Wednesday, July 8, 2009, Glenn Research Center – Cleveland, OH
Following the loss of the Space Shuttle Columbia in February of 2003, NASA appointed the Columbia Accident Investigation Board to carry out an investigation on the accident and its causes. Dr. Matt Melis, a ballistic researcher at NASA’s Glenn Research Center, was part of that committee and carried out the fundamental testing in the investigation. His talk centered on the story of this tragic event and how his work was used to determine why the shuttle was lost.
To preface his work, Dr. Melis began with a brief overview of the shuttle system and its various components. The shuttle is comprised of the external tank, the orbiter, and two solid rocket boosters. All together these components are called the stack.
The two white rockets that flank the stack are the solid rocket boosters. The boosters provide approximately 3.3 million pounds of thrust and burn for the first two minutes of flight. They are connected to the external tank through two aft connections.
The external tank holds the hydrogen and oxygen fuel for the shuttle engines. It is sprayed with a one-inch thick foam to keep the cryogenic propellants cold. Structurally, it is made of an ortho-grid to hold up to the stress of space travel. The tank holds the first forty minutes of fuel for the flight and is the only non-reusable part of the shuttle system. Lockheed Martin manufactures the tank at the Michoud Assembly Facility in New Orleans.
The final component of the shuttle is the orbiter, a seven-passenger space place with its own (albeit less powerful) solid rocket boosters. The rocket boosters are ten times less power per engine than the solid rocket booster. They burn 750 and 250 gallons per second of H2 and O2. These engines ignite six seconds before launch because they take longer to stabilize. These engines can be shut down while in use unlike the solid rocket boosters. The orbiter has one fuel supply to stabilize and circularize the orbit. There is nothing frivolous on the vehicle; each pound of payload requires four pounds of fuel. NASA has constructed five different orbiter crafts: Atlantis, Endeavour, Challenger, Columbia, and Discovery, but lost Challenger in 1986 and Columbia in 2003.
After the briefing on the shuttle system, Dr. Melis began his discussion on the Columbia accident. The accident was caused by a piece of foam impacting the leading edge of left wing of the orbiter. The foam broke off the external fuel tank at mach 2.46 near an altitude of 65,800 feet. The event was captured on film, but NASA decided to continue the mission. A team at Kennedy Space Center sent the footage and other data to various NASA centers for analysis. The researchers concluded the shuttle was still fine for re-entry, citing the fact the piece of RX 250 foam from the external fuel tank that impacted the wing was much softer than the reinforced carbon-carbon tiles that make up the wing itself.
Ultimately, the NASA researches reviewing the footage did not understand the problem and the orbiter was destroyed upon re-entry. The foam had broken off of the strut holding the shuttle to the external fuel tank. Thermal stresses in the strut caused by the connection between the cool tank and the warm shuttle led the foam to fracture off the tank.
In the end it was determined this strut had been designed incorrectly, an ironic conclusion because the piece had functioned properly for twenty-five years before it failed. Dr. Melis was very poignant in reiterating that this design was a huge problem, even thought it had done its job for a significant portion of the shuttles’ lives.
After the Columbia catastrophe, NASA assembled a team of 5,000 researchers to comb the debris field shoulder to search and recover as much of the shuttle as they could. The field was so large that the raining material could be seen on weather radar. The process was long and tedious as the team combed parts of Texas, Louisiana, and Arkansas for nearly three months. Each piece of material was tagged and taken to a hanger at Kennedy Space Center for analysis. The team was eventually able to recover and piece together nearly 38 percent of the shuttle.
Upon investigation of the pieces of Columbia, the disaster team discovered a white hazy substance on the inside of the left wing leading edge panel. This substance was determined to be slag, a metal that vaporizes and re-condenses. It was determined a crack must have formed in the wing, allowing the reinforced carbon-carbon of the orbiter wing to heat up and melt. This lead to Dr. Melis’s work on the project of utilizing the resources of the Glenn Ballistics Impact Laboratory to determine what could have cracked this leading edge.
The Glenn Ballistics Lab looked at how foam and reinforced carbon-carbon reacted on impact. The lab first ran a number of small-scale tests before conducting a full-scale test of the impact. The team also worked on validating their results with computer models, a process that took five years of research time. The full-scale test involved the construction of a mock-up orbiter wing and a large pneumatic cannon, capable of blasting a piece of the BX-250 external tank foam at Mach 2.46.
Dr. Melis said the engineers and technicians gasped when they watched the foam fire. The foam glanced the wing leaving a hole the size of a pizza box on the lading edge of the orbiter. It was determined from these tests definitively that the foam was capable of seriously damaging the orbiter body and surely left a crack in the wing when it fell off the external tank.
It would not be until July 26, 2005 that NASA would return men to space after the Columbia disaster. From Dr. Melis’ testing, NASA developed new in-flight procedures to determine if the shuttle was safe for re-entry including a full 360 degree flip under the international space station and use of the Canada Arm to inspect the craft for any damage. The Columbia accident was a tragic price to pay for such a small design flaw, but the lessons learned will help ensure safer space travel and spacecraft design in the future of NASA.