Learning Forum

[bahar27]

hello,

 

I  want to use inflation in multizone, I used edge sizing, so the inflation does not work. If I remove edge sizing, the mesh does not look good and changes from hexa to prisms. What can I do?

[peteroznewman]

Slice the geometry up into six-sided bodies and you can keep the hex elements.

[bahar27]

I divided it into six-sided bodies and the inflation worked, but when I tried to do it for the second time it did not work anymore and the shape of inflation layers gets weird when I want to repeat the process or change something.

the procedure is as follows:

 edge sizing, then the method which was multizone, and then the inflation and after all the face meshing for the center.

 

 

.

[bahar27]

I have also tried to divide the geometry the other way, but inflation does not work or gets weird.

[Rob]

That's an O-grid or butterfly mesh. You may want to use a pave mesh in the middle rather than the decomposition. 

[bahar27]

the inflation is not dependant on the face mesh in the middle, the inflation behaves weird when I change the order of edge sizing. Is multizone a good solution? and should I divide the geometry into six-sided bodies? is the order of meshing right? why the inflation gets weird(in pic 2)when I change something such as edge sizing?

 

[Rob]

Edge sizing will influence the face mesh, and is affected by the inflation. If settings conflict you'll get some weird (and undesirable) effects. My approach with Workbench meshing is to use as few local settings as possible, and instead use the global size settings (specifically size functions). More usually, I use Fluent Meshing and/or a lot of tet cells: depends on what you want to do with the mesh. 

[singh33]

Dear rwoolhou,

Does the sequence of the local setting selected do affect the meshing?

Thanks in advance

[Rob]

Potentially, yes. The order you mesh volumes has a greater effect.