CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   How to decide the boundary condition for free surface? (

dut_thinker February 5, 2011 22:29

How to decide the boundary condition for free surface?

I know that at the bottom we can use the wall function to determine the BC for turbulent kinetic energy k and dissipation e.
So how about the free surface?
Does the wall function still apply to free surface or we can use something else?

Thank you in advance

hadian February 6, 2011 03:10

at the free surface the velocity vector normal to the free surface is zero and the zero gradient is applied for tangential velocity vector.

dut_thinker February 6, 2011 19:10

Thank you for your explainnation.
I know how to set velocity at the free surface.
I mean, at present i used k-e model to simulate the turbulence.
I want to know how to set k and e at the free surface.

hadian February 7, 2011 01:50

you can use zero gradient for k normal to free surface and some researchers apply a damping factor to it too.
for \epsilon, Celik and Rodi (1988) suggested \epsilon=k^{3/2} /(0.43*h) at free surface. but Nezu and Nakagawa (1993) mentioned \epsilon=k^{3/2} /(0.18*h).
h in both equations is water depth.

dut_thinker February 7, 2011 02:16

I have got it
Thank you again

newcomer February 16, 2011 07:48


I am a bit confused about how to specify boundary condition for the top surface. I have seen many ppl set the surface as symmetry. I am wondering when such boundary condition can be applied and when do we need to use other types of boundary conditions?

Specifically, if we are modeling the flow crossing a weir, what type of boundary condition should I specify for the top surface (both in front and after the weir)...


All times are GMT -4. The time now is 16:34.