Fatigue Analysis of Bolted Joints

[wnmascare]

Hi everyone!

I'm trying to run a high cycle fatigue analysis of a bolted flange submitted to bolt preload and a bending moment varying harmonically. The objective is to analyze only the flange, not the bolts.

The first load step is the preload and the second is the application of a bending moment of 1 kN.m. I used Abaqus to run them. The figure attached illustrates the Mises stress distribution at the end of the second load step, which accounts for the stress state generated by the preload step. The bending moment was applied at the right end of the flange.

For the fatigue analysis, I organized a spreadsheet to perform the calculation, based on the stress results of the preliminary FEA analysis. Thus, for the fatigue analysis, should I subtract the stresses of the preload step from the stresses of the bending moment step and use it for the calculation of the damage?

Any advice will be appreciate.

Thank you very much.

 

[rb1957]

aside from all the obvious questions to give you a sensible answer (material, mesh, etc ...)
you're interested in the bolted flange (and not the obvious Kt on the shoulder (away from the flange joint) ?
Von Mises is not the best stress formulation to use ... you lose sight of the compression in the flange (due to preload). The flange should see tension stress only after the joint gaps.
if you like canned solutions, possibly investigate "ncode" (which does fatigue calcs on FEMs).

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

 

[FEA way]

Check the article "Analysis of Additional Load and Fatigue Life of Preloaded Bolts in a Flange Joint Considering a Bolt Bending Load" by I. Okorn et al. It can help you, especially since its authors also use Abaqus. And for that software, there's a dedicated tool called fe-safe that can handle advanced fatigue calculations.

 

[wnmascare]

Guys, thank you for your prompt replies.

The company I work for doesn't have nCode or Fe-Safe licenses, then we have to get by on our own.

Yes, I'm interested in determining the fatigue life of the flange only, not the bolts. I took a training on nCode almost ten years ago and I vaguely remember that the instructor quickly mentioned that, when dealing with bolted flanges, the stress from the bolt preload step must be subtracted from the varying stress state. We didn't do any example of bolted flanges at that time. And now I'm facing this kind of problem.

The article "Analysis of Additional Load and Fatigue Life of Preloaded Bolts in a Flange Joint Considering a Bolt Bending Load" deals only with the fatigue strength of the bolts.

 

[rb1957]

but the flange should be in a state of compression (making fatigue unlikely) ... no?

what material ? steel ??

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