Weird behavior with cyclic symmetry, plasticity and contact

LucasGLucasG
Hello!

I've been trying to run a 1/12th model of a bolt tightening + pressure loading simulation of an API flange.

I've previously managed to run it without symmetry and the run times weren't satisfactory plus I was running into non-convergence so I decided to switch to the 1/12th model to troubleshoot faster.

I've since run into the weird phenomena below, and haven't found yet what could be causing it. I was wondering if anyone could shed some light in how to properly set up this model and/or what I might be doing wrong?

1) Whenever I set the output to anything more than 12 segments, I get a MecWay crash once the very first increment is done.



2) I get this weird stress concentration on the ring gasket and its channel at the symmetry faces. See picture 1. If I run with sliding support it seems to behave properly however I'm not sure if this is the proper boundary condition choice.




I've tried switching solvers, refining and coarsening the mesh in the circumferencial direction, unstructured mesh on the blind and weld neck flanges, changing contact settings, removing plasticity, different boundary conditions... Nothing seems to have an effect.

Thanks in advance!

Comments

  • VictorVictor
    I'm not sure why but it doesn't happen with linear elements. You're fortunate to have a hex mesh so they should perform pretty well.


  • disladisla
    -Isn't Cyclic symmetry and elastic supports on both symmetry faces redundant BC?.
    -I miss the internal pressure on the gasket it self.
    -There is an elastic support applied to the inside area of the gasket and bolt which seems unrealistic.
    -Symmetry planes are not planes sharing common z value.

    In my opinion all those small details may be giving an unsymmetryc prefered direction to the displacements.


Sign In or Register to comment.

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In Register