[BenG]

Dear all,

I am using ANSYS Fluent Release 17.1 to simulate a compressor wind back seal. I have extracted the cavity in the seal filled with oil and try to mesh it in Fluent. Here below you can see an extract of the model (volume shown is filled with oil).
https://www.dropbox.com/sh/gks819w6p...TBaIlp-ia?dl=0

The seal diameter is 80mm. The total thickness is about 0.3mm and the area with the groove is only 0.1mm (highlighted green). As the inner diameter of the seal is rotating, I would need about 10 layers in radial direction for a reasonable CFD simulation. So far I could not find a way to achieve that. Does anybody have any ideas or suggestions how to achieve that?

Many thanks in advance for your support
Ben

[BlnPhoenix]

Why is it not achieveable? Is the amount of resulting cells to high?

[BenG]

Originally Posted by BlnPhoenix

Why is it not achieveable? Is the amount of resulting cells to high?

Hi Phoenix,
Thanks for your reply. The problem is that the min. element size where I would get good results is around 0.01mm. This would give me theoretically 10 elements radially at the narrow clearance of 0.1mm, but the total number of elements explodes to an almost infinite number. So far, I only achieved 3 elements radially and ended up with a number of elements around 20mio. Any finer mesh results in my PC not being able to build the mesh even after several hours.

Ideally, I would like to use a hexahedral mesh similar to the example shown here:
https://www.youtube.com/watch?v=0j915gwpq4A

This mesh reduces the number of elements a lot, but should still provide good results. However, due to the fact that the model has this helical groove, I did not manage to achieve a hexahedral mesh and had to use a tetrahedral mesh instead.

At present, I have only one potential solution to achieve a reasonable mesh, which I did not yet try and this is to cut the model in two parts: helical and thin walled pipe. The individual components could most probably be meshed much simpler and I guess there is a command to combine the two meshes afterward. This solution, however, seems to me a rather difficult one and I don't know if it works.

Do you have any ideas/suggestions?

Thanks and regards
Ben

[BenG]

Does nobody have any ideas how to get a decent mesh on this wind back seal? I would really appreciate your support :-)

[alibomayaye]

Maybe do great convergence mesh analysis?


Sent from my iPhone using CFD Online Forum mobile app