Convergence warnings in CFX transient simulation

- February 16, 2022 at 8:53 pm
  
  vinnakbm
  Subscriber
  
  Hi, I'm doing a transient simulation in CFX and in the monitor data output of solver I am seeing a convergence warnings summary table. I could not find this in the documentation and would like to understand it better to gain confidence about the results. Can someone point me in the right direction. Attaching a screenshot.
  Appreciate any help. Thank you!
- February 17, 2022 at 6:14 pm
  
  Surya Deb
  Ansys Employee
  
  Hello,
  Can you check the other convergence parameters in the out file?
  It seems like the momentum and continuity equations are not getting converged properly.
  Check the convergence history for these 2 equations in the out file to see how many times they failed to achieve convergence.
  Regards SD
- February 18, 2022 at 4:44 am
  
  vinnakbm
  Subscriber
  
  Thank you for your response. I am not familiar with where to look or how. Using this tool for the first time. Can you please suggest or point me to any help links? I could not find any reference to this convergence summary table in the solver manual. I set an RMS residual cut off of e-6 and 30 coeff loops. I randomly sampled the solver monitor display (total 3000 timesteps) and noticed that the residuals are converging below 20 coeff loops. The max residual is in the order of e-5.
  Thank you again!!
- February 22, 2022 at 2:36 pm
  
  vinnakbm
  Subscriber
  
  Hi, the out file is very large and I randomly sampled some timesteps. Since it is only 3 to 5 convergence failures, it is like looking for a needle in a haystack. What is the error message if it doesn't converge? (couldn't find this in the manual) I can use that phrase to search the outfile for failures. I tried searching using "NAN", "F". Please suggest.
  Thank you
- February 22, 2022 at 11:40 pm
  
  Surya Deb
  Ansys Employee
  
  Hello,
  You will find complete details of OUT file output in the link below.
  Yes, you can search using "F" as it denotes the equations that fail to converge.
  Also check for imbalances in domains to make sure the imbalances are sufficiently low.
  Regards SD
- February 23, 2022 at 3:15 am
  
  vinnakbm
  Subscriber
  
  I did check the mass imbalances and it is negligible between inlet and outlet (difference in the magnitude of e-4). For some reason the link in your post is not visible. Could you post the link again? Thank you
- February 23, 2022 at 1:57 pm
  
  Amine Ben Hadj Ali
  Ansys Employee
  
  3.2.1. CFX-Solver Output File (ansys.com)
- February 23, 2022 at 4:54 pm
  
  vinnakbm
  Subscriber
  
  When I searched in the out file with search string "F" & "NAN" there are no instances. Where else can I look?
  Thank you