CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   SU2 (https://www.cfd-online.com/Forums/su2/)
-   -   Problem on Nasa stage 35 simulation using su2 7.3.1 (https://www.cfd-online.com/Forums/su2/242355-problem-nasa-stage-35-simulation-using-su2-7-3-1-a.html)

wkholy April 17, 2022 23:57

Problem on Nasa stage 35 simulation using su2 7.3.1
 
1 Attachment(s)
I tried to simulate Nasa stage 35 using su2 7.3.1.
It can run in first 10 steps when the rotational speed is set much smaller. However, it will get NaN for residual computation when the rotational speed gets larger.

Error in "void CSolver::SetResidual_RMS(const CGeometry*, const CConfig*)":
-------------------------------------------------------------------------
SU2 has diverged (NaN detected).
------------------------------ Error Exit -------------------------------

Please see the attachment for the settings.

Hope it can be resolved.

koray May 27, 2022 10:51

Hi Kyle. I think your setup is ok. Is it possible for you to solve both rotor and stator separately first and then solve it together starting from the converged data. Convergence of compressor flows are very sensitive to initial data thanks to complex 3D flow field and adverse pressure gradients. With the same logic, you may want to keep CFL low along the computation. And I suppose your mesh is ok. Good luck.

wkholy May 29, 2022 22:37

Quote:

Originally Posted by koray (Post 828778)
Hi Kyle. I think your setup is ok. Is it possible for you to solve both rotor and stator separately first and then solve it together starting from the converged data. Convergence of compressor flows are very sensitive to initial data thanks to complex 3D flow field and adverse pressure gradients. With the same logic, you may want to keep CFL low along the computation. And I suppose your mesh is ok. Good luck.

Thanks, Koray. I have successfully run the case. Just the efficiency cannot converge but increase to 1000 and more. I have tried the tutorial case. Yet the efficiency is not stable as I change the time step from 3000 to 5000. SU2 CFD version is 7.3.1.

ashvin September 28, 2022 10:02

Quote:

Originally Posted by wkholy (Post 828877)
Thanks, Koray. I have successfully run the case. Just the efficiency cannot converge but increase to 1000 and more. I have tried the tutorial case. Yet the efficiency is not stable as I change the time step from 3000 to 5000. SU2 CFD version is 7.3.1.

Hello Kyle,
I stumbled upon this thread looking for why my efficiency values are all zero in my history.csv for my rotor37 case. I am using V740 and I am trying to figure out how to get efficiency data in history.csv file. Is there an option I need to add in SU2 card?
Thanks
Ashvin

wkholy September 29, 2022 05:10

Quote:

Originally Posted by ashvin (Post 836628)
Hello Kyle,
I stumbled upon this thread looking for why my efficiency values are all zero in my history.csv for my rotor37 case. I am using V740 and I am trying to figure out how to get efficiency data in history.csv file. Is there an option I need to add in SU2 card?
Thanks
Ashvin

Hi, Ashvin. I think the 7 version of SU2 cannot directly output turbo parameters in history.csv. But you can conveniently revise output_physics.cpp for a new output file of turbo efficiency or collect screen output and extract the turbo data using python. If the optimization is your goal, then SU2 version 7 can only deal with single zone problem of turbomachinery from my understanding.

ashvin October 21, 2022 09:07

Hello Kyle,
Thanks! I will work on that! I am only using SU2 to perform the CFD solutions.
Thanks
Ashvin


All times are GMT -4. The time now is 07:07.