turbulence
Turbulence model configuration
RANS turbulence model (sst = Menter SST k-ω, sas = Scale-Adaptive Simulation, sa-neg = Spalart-Allmaras Negative, sst-transition = γ-Reθ transition)
| Value | What it does |
|---|---|
| sst | The Menter Shear Stress Transport Turbulence Model (https://turbmodels.larc.nasa.gov/sst.html) |
| sas | Scale-Adaptive Simulation, run as a standalone RANS model |
| sa-neg | Negative Spalart-Allmaras turbulence model (https://turbmodels.larc.nasa.gov/spalart.html) |
| sst-transition | The Langtry-Menter 4-equation Transitional SST Model (https://turbmodels.larc.nasa.gov/langtrymenter_4eqn.html) |
| spalart-allmaras | Spalart-Allmaras turbulence model (https://turbmodels.larc.nasa.gov/spalart.html) |
Hybrid RANS-LES approach (DES = Detached Eddy Simulation, DDES = Delayed DES, IDDES = Improved Delayed DES, SAS = Scale-Adaptive Simulation)
| Value | What it does |
|---|---|
| DES | Detached Eddy Simulation |
| DDES | Delayed Detached Eddy Simulation |
| IDDES | Improved Delayed Detached Eddy Simulation |
| SAS | Scale-Adaptive Simulation run as the hybrid branch of |
Limit eddy viscosity in SST model
Gradient limiter coefficient for turbulence kinetic energy k
QCR variant year (2000, 2020, or 2021)
Guidance: Choosing a turbulence model →Enable the rotation correction. Read by both turbulence models, in their own forms: SST applies Hellsten's f4 = 1/(1 + Crc*Ri) to the omega destruction term, and SA adds Crot*min(0, S - Omega) to production (Dacles-Mariani / Spalart-Shur).
Production term model for SST (vorticity = SST-V, strain-incompressible = strain-based for incompressible, strain-compressible = strain-based for compressible, kato-launder = Kato-Launder). When omitted the solver build chooses: kato-launder for sst-transition, vorticity otherwise (the solver's Menter SST defaults) — a schema default here would silently override the transition build's choice.
| Value | What it does |
|---|---|
| vorticity | The vorticity based production term (SST-V) |
| strain-incompressible | The strain based production term for incompressible flows |
| strain-compressible | The strain based production term for compressible flows |
| kato-launder | The Kato-Launder source term |
Constant in the dissipation term on the k equation in the SST model
The k-omega part of the blended C_DES in SST
Guidance: Choosing a turbulence model →The k-epsilon part of the blended C_DES in SST
Guidance: Choosing a turbulence model →Constant in the f_dt blending function in DDES and IDDES
Guidance: Choosing a turbulence model →Constant in the f_dt blending function in DDES and IDDES
Guidance: Choosing a turbulence model →Constant in the length scale calculation in IDDES
Guidance: Choosing a turbulence model →Constant in the eddy viscosity limiter in SST
Guidance: Choosing a turbulence model →C_DES constant used for SA-neg based hybrid RANS/LES models
Guidance: Choosing a turbulence model →Constant in the production of the gamma equation in the transition model
Guidance: Choosing a turbulence model →Constant in the destruction term for gamma equation in the transition model
Guidance: Choosing a turbulence model →Constant in the production term of the gamma equation in the transition model
Guidance: Choosing a turbulence model →Constant in the destruction term for gamma equation in the transition model
Guidance: Choosing a turbulence model →Constant in the Re-theta transition correlation in the transition model
Guidance: Choosing a turbulence model →Constant in the transition model in the compressible solver
Guidance: Choosing a turbulence model →Constant in the transition model in the compressible solver
Guidance: Choosing a turbulence model →Activates the separation correction for the transition model
Guidance: Choosing a turbulence model →Restrict the LES branch of the DES/DDES/IDDES length scale to cells whose nominal CFL is at or below this limit. 0 (the default) disables the shield, which is also always inactive unless the run is time accurate
Guidance: Choosing a turbulence model →Width of the blend above les_cfl_limit. The shield ramps from LES at les_cfl_limit to full RANS at les_cfl_limit plus this width.
Guidance: Choosing a turbulence model →Weight on the eddy turnover contribution to the shielded CFL. The time step has to resolve both the transport of an eddy across the cell and the eddy's own turnover, and the shield takes whichever is larger. 0 leaves the transport only criterion, which cannot see a cell whose speed relative to the mesh is near zero. Defaults to 1, and has no effect unless les_cfl_limit is set
Guidance: Choosing a turbulence model →LES subgrid-scale model (wale = Wall-Adapting Local Eddy-viscosity, deardorff = 1.5-order prognostic SGS-TKE with a stratification-limited length scale, none = no SGS model)
| Value | What it does | When to use it |
|---|---|---|
| none | Implicit LES (no explicit SGS model). | |
| wale | The Wall-Adapting Local Eddy-viscosity model. | |
| deardorff | The Deardorff 1.5-order closure, which transports the subgrid turbulent kinetic energy and limits the mixing length under stable stratification. The eddy viscosity is and the heat diffusivity is times it. | For stratified atmospheric boundary layers with the incompressible solver. The compressible solver does not implement it. |
Enable wall damping for LES model
Under-relaxation factor for the SGS turbulence equation. Read UNCONDITIONALLY by the C++ (IncompSolverData.h initKernels, turbulence.relax), so it must always be present in the emitted dict -- it carried a default of 0.5 in the legacy schema and was lost when the schemas moved, leaving the deardorff path with no way to supply it.
von Kármán constant for WALE near-wall scaling