Mesh refinement with FEMAP V9.2

[adnanh]

Hi,

I have been busy doing some contact analyses, however each time I refine the mesh by specifying the maximum element size, the maximum contact pressure increases. Does somebody know any tutorial or example for FEMAP on how to refine the mesh at a specific location. If I specify the maximum element size only, the mesh is refined as a whole which results in a much heavier model than necesary. It would be more efficient to just locally refine the mesh (in the neighbourhood of contacts).

Thanks in advance.

 

[adnanh]

By the way, it would be very helpful to have an option for a sort of autmatic mesh refinement near contacts. There is an option for automatic contact detection between various parts in an assembly, but automatic mesh refinement near these contacts would make the whole thing much more complete. Is there such an option perhaps?

Regards.

 

[ProbaSci]

I've been using v8.3, so this may not exactly apply to your version.

You can use the Mesh > Mesh Control > Size along curves or Size on Surface commands to locally refine the surface mesh before extending it into 3D. Be careful that you keep mesh continuity with adjacent curves/surfaces.

I've also had luck specifying a very fine mesh in my area of interest, a larger mesh away from my area of interest, then allowing the auto mesh to transition between the two.

http://www.probasci.com

-
Implantable FEA for medical device manufacturers

 

[adnanh]

Hi,

I have been trying using Control Mesh -> Size along Curves. The problem is that FEMAP sees two curves at an edge where two parts meet (contact region), i.e one curve for one part and one for the other part. Especially, when you apply biased curve devision, the mesh continuaty between the two parts is not guaranteed, because the two curves have often different length. How to address this problem? Or is mesh discontinuity allowed in contact regions? (I think I have read somewhere that the two meshes don't need to be continous in region of contact).

Regards,

Adnan

 

[ProbaSci]

Have you tried selecting both curves at the dividing line and use a constant mesh size?

You could also try slicing the solid into a smaller section at your region of interest and treat as two solids with matching meshes at the interface.

http://www.probasci.com

-
Implantable FEA for medical device manufacturers

 

[R1100S]

Yes, the newer versions of NEi Nastran (I'm assuming that is what you are using) allow for mesh discontinuity between contact surfaces.