CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   convergence problem (http://www.cfd-online.com/Forums/main/2823-convergence-problem.html)

manjgi November 22, 2000 12:51

convergence problem
 
Dear all,

The following question is related to the using of SIMPLE method in incompressible steady 2-D channel flows with a stenosis and is about (what a surprise!) convergence.

Could someone advise why the residuals for the y-momentum equation would be the most difficult to converge. My grid is very fine and I have already tried CDS ans 1st order UDS for convection.

thanks a lot

manjgi

I. Dotsikas December 22, 2000 10:00

Re: convergence problem
 
Hallo, I suppose that your main flow direction is the x and the y side is getting narrow. You might use struckturred grid.

At your outflow you have, in extreme case, a very small opening with a very fine discretisation, as you have already said. If you use a strucktured grid you have at the beginning of your computational area "big" distances between your grid points compared with the small distances at your outflow region. So you have big differences between the width of your kontrollvolumes (because of the stenosis). That makes your convergence very slow.

Try to devide your computational area in more that one regions and try to use kontroll volumes that are not very different. (Might help ! keep me in touch.)

best regards

Jannis


All times are GMT -4. The time now is 09:40.