Ensight Gold mesh format import in Code_Saturne

Questions and remarks about code_saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
Axel Cablé

Ensight Gold mesh format import in Code_Saturne

Post by Axel Cablé »

Hi,
 

I built my mesh in Star-ccm+ and exported it in Ensight Gold format
(.case).
I am trying to import it in Code_Saturne but the boundary regions
aren't detected, yielding the following error when I run the
preprocessor:

node id : off
element id : off

part 1 : Fluid
81208 noeuds
51912 éléments hexa8
4654 éléments nfaced
Remarque : pas d'id de sommets précisés
--> impossibilité de fusionner les "parts" EnSight ,
donc "parts" suivantes ignorées.
 
The part1 (fluid) is recognised but it doesn't recognise the
part2 (default wall) part3 (inlet) and part4 (outlet), because "no
nodes' ids are defined".

Any idea how I could solve this problem would be greatly appreciated.
Also, would you know about a software that could convert my mesh from Ensight Gold format to .unv format for example?

Regards,
Axel
Yvan Fournier

Re: Ensight Gold mesh format import in Code_Saturne

Post by Yvan Fournier »

Hello,

If Star-ccm+ gives you an option to set vertex ids when exporting EnSight Gold files, using it will solve your problem, but the EnSight format is designed as a visualization format, not as a mesh format; when vertex ids are present, we assume that all parts are pieces of a same mesh and read them all, but this is a convention we adopted (and which works well with files output by EnSight), but is not specified in the format documentation. In EnSight Gold format, all parts have independent vertex definitions, so there is no easy way to match vertices on boundary parts to those on volume parts (a geometric test would be dangerous for cases with thin walls, so it would not be general).

A better solution is to ask CD-ADAPCO for the libccmio library, (which they provide upon request, but which we do not seem to be allowed to distribute ourselves), recompile it, and reconfigure/rebuild the Code_Saturne preprocessor with libCCMIO support.
Axel Cablé

Re: Ensight Gold mesh format import in Code_Saturne

Post by Axel Cablé »

Hi Yvan,

Thanks for you answer. The problem is that the libCCMIO is not compatible with version 1.3.3 of the preprocessor, which is the only one I have access to on my lab's calculation station. I may end up using gambit instead.
Chebli Rezki

Re: Ensight Gold mesh format import in Code_Saturne

Post by Chebli Rezki »

Hello,

I have same problem with Star-ccm+ and it don't give an option to set vertex ids when exporting EnSight Gold filesI , I use the 2.0.0-rc1 saturne version.

If the libCCMIO is not compatible with this version of the preprocessor, how I could solve this problem :(

The mesh quality file is attached.

Best regards.

Rezki.
Attachments
mesh_quality.txt
(5.58 KiB) Downloaded 209 times
Yvan Fournier

Re: Ensight Gold mesh format import in Code_Saturne

Post by Yvan Fournier »

Hello,

libCCMIO is compatible with preprocessor versions 2.0 (with 2.0-rc2 being recommended), so you are not in the same situation as Axel, who is stuck with version 1.3.

Finally, it would seem that Star-CCM's EnSight output is buggy, as prism (penta6) elements are badly oriented. Just use the --reorient command-line option for the Preprocessor to compensate for this.

Best regards,

  Yvan
Chebli Rezki

Re: Ensight Gold mesh format import in Code_Saturne

Post by Chebli Rezki »

Ho Yvan,

Thank you for help.

I used --reorient command and it works ( java_hydro/saturne_java_layer/java/MESH$   cs_preprocess -m java_hydro.case --reorient ) and preprocessor_output  file has been created in MESH file

Please, how to take this correction in count  from code before setting boundary conditions and launching calculation (from GUI) ?

NB: I created several groups in Star-ccm+ (inlet, outlet,....) but I don't find them when I checked the mesh, what can be reason for that ?

Best regards.

Rezki
Attachments
reorient.txt
(5.83 KiB) Downloaded 218 times
Post Reply