[BahaZero]

Hi,

I'm having some difficulty in achieving a 1:1 translational periodic mesh in ICEM using an unstructured hex dominant method. My current geometry consists of a hot side fluid volume, two rows of cooling holes, and a cold side fluid volume. The holes are oriented in a very shallow angle to the surface (approx 25 deg) and in order to not have my cutting plane slice through the holes, I have offset the cold fluid side in the same direction as my periodic normal. This way, I can maintain higher quality elements all the way through the hole.

I set my periodic condition using the "global mesh setup" as per the manual and other forum posts. My offset is at 0 0 24.2 where 24.2 is the offset in the Z direction. Since my mesh is unstructured, I do not believe there are any additional settings to make.

After I generate my surface mesh, the mesh checker says "shell XXX has node XXX which has no twin". On close inspection, both my periodic pairs (hot and cold fluid regions) are indeed not matched.

Can anyone shed some light on what the cause is? Could it be due to my cold/hot zone offset?

Here are some things that I have examined:
- duplicate points or curves on periodic planes
- used "show point info" to check my periodic offset
- grouping associated points and curves to the periodic faces
- separating each periodic face into different parts

I have read that one could generate the mesh on one side and then copy it to the other. Can someone also explain how this works with inflation layers? Am I limited to certain methods in generating my volume mesh afterwards?

Sorry for asking so much at once but any comment would be greatly appreciated. I have been stuck with this problem for days... T_T