[niriosrat]

The main purpose of the current simulations is to estimate the velocity at which a bubble of gas (known as Benjamin bubble) propagates along a liquid phase in a horizontal pipe considering a slug flow pattern. This velocity is known as the drift velocity and it is of interest to understand and predict its behavior under the variation of several parameters such as viscosity, density, surface tension (related with the Froud and viscosity dimensionless numbers) of the substance considered or diameter of the pipeline.


The predicted bubble drift velocity aggress with the experiments and literature data for water (0.001 Pa s) and oil viscosities up to 0.15 Pa s with an error up to 5%. When the viscosity increases at 0.3 Pa s the error increases around 30%. We are using STAR-CCM+ with VOF. Any suggestion on how to improve the results? We tried modified the sharpening factor, the angular factor and the Interface Momentum Dissipation (IMD), but with no luck. Any recommendation?

[Deranda]

Hi

maybe you should work on your mesh and time-step.
The VoF Model needs a finde mesh at the interface of your 2 face flow.
And to avoid inadvertently surface convection, such as u have no sharp interface.
Another adjustment is to set the integral order to 2nd order in the solver settings