CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Steady needs unsteady. (http://www.cfd-online.com/Forums/main/14128-steady-needs-unsteady.html)

nico September 21, 2007 04:50

Steady needs unsteady.
 
The flow i am looking at has a fair bit of separation on a foil (3D) (around 15 % of the foil).

If solved with a steady solver, * k-eps, 2nd o on momentum: no convergence on the residuals, 'convergence' of the forces * SST, 2nd o on momentum: no convergence on the residuals, oscillation of the forces * k-eps, 1st o on momentum: convergence on residuals and forces.

So far so good, it's was I expected. All forces results look similar. (average of oscillation for SST)

Now if i switch to the unsteady solver, after around 20s the flow becomes steady: forces are constant, and flow characteristics are steady.

Obviously my unsteady calc takes 'much' longer to compute, why do I need to run this case in unsteady to reach steady state? I have also tried to start from the 1st order case and to go 2nd order when converged without better success.

Any comments are more than welcome.

Regards,

Nico


All times are GMT -4. The time now is 14:42.