A few days before the Columbia shuttle disaster, NASA noted that debris striking the left wing after takeoff might have "the potential for a large damage area to the tile."
Tile damage stemming from the launch "may certainly be the leading candidate" in the search for the cause of the disaster that claimed seven lives Saturday morning nearly 40 miles above Texas, according to the agency.
The memo, written on day 12 of the 16-day mission, noted that something hit the tiles during the shuttle's launch January 16, most likely foam from the craft's external fuel tank.
An analysis of the impact showed that multiple tiles were missing over a 7-inch-by-30-inch area, but that the loss shouldn't be a problem, NASA official Don L. McCormack wrote in the memo, dated 17 years to the day after the shuttle Challenger disaster.
"Thermal analyses indicate possible localized structural damage but no burn-through, and no safety of flight issue," the document said.
In other words, a few tiles might have been knocked off, which has happened before, but the overall integrity of the mosaic of tiles, which protects the underbelly of the shuttle from temperatures of more than 3,000 degrees Fahrenheit during re-entry, was thought to be sound.
"We looked at the two worst cases: the loss of a single tile near the main door and the loss of multiple tiles in a larger area," Ron Dittemore, manager of NASA's space shuttle program, said Monday.
He and many NASA engineers determined at the time that there was no reason for worry.
"It's not the first time we've had debris generated from an external tank strike [the] underside of the wing," Dittemore said. "Were there any concerns expressed by any single individuals? At the time, I was not aware of any reservations."
Foam debris had nudged shuttles and knocked off tiles during previous launches, one as recently as October. A similar incident took place on a flight in 1985, which included then-astronaut Jake Garn.
"We didn't know about it until after landing," Garn told CNN. "There was a foot-sized hole on the bottom right wing."
Sensor readings in the final minutes of Columbia's flight showed heat spikes and unexpected drag resistance, which would be consistent with damage to the heat-resistant tiles that protect the aluminum shuttle body.
Some of the last data from the doomed orbiter also revealed rising temperatures inside the left wheel well, which houses the landing gear, vital wiring and sensors.
But the minor heat spike there likely reflected a much hotter problem somewhere else, not a breach in the wheel well, Dittemore said.
"It does not seem logical that the wheel well was the source of the problem," he said.
However, by studying more data transmissions from the shuttle in its final seconds, looking at how the onboard automatic flight computer tried to deal with unexpected drag on the left wing and gathering more sensor readings from the craft, NASA investigators hope to learn more about what went wrong.
On the ground, the search picked up for pieces of the shuttle, which investigators are using to plot a debris-field map and possibly to reconstruct the shuttle. In Sabine County, Texas, officials said they found the shuttle's nose cone.
Moreover, the space agency is looking at what might have caused the insulation to fall off the external fuel tank. (Learn more about the external tank)
In Louisiana, investigators turned their attention to Lockheed Martin's Michoud Assembly Facility in New Orleans, where the tank was built.
The tank was a version that NASA is phasing out in favor of a lighter model, which has been used recently on shuttle missions to the international space station. NASA said no safety concerns had been raised with the older, heavier model.
Dittemore said NASA is assuming that foam debris from the external tank is the "root cause of the problem that lost Columbia."
That prospect is a "sobering" one, he said, because it implies a fundamental problem with space shuttle flight design.
