CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Free surface flow with air entrainment (

Ken October 14, 2005 13:08

Free surface flow with air entrainment
Any hints to simulate using CFX free surface flow with air entrained into water and released at downstream? thanks.

Bak_Flow October 15, 2005 15:49

Re: Free surface flow with air entrainment
Hi Ken,

the "normal" free surface model (homogeneous multiphase of VOF) is not the way to go for this problem. It assumes that the 2 fluids share the same velocity field (ie locked or homogeneous) therefore the entrainment and degassing will not be well predicted.

You will have to go to something like an algebraic slip or full Eulerian model. Here you still have to be careful as these models assume that all of the dispersed "phase" (although there can be more than one of them) has the same diameter which is used to calculate all of the exchange terms. In many cases it is not easy to come up with a representative diameter due to the break-up and coalescence processes...then you have to go to a more complex model! Something like MUSIG or population ballance!

Good Luck


Jon October 17, 2005 10:56

Re: Free surface flow with air entrainment

You can do this with the inhomogeneous free-surface model.

i.e. set two continuous phases, set the model to be free-surface, but do not set homogeneous.

This will use the free-surface model (giving a sharp interface), but solve for two seperate velocity fields.


Ken October 17, 2005 14:59

Thanks.Re: Free surface flow
Thanks for all.

Jon, I tried that (free surface inhomo). but i really got confused with the drag coeff. I need a big Cd to initiate air entrainment from water surface, but I doubt the value is valid for air bubbles transporting in water. If I just simply set Cd such as 0.44, there's no air entrainment at all. Any inputs?

Bak_Flow October 17, 2005 23:01

Re: Thanks.Re: Free surface flow

this is what I meant earlier about the challenges in modeling the phase interaction terms in such a simplified fashon. ;-)

You can play with the dispersed phase diameter and the drag correlation a bit. However, the physical processes of entrainment, break-up and coalescence, degassing etc will require more details in the model.

Best of Luck,


All times are GMT -4. The time now is 17:13.