Page 2 of 2

Re: Electrodes inside of the domain and its properties.

Posted: Thu May 04, 2017 1:48 am
by rodion
Yvan, could you please explain, if I want to initialize enthalpy or impose thermal conductivity, I have to use pointer ivarfl(isca(ihm)) or ivarfl(isca(iscalt))? What is the difference for the case with electric arc simulation?

Thanks in advance.
Best regards,
Rodion

Re: Electrodes inside of the domain and its properties.

Posted: Fri May 05, 2017 1:54 am
by Yvan Fournier
Hello,

For electric arcs, ihm and iscalt should be the same (I recommend checking though, I am not 100% sure).

In general cases (i.e. without a specific model such as electric arcs or combustion), the ihm value is not used or defined, even when solving for enthalpy (the C field pointers are more consistent for this than some older Fortran variables).

Best regards,

Yvan

PS: please do not address questions specifically to me, even if I am probably the person who will try to answer. I am trying to encourage others to answer questions, not handle this forum by myself...

Re: Electrodes inside of the domain and its properties.

Posted: Fri May 05, 2017 1:59 am
by Yvan Fournier
Hello,

I need to check this, but I suspect we may have a "Temperature"/"temperature" upercase/lowercase issue which might be specific to Windows (on which file names may be case insensitive, though I am not sure of the subtleties of the Windows filesystem, which may try to be compatible with several approaches simultaneously). I'll check in the code for this.

If in doubt, and you use the SALOME platform, you may try to set preprocessing soutput to the MED format: since all fields are in a single file, this issue should dissapear in this case (though you then need SALOME's PARAVIS moduel to view your files).

Best regards,

Yvan

Re: Electrodes inside of the domain and its properties.

Posted: Fri May 05, 2017 10:08 pm
by rodion
Thank you, Yvan.

I have checked the problem with output files: Windows is case insensitive and cannot create two files with names results.temperature.00001 and results.Temperature.00001 (only t and T are different).
Thank you for the idea about MED format, it has to work too.

I have checked the pointers and I see that pointers ivarfl(isca(ihm)) and ivarfl(isca(iscalt)) are the same (ivarfl(isca(iscalt))=ivarfl(isca(ihm))=2) and there is only one field called Enthalpy.

Best regards,
Rodion

Re: Electrodes inside of the domain and its properties.

Posted: Tue May 09, 2017 11:15 pm
by Yvan Fournier
Hello,

The default labels for temperature and boundary temperature should differ by more than capitalization, so we'll need to check those defaults (but can't easily ensure modified values are compatible).
In any case, for the EnSight format, labels (and not fixed, "canonical" names) are used, so using the GUI, in the surface and/or volume output settings, you can modify temperature and/or boundary temperature labels so that you either have the same name (including capitals), or a name differing by more than capitals.

In the future, I'll try to add a Windows-specific check in the EnSight gold output so that on case-insensitive systems, labels are also case-insensitive.

Best regards,

Yvan