Fastener Proof Testing

[spacecase]

Greetings all,

One of our project teams has expressed interest in performing proof testing on single point failure fasteners to verify structural integrity rather than choosing other nondestructive evaluation methods. Our written standards for this testing specify that fasteners be proof tested to 75% of the MTR utlimate tensile load. However, no one seems to know where this 75% came from. Does anyone else you this standard or think they can shed some light on this mysterious 75%?

Thank you!

 

[TVP]

The standard meaning of "proof load" in the fastener industry means loading a fastener to a point where the stress is 90% of the yield strength. See ISO 898-1 or similar industry standard. Usually high strength fasteners have a yield-to-tensile ratio of 0.9, which means that 75% of UTS should be below the yield stress of the fastener, and only slightly lower than a typical proof load (0.9*0.9 = 0.81 of UTS).

 

[rb1957]

i would express some caution with this approach. maybe different industries appraoch the problem differently, but in aerospace, because of some very painful leasons, real "proof testing" requires a fatigue test with the proof test loads included. all too often we've had in-service failures because we've applied too high a proof load (how ever you define it) or too low a proof load, or too long a service cycle (ie time between proof loads).

 

[MintJulep]

Fix the design so that you do not have fasteners where a single point failure leads to undesireable consequences.

However, if you won't do that, then test the joint - as it is used in your appliation. Do not test only one component of the joint (the fastener).

 

[GregLocock]

The annual proof testing of man rated lifting gear, to twice its SWL, does more fatigue damage than the rest of its service life, for most of our gear.



Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Screwman]

I agree about not liking this concept. The origin of proof testing was simply a pass-fail test to show if parts had gone through heat treatment or not. If they had not beeen heat treated then they would yield during the prooof load test.
Using proof testing for anything beyond that simple purpose is not a good idea IMHO. Try to redesign the structure to elimiminate the single point failures. Threaded fasteners have so many stree risers on them anyway that they are notorious failure points. (I keep trying to eliminate the stress risers, but every time that I do, the nuts slide off....a little fastener humor there). The other possibility is to grossly overdesign that portion of the structure so that in-service failure just isn't going to happen.

Dick

 

[israelkk]

See FF-S-86E SCREW, CAP, SOCKET-HEAD available at

http://assist.daps.dla.mil/quicksearch/

as an example of how things are done in the Federal and military business. When you buy fasteners make sure that they are from a respected company and come with a valid C.O.T that shows that they are according to the spec.

If you know the batch number you can double check according to the spec.

 

[spacecase]

Thank you all for your insight.

Several of us here question this testing method as well. But of course before discrediting it completely, we have to figure out where it came from and why it was initially considered acceptable for these purposes.