FEA constraints :(

I am new to FEA constraining and I can not seem to properly constrain my model.

I have a public project with a lot of confusing errors for all the sim runs I have tried for my static nonlinear simulation with Pressure following.

The object is basically a pressurized balloon.

I have made the geometry a 1/4 shape with two symmetry planes in the X and Y directions.

I have not made a Z symmetry plane because gravity in the Z direction will make the balloon unsymmetrical in the Z direction.

In ‘Run 3’ I simply used the following boundary conditions:

  • Symmetry Plane X (all faces on the X symmetry plane)

  • Symmetry Plane Y (all faces on the Y symmetry plane)

  • Pressure Follower INTERNAL (all internal faces selected)

  • Fixed support node (a single node at the very top of the balloon is selected)

Here is a shot of the node I selected:

On ‘Run 3’ I get this error:

But when I click on ‘View’, to see the error, I am unable to find any Event Log or Solver Log to find out why the sim run failed.

The only run where an error is actually visible is ‘Run 1’ where it points to constraining error, which is why all my other runs attempt to constrain it properly, but they have all failed without being able to see any error text.

Any help would be appreciated…

Dale

Hi @DaleKramer,
there seems so be an issue specific to constraints on single nodes - until it’s fixed I would propose a setup that is similar as in my project - just with an additional symmetry but still with the remote displacement on the node - now only constraining Z.

Best,
Richard

Thanks Richard.

Your constraints still confuse me.

You don’t seem to have a symmetry plane boundary condition at all, you simply constrain the faces on what should be a symmetry plane such that the faces can not move off that plane. What is the purpose of the ‘Symmetry plane’ boundary condition if you don’t really need to use it? Also, what is the geometry primitive ‘Half space’ used for?

Also, your ‘Remote displacement 3’ acts on faces, not nodes as you suggest above.

I will keep trying.

Dale

Hi @DaleKramer,
I try to avoid the symmetry BC as it sometimes can lead to numerical troubles together with other BCs on shared nodes. So as long as the faces which I want to put a symmetry BC on are orthogonal to one of the coordinate axes I tend to use an equivalent fixed value BC and only use the symmetry BC if this can’t be used e.g the surface normal is not aligned with any of the axes.
The half space geometry primitive can not be used for anything explicitly - it is used to define the half space for the mesh clip.

still with the remote displacement on the node - now only constraining Z.

Sorry, I wanted to say “nose” instead of “node”.

Best,
Richard

@rszoeke

I finally got a sim run to finish (‘Run 5’) without errors, but something is still not quite realistic in the constraints:

Any ideas?