Create coarser mesh for post-processing

Horus · March 2, 2012, 17:52

Hello!

My current mesh has about 23 M cells. It's fine for solving but somewhat too large for post-processing. Is there any possibility to coarsen an existing OpenFOAM mesh including results?

Thanks!

wyldckat · March 2, 2012, 18:51

Greetings Florian,

mapFields perhaps? http://www.openfoam.org/docs/user/ca...5-280002.1.6.3

Although I'm not sure that it can work from a high-res mesh to a coarse mesh, since it's usually done the other way around.

Best regards,
Bruno

gschaider · March 3, 2012, 09:36

Quote:

Originally Posted by wyldckat

Greetings Florian,

mapFields perhaps? http://www.openfoam.org/docs/user/ca...5-280002.1.6.3

Although I'm not sure that it can work from a high-res mesh to a coarse mesh, since it's usually done the other way around.

Best regards,
Bruno

I think that shouldn't be the problem. What he is asking for is an autoUnrefineMesh-utility. To my knowledge there is no such thing. What I'd do is: write the patches as STL (one of the surface-utilities should do that). Then use snappyHexMesh to generate you a coarse mesh (doesn't have to be high quality too)

Horus · March 5, 2012, 04:19

I think mapFields would be so way to go. I've created a coarser mesh by setting the base size in my mesher (Spider) to a higher value. Curiously the mesher then avoided creating some patches it should have and consistent mapFields failed.

Ok, I'll be investigating that issue whith the mesher, thanks so far!

(if anyone knows about that behavior with Spider I appreciate a reply of course)

LVDH · March 5, 2012, 08:03

Hi,
I frequently do this.

Here is my approach:

1. I copy the case and call it _CM at the end.
2. I modify the blockMeshDict so it encloses a smaller volume. I do not need the farfield for visualizations.
3. I modify the snappyhexMeshDict. Here I keep a high refinement on the surfaces. I lower all other refinement levels by 2 (eg from 7 to 5). I only keep one cell between levels and a switch off the surface layers. This gets me from about 70e6 to about 8e6 cells.
4. I map the fields from the original case to the _CM case.
Done!

March 2, 2012, 17:52	Create coarser mesh for post-processing	#1
Horus Member Florian Join Date: Nov 2009 Posts: 59 Rep Power: 16	Hello! My current mesh has about 23 M cells. It's fine for solving but somewhat too large for post-processing. Is there any possibility to coarsen an existing OpenFOAM mesh including results? Thanks!

March 2, 2012, 18:51		#2
wyldckat Retired Super Moderator Bruno Santos Join Date: Mar 2009 Location: Lisbon, Portugal Posts: 10,975 Blog Entries: 45 Rep Power: 128	Greetings Florian, mapFields perhaps? http://www.openfoam.org/docs/user/ca...5-280002.1.6.3 Although I'm not sure that it can work from a high-res mesh to a coarse mesh, since it's usually done the other way around. Best regards, Bruno __________________ OpenFOAM: FAQ \| Getting started Forum: How to get help, to post code/output and forum guide Read this before sending me PM

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[snappyHexMesh] Number of cells in mesh don't match with size of cellLevel	colinB	OpenFOAM Meshing & Mesh Conversion	14	December 12, 2018 08:07
Gambit problems	Althea	FLUENT	22	January 4, 2017 03:19
Converting Starccm+ mesh	Ladnam	OpenFOAM	0	September 14, 2011 06:30
Help ! Create surface for sliding mesh.	YH Tan	FLUENT	0	October 5, 2008 04:21
unstructured vs. structured grids	Frank Muldoon	Main CFD Forum	1	January 5, 1999 10:09

March 5, 2012, 04:19		#4
Horus Member Florian Join Date: Nov 2009 Posts: 59 Rep Power: 16	I think mapFields would be so way to go. I've created a coarser mesh by setting the base size in my mesher (Spider) to a higher value. Curiously the mesher then avoided creating some patches it should have and consistent mapFields failed. Ok, I'll be investigating that issue whith the mesher, thanks so far! (if anyone knows about that behavior with Spider I appreciate a reply of course)

March 5, 2012, 08:03		#5
LVDH Member Andre Z Join Date: Dec 2009 Posts: 75 Rep Power: 16	Hi, I frequently do this. Here is my approach: 1. I copy the case and call it _CM at the end. 2. I modify the blockMeshDict so it encloses a smaller volume. I do not need the farfield for visualizations. 3. I modify the snappyhexMeshDict. Here I keep a high refinement on the surfaces. I lower all other refinement levels by 2 (eg from 7 to 5). I only keep one cell between levels and a switch off the surface layers. This gets me from about 70e6 to about 8e6 cells. 4. I map the fields from the original case to the _CM case. Done!