Published in

Volume 1B, Symposia: Fluid Machinery; Fluid-Structure Interaction and Flow-Induced Noise in Industrial Applications; Flow Applications in Aerospace; Flow Manipulation and Active Control: Theory, Experiments and Implementation; Multiscale Methods for Multiphase Flow; Noninvasive Measurements in Single and Multiphase Flows

DOI: 10.1115/fedsm2014-21565

Links

Tools

Export citation

Search in Google Scholar

Interface Tracking Methods Applied to Phase Separation

This paper was not found in any repository, but could be made available legally by the author.
This paper was not found in any repository, but could be made available legally by the author.

Full text: Unavailable

Green circle
Preprint: archiving allowed
Green circle
Postprint: archiving allowed
Red circle
Published version: archiving forbidden
Data provided by SHERPA/RoMEO

Abstract

We are interested in the phase separation of a light fluid in a heavier fluid in a closed box. A simple initial configuration is considered. A cubic volume of light fluid with a characteristic length H/2 is contained in the bottom part of a cubic box of side H filled with a heavier fluid. This problem is typical of multiscale interfacial flows in which large interfacial scales coexist with small dispersed droplets interacting with larger drops according to coalescence or rupture of interfaces. A benchmark of phase separation is proposed here. Several CFD codes devoted to multiphase flow simulations are evaluated on DNS simulations for different meshes. The VOF method, the level set method and the AMR method are used for tracking interface, the final objective being to see if codes using different numerical methods converge to the same macroscopic quantities. Time evolution of potential energies, kinetic energies, and interfacial area are computed for each simulation and used for codes comparisons. These physical quantities are also used to ensure that grid convergence is achieved. The phase separation also induces turbulence effects. In order to quantify it, time evolution of enstrophy is evaluated for each simulation. It is shown that all the codes converge for all macroscopic quantities for two turbulent Reynolds numbers, except for enstrophy. This difference is due to an under resolution of the vorticity.