CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Open Source Meshers: Gmsh, Netgen, CGNS, ... (http://www.cfd-online.com/Forums/openfoam-meshing-open/)
-   -   Null determinant cells in Netgen (http://www.cfd-online.com/Forums/openfoam-meshing-open/112020-null-determinant-cells-netgen.html)

danvica January 20, 2013 11:55

Null determinant cells in Netgen
 
4 Attachment(s)
Meshing a simple geometry using Netgen I obtained some null-determinant cells that crash a solver (see http://www.cfd-online.com/Forums/ope...tml#post402809)

Enclosed you can find pictures of the mesh and the file of the .brep geometry.

The white coloured cells are the null-determinant ones.

Is there anyone able to solve this problem ? Is there any guideline to follow to avoid this cells ?

More: Why did they come in such a simple case ?

Thanks.

skeptik January 24, 2013 07:23

Success in SALOME
 
2 Attachment(s)
I generated both meshes in SALOME without any problem.

1. Simple mesh NETGEN 1D-2D-3D with very fine settings.
Attachment 18432
2. Mesh with refinement on edges.
Attachment 18433

I did not mesh conversion due to i have not OF on my working computer, but SALOME didn't detect zero volumes so i think there is no elements with negative determinant.

P.S. I use SALOME on Win7

danvica January 24, 2013 09:34

Hi,
Thanks for your time. Have you tried to check your mesh with:

checkMesh -allGeometry -allTopology

and see how many (and if...) undeterminet cells it has ?


BTW: I use Salome but just to preprocess existing STEP geometry. Then usually I use SHM. I tried several times meshing with it but I found it quite instable (i.e. lots of crash), above all making submesh.

Which version are you using ? Do you download it from the official site ?

skeptik January 24, 2013 11:55

Mesh OK
 
Here my checkMesh output:

Code:

Checking geometry...
    Overall domain bounding box (-0.05 -0.05 0) (0.05 0.05 0.375)
    Mesh (non-empty, non-wedge) directions (1 1 1)
    Mesh (non-empty) directions (1 1 1)
    Boundary openness (-8.45579e-18 -2.98649e-18 -1.5611e-17) OK.
    Max cell openness = 2.74985e-16 OK.
    Max aspect ratio = 4.80757 OK.
    Minumum face area = 6.88965e-07. Maximum face area = 0.000152252.  Face area magnitudes OK.
    Min volume = 3.02192e-10. Max volume = 6.23914e-07.  Total volume = 0.00198994.  Cell volumes OK.
    Mesh non-orthogonality Max: 50.6206 average: 15.2056
    Non-orthogonality check OK.
    Face pyramids OK.
    Max skewness = 0.557965 OK.
    Coupled point location match (average 0) OK.

Mesh OK.

Mesh is good. In this case i checked mesh with refinement. I think the problem in your mesher. However SALOME uses NETGEN algorythm. May be you did something wrong in NETGEN settings.

Quote:

I tried several times meshing with it but I found it quite instable (i.e. lots of crash), above all making submesh.

Which version are you using ? Do you download it from the official site ?
Yes, It could be unstable due to big mesh and unsufficient memory. So i just restart it and save my projects before each operation with mesh.

I use version 6.4 32-bit from official website.

danvica January 24, 2013 13:34

Can you please execute checkMesh with -allGeometry -allTopology options ?

PS. I really hope you had find a good way to avoid these cells...

Thanks.

skeptik January 25, 2013 10:19

1 Attachment(s)
OK!

I executed it with options -allGeometry -allTopology and got an undetermined cells as you describe.

It was pretty strange so i did checkMesh with the same options but on my nozzle head case. The mesh was also generated using SALOME.

So i got same cells in the similar positions as in your case.

There is:

Attachment 18485

So i think you need keep it simple) it's not your fault


All times are GMT -4. The time now is 16:58.