Learning Forum

[Stefan_Panajotovic]

Hi everyone,

I created a system coupling between a FLUENT and a steady-state thermal modules. Two of them are 2-way coupled using temperature, film coefficient and reference temperature. Initially, I turned off energy equation in FLUENT and let the system coupling run for 5 coupling iterations, just to see how it behaved. It worked nicely, only the FLUENT model didnt converge but I did not expect it to. Then I turned on the energy equation and repeated the system coupling, when following errors occured:

[Rob]

Check what Mechanical is seeing re the temperature. It looks more like an issue with the solver than system coupling. How does it look in 2020R2 as R17 is about 4 years old. n

[jenja1711]

Hey Stefan,ndid you solve the Problem? I've got the same one. I'm trying to simulate a System coupling between fluent and transient thermal Analysis, each Simulation works for itself but when coupiling i get the Error: n(DP 0) Participant Solution 1 encountered a fatal error. Message: A large negative pivot value ( -1736.84052 ) has been encountered in the global assembled matrix at the TEMP degree of freedom of node 26524. This may be caused by a bad temperature-dependent material property used in the model.I use Ansys Workbench 2020R1. Thanks in advance,nJennyn

[Stefan_Panajotovic]

Hi Jenny,Indeed I have. I did not work on 2020 version but in 2017 version you cannot couple the transient thermal and FLUENT. You control if your analysis is steady or not in the system coupling block through the number of steps option. Maybe that is possible in 2020, you have to check that. nRegardless of that, what solved my problem is that I made a pre-calculation in FLUENT until I got a convergence of residuals and only then I connected it to the system coupling block. Once you start your system coupling calculation it will start from where your pre-calculation stopped and you will even see the pre-calc residuals. Once I did that I worked like a charm and I think it should work for you as well. One downside is that you have to start the process from the beginning every time you want to re-calculate. You have to decouple the system coupling block, solve in FLUENT again, and then recouple it again. This is because 2017 does not allow the restarting of the two-way system couplings. Again, this could be different in 2020 (I hope it is). nTry it out and please let me know if it works, especially with the transient analysis, because I would like to do it in the future as well. If it doesn't, we'll think of another solution why it worked for me nKind regards from GermanynStefan n

[jenja1711]

Hi Stefan,nthank you for your quick answer.nIn Ansys 2020R1 it's possible to couple transient thermal Analysis with fluent. Your advice actually helped, the coupled Simulation converged successfully, thank you very much!nIn Addition I saw a tutorial where they explain a Restart of a two-way System coupling. I'll try it the next days and give you feedback.nnViele Grüße aus Dortmund nJenny

[Stefan_Panajotovic]

Glad to help! I would be interested to know how to restart them, thanks a lot! Do you know from which version of Ansys is it possible to do so? nViele Grüße aus Aachen nStefann

[jenja1711]

Just watched the Video again and it's about restarting a two-way coupling between transient structual and fluent. Thought it shouldn't make a difference but in thermal Analysis there is missing the Restart Control in the Analysis Settings. Sorry for raising false hopes nn