[rokugo0512]

Hello everyone.

I am trying to test the mechanical properties of a closed-foam porous solid.

I have created a 10x10x10 hollowed tetrakaidecahedron stack. The thickness of each tetrakaidecahedron shell is only 0.25mm (as provided in attachment).

When I try to mesh this model using Ansys workbench's automatic mesh settings, I always stuck at the progress when it shows "Meshing volumes...".
I waited once for at least 48 hours but it just won't go through. But if the shell thickness is 1mm, the meshing volume process will finish in around 15 min.

Is there anyway to solve this problem?

[amin.z]

Originally Posted by rokugo0512

Hello everyone.

I am trying to test the mechanical properties of a closed-foam porous solid.

I have created a 10x10x10 hollowed tetrakaidecahedron stack. The thickness of each tetrakaidecahedron shell is only 0.25mm (as provided in attachment).

When I try to mesh this model using Ansys workbench's automatic mesh settings, I always stuck at the progress when it shows "Meshing volumes...".
I waited once for at least 48 hours but it just won't go through. But if the shell thickness is 1mm, the meshing volume process will finish in around 15 min.

Is there anyway to solve this problem?

I guess you’re encountering this problem deu to size of your elements, specially cause you’re using automatic mesh, ansys probably trying to generate a relatively fine elemnts in your domain, and since your part is too thin, the elements have to be too fine as well,
I can now think of two ways,
First , if possibel, would suggest to model a smaller part of your domain, considering a periodic or symmetry boundary conditions, the smaller you bring in meshing, the better you can estimate and solve your problem, e.g. bring a 1/8 of your whole domain, mesh it with the same method you were using earlier, now i guess the mesh would be generated, check the size and number of elements, I guess it’s probably too fine which causing a problem when you’re trying to mesh the whole domain. Does your computer have enough resource to generate the same quality for the whole domain which has for instance 8 times more cells?
Second idea is forgetting using the default value and auto mesh, go ahead and either give an elemnt size to your domain, or use a coarser size in default values of auto mesh, which probably gonna fail if you have a complex domain with curvature and narrow gaps

[rokugo0512]

Thanks a million!!!
I will try it right away.