[davod]

What do the arguments <root> <case> and <fileprefix> mean in the mesh conversion tools? I probably should know, but I don't, and can't figure it out.

[chris]

root and case are OpenFOAM basics:
http://www.opencfd.co.uk/openfoam/doc/userse6.html

fileprefix means the prefix to the mesh file names. There's something similar in:
http://www.opencfd.co.uk/openfoam/do...-1800006.4.2.7
(I got that by typing 'fileprefix' in the Search on http://www.opencfd.co.uk)

Hi,

does anyone know, why or wheather the tetgenToFoam-utility needs a tetgen-mesh, which is generated with the f-Option?

The reason why I'm asking is, that in my first test of the tetgenToFoam-utility I didn't use this option (tetgen -qa0.05 file.poly) and it seemed to work very well. checkMesh said my mesh is ok. Unfortunately all simulations I tried to do with such a tetgen-mesh lead to a diverging solution. So I thought, the mesh isn't really ok, perhaps.

I looked in the sourcecode of tetgenToFoam and read that you should use the f-option. But with this option I get boundary-marked internal faces in the .face file. So I had to use the -nofaceFile option for tetgenToFoam and so I get of course only one defaultpatch in the boundary-file, also a warning ("found 68 undefined faces in mesh; adding to default patch"). It seems not to work how it should be work. Also I need some different boundary patches.
Without the -nofaceFile option tetgenToFoam doesn't work at all (because of the boundary marked internal faces I think -> FOAM FATAL ERROR: Trying to specify a boundary face ... on the face on cell 0 which is either an internal face or already belongs to some other patch).
For all of that I used OpenFOAM 1.4.1.

Is there anybody who knows something about that? Have I really to use the f-option for generating the tetgen-mesh?

Thanks in advance!
Karen

[mattijs]

You'll have to use the -f option to output the region information so OF can sort the boundary faces into patches.

Problem is that tetgen seems to include internal faces if they have all three points on the boundary and this is giving problems in the converter.

Can you send me (m dot janssens at opencfd dot co dot uk) some sample files (generated with -f) that don't convert and I'll have a look at it.