CAT Scans Suggest Cause of Columbia Disaster

Kathy Miles writes "The latest information from Columbia's flight data recorder, along with CAT scan results from RCC layers from shuttle Atlantis' leading wing edge, may give clues to what really happened to Columbia. The flight data recorder shows that there was likely structural damage before Columbia began re-entry. Investigators have been looking at the remaining shuttles and have done CAT scans on Atlantis' reinforced carbon-carbon layers, which show gaps that should not be there. If Columbia had similar gaps, it could have doomed the orbiter."

Comments

I wished I could trust NASA...

[HotNeedleOfInquiry]

But I strongly suspect that any cause of the shuttle loss will be phrased so that Ron Ditmore doesn't have to lose his job.

He lost tons of credibility in the beginning when he stood up and said that it couldn't be the foam that caused the problem. Soon after, we learned that he refused the request of NASA engineers to have pictures taken of the craft while in orbit.

A failure of imagination

[geoswan]

Someone else mentioned those "s" turns, I think it was with the idea that they could have somehow changed those to favor the left wing. Those s turns are very precise and are designed to reduce orbiter speed. They already plot a course for maximum reduction of speed and therefore exposure to higher temperatures.

I am sorry, but I don't think you have given the suggestion of abandoning the "S" turns a real answer. Answering that NASA has plotted the descent profile with great precision is not a meaningful answer at all.

Yes, I am sure that NASA has some of the best engineers, technicians and technologists working in America. I read the assertion that the programs that control the orbiter's descent are among the most reliable realtime control software ever written -- that they have terrificly well thought-out test-suites, bench-checks, and so on. I am ready to believe that.

But, none of that prevented Columbia sticking with a descent profile that was wildly inappropriate for an Orbiter with a damaged wing.

Let me suggest that what we saw was a failure of imagination.

What did those famous programs do? Information flowed from the Orbiter's sensors that it was having trouble following the descent profile. Information flowed from those sensors that could have been interpreted to mean, "The left wing is experiencing unexpected and alarming drag". But the famous programs apparently were not written to recognize the implications of this unexpected drag. So the program doggedly kept trying to fly level, and bring that lazy left wing back into line, so it was carrying its fair share of the burden of re-entry.

This was exactly what it should have been doing if the wing was undamaged, and the primary goals of the descent profile were to maintain the orbiter's working life to the 100 missions it was designed for, and to make sure it arrived at Kennedy.

In a press conference a week or two after Columbia's loss I heard Ron Dittemore speak about Columbia's reaction to that increased drag. I heard him say that when the flight control surfaces were insufficient to deal with the mounting extra drag on the left wing, the control programs would have kicked in with the attitude control jets, until they were exhausted.

And I thought to myself, "And when the jets are exhausted, boom, Columbia tumbles out of control."

This reflects a clear failure of imagination during the specification of the descent control software. I am ready to believe that NASA engineers made sure that the Orbiter's descent software is marvellously within spec. But, it seems to me, that the specifications were deficient, and that no shuttles should fly until the software can cope better with this kind of damage.

Would Columbia have survived if the control software had been prepared to recognize that the unbalanced drag implied serious damage portside, and adapted to it? This seems unlikely. It looks like the damage was too profound.

But I remain disturbed that the software didn't try adapting.

And trying to reduce heat by keeping the orbiter in a less steep descent would have been just as disasterous because it would have meant a longer period of time, even if in slightly lower temperatures. That was just as risky, and possibly more, than the descent carried out.

Excuse me, you expressed this opinion very authoritatively. Can you justify this opinion?

I am not a rocket scientist. Nor do I claim to be. I give my opinions here. And I have explained my reasoning. You haven't.

"Just as risky" you say? Risky to the 100 mission goal of the shuttle design team? Or risky for the life of the crew?