Hi,

I have one question regarding free surface modelling in starcd. Q: There is a limit of courant no (0.3). That means coarser the mesh bigger the time step for this courant no. Why there is a need of refining the mesh, especially in flow with cavitation.

Thanks

a) there is indeed a limitation on the Courant number *near the liquid-gas* interface. This is inherent to the interface-tracking algorithm. You cannot allow one cell to empty completely within one time step, whereas the cell next to it is still filled (this can happen if you have the Courant number is approx. 1). To understand the algorithm, I would recommend you to take a look to the PhD thesis of Uno Ubbink (Imperial College, 1997), which you can donwlaoad from the internet.

b) For simulating cavitating phenomena, you need to have the mesh fine in the regions where cavitation appear for 2 major reasons: 1) you have to resolve the cavitation zone accurately enough 2) the cavitation inception and development depends also on the behaviour of the flow in the boundary layer, so the mesh has also to be fine there.

Hope this helps.

Let me know if star-cd can solve cavitation. STAR-CD support people say it cannot.

I am sorry to disagree with you on that point. The STAR-CD code *can* solve cavitation problems, of course up to the limits of the model used. See also the CD on the cavitation workshop held in Hamburg last year which you can obtain from your local support office.

I did ask support office last year and they said 3.15 cannot do it. They were expecting 3.2 would do it, but now they are saying 3.2 cannot do it either.

4xF is right, STAR-CD can handle cavitation. I do not know how accurate, but it can do. Maybe you should complain at London office for the poor support.