Chaining Aster and Saturne

[David Monfort]

You need to have the libmedc-dev package (or libmed-dev...) installed on Ubuntu (have a look at Synaptic if not). If you installed Code_Saturne with MED support then it should also work for SYRTHES with the following command line:

gcc -o syrthes2med /path/to/syrthes3.4.2/src/util/syrthes2med.c -lmedC

David

[Alicia Consigny]

Hi Claus,

I'm following your issues as I'm going to have to run coupled calculations with Saturne/Syrthes, so any information is interesting !!

I had a problem too with the ensight files created with syrthes2ensight. The files were created but I couldn't read them in Paraview. I don't know if it is the same problem you are experiencing, but if it is you can have a look at this topic Syrthes : post processing format / tutorial for coupling with Saturne . David fixed my problem with a modified version of syrthes2ensight.c that you can find attached on one of the posts. Hope this helps !

Regards,

Alicia

[Claus Andersen]

I had the libmed-dev installed, but not the libmedc-dev - running GCC gives this warning:

claus@claus-desktop:/usr/local/syrthes3.4.2/src/util$ gcc -o syrthes2med23 syrthes2med23.c -lmed syrthes2med23.c: In function ‘entete_resu_syr’: syrthes2med23.c:437: warning: too few arguments for format

libmed et al, is version 2.3.5-2build1.

However, it does seem to work all right - even the time-dependent part.

If the SYRTHES folks don't have a easy fix for the Ensight problem, at least I can check the results in Salome. (.med file wont display in GMSH however) I guess you can't have it all - yet :)
 
Alicia: I read the post before I started messing around with SYRTHES, just to avoid to most common mistakes, so unfortunately I already have the version of syrthes2ensight you're referring to - the util. works fine for extracting the result at the end of a calculation, but not the time-dependent ones.

[David Monfort]

Just to precise that I've just edited my previous post to correct the command line to compile the syrthes2med tool... though the former command was also working ;)
Thanks for the noting warning, Claus. I'll pass this peace of information to the developers of SYRTHES.

[David Monfort]

Ok. I've talked with the SYRTHES developpers regarding your problem and the warning.

the syrthes2med warning (l.437) can be (probably safely) ignored, but you can suppress it by removing one of the %d of the scanf format. the syrthes2ensight issue is a bug in the the code. You asked for the min/max in time of the temperature (this feature is not often used) but it is not post-processed in the time-dependent results (because it is less relevant) though the number of scalar remains 3 (should be 1). A patch is proposed attached to this post, to be tested... Let us know if it works ;)

David

[Claus Andersen]

I can confirm it works! I wouldn't have pegged it down to having MIN/MAX enabled in the calculation and I agree on its (lack of) relevance, however, now it works even if it's enabled.  The old syrthes2ensight also works without having MIN/MAX enabled, just FYI.

Only Paraview 3.4 can read the output from SYRTHES though, whereas the output from Code_Saturne can be read by both P 3.4 and P 3.8 - this is just another FYI :)

Thanks for the help.

Regards,

Claus
 
Ps. attached some warnings that were produced while compiling the syrthes2ensight - so far I've had no problems with the util.  though.

[David Monfort]

Hello Claus,

Thanks for the feedback, I think that most of the warning can be safely ignored. As SYRTHES team is now working on SYRTHES 4, I think they will focus on the new syrthes2ensight converter. And regarding ParaView difficulties, this may be due to the refactoring of EnSight readers (ParaView 3.9 should be better with respect to this).

David

[Luigi Giaccari]

GuidHow to chain Aster and Saturne can be found here
 
http://www.advancedmcode.org/open-source-fluid-structure-interaction-chaining-code_aster-and-code_saturne.html

[Alexandre Douce]

Hello Luigi,
Thank you very much for proposing this how-to!  :)