Bolt shear capability independent of preload? 5

[tomsing]

I've recently heard of some testing that indicates that the shear capability of a bolt is effectively independent of the applied preload, for "normal" values of preload. The tests have been done with a bolted double shear joint, with the faying surfaces of the clamped members lubricated to minimize the clamping friction. They've found that varying the preload between 50% and 85% of yield has around a 1% influence on ultimate load capability.

I haven't thought it through yet, but on the surface, this doesn't make a lot of sense to me. Clearly, the body of the bolt is under combined tension and shear stress, and that should reduce the capability. Maybe it's the little bit of friction remaining between the plates, maybe there are some nonlinearities that are playing havoc with the linear lens through which I view the world.

Does anybody have any insight on this? I apologize, I don't have a reference for the testing.

--Tom

 

[israelkk]

This is not a little friction. Even with lubricated joint the friction coefficient can be ~0.2. Recall that it is on both sides of the lug therefore, the total friction force is doubled. In case of 0.2 the friction force is 40% of the clamping force.

 

[CoryPad]

The reference you need is Guide to Design Criteria for Bolted and Riveted Joints, available at no-cost here:

http://www.boltcouncil.org/files/2ndEditionGuide.pdf

Go to page 47 and read from there.

 

[tomsing]

Thanks for the link. It seems like it's saying that the shear stress on the bolt near the ultimate failure elongates the bolt sufficiently to reduce preload. This seems like it would be sensitive to the relative stiffness of the bolt and the joint - a stiff bolt and a compliant joint would get more benefit from this effect than the reverse. It also seems like a short grip length would be more beneficial, because a short region near the joint winds up in shear, and whatever strain you get in this region, the actual change in length will be small; a long joint would require more actual change in length to alleviate preload.

I guess I need to look up the references cited there.

 

[CoryPad]

I think they way to envision this is that fastener interaction between axial and shear stress is important up to the yield point (e.g. von Mises equivalent stress), which is widely documented for fasteners (see same reference above and no-cost MIL-HDBK-5). After yielding, the stress distribution in the entire joint has changed so much that the largest principal stress is the important factor.

 

[connectegr]

Geoff Kulak spoke last week at the NASCC conference. He had some slides related to the shear strength of bolts regardless of the pre-tension. I have attached the presentation.

http://www.FerrellEngineering.com

 

[slickdeals]

The same (or almost) presentation is available at AISC's website

http://www.aisc.org/content.aspx?id=4502