decomposePar changes sign of surfaceVectorField at processor boundary

hertzsprung · June 16, 2014, 08:45

Digging through the source code I found the a comment in src/parallel/decompose/decompose/fvFieldDecomposer.H:

Code:

//- Processor patch field decomposer class. Surface field is assumed
//  to have direction (so manipulates sign when mapping)

which is the wrong assumption for my wind vector field. What's the best way around this?

hjasak · June 17, 2014, 02:46

Are you sure you know what you're doing? This works on surface fields such as flux, where the neighbour processor geometry flips the face area vector. This is the reason for sign change.

If you have a surfaceVectorField, simply re-interpolate in parallel or evaluate eg. Sampling in parallel. Without the change of sign, fluxes would be wrong.

Hrv

June 16, 2014, 08:45	decomposePar changes sign of surfaceVectorField at processor boundary	#1
hertzsprung New Member James Shaw Join Date: May 2014 Posts: 8 Rep Power: 11	Digging through the source code I found the a comment in src/parallel/decompose/decompose/fvFieldDecomposer.H: Code: //- Processor patch field decomposer class. Surface field is assumed // to have direction (so manipulates sign when mapping) which is the wrong assumption for my wind vector field. What's the best way around this?

June 17, 2014, 02:46		#2
hjasak Senior Member Hrvoje Jasak Join Date: Mar 2009 Location: London, England Posts: 1,905 Rep Power: 33	Are you sure you know what you're doing? This works on surface fields such as flux, where the neighbour processor geometry flips the face area vector. This is the reason for sign change. If you have a surfaceVectorField, simply re-interpolate in parallel or evaluate eg. Sampling in parallel. Without the change of sign, fluxes would be wrong. Hrv __________________ Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
decomposePar 4-core warning/error?	Boloar	OpenFOAM Bugs	23	April 8, 2014 08:57
Herschel-Bulkley non-Newtonian viscosity model has term with sign error	pbryant	OpenFOAM Bugs	5	June 18, 2013 23:53
decomposePar gives errors	of_user_	OpenFOAM	1	July 4, 2011 05:27
decomposePar: can use this decomposition method only for the whole mesh	aloeven	OpenFOAM Bugs	0	March 16, 2011 10:15
DecomposePar links against liblamso0 with OpenMPI	jens_klostermann	OpenFOAM Bugs	11	June 28, 2007 17:51