CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Multiphase Problem Settings (http://www.cfd-online.com/Forums/fluent/102779-multiphase-problem-settings.html)

aniruddha.nasalapurkar June 2, 2012 02:08

Multiphase Problem Settings
 
I am student doing my project on FLUENT, piston analysis for oil cool gallery application.
I am using VOF model. In that my solution is diverging. Can anyone help me out with the appropriate settings?
My objective is to find out the heat transfer and temperature distribution in the cooling oil gallery.
I have tried with SIMPLE and PISO but still the solution is diverging.

sicfred June 5, 2012 03:55

Hi,
Is your simulations steady or transient? wich method are you using for VOF (implicit or explicit). If you are in transient and explicit, you should decrease the time step (check the courant number), Piso is a good option, but I do not think it is your problem.

aniruddha.nasalapurkar June 11, 2012 03:13

Thank you sir for yor reply.

Avtually i am using Transient, VOF (implicit) settings. Still the solution is getting diverged in the "epsilon".

Please help me for the correct settings.

Jabba June 12, 2012 14:09

Quote:

Originally Posted by aniruddha.nasalapurkar (Post 365753)
Thank you sir for yor reply.

Avtually i am using Transient, VOF (implicit) settings. Still the solution is getting diverged in the "epsilon".

Please help me for the correct settings.

I believe that for transient calculations, the VOF explicit option would be more consistent. Also you should use the "Geo-reconstruct" discretization for the volume fraction. I believe that the "first-order" is the default for the implicit method and this is not good since the interface between the phases is not well defined through this discretization.

I have seen in some works the application of PISO method with "0" value set in the "Skewness correction" option also.

You should also check if your mesh meets the requirements of turbulence model and wall functions in use.

Good luck.

sicfred June 13, 2012 13:11

Looks like your problem is more with turbulence model (epsilon divergence) than the VOF, so maybe you should check this part.
As Jabba wrote, the best option of VOF is the explicit with geo-recronstuc, specially when you have to simulate an interface (free-surface), but also can be very unstable and you have to be care of the courant number or use the NITA option. Another option of VOF (no so good as explict-geo) is use the VOF implicit with "modified HRIC" that has a little diffusion in the interface but is more stable


All times are GMT -4. The time now is 09:10.