GebhardtFactors binary

[cemg]

Hi Elmer team,

I've tried the GebhardtFactors program (just for testing its funcionality) with different meshes (with updated view factors) and correct sif files (tested launching ElmerSolver with same mesh) and I always get a seg fault. What is this binary intended for? Can I use it as a stand-alone Gebhart factors solver?

Regards,

Cesar

[raback]

Hi Cesar

This used to be part of the radiation workflow. First View Factors where computed and then Gebhardt Factors. In case emissivity is temperature dependent the latter need to be recomputed every iteration. Hence the disk I/O becomes annoying. Since quite many years the Gebhardt factors are computed internally, and recomputed when needed, with quite a lot more intelligence to control the matrix filling, for example. So this routine is basically obsolite but could still work in the old way which involved defining the file with the factors.

-Peter

[cemg]

Hi Peter,

thank you very much. That was what I supposed.

If someone cares, the segmentation fault disappears changing the line 132 of the GebhardtFactors.src file. Where it says:

Code: Select all

     Model => LoadModel( ModelName,.FALSE. )

it must say:

Code: Select all

     Model => LoadModel( ModelName,.FALSE.,1,0 )

Besides, this stand-alone Gebhardt factor solver is pretty slow. The new implementation in the RadiationFactors solver does the trick much faster.

Regards,

Cesar