CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Meshing & Mesh Conversion (https://www.cfd-online.com/Forums/openfoam-meshing/)
-   -   [snappyHexMesh] Mesh becoming coarser during addLayers in sHM (https://www.cfd-online.com/Forums/openfoam-meshing/83060-mesh-becoming-coarser-during-addlayers-shm.html)

Arnoldinho December 13, 2010 11:49

Mesh becoming coarser during addLayers in sHM
 
1 Attachment(s)
Hi,

I set up a test case for testing snappyHexMesh and especially the layer addition possibility.
I noticed that after the addLayers process, the surface mesh of my structure has become coarser, i.e. points and edges from the fine surface mesh (time 0.06) seem to have been used for the layer creation (time 0.09, vertical edges). Interestingly, this only happens on the cylinder part, not on the box part.

How can I stop this, as it is destroying my mesh in certain locations of the 'real' structure?

Pictures. left: after snapping procedure 0.06 http://img3.imageshack.us/img3/9049/time006.th.png right: after addLayers procedure 0.09. http://img153.imageshack.us/img153/2715/time009.th.png

The sHM dict is attached.

Arnoldinho December 14, 2010 03:04

Ok, let's ask another question: Is this a 'normal' behaviour of sHM, i.e. does this happen in a similar way in your cases, or not?

JimKnopf January 21, 2011 08:17

Do you have some pictures? I'm not sure about the phenomena you're describing

THX
jim

shankar1221989 February 22, 2011 06:44

Hello Arnoldinho,
Your observation is right, i too have seen this strange behavior while generating boundary layer padding.
As of know i am also not aware of the real reason.

eugene April 6, 2011 12:20

The volume mesh isn't getting coarser. You are just observing a surface effect. Before the layers are added, SHM tries to merge all boundary faces in a cell that are close to co-planar to improve quality and reduce the number of added layer cells. The "coarsening" you are seeing on the cylinder is just this. Make a cutting plane normal to the cylinder and you will see. The reason you don't observe this on the box is because there much fewer same-cell co-planar faces on a coordinate aligned box.


All times are GMT -4. The time now is 06:55.