Hello,

Well, I have a problem... I am working on a two-phase flow simulation in a long pipe. I am using the VOF model(steady). The boundary conditions I use are the following: 1) velocity inlet for phase 1 2) velocity inlet for phase 2 3) outflow The simulation seems OK and converged. However, I don't get mass conservation even when I decrease the residual. Is there something wrong with my simulation or is it something to do with the VOF model?

Here is what I obtain:

mixture Mass Flow Rate (kg/s) inlet_inf 45.185421 inlet_sup 217.28444 outlet -222.14208 Net 40.327781

Any idea would be great. Many thanks!

pat77

[kenan]

Hi,

I have the same problem. I'm simulating the two-phase air-water flow in a corrugated pipe of 60 m long but it turns out to be a significant mass imbalance between inlet and outlet. May the problem be due to curvature of corrugations? Corrugations are annular and quite frequent with a depth to pitch ratio of 1/5. (Depth is height of corrugation, while pitch is distance between each corrugation element). I haven't had any problem with smaller ratios using the same model.

I modeled only the quarter pipe, assigning symmetry condition to the longitudinal mid-plane of the pipe. The top boundary is defined as wall, except for a small face on it near the pipe inlet, to allow air in where velocity inlet BC is applied. Pipe inlet is defined as mass flow inlet for water. The discretizaton scheme i employed for volume fraction is Modified HRIC.

Any help would be appreciated.

Regards,
Kenan

[kenan]

I got the solution. I found out that it was just because of coarseness level of the grid, as the VOF method needs finer grid to resolve the flow structures in curvature-dominated flows. So i created a finer mesh especially in the region where free surface to be formed. Then mass gain / loss has dissappeared.

[Hossein1]

Hi Kenan,
Could you please let me know how you check 'mass conservation' of VOF method in Fluent?

Thanks

[kenan]

Hi Hossein,

If you compare the inlet mass flow rate (which you define by fixing velocity etc.) and the mass flow through outlet boundary, then you can see if there is a difference. If they are roughly the same, then you can say grid resolution is adequate.

[Hossein1]

I'm using axisymmetric solution. Instead of an inlet mass, I have a semi-circle in front of the particle representing my droplet before the impact. After the impact, the droplet deforms to a thin liquid film around the particle as shown in the figure. My question is that how I can calculate the volume (or mass) of the droplet before the impact (and also mass of the liquid film after the impact). Which command or tool does this in Fluent?

[KevinZ09]

How about using a volume integral of your density? I'm not sure if you can integrate only over your liquid volume, but if one of your fluids is air and the other is water, integrating over the whole volume is fine too I'd think.