Modeling Phasic Turbulence
You can model the effects of turbulence on both the continuous and dispersed phases.
Simcenter STAR-CCM+ allows you to model the effects of turbulence on continuous and dispersed phases in several ways:
- Continuous and dispersed phases can each be modeled with their own set of equations for the turbulence energy and dissipation. You can model phases independently as laminar, turbulent or with different turbulence models.
The turbulence equations are identical to the single phase formulation, except each phase is scaled with a factor of the volume fraction of that phase. The same phase volume fraction also modifies convection and diffusion fluxes of phase-turbulence quantities.
- The turbulence of the dispersed phase can be calculated from the turbulence of the continuous phase using the Turbulence Response Model. This algebraic model reduces the computations that are required as only the continuous phase equations are solved. However, the model couples the turbulence of dispersed phases to the continuous ones.
- The phases can also be coupled through bubble/particle induced turbulence of the dispersed phase which adds more source terms to the continuous phase turbulence energy and dissipation equations.
- In addition to its contribution to the model turbulence, particle-generated turbulence can also be dissipated locally at the particle scale. This effect can be thought as a local enhancement of the mixing and is accounted for by the Particle Induced Mixing Model or Particle Induced Turbulence Source.
- The transfer of turbulence from one phase to another is accounted for by the Interphase Turbulence Transfer Model.
Simcenter STAR-CCM+ supports Large Eddy Simulation (LES) turbulence in multiphase flows. Large Eddy Simulation is a technique intermediate between the direct numerical simulation of turbulent flows and the solution of the Reynolds-Averaged Navier-Stokes (RANS) equations. In LES the contribution of the large, energy-carrying structures to momentum and energy transfer is computed exactly, and only the effect of the smallest scales of turbulence is modeled. Since the small scales tend to be more homogeneous, isotropic, and universal, and less affected by the boundary conditions than the large scales, the LES models can be simpler and require fewer adjustments when applied to different flows than similar models for the RANS equations. A good application for LES in a multiphase flow is a bubbly flow.
Simcenter STAR-CCM+ supports the following Sub-Grid Scale models for LES:
-
Smagorinsky Sub-Grid Scale
-
Dynamic Smagorinsky Sub-Grid Scale
-
WALE (Wall Adapting Local Eddy-Viscosity) Sub-Grid Scale
The dispersed phase can use the Laminar model instead of a turbulence model. However, if you expect the dispersed phase to carry a significant proportion of the wall stress or wall heat transfer, do not use the Laminar model.
The steps in this procedure are intended to follow on from one of the following:
To model phasic turbulence:
- In the Phase Model Selection dialog, in the Viscous Regime group box, select Turbulent.
-
In the
Turbulence group box, select one of the following:
-
Large Eddy Simulation
This option is available only for three-dimensional, implicit-unsteady simulations.
For more information, see Large Eddy Simulation (LES).
-
Reynolds-Averaged Navier-Stokes
For more information, see Reynolds-Averaged Navier-Stokes (RANS) Turbulence Models.
-
-
If you selected
Reynolds-Averaged Navier-Stokes, choose one of the following from the
Reynolds-Averaged Turbulence group box:
-
K-Epsilon Turbulence
This model provides a good compromise between robustness, computational cost and accuracy. It is appropriate for industrial-type applications that contain complex recirculation, with or without heat transfer.
See K-Epsilon Turbulence.
-
K-Omega Turbulence
This model is similar to the K-Epsilon model in that two transport equations are solved, but differ in the choice of the second transported turbulence variable. This model is suitable for aerospace applications.
See K-Omega Turbulence.
-
For situations in which the turbulence is strongly anisotropic, such as the swirling flow in a cyclone separator, select Reynolds Stress Turbulence
-
To calculate the turbulence of the dispersed phase from the turbulence of the continuous phase, select Turbulence Response.
-
-
Choose the models that are appropriate to your selection.
Each model has its own collection of sub-models. For more information on the available models, see Reynolds-Averaged Navier-Stokes (RANS) Turbulence Models and Subgrid Scale Turbulence Models.
When the Tchen Turbulence Response model is selected for a phase, you must ensure that the Virtual Mass Coefficient model is selected under the corresponding phase interaction. This model is usually selected automatically in a turbulent flow, since the Virtual Mass Coefficient is also a required model for the Turbulent Dispersion Force model.
The Virtual Mass Stress model is normally used together with one of the Reynolds Stress turbulence models. If a Reynolds Stress Turbulence model is selected for the continuous phase, and a Turbulence Response model is selected for the dispersed phase, then the Normal Stress Term property should be activated on the Turbulence Response model.
The Virtual Mass Stress model can also be activated together with K-epsilon or K-omega turbulence models, although this is not recommended. For reasons of balance, whichever combination of turbulence models you use with Virtual Mass Stress, you should activate the Normal Stress Term property either on all of the models or on none of the models. In this context, full Reynolds Stress models always include normal stress terms in the momentum equation (unless the Use Boussinesq Approximation property is activated for initialization purposes or for diagnostic reasons).
-
If you activated K-Epsilon or K-Omega turbulence, and you want to apply a custom scaling factor to the turbulent viscosity, select
Turbulent Viscosity User Scaling from the
Optional Models group box.
Use this model when the turbulent flow includes a range or mixture of densities.
- If you want to suppress turbulence in a particular region, select Turbulence Suppression from the Optional Models group box.
- If you want to model the transfer of turbulence between the phases, select Interphase Turbulence Transfer.
- If you want to model the interaction between the dispersed phase and the surrounding turbulent eddies, select Turbulent Dispersion Force.
-
If you want to model the turbulence due to the presence of bubbles, droplets, or particles in the continuous phase, select one of the following:
- Particle Induced Turbulence
Source or LES Particle Induced
Turbulence
Particle Induced Turbulence Source is available when Reynolds-Averaged Navier-Stokes is activated in the continuous phase. LES Particle Induced Turbulence is available when Large Eddy Simulation is activated in the continuous phase.
- Particle Induced
Mixing
This model contributes to the effective viscosity of the continuous phase.
- Particle Induced Turbulence
Source or LES Particle Induced
Turbulence
-
Set the appropriate solver settings.
-
If you activated K-Epsilon Turbulence, K-Omega Turbulence, or Reynolds Stress Turbulence in a phase, set the corresponding solver properties.
See K-Epsilon Solvers Reference, K-Omega Solvers Reference, or Reynolds Stress Solvers Reference.
-
If you activated a turbulence response model in the continuous phase, set the Response Coefficient Turbulent Viscosity solver properties.
-
If you activated a Large Eddy Simulation model in a phase, set the corresponding solver properties.
-
Return to the appropriate workflow:
-
Modeling Bubbly and Droplet Flows and continue with Step 2.
-
Modeling Multiple Flow Regimes and continue with Step 2.
-
Modeling Particulate Flows and continue with Step 4.