CGNS error:Invalid index ranges.

Questions and remarks about code_saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
Ionut G
Posts: 43
Joined: Fri Apr 20, 2018 2:43 pm

CGNS error:Invalid index ranges.

Post by Ionut G »

Hi all,

I want to run an k-epsilon turbulence model simulation with Code_Saturne 5.0.7, windows version.

The mesh that I use for simulation is made using ICEM CFD. The mesh file si saved as CGNS, version 3.1 ADF.
In the past I use the same CGNS file type having no error.
Now when I use the mesh, the simulation stops at the beginning of the computation resulting in error: "CGNS error:Invalid index ranges."

Can somebody help me? I attached the listing file.

Thank you,
Ionut
Attachments
.listing
(14.87 KiB) Downloaded 155 times
Yvan Fournier
Posts: 4077
Joined: Mon Feb 20, 2012 3:25 pm

Re: CGNS error:Invalid index ranges.

Post by Yvan Fournier »

Hello,

I do not understand what has changed in your setup? Are you using a new code_saturne version, or a newer CGNS file ?

When running a version of code_saturne build with a recent (3.4+) CGNS version, it might be necessary to run cgnsupdate on older CGNS files.

If on the contrary, you generated your file with a recent ICEM CFD version but have an old code_saturne version, you may need to upgrade code_saturne. I know ICEM CFD allows you to save CGNS files as older versions, but in past versions, this seemed to mainly "cheat" on the version number, rather than really use an older version. I am not sure about current versions.

Best regards,

Yvan
Ionut G
Posts: 43
Joined: Fri Apr 20, 2018 2:43 pm

Re: CGNS error:Invalid index ranges.

Post by Ionut G »

Thank you Yvan for the quick reply!

I am using ICEM CFD 16.2.
I did not updated the version of any software that I am using.
I am using the same software versions as in the past, Code_Saturne 5.0.7 and ICEM CFD 16.2, and everything was fine on the same geometry.
When I am using a mesh that I made with y+ up to 1 the simulation runs ok, but when I wanted to generate a mesh with y+ between 2 and 5 the simulation won't start.
I thought that the mesh is not good but, I checked the mesh criteria from ICEM CFD and the quality is good, even in Code_Saturne the quality is ok.

I use the mesh with y+ from 2 to 5, in CFX 16.2 and the simulation runs, having no issue.
Maybe something is wrong with the mesh that I want to generate (y+ from 2 to 5) and is not accepted by Code_Saturne.

I don't understand what "CGNS error:Invalid index ranges." error is representing.
Do you have any recommendation where to look at in the mesh that I want to create, that can fix this issue?

Thank you very much,
Ionut
Yvan Fournier
Posts: 4077
Joined: Mon Feb 20, 2012 3:25 pm

Re: CGNS error:Invalid index ranges.

Post by Yvan Fournier »

Hello,

It seems the CGNS error message you get is from the CGNS library itself, not "code_saturne" itself.

So I am not sure in your case why some refinement seems to lead to a usable file and others do not.

Could you test pure cgns tools such as "cgnscheck", "cgnsplot", or "cgnsview" (from the CGNS bundle) on your mesh ?

Best regards,

Yvan
Ionut G
Posts: 43
Joined: Fri Apr 20, 2018 2:43 pm

Re: CGNS error:Invalid index ranges.

Post by Ionut G »

Hello Yvan,

I could not find the tools for testing the CGNS file format mesh for the windows platform, I am using windows 10.

I managed to solve the problem by reinstalling all the software, Code_Saturne and Icem CFD.

Now the simulations are working again.


Thank you very much for your support.

Best Regards,
Ionut
Yvan Fournier
Posts: 4077
Joined: Mon Feb 20, 2012 3:25 pm

Re: CGNS error:Invalid index ranges.

Post by Yvan Fournier »

Hello,

Good to know it is working again.

For your info (and that of future readers searching the Forum for similar issues), on Windows 10, if you have the Linux subsystem for Windows with a recent Ubuntu or Debian Linux distribution, you can probably simply install those tools with the cgns package. For a native Windows version, I do not know if there is a binary installer.

Best regards,

Yvan
Post Reply