On the 1st February 2003, space shuttle Columbia fell from the sky. It was NASA's worst disaster for 17 years. Should flight STS107 have launched at all?

by Stuart Carter and Nigel Henbest

Today’s shuttle system is outmoded and costly. It is also potentially lethal. Space Shuttle Columbia disintegrated into thousands of pieces as it fell to Earth, killing all seven crew. Was this a freak accident or a disaster waiting to happen?

We need to look beyond the immediate technical failings of flight STS 107 and uncover the truth about compromises made in the shuttle design over 3 decades ago. In the early 80s, the space shuttle system was widely hailed as the ultimate access to space. In fact, its design is flawed – and that may have led to two disasters. The design is an awkward muddle of civilian and military requirements, further confused by politicians and budgets. It’s a compromise that dates back to the height of the cold war.

In the 1960s, America won the space race by getting to the Moon before the Soviet Union. But it wasn’t cheap. With every flight, NASA had to build a giant, disposable Saturn 5 rocket. Hundreds of millions of dollars of equipment were thrown away every time a mission was launched. NASA’s dream was to build a reusable launch vehicle, which would make space flight cheap and commonplace. NASA’s original plan was for a small, lightweight vehicle that could be flown into space on top of a fully reusable, piloted aircraft. The shuttle would peel off into orbit as the winged fuel tank flew back down to earth. NASA had wanted to build the ultimate dream space vehicle – but what they got was a jumble of tanks and rockets with an aircraft (the Orbiter) stuck on the side.

To escape the Earth’s gravity, each Orbiter has to carry over half a million gallons of fuel in an enormous External Tank. Even though it’s burning one and half tons of hydrogen and oxygen every second, the Orbiter is still too heavy to blast off from the ground. For extra lift, two powerful rockets are strapped to the side – the Solid Rocket Boosters. Lift off is the most dangerous time to be on a rocket. With so much explosive energy around things can go violently wrong. And solid rockets add to the potential danger: unlike liquid-fuelled rockets, once you’ve lit the blue touch paper, they can’t be switched off.

STS107-S-002 (October 2001) - The seven STS-107 crew members take a break from their training regimen to pose for the traditional crew portrait. Seated in front are astronauts Rick D. Husband (left), mission commander; Kalpana Chawla, mission specialist; and William C. McCool, pilot. Standing are (from the left) astronauts David M. Brown, Laurel B. Clark, and Michael P. Anderson, all mission specialists; and Ilan Ramon, payload specialist representing the Israeli Space Agency. We remember them.

When Columbia blasted space-wards on the very first shuttle mission in 1981, NASA continued the tradition of safety first - the two-man crew had the benefit of ejector seats built into the cockpit. But by the fifth shuttle launch, ejector seats were removed, for good. For the first time in American space history, crews were being blasted into space with no way of escaping from a launch emergency.

Space Shuttle Columbia STS-1 lifts off from Complex 39A of the Kennedy Space Center during the first launch of the space shuttle in 1981

When NASA had first proposed their plans for a space shuttle, they were going to need big money to get it off the ground. This was the era of Richard Nixon. He appeared to be a space enthusiast but the reality was rather different. The only way NASA would get their shuttle was if they co-operated with the Department of Defense. The CIA and the US Air Force were pushing for a large payload capacity, big enough to house their 60 foot spy satellites. But a large payload bay meant a heavy vehicle - and a heavy vehicle requires more power and fuel for lift off. NASA was also under pressure from Nixon’s Office of Management and Budget. This began a decreasing spiral of short-sighted cost-cutting measures. One proposal suggested doing away with the reusable launch vehicle. Instead, the system would use a disposable liquid fuel tank and 2 solid rocket boosters.

Initially NASA dismissed the idea out of hand - it might be cheaper in the first instance, but it would add to the cost of each successive flight. But when the Office of Management and Budget heard of this design, they insisted NASA accept it. Still sceptical, NASA researched the failure rate of solid fuelled rockets and identified one potential weakness: hot gas burning through the rubber O-ring seals at the joins between segments of the solid rocket boosters. Their fix was to use two sets of O-rings for safety. But still, it wasn’t enough

Challenger STS-51L Explosion - Hurtling out of the conflagration at 78 seconds are the Challenger's left wing, main engines (still burning residual propellant) and the forward fuselage (crew cabin).

On a frosty January day in 1986, as Challenger was about to launch, the O-rings were frozen stiff with the cold. Two minutes after lift-off, at 48,000 feet, hot gases leaking from the faulty seal burnt through the connecting struts. In an instant, the fuel tank exploded. With Challenger, the flawed design of the Space Shuttle system had claimed its first seven victims.

NASA’s initial fears had come true. Since the Challenger disaster, every shuttle astronaut flies equipped with a pressure suit and a parachute so that they can bail out in an emergency.

But NASA’s problems with the shuttle continued. In May 1995 the launch of Discovery had to be abandoned when woodpeckers pecked 200 holes into the foam coating on the fuel tank. To scare them off, NASA employed a foghorn - and six life-size owls, bought from the local Wal-Mart. And when Atlantis launched in October 2002, eyewitnesses reported insulating foam falling from the fuel tank during lift off. It severely dented the skirt on one of the Solid Rocket Boosters. But NASA reckoned the foam was not a danger and there was no real attempt to stop it happening again.

When Columbia launched in January 2003 and foam again fell from the tank, NASA engineers examined film footage of the falling debris hours later. There seemed little cause for concern. Crucially, the original shuttle design would have avoided this problem altogether - the Orbiter would have had much smaller wings and sat in front of any falling debris. The exact reason why Columbia disintegrated is still a mystery, but damage to its wing during launch remains at the heart of the investigation.

And any damage to the Orbiter could prove fatal to the crew during the critical few minutes of re-entry. Again, the compromise design of the shuttle system laid its crew open to danger. Originally, NASA’s ideal space craft was a small orbiter with stubby wings - similar to the X15 rocket plane of the 1960s. But, starved of funds by Nixon’s administration, NASA had to share its dreams with the military. And they had their own ideas. The Pentagon’s dream machine was delta-winged, capable of gliding long distances from its main orbital path. But this design presented NASA with a new challenge. The old style capsules only had to withstand the inferno of re-entry for 4 minutes. But the shuttle, with its delta wings, glides back to earth at a much slower, shallower angle, so its ‘hot’ period lasts more than 12 minutes. And the underside area exposed to re-entry is far greater than their original smaller space-plane.

The X-15

NASA was faced with developing a totally new heat shield - something both light and re-usable. They opted for a jigsaw of ceramic tiles. Right from the inception, the tiles were the shuttle’s greatest weakness.

NASA blasted the first shuttle Columbia into space in 1981. But in orbit, the crew reported the loss of several tiles. Luckily they weren’t in a crucial area. But the loss of any tiles exposes the airframe to the dangerous hot gases of re-entry. On the 16th January 2003, the crew - under the command of Rick Husband - prepared to launch on Columbia’s 28th mission. For four of the astronauts, this would be their first flight into space. As Columbia blasted into orbit, NASA’s cameras tracked the launch. They recorded foam insulation hitting the leading edge of the left wing.

This came as no surprise to independent consultant Paul Fischbeck. He had already published a report showing that in 27 previous launches falling debris had caused damage to thousands of tiles. But the debris that hit Columbia this time was bigger than anything Fischbeck had seen before.

After 17 days in orbit, the crew prepared Columbia for re-entry. On the 1st February at 7.50am the orbiter punched through the upper atmosphere. Travelling 25 times faster than the speed of sound, the plasma shock-wave began heating the underside of the orbiter. With the temperature’s reaching over 1600 degrees Celsius, the only thing between the crew and the extreme heat of re-entry was the fragile silica tiles. As Columbia roared through the atmosphere, mission control monitored the re-entry. Data from the shuttle began to show abnormal temperature rises inside the left wing.

In less than 5 minutes it climbed by more than 60 degrees. The internal structure of the Shuttle’s wings are made up of hundreds of aluminium cross members wired with thousands of heat sensors. These struts bear the load of the wing – giving it rigidity for the high stresses of launch and re-entry. If the leading edge had been damaged during launch, hot gases might be able to blast into the wing cavity.

The Insignia of the STS-107 Crew

We now know that 5 pieces of the Orbiter broke off over California. As the shuttle began to break up its wreckage spread over thousands of square miles. The small town of Nacogdoches in Texas was directly in the line of fall out. Reports began to come in that pieces of debris were literally falling out of the sky. Some of them clearly segments of the heat shield. A massive recovery operation got under-way. NASA wanted answers. The mystery as to why the shuttle broke-up could lie in any one of these pieces of wreckage.

What we do know is that data reveals that in the last few seconds the pilots may have been fighting to switch off the autopilot and get the shuttle under manual control. So was this horrendous accident avoidable? Could NASA have investigated the possible damage to the wing more thoroughly before re-entry? And how can they prevent future tragedies?

Even before the Columbia accident, NASA was starting to look for a future shuttle replacement. Ironically, some of the design concepts are looking more like their original vision from the early 1970s - a smaller stubbier winged Orbiter, mounted above the launch system. Whatever the final design, this tragedy must lead to a safe and reliable launch system that matches the bravery of the humans who willingly face the dangers of space.