SpaceX Launch Not So Perfect After All

First time accepted submitter drichan writes "Those of us who watched the live feed of last night's Falcon 9 launch could be forgiven for assuming that everything went according to plan. All the reports that came through over the audio were heavy on the word "nominal," and the craft successfully entered an orbit that has it on schedule to dock with the International Space Station on Wednesday. But over night, SpaceX released a slow-motion video of what they're calling an 'anomaly.'"

Comments

Re:Whats the problem?

[residieu]

Just because you have a backup plan, and it works, doesn't mean the launch was perfect.

not really a bad thing

[wierd_w]

The engine failure of the falcon 9 engine #1 is not really a bad thing. It served to prove the reliability of the shutoff system, and flight control hardware.

Considering the horrendous failure rate of NASA's early engines, (the kind that explode spectacularly), this managed failure situation is very promising.

Rest assured, there will likely be a strong inquiry concerning the manufacture and design of the engine fairing that failed, causing the pressure drop, and engine shutdown.

Managed failures like this one don't speak poorly of spacex. On the contrary. They show spacex planned ahead, and the failsafes they built actually work.

It was a good launch

[Maximum Prophet]

Pilots say any landing you can walk away from is a good landing.

In space, any launch that accomplishes its goals is a good launch. If good costs 10% of perfect, go for good.

Re:It was a good launch

[TWX]

Obviously SpaceX wants to achieve man-rating so that they can launch and return personnel in addition to the cargo runs they're currently beginning. I'm curious as to how this moderate malfunction will impact the rest of the program.

Bearing in mind, of course, the deaths of Chaffee, Grissom, and White in the Apollo 1 accident, the launch-time engine failure and later unrelated catastrophic failure for Apollo 13, the Challenger disaster, and the Columbia disaster, it's difficult to call SpaceX's anomaly as being any worse than those. If SpaceX manages a series of cargo deliveries without any loss of the capsule or with complete success on delivery then even with this anomaly they're arguably no worse off than any of the previous space programs were, as far as reliability and safety goes.

Re:It was a good launch

[Electricity Likes Me]

Obviously SpaceX wants to achieve man-rating so that they can launch and return personnel in addition to the cargo runs they're currently beginning. I'm curious as to how this moderate malfunction will impact the rest of the program.

Bearing in mind, of course, the deaths of Chaffee, Grissom, and White in the Apollo 1 accident, the launch-time engine failure and later unrelated catastrophic failure for Apollo 13, the Challenger disaster, and the Columbia disaster, it's difficult to call SpaceX's anomaly as being any worse than those. If SpaceX manages a series of cargo deliveries without any loss of the capsule or with complete success on delivery then even with this anomaly they're arguably no worse off than any of the previous space programs were, as far as reliability and safety goes.

The important thing is whether they can successfully determine what actually happened, and why it happened (i.e. replicate the malfunction on a test bed engine). This was the thing Feynman was most critical of NASA for post-Challenger - that the whole disaster was caused by this faulty assumption about engineering risks on the O-Ring seals (i.e. the seals were getting eroded by exhaust during launch, but the question posed was "is this dangerous" not "why is this happening" - the former being foolish since the system was not designed to cope with this, and it's true cause was unknown).

It's a triumph that the launch still succeeded, but having averted an unforeseen consequence the only safe thing to do is make sure it's both forseen and mitigated in the future.

Re:An

[the gnat]

Libertarians rode the back of this and shouted about how much better it would be to privatise space. But in fact we're just right (*) here again, with SpaceX substituted for Boeing.

I think you'll find it's not just libertarians cheering for this - after all, privatizing the launch infrastructure has been a key element of Obama's space plans. The difference from the previous situation, where NASA relied on bloated defense contractors, is that SpaceX and its competitors will have to enter fixed-price bids, instead of the old cost-plus contracts which gave the contractors zero incentive for efficiency. Whether this will actually work in the long run remains to be seen, but it's hard to see how this is worse than the old system, and putting the federal government into the launch vehicle business sounds like a spectacularly awful idea.

Re:Whats the problem?

[rickb928]

If that's accurate, then SpaceX is looking into a shutdown event, a LOT different than a destructive failure. The fairing imploding will either be the anticipated result, or a new issue to understand and resolve/document.

Shutdown may be accompanied by data, and there is a fix. Valves, pumps, all kinds of fairly well understood stuff to analyze and resolve. Destructive catastrophic failure would be much more disturbing.

So far, they seem to be doing at least as well as NASA did in the early days. Mercury was a real crap shoot, and early Saturn development was exciting to say the least. I filled a few scrapbooks with notes on those faiures. Fun times...

A statistical analysis:

[mosb1000]

They've launched 4 Falcon 9 rockets. One engine has failed, so that's an observed failure rate of 1/36 or about 3%. The means the odds of 0 or 1 engine failing (a successful launch) is 97.6% and the odds of more than one failing is 2.4% assuming the currently observed rate is representative of the actual rate. 2.4% would be an excellent failure rate for any rocket launch system. In fact, no one has achieved a failure rate that low. And bear in mind this rate includes 3 experimental launches and only one production launch. Of course, a launch failure can be brought about by more than just engine failures, so 2.4% is really a minimum and other factors which haven't yet manifested themselves would add to it.

Space X is saying that this is probably a failure in the aerodynamic structure of the rocket, not the rocket engine itself. If that's the case, the above statistical analysis is invalid because it assumes no interdependency in engine failures. A structural failure could lead to more than one engine failing. It would also be problematic in assessing the future failure rate because the engine configuration is going to change in their 1.1 version. The outer engines will be circularly arranged in future versions while in current versions they're arranged in a square.