easily repaired. The reinforced carbon on the leading edges was typically very resistant to strikes.

On 17 January, engineers reviewing the launch tapes saw something break off the tank and end up as dust after hitting the shuttle. The foam came from the bipod ramp and struck the port wing with a relative velocity of about 500 mph, which one engineer compared to a jackhammer. Only two of the 12 launchpad cameras captured the falling foam, so engineers could not identify the exact spot of impact. Early analyses suggested it hit the tiles near the landing gear.

Requests and assessments

On day three an army radar detected something that had fallen off the shuttle, but the engineers did not receive that data at the time. On 21 January, technicians in Houston warned their boss, who emailed his superiors that without a better photo the damage could not be determined. He asked permission to use a military satellite. NASA reacted with skepticism, since the bipod ramp foam had come off about 10% of the time and was seen as low risk.

The engineers used a software tool called Crater to assess tile damage. It was designed for debris several millimetres wide, 400 times smaller than the foam that hit Columbia, so the documentary calls it an inappropriate application of existing software. On 24 January a meeting concluded that there was no safety-of-flight issue and that it was essentially a maintenance issue. Everyone during the mission thought the likely impact was on the tiles, not on the reinforced carbon panel, where the foam in fact hit (panel 8).

On 23 January the crew was told by email that the foam strike was inconsequential. Jonathan Clark, Laurel Clark's husband and a flight surgeon at mission control, heard colleagues' worries. He decided to follow the rules and not talk to her about them.

Re-entry

On 1 February the de-orbit burn went well. Columbia entered the atmosphere at 8:44. At 8:54, flight controllers lost four temperature transducers on the left side. Hot plasma was coming through a hole of about 25 cm in the wing's leading edge, and the aluminum inside melts at 600°. Soon after came the loss of tire pressure readings on the left side. At 8:59 the capcom's call about those messages was cut off. The crew tried to restart the auxiliary power units, and the left wing then failed structurally and ripped off. Columbia disintegrated at 60 km altitude, three seconds past 9:00.

Debris fell over Texas and was spread across 300 km. About 30,000 people took part in the search, which cost $300 million. Six months later, 82,000 pieces, 40% of the shuttle, had been recovered.

Could the crew have been saved?

Some proposed launching Atlantis to meet Columbia and moving the crew across by tether. The chances were very low, no two shuttles had ever been in space together, and a failed rescue could have cost 14 lives instead of seven. Shuttle flights were suspended for 18 months, and the program ended on 8 July 2011.