CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   VOF Model (http://www.cfd-online.com/Forums/fluent/29598-vof-model.html)

 Arman Altincatal April 8, 2002 12:59

VOF Model

Hello, I am working on a project where I have to use VOF model to analyze the motion of a fluid that is sandwiched by another fluid. The channel is rectangular and has three domains: 1st and 3rd for ex. water and 2nd is oil. I am applying pressure at one boundary and letting the other boundary to be pressure outlet. I was able to converge the model and run a 24 time step with 1800s each. I am wandering if there is a way to determine the motion of the distinc fluids in Fluent. My time step could also be very large for this application too. Anyway, I would appreciate any help on this matter. Best regards,

Arman

 Laika April 9, 2002 04:00

Re: VOF Model

Hello,

what do you mean with: "I was able to converge the model and run a 24 time step with 1800s each"?

Do you have a time step of 1800sec?

How did you initialise your solution?

Are you including gravity?

greetings,

Laika, still orbiting

 Arman April 9, 2002 12:09

Re: VOF Model

Hi, Laika. I did initialize my solution with the pressure inlet boundary conditions(30psi,normal to boundary). I did turn on the gravity but what I am missing is I think Volume fraction residual to monitor the fraction of the fluid at different time steps. I must not have the appropriate conditions. What I meant by converging the model was that I have converged the steady state solution and then tried to run VOF but was not successful.I am also trying to read some examples about VOF modeling to make sure I assign the correct boundary conditions. Let me know if there are key steps to VOF modeling. Thanks.

Arman

 Laika April 10, 2002 04:02

Re: VOF Model

Arman,

How did you obtain your stationary solution? With VOF you must use the transient solver. You can use a converged stationary pressure/velocity field, but I suppose you did this with just one phase.

To start your VOF-simulation, you may not forget to tell Fluent what boundary-conditions there are for each phase. If you know what the solution is going to look like, then patch the secondary phase where you expect it to be.

In the solution-controls choose 'body force weighted' for the pressure calculation, and the piso-scheme for the presure-velocity coupling. You probably may set all your underrelaxation-parameters equal to 1.

good luck,

Laika, still orbiting

 Arman A. April 12, 2002 16:26

Re: VOF Model

Laika, how are you? I had a question about the VOF modeling again. I applied the appropriate boundary conditions like the PISO, body force weighted, phase 1-being air and phase 2 being oil. Before running the model I made sure that the volume fraction was 1 for phase 2 at the pressure inlet and 0 at the rest of the region(air). Basically my intention was to apply a pressure at the inlet where there is a block of oil, and adjacent to it is a block of air with a pressure outlet boundary at the far end of the model. I ran the unsteady model until the residuals were converged.I assumed that it was going to update the solution and start running the unsteady analysis at 0.5s for 10 steps but it didn't. It just said iterating. I am not sure if this is a common problem with VOF modeling. I recall from previous temp vs time analyses, that fluent just solved the time steps one after another. I would appreciate it if you could let me know if there is something that I must have forgotten to assign.

Thank You.

 All times are GMT -4. The time now is 12:53.