bolt stress concentration factors 1

[LeeEccles]

Is is realistic to apply both an axial and torsional stress concentration factor to bolt stresses? I'm looking at the head to shaft fillet radius and the threads themselves. I get very high predicted stress values ! Am I doing something wrong or am I being very pessimistic?

 

[Screwman]

In general, almost all of the torsional stress disipates within seconds of completing the tightening process. I have usually only seen an axial analysis. You will get pretty high factors...all those notches make pretty good stress risers.

Dick

 

[LeeEccles]

Yes, I believe this to be true too. I've looked at two stress cases one when tightening the bolt where I calculate torsional stress and axial stress ( with stress raisers applied to both )and determine the von mises stress.Here the axial stress is due to the preload and the torsional stress due to the applied torque minus the friction under the head.

The second case is when the bolt is assembled and subject to operational loads, here I'm looking at just axial stress with concentration factors for the head fillet and the threads, again calculating von mises for both positions. The operational loads are low so the assembly condition is the worst case in this situation.

The stress raisers are typically 3! I end up with really low torque values to reduce stress.

Maybe I just need more /stronger bolts ...

 

[CoryPad]

It is common to use combined tension + torsion during tightening of fasteners. Thus, high friction coefficient leads to higher screw stresses than low friction and this results in less joint preload available. However, it also is common to assume the torsional stresses are relieved during service, so only axial forces are considered. There is good information on this subject in the references shown in faq725-600.

Regards,

Cory

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

 

[strokersix]

I've pondered this question myself and come to the following conclusion:

If the high stress prediction for the underhead radius does not indicate fastener failure then that high stress must be offset by residual compression in the fillet, local yielding resulting in redistributed stress, or a combination of those two effects.

Someone please confirm or refute my conclusion.

 

[CoryPad]

Mostly local yielding that causes strain hardening and stress "redistribution".

Regards,

Cory

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

 

[strokersix]

Thank you Cory.

I suppose then that tensioning a fastener to 110% of design tension, releasing and then tensioning to design tension would improve fatigue life of the fastener?

A similar effect to new pressure vessels pumped full of water to yield, resulting in improved fatigue life?

In addition to seating the nut and underhead surfaces, is this effect part of the rationale of fastener supplier recommendations to tension their fasteners three times?

Thanks!

 

[Kenneth]

LeeEccles:

In additon to the excellent references provided in CoryPad's link, MIL-HDBK-60 (available online at

http://assist.daps.dla.mil/docimages/0000/22/90/70455.PD9)

discusses combined fastener tensile stresses due to this torsion load component in Section 5.5.1 (as well as providing some some nomographs for these additive stresses based on friction coefficients and number of threads per inch).

 

[CoryPad]

strokersix,

It depends if the 110% value takes the fastener into the plastic deformation region and removes some of the fasteners damage tolerance. "Joint conditioning" is one of the names given to "tighten-loosen-retighten". Like you state, this deforms all components (both small scale and large scale) and can improve friction variation.

Regards,

Cory

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

 

[LeeEccles]

Thanks guys for your valuable comments. I think you've confirmed the approach I have been taking.

 

[AJohnson]

LeeEccles,

One additional comment about stress concentration. You mentioned a value of 3 for the threads. In my recent experience, I have found that to be a good value when using rolled threads. I have found many references that use values from 4 to as high as 12 for cut or ground threads.
Also, if ordering a standard rolled thread bolt made from a heat treatable alloy, it will usually come in the rolled-then-heat-treated condition, as it is easier to roll the threads when the material is soft. However, this eliminates the residual compressive stress at the thread roots that is one of the great benefits of thread rolling. To ensure you don't loose this benefit, request heat-treated-then-rolled bolts from the supplier/manufacturer.

Andy