code_saturne User's Forum

Hello,

Thank you for searching errors.
I made the changes that you have mentioned.

The calculation has started but it failed with this error:
"Signal SIGFPE (exception en virgule flottante) intercepté !"

So I did what you said about the same problematic on this link:
https://code-saturne.info/products/code ... /465240901

but nothing changed.

I have attached the study folder.

Fabien.

Hello,

Your time step seems much too big (up to 200, while > 10 can be dangerous), so you should reduce your time step by at least a factor of 10 or 20.

I did not look at your mesh, but if it is not of high enough quality (and specific physics such as combustion can be tricky with tetrahedral meshes), you mau also need to use several robustness-improving options, such as increasing the number of sweeps, switching to least squares gradient reconstruction with extended neighborhood (IMRGRA 2), user relaxation of pressure increase, ...

These recommendations often come up on the forum, qo you may find them in other posts (we need to add a "frequently asked questions" section on this and/or a debugging/troubleshooting howto document, but these are not ready yet, and the reference user documentation will give you plenty of options, but not prioritize which ones to try, so the forum is currently your best bet).

Best regards,

Hello Yvan,

You were right, it was a problem of time step and mesh. It works now!

But now I have a problem on the result, the calculation diverges.
I wish to increase the number of iterations but I'm limited to 20.
In the second part I wish to see the post-processing as I can do with the GUI, to see the calculation's evolution.
How can I do that?

Thank you.

I have attached the study folder.
I made a little burner with a gas inlet and a air inlet.

Best regards,

Fabien

Hello,

Forget the last post, i found where the problem come from.

Now, i can see a flame but the air around this one have a temperature around 3000 K, i attached a picture.
The picture is taken with COD3P adiabatic model. I tried with EBU and LW but the same problem appear.
I only use the adiabatic models because the others models doesn't work for the moment.

The burner is two concentric cylinder, one for gas, the other for air.

Do you know where the problem come from ?

Hello,

You mesh seems quite coarse, and I am not sure how the combustion models behave on a tetrahedral mesh. I would suggest comparing results with those on a finer mesh, but I am not otherwise familiar with the combustion models of the code, so I woul need others to help here.

Best regards,

Yvan

Hello Yvan,

You were right, I made a comparison between Tetrahedral and Hexahedral meshes and
it appears that the Tetrahedral mesh doesn't work correctly, I attached pictures of the test.

I took a cube :
Above : you have the air inlet
Left : the gas inlet
Right : Outlet

I took a cube because this is the only geometry that I succeed to use with Hexahedral mesh.

So, to do things well, I will make a new topic to learn how to use the hexahedral mesh.


Best regards,

Fabien