Hello,

I'm studying the oscillations of a liquid in a U-tube. I use the VOF model with geo-reconstruct. My problem is that I have to report precisely (+- 1mm) the position of the interface during the simulation but my interface is too thick (1~2 cm). I thought at first that it was due to my meshes, but lately I found a simulation where the meshes are as large as mine and the interface is zero-thickness. To define the two phase, I use the patch function as it is described in the user's guide. I also tried another way by defining the zone in Gambit but the result is the same. Does anybody know where I am wrong ? Thanks for your help.

Jean-Philippe.

hi,

Having zero-thickness interface is not possibble with the VOF model since the interface is modeled without applying interface boundary conditions but instead by modelling it as a region through which the properties (density, viscosity) and the forces are changing. In my opinion, you have to have a finer mesh.

regards

özgür

Thanks for your reply,

That was also what I thought at first, but as I said I have been given some of my predecessor's files and apparently he has a zero-thickness interface with a large grid. Unfortunately he didn't explicit how he managed to do it and I have no way to contact him. Do you know where I can find some documentations about how Fluent deals with the interface using the VOF model ? In any case, I will have to explain this problem in my report.

Another question if you don't mind, may I just finer my meshes around the equilibrium level so as to minimize the computation time ? Indeed, with my large grid, the big oscillations fit very well with the experimental data. I have just a problem with the small oscillations, the damping is too weak compared with the experiments.

Jean-Philippe.

abput your question ... Do you know where I can find some documentations about how Fluent deals with the interface using the VOF model ? Simply look into the fluent documentation. Also have you tryied different discretization scheme for space and time step, make sure sure your convergence criteria are low enought ... the best in fact is to disable the convergence criteria checking and fix a large number of iteration per time step, there you are sure that you solution is correctly converged during the time step. Hope this help ! Thomas