CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   initial value file (http://www.cfd-online.com/Forums/cfx/19171-initial-value-file.html)

Olivier September 18, 2002 04:57

initial value file
 
Hello All,

I would like to perform a SST calculation in CFX 5.5.1 The idea was to use a k-epsilon calculation and then use the results of the k-epsilon calculation as initial values for the SST calculation in order to help the convergence. Thus I generate the fine mesh with the first node near the wall for the SST calculation and specify this definition file into CFX Solve. Then I add the name of my k-epsilon result file as the initial value file. But the help in the pdf books says that the mesh is then read from from the initial value file, and apparently it's what happens. So how can I proceed to use my fine mesh refined to the wall and the velocity field comimg from the coarse mesh used for the k-epsilon model ?

Has anyone a suggestion ? Thank you for your help.

Olivier

Dave September 18, 2002 05:54

Re: initial value file
 
HI Olivier, It sounds as though your two models have different mesh. If this is the case you will need to interpolate the results from the Ke results file onto the SST def file. Then all you need to do is run the SST def file without any initial values file specified. To do the interpolation you will find a little gui in the solver manager under tools. Its very self explanitory. Hope this helps Dave

Michael September 19, 2002 00:01

Re: initial value file
 
Did you try starting straight away with the SST model on the fine mesh with a sensible initial guess and no initial results file? The robustness of the SST shouldn't be very different to k-e, so you may not need the initial results file. Mike


All times are GMT -4. The time now is 19:08.