JOHNSON SPACE CENTER, Texas (CNN) -- NASA engineers said they are convinced Columbia's left wheel well was punctured, allowing super heated gas to leak into the wing during the shuttle's last minutes of flight. A preliminary analysis, made public Thursday afternoon, has determined that the temperature inside the wheel well was too hot for it to have been caused by one missing tile. "There is no question that wheel well was breached," said a space shuttle program source. "But who knows how? Did it start there, or was this part of a domino effect?" The super heated gases can reach temperatures approaching 3,000 degrees Fahrenheit. The inner wall of a shuttle wheel well is made of unprotected aluminum, which melts at about 1,220 degrees Fahrenheit. NASA will continue its analysis while looking for scenarios to explain the puncture and show how the gas entered the wheel well or elsewhere in the wing. Earlier reports that Columbia's left landing gear was improperly lowered moments before it disintegrated have been discounted. Only one sensor aboard the shuttle reported that landing gear had been lowered. Other indicators and drag information do not support the scenario. The focus of the investigation remains on the left wheel well. Members of the Columbia Accident Investigation Board do not believe the loss of a single heat-shedding tile would have created a fatal breach. Analysis is under way to consider the possibility of losing multiple tiles near the door; a failure of the seal between the door and the body of the orbiter; or perhaps the loss of a crucial piece of heat protection at the leading edge of the wing, where the temperature is highest on re-entry into Earth's atmosphere. The forward edges of the shuttle's wings are covered in 22 reinforced carbon-carbon caps. The loss of one cap would expose the flat aluminum edge of the wing structure beneath to searing heat that would quickly spread to the wheel well. There are no heat sensors on the leading edge of the wings and thus the temperature spikes recorded in the wheel well could have been the first indication of trouble. "This would explain the data signatures," said one shuttle program source. "But it does not help with the smoking gun." According to a timeline refined and released Thursday by NASA, flight controllers first saw an abnormal increase in temperature in a brake line inside the left wheel well at 8:52:17 a.m. This was seven minutes before the loss of signal from Columbia. Twenty-four seconds later, another brake line in the left main gear well registered abnormal temperatures. At 8:52:59 a.m., a series of temperature sensors at the rear of the wing began failing. NASA concerned about wheel well before descent On Wednesday, the space agency disclosed that an e-mail passed between NASA engineers two days before the fatal breakup of Columbia discussed worst-case scenarios, specifically dealing with failures of the left wheel during descent. Engineer Robert Daugherty outlined several scenarios in his e-mail to David Lechner, who worked with the shuttle's mechanical systems. They were based on possible outcomes of tile damage the orbiter may have suffered only 80 seconds after it lifted off from Kennedy Space Center January 16. When they analyzed video of the liftoff a day later, NASA officials noticed what they believe was a piece of foam breaking off the external tank and glancing off the underside of the shuttle's left wing. "I am admittedly erring way on the side of absolute worst-case scenarios and I don't really believe things are as bad as I'm getting ready to make them out," Daugherty wrote in his January 30 e-mail. "But I certainly believe that to not be ready for a gut-wrenching decision after seeing instrumentation in the wheel well not be there after entry is irresponsible." As the shuttle streaked across the sky February 1 toward its landing site in Florida, controllers at Johnson Space Center in Houston remarked that sensors inside the wheel well showed higher temperature readings. Other sensors were shutting down, one by one. Minutes later, the shuttle broke up, killing all seven crew members. In his e-mail, Daugherty theorized that the shuttle's landing gear may not deploy if there was "a substantial breach of the wheel well." "It seems to me that with that much carnage in the wheel well, something could get screwed up enough to prevent deployment and then you are in a world of hurt," his missive said.
