Shock Testing Requirement Reasoning 1

[Transient1]

MIL-STD-1540 states that 50% of the spectrum points during a shock test must be above the nominal test value. Can anyone provide some insight into how this requirement came about?

50% of the points above nominal doesn't seem to mean much because you could still test below nominal (within the tolerance band +6dB/-3dB) in the frequency range most likely to damage the device under test.

 

[jimkinney]

That's a good question, and so far the only answer that I can come up with is that we have always done it that way, which isn't always a good answer. It's in MIL-1540C as you mentioned, but it's also in NASA-STD-7003, IEST-RP-DTE032.2 and even ESA standards. I haven't confirmed it yet, but I think it has to do with one sided tolerance limits and the fact that stress cycles below the mean don't lower the fatigue life. Having 50% of the spectrum adds conservatism to the test.

As to the issue of tested levels being below the nominal level but still in the tolerance band, the nominal test level is defined as the maximum predicted environment plus some amount to get you to a statistically derived confidence level. That amount is usually the same as the test tolerance, so as long as you don't drop below the lower tolerance limit, your test is still adequate. The amount of confidence that you need will set your limits, if you absolutely can't fail, then you test the hell out of it.

I will keep digging to find the actual reason behind the 50%, and pass it along if I find it.

Jim

Jim Kinney
Kennedy Space Center, FL

 

[Transient1]

Jim,

Thanks for the response. If you do find out any more, I would love to hear about it. As I try to find out mroe about it, I have seen some programs(long ago) that used 30% as the minimum above the nominal SRS. If I was to hazard a guess, it feels like that number is simply a safety factor that has creeped up slowly over time. Probably (I would hope) there is some data supporting units that failed during actual use versus the SRS experienced during shock testing.