It sounds like the tiles are the Achilles heel of the Shuttle. However I think we have made progress in chemical engineering since the Shuttle tiles were developed. It seems really stupid, to me, to send even one more person into space with the Shuttle the way it is now.
Also, as far as temporary repairs go: Shuttles can survive with some tiles missing. We need to determine as much as possible, which tiles are critical to survival. Those areas would get inspected, if any Shuttles ever fly again, after achieving orbit. I still think it is possible to develop some type of repair that would give our astronauts a better chance to survive. It’s starting to look like the astronauts in Columbia had no chance. That’s wrong.
Ok, here’s the story. It’s actually pretty complex.
The Apollo module used a 28VDC main electrical bus. All the systems on board were originally rated for that voltage.
At one point early in the program, it was decided that for on-pad testing and servicing, a 65V power supply would be provided from the ground to power some of the Apollo’s systems. A lot of the subassemblies were already made, so electronics parts rated for 28V had to be removed and replaced with parts rated for 65V. Almost all of them were.
Inside each oxygen tank was a heater core, to force the oxygen to go from liquid to gaseous to get it out. Attached to each heater was a thermostat, designed to keep it from getting too hot. Under normal conditions this thermostat would never be used - it was more of an redundancy safety device than anything else. They forgot to replace this 28V thermostat with a 65V rated one. Apparantly this mistake was made on most or all of the early oxygen tanks.
The tank that would eventually go into Apollo 13 was first assembled into an earlier Apollo. At one point, as happened often, there was an engineering change made that required them to remove the oxygen tank assembly form the half-built Apollo capsule. The guy removing the bolts that held it in lace missed one. When they tried to lift the oxygen tank shelf out by crane, it raised up an inch, then snapped the lifting fixture and whanged back into place.
The tank was properly removed and given a detailed examination. Another tank ended up being used for that flight, and the dropped tank, after being declared safe to use, was installed into Apollo 13.
Each tank has three pipes going into it. There’s the supply line, which feeds all the spacecraft’s systems, the fill line, for filling it up on-pad, and the drain line, for draining it on-pad. The fill and drain lines are only used pre-launch. They go through a common fitting, through a coupling that it was later determined was not well fitted to begin with. When the Apollo 13 tank was dropped, this fitting was slightly dislodged.
Fast forward. Apollo 13 is sitting assembled on the pad. NASA is doing tests, one of which involves filling the cryo tanks and testing the electrical and life-support systems. All goes well, the test is done, and they’re ready to drain the tanks. Draining the oxygen tanks is accomplished by pumping gaseous oxygen into the fill lines, forcing the liquid oxygen out. Problem is, oxygen tank number 2 doesn’t want to empty. The dislodged fitting means the gaseous oxygen is going straight from the fill line to the drain line, getting very little of the tank’s contents out in the process.
Well, taking the Apollo apart to fix the tank would have pushed the flight back months. Some clever fellow suggested that, since the drain and fill lines are only used pre-launch, it really didn’t matter if they’re not working properly. All they have to do is figure out how to drain the tank. They came up with the idea of simply leaving the heater on, to boil off the liquid oxygen. After all, the thermostats will prevent the tank from getting hot enough to be damaged, right?
Well, with the 65V on-pad voltage, the 28V rated thermostats welded on, and the tank baked like an oven. Teflon insulation on exposed fan motor windings inside the tank crumbled and cracked.
Several days into the flight, the crew is having trouble with the cryogenics. The dislodged tank fitting came loose enough to short out the capacity probe, and they couldn’t determine how much oxygen was in the tank. Then a liquid-hydrogen tank registers an underpressure, which triggered the cryo system master alarm. The crew canceled the alarm, then decided to switch on the cryo stir fans - located inside each oxygen and hydrogen tank - to see if that would help the problem.
Initially, it did. They hydrogen tank evened out, and the capacity probe on the number 2 oxygen tank started working again. Then the damaged fan windings shorted out. Circuit breakers shut down the fans, but not before a fire had started inside the oxygen tank.
Pressure in the oxygen tank rose to 1000 PSI, at which point the overpressure valve opened to keep the tank from exploding. This would have triggered a master alarm, but the crew had cancelled the previous hydrogen tank alarm, which (being part of the same alarm as the oxygen-pressure tank), caused the oxygen tank pressure to not trigger an alarm and go unnoticed.
Still with me? The fire burned inside the tank for maybe a minute, spreading along the wiring harness inside the tank. The fire heated up the neck of the oxygen tank, where the wiring and pipes went in, weakening the metal until the tank blew apart. You know what happened after that.
How would NASA have “checked the tiles after witnessing … impact?” The liftoff can’t be aborted after the solid rocket boosters ignite. My understanding also is that there was no equipment available on the shuttle to allow the underside to be checked. The shuttle was configured for a particular mission in orbit and that didn’t include getting around on the underside. Carrying along extra gear to allow for any of the many possible mishaps would be prohibitive because of weight limitations.
My only point is that in plain old ordinary aircraft flights, inflight repairs of damages occuring on takeoff can’t be done.
Sometimes the dragon wins and that seems to be something that many are unable, or unwilling, to face.
As has been noted, the Shuttle has a lower failure rate than any other space launch vehicle. It has roughly one third the failure rate of the Soyuz, for example. The only way we can get a safer vehicle is to wait a decade or more for a replacement (Which only might be safer).
I do think doing an inspection of the vehicle in orbit may be a good idea, but I doubt it would have changed anything. Carrying along supplies to perform repairs on the orbiter is somewhat absurd, though. It’s like trying to use a tire patch kit to save the Hindenburg.
My understanding, from listening to a variety of experts, is that the Shuttles have a cargo capacity of 25 tons. They apparently often only carry a cargo of half that weight. It doesn’t sound like weight limitations would be a problem on most flights. I don’t know the cargo weight of Columbia’s last flight.
Inspect: Yes. Change anything: On Columbia you’re probably right. There was apparently nothing that could be done.
It might be absurd, but sometimes even absurd ideas (like the lightbulb and the electric starter on cars) work if enough time, creativity and effort are put into them…
My guess is that Columbia was probably pretty much at her load capacity. The space-lab module they used on the mission was $100 million, even at $500 for a toilet seat (trivia note: In shuttle toilets, the shit’s supposed to hit the fan!), you’re going to get something pretty heavy for $100 million. Don’t forget for the extra gear you carry, you’re going to have to carry extra fuel, and that adds weight, so you’ve got to carry more fuel to compensate for the additional weight. Besides, how do you determine what they should and shouldn’t carry as “emergency supplies”?
Face it, folks, no matter how well you plan, Murphy’s Law get’s you sooner or later. Besides, we’re still not even certain that it was missing tiles which caused Columbia to break apart. So what if you prep the next shuttle with the Ultimate Tile Patch-Kit™, and that shuttle ends up suffering the same fate as Columbia because it wasn’t the tiles that caused Columbia to be lost, but a software glitch which opened the left gear doors too early!?!
That’s true. However, in the case of the Shuttle, we have not planned well. I think it’s time to ground the remaining three Shuttles. An interesting quote is here:
"in financial terms and safety terms no project has done more harm to space exploration."
2. We’re not certain, but the evidence leans that way. I believe a software glitch of the type you describe would have been obvious to the investigators by now.
As others have said, you can’t just carry spares for a few possible problems, but for a whole bunch if after a disaster you don’t want to be second guessed by those who are unwilling to accept the reality of accidents. “They could have done something and didn’t so they must have screwed up” is the cry.
You also have to carry the proper tools. And the astronauts have to be trained to use all of them, maybe in a space suit, certainly in a weightless environment, and things like adhesives have volatiles that outgas like mad in a vacuum and so don’t act anything like they do in the atmosphere and on and on.
The idea of in-flight repair of a damaged space ship is pretty much a fantasy. In addition, damage hasn’t yet been proved as the cause. Maybe the computer slipped a belt and didn’t put it in the proper attitude for reentry. Or none of the above.
NASA is doing a good job considering their budget limitations. I would suppose that the astronauts are more aware of the NASA limitations and the risks than most of us and they still go willingly.
A major real enemy in endeavors like this is complacency. Things go along great and gradually the lessons learned from the previous accidents fade into the background. Many of those who participated in the accidents have retired or moved on to other jobs and such things are only theoretical to current workers.
Well, GOM, it looks like NASA considered the idea, and came to the same conclusion; not only would it not help, but it would make things worse. The adhesive used to apply the tiles in space could compromise their performance and cause even more problems, and tiles themselves are easily damaged (They are, IIRC, foamed glass and easily damaged), so they could end up damaging more tiles. See the article mentioned above, about NASA’s options.
Also, that comment about the landing gear door opening early might not be far from the truth. It could be quite possible that one of the landing-gear doors failed durring reentry, and either collapsed in, or was torn off. Then again, a hundred other things could have gone wrong, too.
As for the quote…
Safety terms? Safety terms?? As has already been pointed out, the Shuttle has the lowest failure rate of any launch vehicle, even among one-way rockets that only have to deal with exiting the atmosphere. The Time article is, quite frankly, idiotic. I’m sure NASA would love to get a new launch vehicle that’s cheaper and easier to get into space. Problem is, it’s going to take a long time, because people are already complaining about them spending too much money! And in general, the tradeoff is either economy or safety. To make the systems safer, it’s going to cost more.
People complaing about the Shuttle… First they say it’s not safe, and they should put in fixes to make it safer. Then they say it’s too expensive, and they should cut down costs. Or they say NASA should be scrapped and a brand new agency put in place (Oh yeah, there’s a price-saving endevour if I ever saw one…).
Great article. :rolleyes: I especially love this part:
Totally inaccurate. It wasn’t throttle up that killed the astronauts, it was the hot jet from the solid rocket booster (which can’t have it’s “throttle” adjusted) which had finally burned through the insulation on the external tank and ignited the hydrogen stored inside. Even if the shuttle hadn’t applied maximum throttle at that point, she still would have exploded.
Yeah, the shuttle’s a creaky, antique design, but we’ve got nothing else available to us at the moment. The Space Launch Initiative was cancelled because it wasn’t yeilding the results NASA needed. Had NASA not cancelled it, you can bet some yammerhead would be bitching that the SLI was diverting necessary funds from some other program.
There’s also this little number from the article:
I bet that Mr. Easterbrook would be shocked to find out that the Hubble telescope only has a low-speed 486 processor on-board and not a Pentium or better processor! Why? Because it takes a long time to design something to withstand the rigors of being in space! So you can’t change the design parameters just because someone’s come out with a faster chip and still meet your deadline. Besides, when it comes to things like the shuttle (where one can’t exactly “get out and get under” to borrow a phrase from the Model-T era), you want older technology in certain areas. Why? Because if it’s old stuff, then you’ll know ahead of time the kinds of things that can and will go wrong with it. Laboratory testing can only reveal so much. (And if you don’t test for it, you probably won’t find it.)
As for his claim that the shuttle has done more damage in regards to safety, that’s a bit of “creative accounting,” IMHO. The shuttle carries more people at one time than any other spacecraft, so naturally, a fatal accident with one is going to cause a bigger spike in the numbers than the loss of a Soyuz capsule. That’s akin to saying cars are safer than 747’s, since fewer people die in cars than on 747’s at any one time. (IOW car crash =10 dead, 747 crash = 200 dead.)
The Apollo program had 3 fatalities (and one near triple fatality with Apollo 13) in 21 flights (that’s counting the three Skylab missions and the Apollo/Soyuz flight), the shuttle’s had 14 fatalities in 113 flights. So 4.6666 times the fatalities in just over 4 times the number of flights. One total loss of the crew for every twenty-odd flights, not bad odds, if you ask me.
Also, assuming that the investigators would have found a software glitch by now is granting godlike powers of divination to mere mortals. The disaster happened Saturday, they haven’t even found all the pieces of the spacecraft, muchless inspected them. To say that preliminary evidence indicates what the final conclusion will be so soon after the event is a bit much in this case. There are no black boxes on the shuttles, and NASA hasn’t had much time to examine all the telemetry data as of yet. It may have been a tile problem, or it might have been something else. Even the folks at NASA say they’re not leaning in any particular direction at the moment because they don’t want to overlook anything important.
Sorry. I must have been unclear. No way anyone could find the software glitch in this amount of time, if it exists. What I meant is that they should have had time to determine if the left gear doors had opened too early. We already know about the excess heat on that side. I must assume, travelling at that kind of speed, that we monitor the opening of the left gear doors.
They know that there was an increase in drag in that area of the shuttle, they don’t know what caused it. If they haven’t studied all the telemetry data, they don’t know if the doors were open or not. Nor can they be certain that the data is accurate until they’ve had a chance to examine the remains. It’s that simple.
I recently read (but can no longer find) an article referring to a “tile patching kit” that was created for earlier flights. It consisted basically of a calking gun, but after furthur testing it was found that applying it didn’t really help matters much, so it was removed on later flights.
I’d like to reply to the earlier incredulity on not even trying to examine the tiles. As people have mentioned, maybe they were damaged: it’s happened before, it didn’t (and as of today, still doesn’t) seem like a big deal.
When flying airplanes with retractable gear, sometimes the gear gets stuck or indicates that it’s not locked down. Current wisdom is that you should be careful, but land anyways.
You could try violent manoevers, but they’re dangerous. Perhaps bump the landing gear on the runway: more so. The biggest is flying low and getting the tower to look. This is also dangerous, and still useless. They’ll say: “No, it’s stuck up”, in which case you land carefully. Or they’ll say “Looks down, but who knows?”, in which case, you land carefully. Just like trying a dangerous excursion to check the tiles, it’s just not useful.
This isn’t on topic but what the hell, many posts aren’t.
NASA might be open to criticism for the repeated occurence of pieces of the foam insulation and tiles coming off.
Going back to the Challenger investigation, Richard P. Feynman was quite critical of NASA decisions as to repeated difficulties with the booster motor O-rings.
In his book What Do You Care What Other People Think? he said that on some previous flights the rings were partly burned through. In one case the burn-through was about one third and that was interpreted as there being a safety margin of three.
However, Feynman pointed out that the O-rings weren’t supposed to burn at all and this pointed to a design problem that should have been addressed as soon as it was identified.
Now as I understand it piece of the foam insulation have fallen off during takeoff before. And according to the reports tiles have been dislodged.
These are design weakenesses that should have been investigated and corrected. Random falling off of tiles surely can’t be a part of a well understood design nor can pieces of the foam coming loose. If these things happened and * no fix was tried* then it seems to me that criticism of that is in order.
Whoa. Seems the press read the same article I did… Today google for nasa, shuttle, and “tile patching kit” and you get tonnes of stuff. They all seem to be based off the first article I can’t find, but they all say that it never actually flew, as it turned out to compromise tile performance.