Good article about the challenges of a proposed mission to rescue the Columbia astronauts by fast-tracking Atlantis preparation to go get them.
Anyways, the article says that once the astronauts were safely aboard Atlantis, the damaged shuttle would have been allowed to re-enter and burn up over the Pacific.
Would it have been possible to leave the empty Columbia in orbit and plan a later mission to repair the damaged tiles? Shuttles were expensive. Or is getting the tiles to adhere properly something that requires very controlled conditions on Earth?
One more question. The article implies the NASA controllers were initially unsure what was happening when the breakup started. But if they had already held brainstorming sessions over the feasibility of a rescue in orbit, how was the breakup a surprise to anyone?
That, at least, is easy to answer. Space missions require extensive contingency planning, including abort modes, well before launch is authorized. A scenario where the vehicle is in orbit but too damaged to survive re-entry was certainly envisioned, but it could still be the case that the extent of the damage to Columbia came as a surprise, that they didn’t think the damage was that bad and allowed the mission to continue.
Now, whether it should have been a surprise is an incisive question, and one may legitimately criticize NASA’s management.
Interesting trivia: in Feynman’s account of the Rogers Commission he reports that the Shuttle Main Engine Project Manager estimated a 1-in-200 probability of a mission failure due to an engine failure. Historically, 2 out of 135 missions failed, so that number turned out to be a pretty good estimate of the reliability of the entire program.
Except none were from engine failure… depending on how you see leaky side boosters. The main engines turned out to be quite reliable, although IIRC during initial design and testing they blew up more than one from major pump failures.
Certainly. I just thought it was an interesting coincidence- or is it a coincidence? Maybe, for some reason, about 1% is the right order of magnitude of the failure rate of space launches today.
NB to be fair, there was an official engineering document, also mentioned by Feynman, which gave a 1 in 100000 failure probability for the main engine although “the approach of determining when a vehicle is safe to fly based on a well evaluated and documented program with attention to details is superior to relying on a reliability number derived from an insufficient data base.”
Perhaps they crossed their fingers on re-entry. But the fact that they (according to the article) came up with a fairly detailed outline of an emergency rescue while Columbia was in orbit, says to me that they were holding their breath when they decided not to go with it and would have known exactly what happened when it broke apart.
Where does it say that? On Page 1 it says the Columbia Accident Investigation Board asked them to come up with a detailed plan, and that plan appeared in an appendix of the report. The plan did not exist while Columbia was in orbit (which may itself indicate a fuck-up).
I’ll have to re-read. Maybe I missed a key section, but my impression of the article and therefore my question was that this was discussed while in orbit, not a shoulda-coulda- what-if after the accident.
I DID read it at work during a slow period, so cut me some slack.
But regardless, im still curious about my main question. IF a successful rescue mission had taken place. Was trashing the now empty Columbia the only option or would a later repair mission in orbit even be theoretically doable?
The foam strike was thought to be “acceptable.” No imaging of the leading edge of the wing was attempted, so the mission proceeded a planned. If the damage was observed, it might have prompted a discussion as to whether the shuttle should try to de-orbit, or if some emergency rescue should be attempted.
I imagine you would have to attach some sort of periodic boosters to push Columbia back out into a higher orbit any time it started sinking too close to Earth. Then find volunteer astronauts who would be willing to fly a repaired, but dicey, Columbia back through reentry and to the landing strip at Florida. And even then, astronauts would always be wary of flying aboard the repaired Columbia in all future missions.
This comment by an anonymous engineer seems relevant:
This may also answer your question about whether it would have been possible to park Columbia in some kind of orbit (for 18 months!?) while a mission was drafted, then repair it, transfer crew back on board, and land it without incident.
NASA identified the foam impact early on. When asked repeatedly if this would cause a reentry problem, they said no. They claimed that it would create a longer turn-around and reset on the ground while the damage was repaired, but that it did not constitute a flight safety risk.
I think that while they were running their analysis, a rescue contingency was in the works. However, since the engineers concluded the shuttle would be fine, NASA went ahead with reentry.
I’m assuming that the attach ring is a pretty significant chunk of steel - and it took a really nasty dent. Yet no-one could envision the scenario where the leading edge of the wing gets the hit. The leading edge was made of reinforced carbon–carbon (RCC). From another Wiki page: “Carbon–carbon is well-suited to structural applications at high temperatures, or where thermal shock resistance and/or a low coefficient of thermal expansion is needed. While it is less brittle than many other ceramics, it lacks impact resistance; Space Shuttle Columbia was destroyed during atmospheric re-entry after one of its RCC panels was broken by the impact of a piece of foam insulation from the Space Shuttle External Tank.”
The fuel cells were only good for a few weeks. I’m pretty sure an orbiter left in orbit for months would have been unpowered and uncontrolled. Like, tumbling and hard to find.
So no, I don’t see another spacecraft, especially another orbiter, trying to approach it. Imagine trying to repair the damage, refuel the fuel cells and do a cold (very cold) start of everything. It would have been much more difficult than Skylab 2.
Although it sounds a waste - if they had managed to rescue the crew, the cost of retrieving the marooned orbiter, along with the risk involved, would probably have added up to a waste of money. Launches were costing $500 million each. Add on the cost of developing a retrieval mission, with (as described) very detailed and difficult planning. The mission would need to work out how to repair the orbiter in space, bring the systems back up, which would probably involve procedures that were never intended to happen except on the ground with specialised equipment. You could easily spend another $500 million. And you have a very real risk the orbiter fails and kills everyone onboard anyway. A second hand orbiter just isn’t worth it.
The story of the Columbia accident makes for worthwhile reading. There were lots of issues. But the creep of acceptable failures was at the core of it. NASA got away with a failure that the original mission rules dictated was a mandatory grounding of he fleet until solved. But since they had been getting away with it, they continued to launch, even when it was listed as a critical failure that needed solving. There were contributing issues. The RCC sections degraded over time. The leading edge sections on Columbia had done a lot of missions, and when they tested a new section versus one with a similar number of missions taken from another orbiter the difference in strength was startling. A foam block of the same size and composition hitting the new section at the speed the one hit Columbia’s wind left minimal damage on the new section, and punch a hole big enough to crawl into in the old section. Part of the degradation turned out to be dissolved zinc in the rainwater runnoff leached from the launch structure anti-corrosion coatings. But the number of missions and re-entry cycles was the big contributor.
The investigation uncovered other failures, for instance the explosive bolts that held the top of the SRB to the stack were supposed to be captured in a cup when separation occurred. There were failures of the capture device, and bolt fragments were hurled through the air near the orbiter. For a while this was a credible alternative to foam damage.
What really caused the accident was a systemic failure of safety culture at NASA. The facility building the external tanks had reduced staff numbers. Things like the person responsible for safety oversight was also the guy that signed off on completion. And there were monetary incentives for on-time completion. You can see where this goes…
