CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Fluent VOF Method - At a total loss advice required please (http://www.cfd-online.com/Forums/main/63526-fluent-vof-method-total-loss-advice-required-please.html)

LSF April 11, 2009 09:27

Fluent VOF Method - At a total loss advice required please
 
Hi,

This is my first post so please be kind :)

I've used this forum a lot for different projects in the past and have always found it to be very helpful in the past without ever even having to post, so thanks to everyone on here!

I have an issue that I was wondering if you could help me out with....

I am trying to analyse fluid dispersion from a tube in a free stream of air.

I have set up a control volume with the tube in it...In Fluent I then selected the VOF (volume of fluid) option, unsteady state problem, two phases - air and water, with turbulent flow and to solve explicitly.

Plotting a filled contour plot of the system before I began iterating suggests both fluids are in the correct position etc.

I've set the time step as 1E-6 based on the cell volume to velocity ratio etc.

The problem however is no matter what I try the solution only ever diverges and the liquid flow doesn’t leave the tube into the air flow.....

Am I using the correct methods? Should I try solving it implicitly?

I'm really sorry for the long first post, if you're still reading, thank you. I am at a total loss here.

Any advice would be very much appreciated. Or even better if someone could look at it for me, that would be the most amazing thing ever!

Thanks

panda April 12, 2009 20:50

Is there any problem with your boundary conditions? the liquid at the tube may not be correctly initialized.

LSF April 13, 2009 05:38

Quote:

Originally Posted by panda (Post 212653)
Is there any problem with your boundary conditions? the liquid at the tube may not be correctly initialized.

I too have had this thought but I dont know if I have done it correctly....would someone mind having a look at my case file for me? Please :):o

radiatn April 13, 2009 05:46

this is main forum so u should have posted in Fluent forum. But what time step did you use, i have experienced divergence when time step is too big.

LSF April 13, 2009 06:13

Yeah i realise....sorry.

I tried using a time step of 1.0e-6 first but have also tried 1.0e-8 as well.

Both did not converge...the model is just a box with a tube in it meshed as coarsely as possible just to get me started with some results....can refine it later.

I run the iterations but nothing ever seems to happen to the water....when i display a contour phase plot it shows both phases but the water wont exit the tube....???

radiatn April 13, 2009 21:56

1e-08 is small enough! If you still still face problem it must be the model setup. Check back your scalar BC as well as region BC. Redo the Fluent totorial until you master it before moving on to your own model.


All times are GMT -4. The time now is 04:35.