CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   SU2 (https://www.cfd-online.com/Forums/su2/)
-   -   3d case in SU2 (https://www.cfd-online.com/Forums/su2/200871-3d-case-su2.html)

sbista21 April 15, 2018 14:33

3d case in SU2
 
Hey.
I have been performing simulations and validating results using SU2. I have successfully carried out 2D cases. However, I am having a hard time while doing 3D cases, as they are not converging.
Can someone help me by sharing an example of a 3D case config file dealing with Euler and Navier-Stokes equation?
Thank you.

hlk April 29, 2018 20:44

There are some examples of 3D cases in the tutorials, as well as in the testcases repository. Those in the TestCases repository generally use a very small number of iterations, as they are for code regression testing purposes.

It is usual for 3D cases to be more difficult than 2D cases, and will take more iterations to converge. If you have not already tried, things like lowering the CFL number (if the residuals diverge) or raising the CFL number (if they are just converging slowly) and trying different limiters and numerical methods can often help.

Quote:

Originally Posted by sbista21 (Post 688967)
Hey.
I have been performing simulations and validating results using SU2. I have successfully carried out 2D cases. However, I am having a hard time while doing 3D cases, as they are not converging.
Can someone help me by sharing an example of a 3D case config file dealing with Euler and Navier-Stokes equation?
Thank you.


sbista21 May 2, 2018 11:09

Quote:

Originally Posted by hlk (Post 690701)
There are some examples of 3D cases in the tutorials, as well as in the testcases repository. Those in the TestCases repository generally use a very small number of iterations, as they are for code regression testing purposes.

It is usual for 3D cases to be more difficult than 2D cases, and will take more iterations to converge. If you have not already tried, things like lowering the CFL number (if the residuals diverge) or raising the CFL number (if they are just converging slowly) and trying different limiters and numerical methods can often help.


Thank you. I did try that and it has converged. Much appreciated anyways.
I wonder if you could help me with paraview. How can I calculate Cp and do sensitivity analysis.

aero_amit July 2, 2018 12:28

Try plot over intersection curves in ParaView for sectional extractions


All times are GMT -4. The time now is 20:26.