SoylentNews is people
SoylentNews
from the we-don't-know-what-we-don't-know dept.
After a decade, NASA's big rocket fails its first real test:
For a few moments, it seemed like the Space Launch System saga might have a happy ending. Beneath brilliant blue skies late on Saturday afternoon, NASA's huge rocket roared to life for the very first time. As its four engines lit, and thrummed, thunder rumbled across these Mississippi lowlands. A giant, beautiful plume of white exhaust billowed away from the test stand.
It was all pretty damn glorious until it stopped suddenly.
About 50 seconds into what was supposed to be an 8-minute test firing, the flight control center called out, "We did get an MCF on Engine 4." This means there was a "major component failure" with the fourth engine on the vehicle. After a total of about 67 seconds, the hot fire test ended.
During a post-flight news conference, held outside near the test stand, officials offered few details about what had gone wrong. "We don't know what we don't know," said NASA Administrator Jim Bridenstine. "It's not everything we hoped it would be."
He and NASA's program manager for the SLS rocket, John Honeycutt, sought to put a positive spin on the day. They explained that this is why spaceflight hardware is tested. They expressed confidence that this was still the rocket that would launch the Orion spacecraft around the Moon.
And yet it is difficult to say what happened Saturday is anything but a bitter disappointment. This rocket core stage was moved to Stennis from its factory in nearby Louisiana more than one calendar year ago, with months of preparations for this critical test firing.
Honeycutt said before the test, and then again afterward, that NASA had been hoping to get 250 seconds worth of data, if not fire the rocket for the entire duration of its nominal ascent to space. Instead it got a quarter of that.
(Score: 0) by Anonymous Coward on Monday January 18 2021, @09:00AM (3 children)
if it wasn't supposed to fail, they wouldn't call it a test.
no test is supposed to fail. especially here, everything about this is very expensive and time-consuming.
yes. sometimes you run a "test" when you're pretty sure failure is the result, because it's simpler to find out the big issues this way.
but protocols that demand tests at various stages are put in place precisely to catch mistakes.
all tests are there because it's conceivable that something will fail.
the only sort of test that's not supposed to fail is something like the "guns are always loaded idea".
if someone passes you a gun, you are supposed to check that it's not loaded, even if you've just seen the other person confirm that it's not loaded.
if that test fails (i.e. you find that the gun is loaded), then you can say that the test was not supposed to fail --- the other person is an idiot, or lying, or something along those lines.
(Score: 3, Insightful) by unauthorized on Monday January 18 2021, @11:19AM (2 children)
You can absolutely expect failures from a test, such as for example when you're testing your failure recovery mechanisms result in a graceful failure instead of a catastrophic one. Validating how a machine performs under failure conditions can be just as important or even more important than validating it under nominal parameters depending on how high the cost of failure is.
(Score: 0) by Anonymous Coward on Monday January 18 2021, @11:40AM (1 child)
I would count "graceful failure" as a success, from the designer's point of view.
I guess in that sense you could say this was a graceful failure: they were able to see that something was going wrong, and they were able to stop it before catastrophic failure (explosion or whatever).
(Score: 2) by theluggage on Monday January 18 2021, @03:01PM
It was only "graceful" because it was a static fire and they had the luxury of responding to a singe engine failure by shutting the whole thing down, which would be problematic if it had already been ten thousand feet over the Atlantic...