CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   STAR-CCM+ (http://www.cfd-online.com/Forums/star-ccm/)
-   -   Problem with implicit unsteady solver (http://www.cfd-online.com/Forums/star-ccm/73242-problem-implicit-unsteady-solver.html)

CCMuser March 2, 2010 18:54

Problem with implicit unsteady solver
 
Hello everyone,
I have been using Star CCM+ for a past few months in order to simulate swirling flows in pipes. The geometry consists of a small rotating pipe interfaced with a honeycomb followed by a very long stationary pipe. The honeycomb is rotated too. For the rotation the "Moving reference frame model" is used.
At low rotation speeds and Reynolds nos. the steady flow model is used and an acceptable degree of convergence is obtained. However at high rotational speed the flow in the stationary pipe is not any more steady hence the steady solver is not of much use.
On using the implicit unsteady solver it is observed that the flow does not develop in the stationary pipe. The tangentioal velocity component shows that the effect of swirl dies out within 0.01D. Even after waiting for a long time there is no change in the profiles. There is something fishy about the unsteady solver since the steady and unsteady solvers show completely different profiles. The solution obtained with unsteady solver, as seen from the velocity contour plots, is such that the solver gets frozen after the interface and there is no development of flow, which is not possible.
I am stuck with this problem and also the CD Adapco Customer care is not able to solve this problem.
Can anyone think of a possible reason for this problem. Is there an error in the solver? Has anyone encountered a similar problem?
SOS!:)

JBeilke March 3, 2010 09:04

Just rotate the honeycomb mesh instead of using mrf.

f-w March 3, 2010 12:20

How large is your time-step relative to the rotation rate? Is your mesh fine enough to capture rotation downstream? What models are you using? Have you tried the Rigid Body Motion solver?

To speed thing up, initialize your unsteady solution with a steady one.


All times are GMT -4. The time now is 12:13.