Flow Models Reference

Flow models solve the conservation equations for mass and momentum.

The Segregated Flow model invokes the segregated solver which solves each of the momentum equations in turn, one for each dimension. The linkage between the momentum and continuity equations is achieved with a predictor-corrector approach. The complete formulation can be described as using a colocated variable arrangement (as opposed to staggered) and a Rhie-and-Chow-type pressure-velocity coupling combined with a SIMPLE-type algorithm. This model has its roots in constant-density flows. Although it can handle mildly compressible flows and low Rayleigh number natural convection, it is not suitable for shock-capturing, high Mach number, and high Rayleigh-number applications.

The Coupled Flow model solves the conservation equations for mass and momentum simultaneously using a pseudo-time-marching approach. One advantage of this formulation is its robustness for solving flows with dominant source terms, such as rotation. Another advantage of the coupled solver is that CPU time scales linearly with cell count; in other words, the convergence rate does not deteriorate as the mesh is refined. This model can also evaluate inviscid fluxes with the optional AUSM+ scheme, which offers advantages for various cases.

Table 1. Models Overview
Model Names and Abbreviations	Segregated Flow		SF
	Coupled Flow		CF
Theory	See Theory Guide—Segregated Flow Solver and Theory Guide—Coupled Flow Solver.
Provided By	[physics continuum] > Models > Flow
Example Node Path	Continua > Physics 1 > Models > Segregated Flow
Requires	Material: one of Gas, Liquid, Multi-Component Gas, Multi-Component Liquid, Multiphase > Volume of Fluid (VOF) (SF) (For Multi-Component Gas or Multi-Component Liquid, further models are required to expose the Flow models.)
Properties	See Properties Lookup.
Activates	Physics Models	Model Groups Equation of State Viscous Regime
	Model Controls (child nodes)	Bounded Differencing See Model Controls.
	Reference Values	Maximum Allowable Absolute Pressure Minimum Allowable Absolute Pressure Reference Pressure See Reference Values.
	Initial Conditions	Pressure Velocity See Initial Conditions.
	Boundary Inputs	See Flow Boundaries Reference.
	Region Inputs	Mass Source Option Momentum Source Option Porous Media Flux Option (SF) See Flow Regions Reference.
	Solvers	Segregated Flow (SF) See Segregated Flow Solvers Reference. Coupled Implicit (CF) or Coupled Explicit (CF) See Coupled Flow Solvers Reference.
	Monitors	Continuity X-, Y-, and Z-Momentum
	Field Functions	Absolute Pressure Absolute Total Pressure Axial Velocity Cell Relative Velocity CFL Number (CF) Density Effective Volume Explicit relaxation factor (CF) Flow Direction Helicity Lamb Vector Lambda 2 Criterion (CF) Local Time Step Mass Flow Rate Mass Flux Mass Imbalance Number of AMG cycles, for Expert-Driver (CF, with Expert Driver enabled) Pressure Pressure Coefficient PressureGrad Q Criterion Radial Velocity Reference Velocity (CF) Relative Tangential Velocity Relative Total Pressure Relative Velocity Static Pressure Tangential Velocity Total Pressure Total Pressure Coefficient Turbulent Charge Velocity Vorticity See Flow Field Functions Reference.

Properties Lookup

The following table shows which properties are used by which Flow model. Use the abbreviations given in the models overview above.


Property	SF	CF
Active Explicit Relaxation The current value of explicit relaxation, either the value computed by the Expert Driver (when the CFL Control Method is set to Expert Driver) or the value you specify for Explicit Relaxation Method (when the Expert Driver is disabled). (Read only.)		✓
Convection Sets the discretization scheme that Simcenter STAR-CCM+ uses for computing the convection flux on a cell face in appropriate transport equations. More information is given in the related topic for the Convection Term: 1st-Order: First-order upwind scheme. This scheme scales the transported quantity by the upstream or downstream mass flowrate depending on flow direction. Only use when a higher-order scheme fails to give convergence, or in order to obtain an initial solution before switching to a higher-order scheme. 2nd-Order: Second-order upwind scheme. This scheme introduces linear interpolation of cell values on either side of the upstream or downstream face. Using this scheme can lead to poorer convergence properties, but gives accuracy as good as or better than the first-order scheme. Central: Central-differencing scheme. This scheme is second-order accurate but is prone to dispersive error and stability problems. However, central-differencing is useful in large eddy simulation (LES) where upwind schemes accelerate the rate at which turbulent kinetic energy decays. Bounded-Central: Bounded-central differencing. This scheme is recommended for large eddy simulation (LES) of complex turbulent flows. This scheme applies a boundedness criterion that makes the scheme more robust than central-differencing alone. Adds the Bounded Differencing sub-node. Hybrid: Hybrid second-order upwind/central. This scheme blends second-order upwind and central differencing, and is appropriate for simulations that use detached eddy simulation (DES). Hybrid-BCD: Hybrid second-order upwind/bounded-central. This scheme is the default scheme for detached eddy simulation (DES). Adds the Bounded Differencing sub-node. MUSCL 3rd-order/CD: Hybrid MUSCL third-order/central-differencing. Adds the Bounded Differencing sub-node.	✓
Coupled Inviscid Flux Selects the discretization scheme for the coupled inviscid flux. Roe FDS Roe's flux-difference splitting scheme, adapted for use with Weiss-Smith preconditioning for all-speed flows. At the incompressible limit, the pressure difference at the flux interface is a blend of reconstructed and average states. This blending results in a less dissipative scheme for low speed, constant density flows. Roe FDS - no blending Same as the Roe FDS option, but without blending at the incompressible limit. AUSM+ FVS Uses Liou’s AUSM+ flux-vector splitting scheme.		✓
Discretization Selects the discretization scheme that is used for evaluating face values for convection and diffusion fluxes. 1st-order Selects the first-order upwind discretization scheme. 2nd-order Selects the second-order upwind discretization scheme. Central Selects the central-differencing discretization scheme (available with Large Eddy Simulation (LES) only). Bounded-Central Selects the bounded central-differencing discretization scheme (available with Large Eddy Simulation (LES) only). Adds the Bounded Differencing sub-node. Hybrid Selects the hybrid second-order upwind/central-differencing discretization scheme (available with Detached Eddy Simulation (DES) only). Hybrid-BCD Selects the hybrid second-order upwind/bounded central-differencing discretization scheme (available with Detached Eddy Simulation (DES) only). Adds the Bounded Differencing sub-node. MUSCL 3rd-order/CD Selects the hybrid MUSCL 3rd-order/CD convection scheme. Adds the Bounded Differencing sub-node.		✓
Delta-V Dissipation When On, this property enables an extra "delta-v" dissipation (in addition to Rhie-Chow) that resembles a term in the coupled flow solver. This dissipation comes from the right-most term in Eqn. (924). This enhancement allows a robust treatment of porous media and similar large body forces by making the segregated solver behave more like the coupled solver, and fixing checker-boarding at a porous/non-porous interface. This property is deactivated by default.	✓
Explicit Relaxation A scaling factor that is used to relax all corrections explicitly before they are applied to the variables.		✓
Flow Boundary Diffusion When activated, this property includes the flow-boundary diffusion fluxes (or viscous fluxes for flow models) as given by Eqn. (899). This property is activated by default.	✓	✓
Integration Switches between implicit and explicit pseudo-time integration. Implicit The coupled implicit solver is invoked. Adds a Coupled Implicit node to the Solvers node with the option to set a Courant number ramp, grid sequencing, and expert driver for the solver. Explicit The coupled explicit solver is invoked. Adds a Coupled Explicit node to the Solvers node with the option to set a Courant number ramp.		✓
Limiting Acoustic-CFL The limiting acoustic CFL, applied for unsteady simulations, to scale the mass flux dissipation. Larger values cause more local dissipation and damp more spurious oscillations, but at the expense of accuracy. Make sure that non-constant values vary smoothly in space. When set to Per-Model, you specify the value in the corresponding child node. When set to Per-Region, you specify the value within the Physics Values node for each region.	✓
Maximum Reference Velocity $U_{r, m a x}$ from Eqn. (954).		✓
Minimum Reference Velocity $U_{r, \min}$ from Eqn. (954).		✓
Positivity Rate Limit For SF, the maximum allowable pressure correction update, expressed as a fraction of the difference between the current pressure and the minimum pressure. Applies only to compressible flows. The default value is 0.2, meaning that any negative pressure correction is limited to 20% of current pressure minus minimum pressure. For CF, the maximum rate by which temperature is allowed to decrease.	✓	✓
Preconditioning Enabled This property includes a preconditioning matrix in the governing equations for the Coupled Flow model, in both original and discretized form. Using this matrix improves convergence and accuracy. This improvement applies to solving both compressible and incompressible flows at all speeds. The default is On.		✓
Pressure Difference Scale Factor $ϵ$ from Eqn. (954).		✓
Secondary Gradients There are two sources of secondary gradients in Simcenter STAR-CCM+ flow solvers: boundary secondary gradients for diffusion interior secondary gradients for diffusion at cell faces Use this property to control which gradients are included in the solver. On gives both gradients while Off excludes them. Interior Only and Boundaries Only select the corresponding gradients.	✓	✓
Unsteady Flux Dissipation Corrections When activated, this property includes mass flux dissipation corrections for unsteady simulations, modifying the coefficients ${\bar{a}}_{0}$ and ${\bar{a}}_{1}$ in Eqn. (923), based on $CFL$ . This property is deactivated by default.	✓
Unsteady Low-Mach Preconditioning * When On, activates additional preconditioning to improve the accuracy in unsteady low-speed compressible regimes. Only active when Preconditioning Enabled is active. The default is On.		✓
Unsteady Preconditioning Max Factor * The maximum factor for unsteady preconditioning, a value from 0 through 1. The default is 0.95. Larger values can reduce robustness. At zero, the solver uses no high Strouhal number unsteady numerical dissipation.		✓

* The Unsteady Low-Mach Preconditioning and Unsteady Preconditioning Max Factor properties control treatment of mass-flux pressure dissipation, $D_{p}$ (described in [31]). When you activate unsteady preconditioning, $D_{p}$ is a function of the ratio of a local time-step to the user-defined time-step. This local time-step is obtained without preconditioning, assuming CFL = 1.

At high speeds, no unsteady modification is done for $D_{p}$ . At low speeds (approaching and in the incompressible regime), $D_{p}$ depends on local flow conditions and also on the local Strouhal number.

At high Strouhal numbers (small time-steps), $D_{p}$ corresponds to the unpreconditioned mass-flux pressure dissipation. At low Strouhal numbers and towards steady-state (when time-step tends to infinity), $D_{p}$ moves towards the steady-state low-Mach preconditioned mass-flux pressure dissipation. This move happens because the dissipation depends on the ratio of the local time-step to the user-defined time-step, which decreases as the time-step increases.

Model Controls

Bounded Differencing

Sub-node that becomes available when you select any of these Discretization (CF)/Convection (SF) schemes:

Bounded-Central (flow solver enabled for LES)
Hybrid-BCD (flow solver enabled for DES)
MUSCL 3rd-order/CD

Upwind Blending Factor: Specifies the proportion of upwind differencing, $ς_{u b f}$ related to Eqn. (891).; For Bounded-Central and Hybrid-BCD, the default value of this property is 0.15. For MUSCL 3rd-order/CD, the default value is 1.0.; The default provides the most robustness for the scheme. Reducing it would, in principle, increase accuracy. However, unless you are thoroughly familiar with the theoretical aspects of bounded differencing, do not change this property. The default value reflects optimization for accuracy and performance.

Reference Values

Maximum Allowable Absolute Pressure: The maximum value allowed for the Absolute Pressure in the continuum, see $p$ in Eqn. (947).
Minimum Allowable Absolute Pressure: The minimum value allowed for the Absolute Pressure in the continuum, see $p$ in Eqn. (947).
Reference Pressure: The pressures referred to by the flow models in Simcenter STAR-CCM+ are implicitly gauge pressures. This is important for minimizing roundoff errors, particularly for incompressible problems where pressure changes might be small. If the Constant Density model is being used, the operating pressure has no bearing on the calculation. However, when the Ideal Gas model is chosen, the absolute pressure is used to obtain the density so that the operating pressure must be specified appropriately.; For more information, see Setting Reference Values.

Initial Conditions

Pressure: Sets the initial working pressure in Pa. The outlet pressure is reasonable to use for the initial value of pressure. If there are no pressure boundaries that are attached to the given continuum, the initial value is arbitrary.
Velocity: Sets the initial fluid velocity in the chosen Coordinate System. A non-zero value is required for inviscid simulations and for initial turbulence intensity.