Single-shot Motor Test Duration

[ScrewYouGravity]

Hi all,

I'm looking for information on how to define the appropriate test duration for a single-shot actuator. The actuator in question is an electric motor which is planned to function for 30s only in orbit. The motor is a single point of failure for the mechanism.

Powering the motor for more than 30s on each test seems sensible, however I'm unsure if adding a simple safety factor is correct of if there's a more scientific way of determining an appropriate test duration. I appreciate any experience and resources you can provide.

 

[GregLocock]

I don't think you'll like the answer, but I suspect for single shot devices you need to test a lot of them. I think your idea of running it for more than 30s is a nice idea, but since success could be all of them failing at 31s, and failure could be an average life of 120 s but with a standard deviation of 60s, you need to understand the failure distribution, probably by using a Weibull analysis.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[TugboatEng]

What aspect of the design leads it to be one shot useful? There is no convective cooling in space so you need to give it enough thermal mass to survive it's expected operating sequence but that doesn't mean it can't be used again after cooling off. My understanding is that microwelding of bearings is a very important issue to consider because a system that survives one operation on earth might not survive one in high vacuum.

 

[rb1957]

the problem is that it Must work when needed, failure is not an option (geeze, I hate that phrase).

You can do as Greg suggests and work up your own reliability, which will be a massive test program.

Or you can buy one ... "surely" there's a proven actuator around, or people who have already done the testing so can make you one that suits you operation. Mind you, it'll be expensive.

What does this electric motor do ? Could you do the same with a simple spring ? (much more reliable)


"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[GregLocock]

Or as we do on cars when something has to work once reliably, explosives.

Cheers

Greg Locock


New here? Try reading these, they might help FAQ731-376

http://eng-tips.com/market.cfm?

 

[ScrewYouGravity]

Thank you for all your answers!

The purpose of the motor is to extend a boom containing sensitive electronics. The boom will be extended at mission start and it will not be required to stow and redeploy it.

I am starting to think that the number of cycles / parts tested is much more important than the duration of the test itself. Would it make sense to have many cycles at close to 30s? And in that case follow wither the ECSS lifetime testing standard or do my own reliability analysis.

We plan on purchasing a motor which is intended for the application, however the purpose of testing is to ensure that it's still functioning as expected after the intended suite of tests, including vibration testing.

 

[rb1957]

you'll be building a test specimen and subjecting the poor thing to all the horrors you can imagine.

If you need a space rated actuator, then buy one with the reliability you require (and preferably one that has a proven track record).
I would still suggest alternative extension methods that are inherently more reliable.

"Hoffen wir mal, dass alles gut geht !"
General Paulus, Nov 1942, outside Stalingrad after the launch of Operation Uranus.

 

[verymadmac]

I suspect that this would be an idea case for an environmental stress screening qualification program.

 

[IRstuff]

Often, people consider building motors such as this with dual windings driven by separate driver circuits to minimize the single point of failure likelihood.

TTFN (ta ta for now)
I can do absolutely anything. I'm an expert!

https://www.youtube.com/watch?v=BKorP55Aqvg

faq731-376 forum1529 Entire Forum list

http://www.eng-tips.com/forumlist.cfm