Check Mesh not functioning

[kennethchaw]

Hi All,

Im new to Code_Saturne and first trying out a tutorial but stuck at the Mesh Quality Check part. I am using Windows 10 to run CS 4.0.4 version. After i clicked the "check mesh" button, it simply do not create any file or prompt any result.

Thanks!

[Yvan Fournier]

Hello,

This may be due to a subtle issue encountered mostly on Windows, but hard to reproduce on Linux developement environments.

Do you have the issue all times you try ?

Could you try with version 4.0.5, which has a tentative fix for this ?

Regards,

Yvan

[kennethchaw]

Yes.. Installed the latest version... the log file came out but it showed:
C:\Program Files (x86)\Code_Saturne\4.0\libexec\code_saturne\cs_preprocess.exe

--post-error ensight --case preprocess_01

E:/PIPESTUDY/CASE1/MESH\Pipe.med



.------------------------------.

| |

| Code_Saturne Preprocessor |

| |

`------------------------------'



code_saturne version 4.0.5 (built 06/27/16 17:26:05)



CGNS 3.1.4 file format support

MED 3.0.7 (HDF5 1.8.12) file format support

Reading of compressed files ('.gz') with Zlib 1.2.8







Case configuration

------------------



Date : 10/15/16 21:13:21

Processor :

Directory : E:\PIPESTUDY\CASE1\RESU\check_mesh



Case name : preprocess_01

Mesh file : E:/PIPESTUDY/CASE1/MESH\Pipe.med







Reading mesh from file in MED (EDF/CEA) format

----------------------







../../../src/misc/MEDversionedApi3C.c [50] : Cette bibliothèque MED n'est pas capable de lire un fichier MED dont le mineur de la versionexcède celui de la bibliothèque.
../../../src/misc/MEDversionedApi3C.c [51] : La version demandée est :
../../../src/misc/MEDversionedApi3C.c [51] : _fversionMMR = 310
../../../src/misc/MEDversionedApi3C.c [116] : Impossible d'obtenir une implémentation de :
../../../src/misc/MEDversionedApi3C.c [117] : key = "_MEDmeshInfoByName"
../../../src/misc/MEDversionedApi3C.c [118] : en version :
../../../src/misc/MEDversionedApi3C.c [119] : _fversionMMR = 310
../../../src/misc/MEDversionedApi3C.c [120] : Vérifiez votre fichier .med
../../../src/ci/MEDmeshInfo.c [83] : Erreur d'appel de l'API
../../../src/ci/MEDmeshInfo.c [83] : du maillage
../../../src/ci/MEDmeshInfo.c [84] : meshname = "Pipe"
../../../src/ci/MEDmeshInfo.c [84] : _meshpath = "/ENS_MAA/Pipe"
../../../src/ci/MEDmeshInfo.c [84] : "MEDmeshInfoByName" = "MEDmeshInfoByName"

Error in preprocessor execution
===============================

../../../preprocessor/pre-post/ecs_pre_med.c:1624: Fatal error.

MED: error reading file "E:/PIPESTUDY/CASE1/MESH\Pipe.med".
Name of mesh read : "Pipe"
Dimension read : "32762"


Is my salome not linking with the code saturne ?

[Yvan Fournier]

Hello,

This is due to the Windows build using MED 3.1, while you probably used a more recent version of SALOME using MED 3.2. We need to rebuild a version of Code_Saturne 4.0.6 using MED 3.2, but you can otherwise try our more recent Windows build of version 4.2.2 or 4.3.1, which should use MED 3.2 if I remember correctly.

Regards,

Yvan

[kennethchaw]

Hi,

Thanks for the prompt reply,

I tried Salome 7.6. It works.
But I wonder why the runcase.bat that supposed to be generated in the SCRIPT folder is not found ?

I'm trying one of the tutorial. I have used the CS GUI to create new case, but eventually, those sub directories also not created automatically.

Thanks

[Yvan Fournier]

Hello,

Did you create just a new file (File menu or icon), or a new case (only in the File menu) ?

Regards,

Yvan

[kennethchaw]

I tried both... but both end up i need to locate the directory myself right ? which I have to select until the case folder... In other words, both the "study" and "case" folder i created myself but not generated....


TQ

[Yvan Fournier]

Hello,

When using "new case", the GUI should create a directory, with appropriate subdirectories.

Regards,

Yvan

[kennethchaw]

Hi,

Thank you, it works after I restart my laptop...

Besides, If now I wish to analyse the temperature change of a box over time, I just have to define the initial six walls boundary conditions right? do I need to define the gravitational force at Z direction?

Thanks.

[Yvan Fournier]

Hello,

You only need to indicate gravity if you are using a variable density, and gravity-indeced effects such as natural convection or stratification may occur in the flow you are modeling.

Regards,

Yvan