CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Convergence Speed (https://www.cfd-online.com/Forums/main/6071-convergence-speed.html)

Maciej Matyka May 5, 2003 15:43

Convergence Speed
 
Hello, I have done three different solvers: SIMPLE,SIMPLER and Vorticity-Stream approach to 2D incompressible... How to test convergence speed? When I am calculating simple test case for Driven-Cavity method (non-dimensional variables, Re=100,200,300 etc.) convergence speed is approximately the same for three methods. For some cases SIMPLER converges faster than SIMPLE, but vorticity-stream approach is between them.For convergence I am calculating (at the end of iteration step) a maximum value of difference between old and new computed velocity (both for u and v component). I am ok with that?Best Regards,Maciej Matyka

Sebastien Perron June 12, 2003 13:20

Re: Convergence Speed
 
To measure convergence, you could also take the norm of the residuals. Personally, I always use both (variation and residuals).


All times are GMT -4. The time now is 19:06.