Orthonogality angle appearing as nan

ari003 · October 3, 2021, 15:14

Hi SU2 user,
I am trying to simulate a compressible case, meshing with gmsh and then doing the numerical analysis with SU2. The simulation is running fine but I observed a weird thing in the orthogonality angle tab of su2 and that is the min and max ortho angle is nan. I am wondering what might be the reason of that? Will that lead to any invalid results? I have attached the snapshot. Any kind of information will be of real help.
Thank you.

TKatt · October 3, 2021, 18:08

Hi ari003,

You can check the Mesh quality statistics in the volume output by adding:

VOLUME_OUTPUT= MESH_QUALITY

to your .cfg file. I guess you will see a few nan-cells but most of it will be fine. So I suspect more a problem in the Orthogonality computation in that case.

I searched quite a few NaN's through the code with some debugging and I always had to get rid of the mesh-quality routines as NaN's are often encountered in those code-bits due to some trigonometry going on in there.

So to some degree I know that code could need a few more error-checks or safety nets but didn't find the time to do so. In case you are interested to take a look into this you are more than welcome

The implementation is in CPhysicalGeometry.cpp CPhysicalGeometry::ComputeMeshQualityStatistics.

Hope that helps, Tobi

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Need Help Here! nan value error	mgkid3310	OpenFOAM Running, Solving & CFD	1	October 6, 2016 01:59
Why simulation results showing "nan" values?	Sargam05	OpenFOAM	4	September 10, 2012 13:37
[Netgen] Import netgen mesh to OpenFOAM	hsieh	OpenFOAM Meshing & Mesh Conversion	32	September 13, 2011 06:50
Cells with t below lower limit	Purushothama	Siemens	2	May 31, 2010 22:58
Could anybody help me see this error and give help	liugx212	OpenFOAM Running, Solving & CFD	3	January 4, 2006 19:07

October 3, 2021, 18:08		#2
TKatt Member na Join Date: Jul 2018 Posts: 90 Rep Power: 8	Hi ari003, You can check the Mesh quality statistics in the volume output by adding: VOLUME_OUTPUT= MESH_QUALITY to your .cfg file. I guess you will see a few nan-cells but most of it will be fine. So I suspect more a problem in the Orthogonality computation in that case. I searched quite a few NaN's through the code with some debugging and I always had to get rid of the mesh-quality routines as NaN's are often encountered in those code-bits due to some trigonometry going on in there. So to some degree I know that code could need a few more error-checks or safety nets but didn't find the time to do so. In case you are interested to take a look into this you are more than welcome The implementation is in CPhysicalGeometry.cpp CPhysicalGeometry::ComputeMeshQualityStatistics. Hope that helps, Tobi