# Turbulence Model Solver

# Type of turbulence model

For simulations where Reynolds-averaged Navier-Stokes equations are solved, turbulence models are needed to calculate Reynolds stresses. See https://www.wikiwand.com/en/Turbulence_modeling#/Common_models (opens new window) for common options.

We offer two eddy viscosity models based on the Boussinesq approximation, i.e.

  • the one-equation Spalart Allmaras (S-A) model,
  • the two-equation k- shear stress transport (SST) model.

Extra partial differential equation(s) of the corresponding turbulence model are solved, together with the time averaged Navier-Stokes equations.

# Turn on the rotation-curvature correction?

The Spalart-Schur rotation-curvature (R/C) correction is implemented in Flow360 for the Spalart Allmaras turbulence model, to account for the effects of flow rotation (e.g., inside vortices) and streamline curvature (e.g., around curved walls). Note that this R/C correction is NOT turned on by default.

# Turn on DDES?

This option is only available when steady is turned off, i.e., simulation is unsteady.

Delayed-Detached-Eddy-Simulation (DDES) turns off the turbulence model away from walls and relies on the mesh and unsteady time steps to resolve the effect of turbulence away from walls. It is one of the most commonly used form of hybrid RANS-LES, where RANS stands for Reynolds-Averaged-Navier-Stokes and LES stands for Large-Eddy-Simulation.

# Tolerance (absolute) of the turbulence model solver

This is the convergence criterion for all the nonlinear residual(s) of the turbulence model equation(s). The nonlinear iterations through pseudo-time steps only complete when:

# Number of linear iterations within each turbulence model nonlinear iteration

Each nonlinear (quasi-Newton) iteration requires solving a system of linear equations, for which we use an iterative method. This parameter specifies how many linear iterations to use when solving the system of linear equations iteratively.

# Kappa parameter in the MUSCL scheme for the turbulence model solver

Specify a number between -1 and 0.333. -1 is more robust and 0.333 is the least dissipative.

# Order of accuracy for spatial discretization of the turbulence model equation(s)

This order of accuracy is referred to the spatial discretization schemes for the turbulence model equation(s), particularly the convection terms. Specify either 1 for first-order or 2 for second-order. You should use second-order solver for an accurate flow solution. But sometimes first-order solutions are useful as initial conditions for cases that are difficult to converge.

← Volume Output →