Columbia

I have a lot of questions about Columbia. The design of the tile system makes me wonder a bit about things like corrosion under the felt (or perhaps on the inside of the skin?). Do they inspect these things regularly--strip off all the tiles and redo them? Could an accident during mounting damage the skin in a way that would not be seen by inspectors? (Or sabotage--a jab with a sharp object between the tiles...)

Do we have some way of simulating the temperatures and pressures and wind speeds of re-entry on a large scale--say 20x20 cm? We need to answer questions like: How big a hole is fatal? If you have a divot in the tiles, what can you patch it with that you can trowel in place with gloved hands? Is there any way of dealing with a crack in the mount for the leading edge? (a sacrificial sheath?) Do tiles unzip?

Ideally we ought to have some sort of scanner on an arm that goes with every shuttle flight, together with a repair kit for use in an EVA. I read about a rather simple scanner in American Archeology today that archaeologists use to map features of sites. The shuttle is chock-a-block with reference locations, so this could be quite accurate. But this scanner isn't a great deal of use if a 1mm puncture between the tiles is lethal--you can't scan for that very well. A missing chunk of tile would show up very quickly, and a loose or cracked tile also (a piece would be at the wrong angle).

If the damage is not reparable, I assume we can dock at the ISS and hold out until the Russians launch a few Soyuz, or use those bubble gizmos if another shuttle is ready. We'd all be a lot more comfortable if we had some ways of testing fixes to know what is reparable and what isn't.

Of course, we then have the question: how confident are you that the fix worked? 95%, 99%, 99.9%? At what point do you bail and call for another shuttle? (That one doesn't have a technical answer--it is more political.) The astronaut is trying to apply the glue in a difficult environment--did he get it in evenly? That's tough to quantify without lots of tests. What does he push against--a keel-hauling rope of some kind, or do you have to carry the big robot arm every single flight? I suppose you can use some kind of bubble-wrap pads to keep the handyman from dinging up other tiles--and even make them stay in place with static charges.
Ok, look at this picture of an old mission.