CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   modeling two-phase flow with VOF solver (

pranab_jha March 8, 2010 17:40

modeling two-phase flow with VOF solver
Hi guys,

I am using VOF solver to model stratified/slug flow in a horizontal pipe.

Here goes my model description:
Dimensions: 3D, Pipe, 50 mm dia, 12.7 m length
BC: inlet- velocity inlet
BC: outlet- pressure outlet (atmospheric pressure)
Phases: water-primary, air-secondary
Unsteady case.

I am specifying inlet velocity as 1 m/s for the mixture and volume fraction as 0.1. But I cannot specify the individual phase velocities. I need to plot pressure gradient vs. both phase velocities. How can I achieve this?


Bak_Flow March 8, 2010 21:01

Hi Pranab,

I think you are trying to DETERMINE 2-phase mixture pressure gradient as a function of air and water flow rates or equivalently superficial velocities of each phase?

The VOF model solves one velocity field and a volume fraction to resolve the interface between the two phases. It is only applicable if you can resolve the details of the interface ie. simple interface structure.

The boundary conditions for phi (phase fraction) should be 1 or zero NOT some mixture volume fraction. The boundary conditions for velocity therefore are the appropriate phase velocity in the regions where that phase exists.

I think you should look at some tutorials or examples of applying the VOF model to horizontal pipe flows before you get too deep into this.



Ganapathy March 9, 2010 00:16

VOF is a lesser model. It is used mainly to track the interface. IF you want to specify different velocities for the phases, use the E-E model.

pranab_jha March 9, 2010 11:15

@Bak_Flow: Did not understand what you are saying. I already specified the volume fraction as I said. My problem is how to specify the superficial velocities of the two phases. As per you, I should specify the fraction as 1 or zero. But then will it not mean that the flow is a single phase not two-phase at all. Is there any way around it?

I am using VOF partly because I want to track the interface in slug flow. What I could understand from the userguide in Fluent was that VOF is the best way to model slug and stratified flow. I will look again at the EE model, but I think VOF is better.

pranab_jha March 10, 2010 00:53

Hey Guys,
I tried the mixture model too... but it asked me for the bubble diameter i think. I want to model stratified flow but I specified it as 10mm. I did not get convergence for continuity in a steady flow.
Any suggestions?

jsm March 12, 2010 00:20


For VOF model, only one phase should be applied in inlet i.e volume fraction should be 1 for any one phase at inlet. For slug flows, just divide the inlet into two parts and define air and water inlet separately. This may be looking like wrong approach. But it is good approximation and just extend the inlet to some length to develop the real flow phenomena correcly. Also patch some initial values for water and air inside the domain to avoid the convergence issue.

All times are GMT -4. The time now is 12:27.