Learning Forum

[Leo Dalla Caneva]

Hello all,

 

I am currently trying to solve a convergence problem. The solver output indicates that some elements have become highly distorded during the simulation. I would like to identify the ID of these elements with the ESCHECK command but i don't know how to use it properly.

I tried to insert a command in the Solution section like this :

Then RMB ----> Excecute post commands

But when i check the Post Output nothing related to this command can be seen.

I precise that the simulation was not performed on my computer but on a HPC server. 

 

Any input would be greatly apprecied,

 

Leo

[Asitang Lahurikar]

Hi,

If you are having trouble with convergence, perhaps due to excessive mesh distortion, you might consider applying load more gradually i.e. by increasing the substeps. If that doesn't work, you should try improving the mesh and checking the stability of the material. Additionally, you may consider using the Nonlinear Mesh Adaptivity (NLAD) approach if the problem persists.

Hope this helps!
Regards,
Asitang

[Ashish Khemka]

Hi Leo, 

 

In addition - please watch this video which might help you in viewing the distorted elements and some suggestions to overcome this issue:

Handling Element Distortion Errors in Hyperelastic Material Models Using Ansys Mechanical — Lesson 3 - YouTube

Ansys Mechanical mesh metrics will help you to see the element quality before solution:

Check Mesh Quality (ansys.com)

Regards,

Ashish Khemka

[Leo Dalla Caneva]

Hello,

 

Thank you to both of you for your suggestions.

 

I am already applying the load gradually, actually my time substeps setting is :

I also refined the mesh in order to have a good mesh quality according to the "element quality" metrics, worst element has a value of 0.1, and i have only 3 elements with a value below 0.2.

What i want to do is identify the location of the distorded elements, to perform more specific corrective actions on the mesh or on the model. It seems like saving newton raphson residuals can help to identify the problematic regions but it is limited to the number of NRR we implemented.

 

Regards,

Leo

[Asitang Lahurikar]

Hi Leo, 

Requesting and checking for Newton-Raphson Residual plots could be the best way to know and locate the regions in your model that led to convergence difficulties. It might be a good idea to request for about 5 Newton-Raphson residuals to begin with. If you haven’t already, I would suggest you to give this a try. Additionally, you may use "Identify Element Violations" option that is available under "Solution Information" to find the problematic elements. To know more about it, please refer to the help document here - 25.2.19. Element n Located in Body (and maybe other elements) Has Become Highly Distorted (ansys.com)

Regards,
Asitang