Velocity Split in Domain for Larger Simulations

pjoynt · November 15, 2021, 15:19

Hi All,

I was wondering if anyone has experienced or might have an explanation for why when viewing x-velocity (or velocity mag.), there is a split / delineation in the middle of the domain? I haven't experienced this for domains that are smaller and I have tried the different domain decomposition options but that doesn't seem to resolve the issue.

The domain is relatively large at 235m in x, 125m in y, and 20m in z. The control files are attached.

Cheers,
Patrick

kamath · November 16, 2021, 05:08

Looks like an artefact from Paraview visualisation at the boundary of parallel processors?

pjoynt · November 16, 2021, 08:33

I've tried paraview 5.09 and 5.10.0-RC1 and seem to have the same problem... I'll have to try some other versions.

kamath · November 16, 2021, 08:39

Well looks like there was a bug resulting in air turbulence damping applied only to proc 0. That resulted in the difference in velocity seen at the processor boundary. This is now fixed.
Please download the latest RC and let us know if the problem persists.

valgrinda · November 16, 2021, 08:47

Hi Patrick,

thanks for the feedback! This is a bug in the use of T 32, leading to different eddy-viscosity in different parts of the domain. I have fixed this in the RC, and it will be fixed in the next master.

pjoynt · November 16, 2021, 08:58

Quote:

Originally Posted by valgrinda

Hi Patrick,

thanks for the feedback! This is a bug in the use of T 32, leading to different eddy-viscosity in different parts of the domain. I have fixed this in the RC, and it will be fixed in the next master.

Much appreciated Prof. Bihs!

November 16, 2021, 05:08		#2
kamath Senior Member Arun Kamath Join Date: Nov 2014 Location: Trondheim, Norway Posts: 265 Rep Power: 14	Looks like an artefact from Paraview visualisation at the boundary of parallel processors? __________________ Arun X years with REEF3D

November 16, 2021, 08:39		#4
kamath Senior Member Arun Kamath Join Date: Nov 2014 Location: Trondheim, Norway Posts: 265 Rep Power: 14	Well looks like there was a bug resulting in air turbulence damping applied only to proc 0. That resulted in the difference in velocity seen at the processor boundary. This is now fixed. Please download the latest RC and let us know if the problem persists. __________________ Arun X years with REEF3D

November 16, 2021, 08:47		#5
valgrinda Super Moderator Hans Bihs Join Date: Jun 2009 Location: Trondheim, Norway Posts: 403 Rep Power: 19	Hi Patrick, thanks for the feedback! This is a bug in the use of T 32, leading to different eddy-viscosity in different parts of the domain. I have fixed this in the RC, and it will be fixed in the next master. __________________ Hans Bihs Team REEF3D www.reef3d.com

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[OLAFLOW] The OLAFLOW Thread	Phicau	OpenFOAM Community Contributions	459	September 12, 2024 12:47
sliding mesh problem in CFX	Saima	CFX	46	September 11, 2021 08:38
CFX - specified domain intialization	qntldoql	CFX	4	September 28, 2020 10:28
problem initializing the domain using velocity inlet	Arvind_CFD	FLUENT	0	October 24, 2014 13:49
Terrible Mistake In Fluid Dynamics History	Abhi	Main CFD Forum	12	July 8, 2002 10:11

November 16, 2021, 08:33		#3
pjoynt New Member PJoynt Join Date: Oct 2021 Posts: 21 Rep Power: 6	I've tried paraview 5.09 and 5.10.0-RC1 and seem to have the same problem... I'll have to try some other versions.