CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Help needed with free surface simulation (

Mark October 24, 2006 08:21

Help needed with free surface simulation
I have a simulation with a open-top container completely full of air (i.e. domain's initial condition of air's volume fraction is 1). This container's walls are spinning, and I am injecting water into the domain through an inlet, such that the water builds up on the walls and exits the container continuously.

Everything seems to solve fine as a steady state problem, but when I look at my results there is hardly any water in the container (i.e. the volume fraction of water is way too low on the walls of the container, where it should be about 1).

Does anyone have any idea what I might be doing wrong?

Johnny October 24, 2006 17:51

Re: Help needed with free surface simulation
What are your imbalances like? Are you conserving mass? What does your BL mesh look like? Do you have buoyancy turned on?

Mark October 26, 2006 14:43

Re: Help needed with free surface simulation
Johnny, Thanks for the response.

I have buoyancy turned on and much of the simulation is based on the tutorial "Free surface flow over a bump".

I have an inflation layer around the outer edge of my device, which is essentially a bowl. So I have inflation elements on the walls of the bowl where water will build up.

I'm not sure what you mean by conserving mass or imbalances. The mass should not be conserved because I have an opening so mass can flow in and out.

If I create an isosurface at "volume fraction of water = 0.5", the only surface I see is right around the inlet. However, if I turn this down to much lower values (<0.1), I begin to see the flow of the water up the edges of the bowl.

It almost seems like the water is vaporizing.

Jason October 27, 2006 02:15

Re: Help needed with free surface simulation

you can create a plane,and select the color as variables,then select the volume fraction --0.5,then you can see what you want,if you had the right setting and solving.

Good luck Jason

Mark October 27, 2006 12:07

Re: Help needed with free surface simulation
I'm now finding that my solution won't converge. What is a typical time frame for convergence of a free surface, steady-state, problem (should I be allowing it to run all day or should I expect accurate results within a few hours)?

All times are GMT -4. The time now is 13:29.