Strange visual artifact in a dynamic simulation

Bug Reports
1

Hello everyone,

First, thank you Matej for the new version and for everything you’ve done for the community.

I tried to evaluate the shape of a belt hanging on a hook. I used a dynamic simulation and looked at the steady-state result, because I think this way we can get around the stability problems of a classic nonlinear static simulation. There was no problem with the calculation, but when I open the results, something strange can be seen at the top. Do you think my model is wrong? The specific element is severely distorted during the simulation. Is that a bug or there is a way to fix that ?

Here is my simulation :

Belt.zip (449.1 KB)

Thanks for taking the time to look at this :slight_smile:

image
image905×692 73.7 KB

image
image740×512 56.5 KB

image
image1025×322 47.9 KB

image
image1120×227 45.9 KB

2

I don’t see so large distortion in your results:

image
image1615×634 40.6 KB

The increments are different though, so it seems to be from a different run of the model.

Anyway, did you try remeshing and using solid element layers ? Shells may not work well in CalculiX, especially for large deformations.

Also, your damping seems very high, you should try reducing it.

1 Like
3

Thanks! Indeed, I had a previous result. But since I added damping, we’re only interested in the steady-state response anyway, so the transient is not physically meaningful, by the way. Why so suggest to put a lower damping ? @FEAnalyst

4

Yeah, but mass-proportional damping coefficient of 6 /s may damp the system so much that it becomes too slow to reach the desired state. Also, it may cause some spurious effects like maybe the issue you observed.

If you need more damping in a dynamic implicit step to run a quasi-static analysis, you should mostly rely on the alpha parameter set independently of the Rayleigh damping. Unfortunately, CalculiX doesn’t provide as many controls for dynamic implict steps as Abaqus (which has a separate quasi-static application type with high damping and some changes in the time integration scheme).

1 Like