[Brunno]

Hello,

I am working on a test case of a swirl atomizer for propulsion applications. Despite those well-known edge problems, snappyHM works fine for the most part. So far, I can't figure out how to make the boundary layer "grow" on the entire cylindrical patch and minuscule tangential channel walls (see attached figure and addLayers settings). It seems that BL attaches only when the mesh is coarse. Does anyone have a clue ?
Any help is greatly appreciated.

[bjr]

I'm working on similar issues, but am a little behind you. I will be investigating some of your layer control settings on my cylinder case asap.

[bjr]

Did you ever get boundary layer resolution and viscous drag results you liked?

[Brunno]

Hello ben,

Thank you for the response.
Boundary layer resolution was acceptable on previous test cases. I am currently running the inviscid, single-phase flow case, so I don't have any viscous results at hand.

[bjr]

Is there any chance you could provide any said test cases? I'm really looking to put together a case that starts from just an stl file, a blockMeshDict, and a snappyHexMeshDict that potentially yield good drag results. I think this would be a highly valuable starting point for many out there without access to commercial gridding tools to start learning quickly.

I have a location that large files can be publicly uploaded and downloaded from if needed.
http://public.me.com/ben.racine

I totally understand if you don't have the chance/time though... days are short.

Many thanks,
Ben Racine
i3enhamin@gmail.com

[bjr]

Brunno kindly supplied me with this orifice case... http://files.me.com/ben.racine/1p3dei

[Mo-ITB]

sorry, i did a mistake in paraview, the mesh was even worse in 1.6.x

[linnemann]

Hi

What is the maximum cell limit in the snappy dict?

I had similar problems once and upping the local and global cell limit made it possible to get the desired mesh.

The entries are there to prevent an explosion in cell count.

[openfoam_user]

Hi Bruno,

did you solve you problem ? How ?

I have encountered the same problem and don't know what to do until now ?

I get layers only where there is more refinement !

Regards,

Stephane.

[Brunno]

Hello Stephane,

Yes, I did solve the problem - not exactly through snappyHexMesh, but using a combination of adaptive mesh (see the dictionaries cellSetDict and refineMeshDict) based on the gradient of velocity (magU) and snappyHM. The results are not always acceptable - it's usually a trial-and-error process, but it worked for me.

Good luck !

[openfoam_user]

Hi Brunno,

thanks for your advice.

I think that the problem for me is a little bit different. I would like to add several layers around my body in order to obtain Y+ near 1, if possible.

Stephane.