"Father of the Space Shuttle" George Mueller Dies At 97 (washingtonpost.com) 75
The Washington Post reports that long-time NASA engineer and administrator George Mueller died on October 12 of congestive heart failure, at 97. Mueller had a hand in NASA programs as Associate Administrator of the NASA Office of Manned Space Flight going back to the Apollo program, but not only as an administrator: he played a large role in the design of Skylab, and in lobbying for the Space Shuttle; this last earned him the (sometimes disputed) nickname of "Father of the Space Shuttle." During his Apollo days, Mueller became well known for his insistence on "all-up" testing, rather than incremental, per-component tests. From the Washington Post's story:
As applied to the space program, [all-up testing] implied specifically such techniques as the testing of all three stages of the giant Saturn V booster rocket while they were coupled together and with a payload attached to boot. It was reported that the scheme had its doubters, among them such leading lights of rocketry as Wernher von Braun. But in time, the forceful Dr. Mueller proved persuasive enough to overcome all such reservations, and it was “all up” for the mammoth Saturn V, the launch vehicle upon which NASA pinned its hopes of sending Americans to the moon.
he's now all-up (Score:2)
Re: (Score:2)
I don't think the future as we know it today would be the same if Zefram Cochrane didn't know about the tall statue in his honor that would marked the first test of a warp drive engine and First Contact with the Vulcans.
http://memory-alpha.wikia.com/wiki/Zefram_Cochrane [wikia.com]
Re:All Up Testing? (Score:5, Informative)
The problem comes when you want to test the first stage of your rocket in flight. It needs to be tested in an environment as close as possible to a real launch. You either spend a lot of money building a fake second and third stage, or you put a real second stage on top. That also has to be tested in an environment as close as possible to a real launch. So you spend a lot of money building a fake third stage, or you put a real third stage on top. Then you need a realistic payload. So you might as well put a real spacecraft on top.
So long as there's a good chance of the first stage working, you probably save money and time in the long run by just launching the whole thing. You can test whichever stages work, and you don't have to spend money building fake stages.
Re: (Score:3)
uh, no. the spacecraft is usually the most expensive part of the launch, especially if it's a man-rated craft.
A Saturn V launch cost about $2,000,000,000 in today's money. So adding another test flight to test the spacecraft after you've already tested the Saturn V would add $2,000,0000,000 to the development cost.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Catering costs more than fuel.
True in this case. Propellant is a minor cost in such rocketry programs.
Re: (Score:2)
Catering [youtube.com] was most of the difference between a fake and real moon landing anyhow.
Re:All Up Testing? (Score:4, Informative)
I work at KSC and here is the issue. Sometime just creating the relevant environment is more expensive then just having a test flight. For example to simulate the aerodynamic and vibration loads on a rocket stage are near impossible. We will shake the avionics or even whole spacecraft. But nobody has a shaker big enough for a whole stage so you need to test it in flight. And for example on the Saturn V are you just going to test the first stage? Then you need to simulate the rest of the vehicle weight, cg, and aerodynamic loads. At that point its just easier to use the second and third stage.
Re: (Score:3)
MSFC has a "shaker" big enough. A whole Saturn V rocket, and then later a Space Shuttle, were tested in the dynamic test stand here.
Re: (Score:2)
From what I remember that was a pretty low frequency shaker looking for modes that would have an impact on the control system. It couldn't simulate the launch environment. But this is a good example of the mentality. You test what you can at a reasonable cost but eventually the tests get more expensive and time consuming than a few test flights.
Re: (Score:1)
As I recall, the second test flight almost failed because of the difference between ground testing and full-up flight testing. It had to do with vibrations, but I don't know if the "shaking" test stand was specifically involved.
During ground testing, the fuel (or maybe it was the O2 lines) were encased in a layer of ice, because of the warm, humid air at ground level. That ice protected the lines from the vibrations associated with firing the engines.
However, when they actually flew, there was no ice, and
KSC? (Score:2)
Huh, very rare to have someone from the Kennedy Space Center here. We usually get input from posters who "work" on KSP instead.
Re: (Score:2)
Why not both? KSP is awesome!
Re: (Score:1)
You would be surprised who reads select click-bait here.
I clicked to see if I remember the guy. I only visited KSC a few times, but worked at JSC.
Even for the space shuttle flight software, we switched from "unit testing" to systems testing and watched our error rate drop over 50%. We were already causing less than 1 SW error per year and that dropped to 1 error every 2 years. It was a difficult transition, but required that we learn the systems from a macro-level, not the micro-level they had been using t
Re: (Score:2)
Re: (Score:2)
Surely you want to do both unit testing as well as integration systems testing in a realistic production environment.
In a software environment those tend to be cheap, especially when compared to the cost of not doing them.
Should've listened to von Braun (Score:4, Interesting)
At least when it came to solid boosters.
Re: (Score:2)
But ... but they have that creamy filling... and are tasty...
Re: (Score:2)
There were plenty of tests of individual boosters before STS-1.
Re: (Score:1)
Or course he was German...
probably should really be Müeller
Re: (Score:2)
Re: (Score:2)
Medical examiner (Score:1)
"He lost too many tiles"
Re: (Score:2)
"He lost too many tiles"
At least he didn't blow his O-Ring out
Re: (Score:2, Insightful)
...shuttles' wasteful irresponsible uselessness, and total failure to achieve almost all of intended goals. everything they did do could have been done more efficiently in other ways.
It was still the most amazing and capable spacecraft ever developed.
Re: (Score:2)
Mixture of approaches works best (Score:3)
Usually, you need a mixture of approaches to get things to work. Idealism in software engineering, or in engineering, works about as well as idealism in politics, ie it doesnt really, it misses key points. But, in both areas, it's much easier to create a platform on idealism, and so people who propose one single idealistic viewpoint often do quite well.
In practice, in software engineering, saying 'all tests must be automated, 100%', misses that some things are really hard to test automatically, but can be tested by hand quite simply. Similarly for creating test harnesses, mocking, which this article is the hardware-engineering equivalent for. Sometimes it's easier to mock, and do real 'unit-testing', and sometimes it isnt, and insisting that every project, and every part of every project, uses the exact uniform standard, might not always work as well as it looks like it will in the Powerpoint presentation :-P
Re: (Score:2)
Comments like that will make you unemployable. Today software is all about continuous integration, automated testing and 100% code coverage. If you test every line of code in your system what could possibly go wrong? No longer the drudgery of major and minor releases, today every nightly build is a releasable product. Even Microsoft has taken this on, Windows 10 will be the "last" version. Customers upgrade continuously secure in the knowledge that bugs are a thing of the past....
Better to have strangled it at birth (Score:2)
The space shuttle had a clear goal, namely to launch stuff into low orbit cheaply. There are various ways to measure the cost per Kg launched, but even if one ignores the huge research cost, the shuttle fails big time. The Russians can launch stuff using relatively simple rockets for a fraction of the cost. And a Saturn V can launch bigger payloads into *low orbit*, I would think.
Worse, having built the wretched thing an excuse had to be found to use it, and that lead to the ISS. A huge white elephant.
I