CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Simulating two fluids in zero gravity environment (

siw February 10, 2010 11:47

Simulating two fluids in zero gravity environment

I have CFX v12.1 and may need to model a liquid flow from a reservoir (which once empty could have air replace the volume previously occupied by the liquid) in an accelerating/decelerating vehicle in a zero gravity environment. This is outside my usual area of expertise - atmospheric external aerodynamics.

Would CFX be able to simulate such a fluid flow?


ckleanth February 10, 2010 12:35



siw February 10, 2010 13:44

ckleanth is this an area you have experience in? If so any words of wisdom.

ghorrocks February 10, 2010 17:44

It sounds like a standard free surface model, but one where the gravity vector is a function of time. This is only a trivial addition to a free surface model. It might get a bit more complicated if your space ship is rotating in addition to accelerating on occasion - then you need to consider whether a rotating frame of reference or moving mesh is required. But if not rotating then you can use a stationary mesh and it is not too complicated.

siw February 11, 2010 11:36

Thanks Glenn, is this something you've done before? I expect the vehicle motion to only be translation.

For free surface simulations are RANS adequate or does LES have to be used? And any thoughts on dealing with the mesh at the phase interface, especially if the interface is unsteady?

ghorrocks February 11, 2010 17:57

Free surface modelling can be done in either RANS or LES.

For free surface models make sure you have a high quality mesh near the interface, hexs if possible. Mesh refinement will resolve the interface better but if the interface moves you will probably have to do global refinement. Remeshing based on free surface motion is tricky in CFX.

ckleanth February 12, 2010 06:38


Originally Posted by siw (Post 245629)
ckleanth is this an area you have experience in? If so any words of wisdom.

sorry for late reply; no mate, have never modeled a zero gravity application but as Glenn said it shouldn't be any different to a standard free surface model. your ship movement should define the gravity vector however I'd think you'd need fine mesh throughout your domain

All times are GMT -4. The time now is 21:37.