Fatal error (divergence) solving for Enthalpy

[husanhao]

Dear all,

I am trying to use the electrical arc module in CS3.0.0, however, I got the fatal error of the divergence for solving for Enthalpy. Could anybody please tell me what is the reason cause this error? Is that because my MESH? Is there anybody can help me to solve this problems? Thanks. I really appreciate it!!!

Please go to the following link to download my MESH and the CS setting, thanks.
https://dl.dropboxusercontent.com/u/776 ... rge.tar.gz

Best Regards
Eric

[Yvan Fournier]

Hello,

I won't have the time to look at this before a few days (though other members of this forum might).

Did you follow most of the troubleshooting recommendations in the documentation section of this web site ? I'm especially thinking of the time step and CFL, because in may cases, electric arcs modeling requires a pretty small time step.

Again, I did not look at your case yet, so this is a quick suggestion, which may be redundant with what you already checked...

Regards,

Yvan

[husanhao]

Hi, Yvan,

Thanks for your reply, I will check the time step and CFL, thanks. Once I get any results, I will post here and let you know, thanks.

Best Regards
Eric

[Christophe Loubet]

Dear Eric Rau,

I know how to solve your problem . I often received this message error. Depends on your thermochemistry data, there are a lot of parameters to adjust, I can help you. Just before, could you explain me how do you use the new code saturne version. I need to know:

-For boundary condition: Do you use both Gui and cs_user_boundary_condition-electric_arcs.f90?
-Do you define your initial hot channel in the GUI or in the cs_user_initialization-electric_arcs.f90?
-Do you define the electical variable scaling in the Gui or by defining a plane or in uselrc.f90?

Best Regards,
Christophe

[husanhao]

Dear, Christophe,

Thanks for your reply very much. I really appreciate your help. Indeed, I am a new user of code saturne, and I am trying to simulate the three phase electrical arc phenomena. The following are what you ask me, if you have any concerned, please feel free to let me know, thanks.

- 1. Boundary Condition:
Yes, I use both GUI and cs_user_boundary_condition-electric_arc.f90.
Based on my understanding, once I put the subroutines (cs_user_boundary_condition.f90) into the /SRC folder, the code saturne will use the setting of the user routines, because the user routine have higher priority than GUI. There are three electrodes in my model, which are phase A, phase B, and phase C of the power system, repetitively. Considering the balance power system, I set the electrical potential of each electrodes have 120 degree phase shift, so I only set the electrical potential of each electrodes.

- 2. Initial condition:
I only use the GUI. I set the Enthalpy of some area is 16.2*10^6;

- 3. Electrical Variable Scaling:
I use GUI. I use the "modele de recalage" as "general case". (please refer the attach file to see the setting in GUI)

Since I am a new user of the code saturne, and I don't have many experiences about using CFD software, the setting that I did may not right. So, if you have any questions or concerns, please feel free to let me know. Again, I really appreciate your help!

Best Regards
Eric

[Christophe Loubet]

Dear Eric,

That is great because I have allready simulated three phase electrical arc phenomena with code saturne 2.0. I will try to help you with code saturne 3.0.

First, could you explain me please, how are you doing to set the Enthalpy in some area with the GUI?Also, I didn't see the attach file to see the setting in GUI. Could you attach it please.

Sincerely.

Christophe

[husanhao]

Dear, Christophe,

Thanks for your help again. I only set the initial Enthalpy in GUI (please refer to the attach file). Also, I attach the setting about the electrical variable scaling in GUI. Based on your experience and professional opinions, in order to simulate three phase arc phenomena, is there any setting I need to consider? If you still have any question, please feel free to let me know, you can either send the email to me (shiuanhaurau@gmail.com) or reply here, thanks!!!

Best Regards
Eric

[Christophe Loubet]

Dear Eric,

I am working on your model and it seems that your initial hot column with 16.2.10⁶ J/kg is not taken into account. You need to define it in the cs_user_initialization.f90 (after hinit l 242) and not in the GUI.

Now, you will reach the difficult part in 3-Phase system simulation. We need to define how to control the simulation (imposed current, power, voltage...). Feel free to wright me a private message (christophe.loubet31@gmail.com) in order to explain how works your system. I need to know if you work with transferred arcs to metal bath or non transferred arcs like my simulations:

[husanhao]

Dear, Christophe,

Thank you very much for your quick and professional reply, I will try to set the Enthalpy initialization condition in subroutine. And I will try to figure out the control of the three phase arc simulation. In my opinion, my work is quiet similar with your simulations. The electrodes position is only the difference between your and what I want to simulate. I will try to do it by myself first, and if I have any questions, I will wright you a private message, thanks.

Best Regards
Eric

[Christophe Loubet]

Ok. Just to simulate this system you need to include the elctrode into the computationnal domain (grey part in the previous picture). In my case, whithout considering the electrode, arcs are not moving outside of the inter-electrode gap.

Moreover, that is maybe better that you use Air thermochemsitry data instead of Azote. Indeed, 16.2*10⁶ J/kg correspond to the necessary 6000 K hot channel to initiate the model but with using air thermochemistry data. I can send you the file.

Sincerely.