Facing issues on running simulation

Project Support
1

Hi,
This project is a nonlinear static structural analysis.

I have tried to constrained my geometry properly. However the simulation failed !
Probably there is an issue in my physical contact. I am not sure how to define the contacts so that no rigid body motion occurs and simulation runs successfully.
Can anyone help me?

corrugated core_uniaxial compression | SimScale Workbench

1 Like
2

Hi, and thanks for using the forum!

Please perform these changes and try again:

  1. The remote displacement BC is redundant, you can remove it
  2. Change the ‘Convergence criteria’ under Numerics to ‘Absolute’ - this helps with the displacement-driven simulation.
  3. You need to constrain the middle part. I would add an Elastic Support BC to do so.

And don’t forget to let me know of your results!

3

Thank you.
If I remove the remote displacement BC, will not the upper plate act as a deformable body when try to compress the middle part? But I want it as a undeformable one.

4

Welcome!

Yes, it will be deformable, but in my opinion that is desired. Anyway it will not bend as you are applying the displacement uniformly. It should compress, reducing the thickness, but as the thickness is so low, you will not notice the deformation.

Also, a rigid part should have special treatment to not cause divergence in the solver. Moreover, we currently do not support rigid parts.

5

Now it runs successfully.
However from the figure below it is visible that the middle part is penetrating the bottom part. How can I stop such penetration?
image

Moreover, I defined elastic support for the only two faces of the middle part, should I define elastic support for other faces too?

image
image841×436 54.8 KB

6

Hey!

You should add the faces at this end to avoid the penetration. Currently they do not belong to the physical contact, thus the interaction is not captured.

The elastic support looks ok to me, specially because it worked.

Also, I wonder why the interaction to the bottom plate is a bonded contact. Is it on purpose or should it rather be a physical contact?

7

Sorry I can not figure out this.
Should I modify the imported geometry?
Probably, I should rather use a physical contact for the bottom part to avoid penetration, is that alright? Is this bonded contact liable for the penetration?

8

Sorry if I was not clear.

You are right, to avoid the penetration you need a physical contact to the bottom plate.

Also, don’t forget to assign the end faces of the middle part to the physical:

image
image819×776 47.9 KB

image
image813×778 67.5 KB

9

The following modifications I have done on the model:

  • removed all the bonded contact
  • redefined the physical contact (add faces as master/slave assignment)
  • increased the penalty coefficient
  • modified the elastic support (add other two faces)
    Despite all, I failed to run this. Following error message appeared:
    image
    Probably there’s issue with my physical contact. How can I fix it?

corrugated core_uniaxial compression - 2 | SimScale Workbench

10

Hi!

I can no longer access your project, so blindly I would say that you can decrease the penalty coefficients to the original value, this could avoid the convergence problem.

11

I have shared the project with you, can you go through it?

I also tried with the original value of the penalty coefficient but it failed.

12

I tried to run it several times with different boundary conditions.
Would you help me to fix the BCs and the contacts to have a solution?

corrugated core_uniaxial compression - 2 | SimScale Workbench

13

Hi!

This is a difficult case. Maybe you need to take a step back and think, what is supposed to happen to the middle plate? As it is right now, it can slide away, maybe even to the side , or deform in multiple ways. This should help you to review your boundary conditions.

14

Some ideas from looking at the results:

  • Make the flat plates bigger (if realistic) so that the middle part does not slip out.
  • Refine the mesh on the middle part.
  • Cut in half around the XY plane, and use a symmetry BC with fixed value Uz = 0 on the cutting plane.
  • Fix one of the ends with Ux = 0 to allow to grow only in one direction.
15

I tried to follow the instructions however it showed error message.
Would you a look at the settings (static 3)?
corrugated core_uniaxial compression - 2 | SimScale Workbench

16

Hey, you did not follow my instruction:

Please just remove the Symmetry Plane BC

17

Sorry for my mistake.
Now, it is showing, occurred problem during material model integration…
I am not sure what would be my next step.

18

This is a hard one.

Please refine the mesh on the middle plate to have at least 2 elements across the thickness. 3 to 4 would be ideal, but I am not sure if the increase in the mesh size is practical.