CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Problem in applying SST k-omega (http://www.cfd-online.com/Forums/main/101478-problem-applying-sst-k-omega.html)

mb.pejvak May 5, 2012 00:54

Problem in applying SST k-omega
 
5 Attachment(s)
I plan to apply SSt k-omega turbulence model in my self-developing code. I wrote a code for laminar flow and obtained reasonable and in comparison with analytical results over a flat plate, it had 4% difference in Re=90000. So I am applying the turbulence model in the code and it start to converge up to 1e-4 in 5000 iteration. but in post processing, results are not acceptable.In that, K and Omega and also U in whole domain have inlet value just in first cell near flat plate. it seems code doesn't work. I changed some part of code such as inlet and freestream value for K and omega, converting convection descritizing from 2nd order (TVD) to 1nd order (upwind), make grid finer near solid surface (y(first row)~0.002) and ... . as it is seen in attached picture K, Omega and U near the wall(knear, omeganear files) only in first row change and almost whole domain have inelt value.

does anyone know why this happen and what wrong exist in the code.

thanks in advance.

mb.pejvak May 9, 2012 21:39

any idea? unfortunately I can fix the problem. Infact, after 4700 iteration, my result in K and omega are still constatnt in whole domain. I change my grid to achieve Y(first row)~0.0001, but nothing changed. I am completely in my wits end.

mb.pejvak May 17, 2012 20:50

finally I can fix it. problem was in under-relaxation factors for turbulence equation.


All times are GMT -4. The time now is 05:28.