code_saturne User's Forum

Hello everybody

I have a problem to use a meteo file in my case using not the GUI.

I would like to define some physical properties at boundary faces according the meteo file. When I edit usatcl and try to run a calculation, an error message tell me that I "must" edit usclim. But if I do that, I will be obliged to impose physical conditions to those inlet boundary faces I am using
Even if the calculation goes to the end, I observe the velocity I set in usclim for exemple and not the velocity expected from the meteo file.

If I try to comment in usclim, velocity conditions or other, in hope to let the meteo file to define them, I get an error message of undefined physics variables...

Thank you for your help

Hello,

Which version are you using?
Regards,

Hello Jacques

I am using code saturne 2.0.4

Hello,

You must define the type of boundary conditions in usatcl (usclim is not needed - I think it's a little incoherency in printings -), and the values (extracted from experimental profile - or not -).

You can look atincl.h in Code_Saturne sources to see pointers associated for each variables (velocity, temperature, ...).

Regards,

Thank you for your reply, but I can't still fix the problem. When I edit Usini1 to place "iihmpr=1" in usipsu subroutine (in order to avoid the calculation stop in usclim), I have a "Signal SIGSEGV (accès à une zone mémoire interdite) intercepté ! " when trying a calculation...

Hello,

You should not edit iihmpr: it is a flag indicating whether an XML file produced by the GUI was used (which implies additional initializations), so if you modify it, you should know exactly what the code execution paths are with or without this condition, and be ready to adapt the code.

Did you force the code into "meteo" mode using the usppmo subroutine ? The first step is doing if you are not using the GUI.

Regards,

Yvan

Thank you Yvan for your answer.

Actually, I would like to use meteo file with the combustion model, and in the code_saturne v2.0.4 I have it is impossible to activate two specific modules at the same time. Maybe using meteo and combustion is impossible, even it seems odd. (I think I should have begin by post this question).

Hello,

Yes, in general, most"specific" physics are mutually incompatible, as both may include specific models for fluid properties, and in any case, are basically never tested / validated in combination.

One exception may be "electric arcs / joule effect" (though I am not 100% sure), but I do not think other models are currently compatible (things may improve with versions 2.2 and above, as boundary conditions are defined in the same file regardless of the specific physics used, though combining specific physics would need testing).

Regards,

Yvan

Thanks Yvan. I will be very interested if an example of combustion with Eddy break up model exists.