CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Issue with volume fraction (

pranab_jha June 7, 2010 15:24

Issue with volume fraction
Hi All,
I am trying to simulate two-phase flow with air and water in a pipe to get slug flow. My issue is that when I initialize the domain with VF = 1 (completely filled with water) and then let the fluids flow into the system, I get the error:
"divergence detected in AMG solver: pressure correction"
But when I initialize with VF = 0 (complete air), I do not get the error.

Can somebody please explain to me the physics behind it? Also, what is the remedy for this? I am guessing that the air is not able to push through the water when the domain is completely filled with water. Am I correct?

My flow velocities are not very high: 1.2 m/s for air and 0.1 m/s for water which leaves me with 0.6 m/s for air and 0.05 m/s for water as the superficial velocities at the inlet (I have split the inlet into two halves- one for each phase). Pipe dia is 50 mm.

Replies and suggestions welcome.


pranab_jha June 10, 2010 16:44

Hi All,
I could get a flow for low inlet velocities of air and water with initialization of volume fraction= 0.5 in the whole domain. But still, the problem persists with higher phase velocities (air = 2m/s and water = 1.4 m/s).
Can anyone tell me how low can I go with the relaxation factor for the pressure correction in the AMG solver?
Any suggestions would be appreciated.

pranab_jha June 25, 2010 17:26

I would myself go ahead here and let everybody know that the AMG pressure correction problem was solved by using 3ddp instead of fluent 3d. I read somewhere on the internet that if your maximum cell volume is less than 1e-6 cu. m., then you have to use 3ddp to get over the pressure correction problem. I cannot validate this claim, but it did work for me.

All times are GMT -4. The time now is 07:21.