snnappyHexMesh with cyclic boundary conditions

longamon · April 11, 2011, 09:20

Hello, I keep getting this error:

Non-uniform separation vector not supported for point or edge fields.

when I try to use snappyHexMesh in parallel. I first decompose my case, have tried several methods (scotch, metis, simple, hierarchical and parMetis) A lot of the times it says a cyclic is being split and it will generate errors, this I understand and the only way I can avoid it is by using hierarchical, however, when the cyclic isn't being split up the Non-uniform separation vector error show up and I don't know what it means.

This error also appear when attempting to use refineMesh in both parallel and non parallel.

When the error is shown in parallel the patch that throws the error is the processor boundary. When the error shows up in serial run the patch that throws the error is the cyclic patch.

thanks in advance
David

wyldckat · April 30, 2011, 07:43

Greetings David,

I guess I'm a bit late answering your question, but I only found about this just now. The answer for me came from this post: http://www.cfd-online.com/Forums/ope...tml#post241413 #8

The usage reference in the OpenFOAM code about this is here:

Code:

applications/utilities/parallelProcessing/decomposePar/decomposeParDict

Best regards,
Bruno

longamon · May 9, 2011, 13:28

This seems to work only with Hierarchical, not with parMetis.

April 11, 2011, 09:20	snnappyHexMesh with cyclic boundary conditions	#1
longamon Member David Aljure Join Date: Mar 2011 Location: CTTC Universidad Politécnica de Catalunya. Spain Posts: 38 Rep Power: 15	Hello, I keep getting this error: Non-uniform separation vector not supported for point or edge fields. when I try to use snappyHexMesh in parallel. I first decompose my case, have tried several methods (scotch, metis, simple, hierarchical and parMetis) A lot of the times it says a cyclic is being split and it will generate errors, this I understand and the only way I can avoid it is by using hierarchical, however, when the cyclic isn't being split up the Non-uniform separation vector error show up and I don't know what it means. This error also appear when attempting to use refineMesh in both parallel and non parallel. When the error is shown in parallel the patch that throws the error is the processor boundary. When the error shows up in serial run the patch that throws the error is the cyclic patch. thanks in advance David

April 30, 2011, 07:43		#2
wyldckat Retired Super Moderator Bruno Santos Join Date: Mar 2009 Location: Lisbon, Portugal Posts: 10,982 Blog Entries: 45 Rep Power: 128	Greetings David, I guess I'm a bit late answering your question, but I only found about this just now. The answer for me came from this post: http://www.cfd-online.com/Forums/ope...tml#post241413 #8 The usage reference in the OpenFOAM code about this is here: Code: applications/utilities/parallelProcessing/decomposePar/decomposeParDict 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
symmetry boundary conditions in cfx	lost.identity	CFX	41	May 22, 2013 08:21
Cyclic boundary conditions in parallel	thibault_pringuey	OpenFOAM Bugs	1	April 2, 2011 16:21
Boundary conditions between blocks	maxNeptune	OpenFOAM	0	March 19, 2011 11:16
periodic and cyclic boundary conditions	Hadi	Main CFD Forum	2	June 29, 2007 08:19
A problem about setting boundary conditions	lyang	Main CFD Forum	0	September 19, 1999 19:29

May 9, 2011, 13:28		#3
longamon Member David Aljure Join Date: Mar 2011 Location: CTTC Universidad Politécnica de Catalunya. Spain Posts: 38 Rep Power: 15	This seems to work only with Hierarchical, not with parMetis.