CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CD-adapco (http://www.cfd-online.com/Forums/cd-adapco/)
-   -   non convergence (http://www.cfd-online.com/Forums/cd-adapco/55187-non-convergence.html)

giorgio March 4, 2006 11:56

non convergence
 
hi all, i have a problem with the convergence of a complex geometry. The program that i use is ccm+ 1.08. i've tried k-e and k-w models, coupled and segregated flow (the velocity of the air is 25m/s) but in any ways after some iterations in the residuals tke and the tdr (or sdr for k-w) go up! what can i do?waht is the problem?maybe numers of cells?kind of cells?

Hubert Janocha March 7, 2006 14:50

Re: non convergence
 
In case of pressure boundaries it is always a good idea to choose the same value for the initial pressure.

Imagine, when the pressure difference is to high, the flow (in both cases steady state, or transient) evacuate the geometry, because of the strong accelaration, causes by the greate pressure difference. This gives of course negative density. Since we don't use "anti materia" for a warp engine like in the Enterprise, this is a problem.

The coupled solver is good, but it has a border, too.

If you are using pressure / pressure boundaries choose the value for the initial pressure from the outlet pressure boundary.

When this not helps, answer to this message. There are some other tricks.


giorgio March 8, 2006 11:32

Re: non convergence
 
hi Hubert, thanks for your help!i've a velicity inlet, a pressure outlet and only wall (slip (for the wall of the virtual wind gallery) and no-slip). In the outlet, static pressure is set to 0 pa, the same that i have set in the initial condition but it doesn't work. In your opinion may be the problem in the turbulence specification values?or in the courant number too high?



All times are GMT -4. The time now is 14:15.