CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Problem with phase separation in t-junction (

estenmark February 20, 2013 04:47

Problem with phase separation in t-junction

I am trying to simulate two phase flow of air and water in a vertical t-junction with a horizontal branch. All parts of the junction have inner diameter 50 mm and are 500 mm long. There is one inlet which is located at the bottom of the junction and two outlets.

I have velocity inlet with known velocities and volume fractions (air: 5m/s and 0.58 volume fraction, water: 6.21m/s). I have a known overall flow split (20/80 in branch/main arm) so first I tried outflow for outlet boundary conditions. I have also simulated using pressure outlet.

As my phases are mixed I am using the Eulerian multiphase model. I'm running transient simulations and have tried both implicit and explicit formulation of volume fraction equation. My primare phase is air. For turbulence I am using the realixable k-epsilon with enhanced wall treatment. I have enabled gravity and surface tension modelling.

I have tried both coupled and segregated solver and first and second order descretization for momentum/energy equations. For volume fraction equation I've been using HRIC.

Now finally to my problem; I get no phase separation in my simulations! There should be more air flowing in to the branch but in my simulations I get almost 50/50 of air and water in the entire domain. All simulations converge (I have 1e-5 for convergence criteria and I run the simulations until mass flow out equals mass flow in).

Does anybody have an idea on what the problem can be? To my knowledge, Eulerian should be the most suitable multiphase model as the phases mix. I have tried refining the mesh but I still get the same problem.

Any ideas/comments would be greatly appreciated!

Sorry for the long post...


sampathevs August 8, 2013 01:17

Did your problem got solved ?

abraham91 June 3, 2014 20:18

i also get the same problem

All times are GMT -4. The time now is 10:48.