Exclusive: Did rough wing break Columbia?

By DEE ANN DIVIS, Science and Technology Editor

WASHINGTON, Feb. 17 (UPI) -- Increased drag -- possibly caused by roughness on Columbia's left wing identified years ago and worsened by time and debris -- may have been enough to pull the shuttle into a fatal sideways flight angle on re-entry, experts have told United Press International.

"It could very well be" that the combined drag from these different sources was enough to cause the demise of Columbia, said John Anderson, a leading aerodynamics expert.


The damage from the insulation that hit the left wing during liftoff "just might have been enough to throw things over the edge," he said.

At least twice before, as Columbia returned to Earth from missions, its left wing experienced a critical aerodynamic shift too early -- prematurely increasing heating and drag on that wing, former shuttle commander Navy Capt. Robert (Hoot) Gibson, now retired, told UPI.

NASA knew about the early aerodynamic shifts at the time and was told by Gibson about a particular roughness he had discovered on the surface of Columbia's left wing. Experts confirmed to UPI that the roughness might have caused the premature aerodynamic shift.

The increase in drag on the left wing, particularly if made worse by tile damage, may have been enough to cause the vehicle to fly sideways, something the vehicle might not survive, said Anderson, the curator for aerodynamics at the National Air and Space Museum.


Columbia disintegrated on Feb. 1 as it returned from STS-107, NASA's designation of the 16-day mission devoted to scientific research. Remains of Columbia's seven-member crew have been recovered, along with thousands of pieces of shuttle debris scattered across the southwestern United States.

The first known aerodynamic shift occurred on mission STS-28 in 1989 and was studied carefully by Gibson, an aeronautical engineer who helped investigate the Challenger disaster and redesign shuttle's solid rocket boosters. The Challenger exploded on liftoff in 1986, killing its seven-member crew. Gibson was also commander during four shuttle missions and piloted a fifth mission.

Gibson told UPI he found the surface of Columbia's wings was two-to-four times rougher than the wings of the three other shuttles -- Atlantis, Discovery and Endeavour -- and that Columbia's left wing was 50 percent rougher than its right. He suspected the roughness caused the 1989 shift and another in 1995.

NASA engineers did not pay much attention to Gibson's concern in 1989, he said, finding another cause for the shift. Other experts told UPI, however, that such roughness could trigger a premature aerodynamic shift, leading to additional heating and drag.

Columbia experienced both additional drag and heating on its left wing before it broke up. By itself such an increase in drag is probably not enough to destroy the vehicle -- explaining why the vehicle returned safely in 1989 and 1995.


During its final, fatal mission, however, the drag could have been worsened by tile damage caused when insulation from the external tank broke off early in the launch, striking its left wing. It is also possible the roughness on the tiles got worse over time, again increasing drag, confirmed Anderson, who has 40 years of experience in high-speed aerodynamics, hypersonic aerodynamics and aerodynamic heating.

Even with replacements over the years, 70 percent of Columbia's tiles were the originals made by Lockheed Martin, according to a document approved by NASA and released by the prime shuttle contractor, United Space Alliance, on Feb. 3, two days after the accident.

Anderson told UPI the combined sources of drag could pull the shuttle enough to the left that it was essentially flying sideways -- at which point the uneven forces on it could break it apart.

A NASA press release issued Feb. 15 said that two more yaw jets than originally thought -- for a total of four -- were firing as the shuttle sped towards its landing site in Florida.

"The flight control system was detecting drag... on the left side of the orbiter," said a NASA spokesman. "To compensate for the drag the automatic computers onboard commanded these jets on the left side of the orbiter to fire. ... Just like if you were driving a car on ice and you started skidding -- you would turn the wheel in the other direction to compensate for the skid."


The three axes of flight are roll (tilting of one wingtip up and the other down), pitch (movement of the nose up or down), and yaw (turning of the nose to the right or left).

"Anything to cause increased drag on that left wing would certainly have caused it to yaw," Anderson said. "The shuttle is designed to fly straight. It is not designed to fly sideways. That would have been absolute disaster if something had yawed it so much that it was basically trying to fly sideways."

As a space shuttle re-enters Earth's upper atmosphere, the initial movement of air over the wings is smooth and orderly -- called laminar flow. At a key point in the flight, called the boundary layer transition, the increasing speed causes the smooth flow to breakup into eddies, becoming "turbulent flow." The shift increases heating and drag on the wings.

The temperature jumps, Gibson said, because "the turbulent flow mixes the air much more effectively at the surface, which brings hotter air in contact with the wing -- so you see higher temperature in the course of the entry.

"That's the way they determined (for STS-28) that they had an early boundary layer transition," explained Gibson, "the temperature profiles (were) hotter than they (were) accustomed to."


Normally the shuttle's wings transition from laminar to turbulent flow at 1,200 seconds into re-entry, Gibson said. "On STS-28, on Columbia, that transition happened at 900 seconds -- 300 seconds early. As you might expect, the left wing saw a significantly higher heating environment than the rest of the orbiter."

Gibson said Columbia experienced another premature transition on STS-73.

"There again, the left wing transitioned ahead of the right wing," he said.

Gibson, who had experienced a close call during a mission less than a year before, took a special interest in an early boundary layer transition on the left wing of Columbia during flight STS-28.

"I pulled together the data from all the orbiters," Gibson told UPI. "I saw that Columbia was two-to-four times rougher overall (than the other orbiters) and the left wing was rougher than the right wing by 50 percent."

Surface roughness is a factor in aerodynamics and, in this case, has to do with the gaps between the shuttle tiles and the "step," or difference in height, between one tile and its neighbor. NASA measured such roughness early in the program, Anderson said.

"A rougher surface on the wings will cause premature transition," said Brian Landrum, a professor of mechanical and aerospace engineering at the University of Alabama in Huntsville.


The roughness of the wing is indicated by a measurement called the K equivalent, derived by combining data on the gaps and steps with information on the airflow, Landrum said. Small differences in the K factor can be significant, he said, comparing the roughness to grains of sand.

NASA spokesmen, citing the crush of news media requests after the loss of Columbia, could not provide information on the roughness of Columbia's wings or an early boundary layer transition during Columbia's last mission.

Gibson said NASA had determined in 1989 that protruding gap-fillers caused the early transition. The fillers are pieces that fit between the tiles to seal all the gaps between them.

"There were a couple of protruding gap fillers," said Gibson, (but) I always wondered if (the gap fillers were) part of the problem and the surface roughness was the rest of the problem."

Gibson said he was not as familiar with STS-73 because he was in the process of leaving NASA at the time. He pointed out, however, that the characteristics of the tiles might have changed and that the early transition did not happen on every flight. He had, however kept his files from STS-28, he said, and had reviewed them before discussing the matter with UPI.


"The one time that I looked at all the data very thoroughly it was for one flight. That was for STS-28 and it happened on the left wing, which was the bumpy wing."

He said he had raised the question of the roughness to NASA after STS-28 but the matter was not examined closely.

"Nobody ever thought a whole lot about the wing roughness except me," said Gibson. "Because I was not a NASA engineer, I was an astronaut, nobody paid much attention to it."

Latest Headlines


Follow Us