Learning Forum

[b.claudet]

Hello,

I am working on a case where i have a very large assembly.

It is essentially a battery pack containing 52 batteries composed by a very long and thin cover with a really complexe geometry, a battery case beneath it  that is just a box but with 52 holes in it in order to fit the batteries that are cylindrical and a thin cover at the bottom just like the one at the top.

I would like to know how can i reduce as much as possible the number of elements generated by meshing given the fact that in order to obtain a good quality mesh, the pack in the middle is the one that is requiring the biggest amount of elements, and that i can not just simplify it into a shell given that i have a large thickness and neither apply a shell to the covers given their complex geometries.

Thank you for any response.

 

[Matt Mehalic]

It is unclear whether the challenge is the complexity of the geometry for converting into a surface body, or whether the "thin cover" is too thick to be modeled with shell theory. If complexity is the challenge, one option is to extract the top (or bottom) surface of the cover (you can do this with a simple copy/paste operation in Ansys Discovery or Ansys SpaceClaim geometry editors) and then use the shell offset feature on the mesh instead of attempting a midsurface operation. If it is indeed too thick to model with shell theory, depending upon the type of simulation you want to run, there may still be some value in starting off with the shell approach at least for a first approximation. Then shell to solid submodeling in key areas of interest may also be an option for gaining the added insights afforded by a solid element mesh without having to use solids throughout the entire cover. If you are forced to model the cover with solids, then geometry simplification and mesh-based defeaturing and virtual topology may be your best approach at reducing mesh count.