Computational fluid dynamics (CFD) draws on a broad set of physical principles and numerical techniques. This glossary defines the foundational terms engineers encounter when building, running, and interpreting CFD simulations in the FLOW-3D family of products, covering fluid physics, numerical methods, turbulence modeling, and free-surface tracking, including the proprietary technologies that power the solver.
A
Advection
The transport of a fluid property (such as heat, concentration, or momentum) by the bulk motion of the fluid. Advection is one of the primary mechanisms by which quantities move through the computational domain and appears as a key term in the governing transport equations.
B
Boundary conditions
Mathematical constraints applied at the edges of the computational domain that define the state of the flow at those locations. Common types include inflow/outflow, pressure, wall, and symmetry conditions. The accuracy of a simulation depends heavily on selecting boundary conditions that reflect real physical conditions.
Boundary layer
The thin region of fluid immediately adjacent to a solid surface where viscous effects are significant and velocity transitions from zero at the wall to the free-stream value. Boundary layer behavior determines wall shear stress, heat transfer rates, and the onset of flow separation.
C
Cavitation
Mathematical constraints applied at the edges of the computational domain that define the state of the flow at those locations. Common types include inflow/outflow, pressure, wall, and symmetry conditions. The accuracy of a simulation depends heavily on selecting boundary conditions that reflect real physical conditions.
CFD
Computational Fluid Dynamics (CFD) is the branch of fluid mechanics that uses numerical methods and algorithms to simulate and analyze fluid flow, heat transfer, and related physical phenomena on a computer.
Compressible flow
A flow regime in which the fluid density changes significantly due to pressure or temperature variations, most commonly encountered in high-speed gas flows. When density changes are negligible, the flow is treated as incompressible.
Contact angle
The angle at which a fluid interface meets a solid surface, measured through the liquid. Contact angle determines how a fluid wets a surface and significantly influences free-surface behavior at walls, particularly in capillary-driven and low-gravity flows.
Continuity equation
A mathematical expression of the conservation of mass, stating that for an incompressible fluid the net volumetric flow rate into any control volume must equal zero. FLOW-3D enforces continuity iteratively at each time step through pressure/velocity coupling.
Convergence
The process by which an iterative numerical solution approaches a stable, self-consistent result. A solution is considered converged when residuals (the imbalance remaining in the governing equations) fall below a specified tolerance at each time step.
Courant number
A dimensionless number that relates the time-step size to the mesh cell size and local fluid velocity. The Courant-Friedrichs-Lewy (CFL) condition requires the Courant number to remain at or below a threshold value to ensure numerical stability in explicit time-stepping schemes.
D
Diffusion
The spreading of a quantity (such as heat, mass, or momentum) from regions of high concentration to regions of low concentration driven by molecular or turbulent mixing. Diffusion terms appear alongside advection terms in the transport equations governing fluid behavior.
Direct Numerical Simulation (DNS)
A CFD approach in which all scales of turbulent motion are resolved directly without a turbulence model, requiring an extremely fine mesh and small time steps. DNS is computationally prohibitive for most engineering problems and is primarily used for fundamental turbulence research.
Discretization
The process of converting continuous governing equations into a set of algebraic equations solvable on a finite mesh. FLOW-3D uses a finite difference approach applied to a structured rectangular mesh, with the FAVOR™ method used to represent complex geometry within that mesh.
Domain
The bounded region in space in which the governing equations are solved, defined by the extents of the computational mesh. Proper domain sizing ensures that boundary conditions do not artificially influence the flow in the region of interest.
F
FAVOR™
Fractional Area/Volume Obstacle Representation. A proprietary method used by FLOW-3D to represent complex geometry on a structured rectangular mesh by computing the fractional open areas and volumes of each computational cell. FAVOR™ allows accurate representation of curved and irregular surfaces without requiring a body-fitted or unstructured mesh.
Finite difference method
A numerical technique that approximates derivatives in the governing equations by computing algebraic differences between values at discrete mesh points. FLOW-3D's solver is based on a finite difference formulation applied to a structured rectangular grid.
Finite volume method
A numerical approach that discretizes the governing equations by integrating them over control volumes surrounding each mesh cell. The finite volume method inherently conserves mass, momentum, and energy across cell boundaries, making it widely used in commercial CFD solvers.
Free surface
The bounded region in space in which the governing equations are solved, defined by the extents of the computational mesh. Proper domain sizing ensures that boundary conditions do not artificially influence the flow in the region of interest.
I
Incompressible flow
A flow in which fluid density is assumed constant throughout the domain, regardless of pressure variations. Most liquid flows and low-speed gas flows are treated as incompressible. The incompressible assumption simplifies the governing equations and is the default in most FLOW-3D simulations.
K
k-epsilon turbulence model
A widely used two-equation RANS turbulence model that solves transport equations for turbulent kinetic energy (k) and its dissipation rate (epsilon) to compute an effective turbulent viscosity. FLOW-3D offers both the standard k-epsilon and the RNG k-epsilon variant, which provides improved accuracy for flows with high strain rates and recirculation.
L
Laminar flow
A smooth, ordered flow regime in which fluid moves in parallel layers with minimal mixing between them. Laminar flow occurs at low Reynolds numbers where viscous forces dominate over inertial forces. The transition to turbulence depends on geometry, velocity, and fluid properties.
Large Eddy Simulation (LES)
A turbulence modeling approach that directly resolves large-scale turbulent eddies while modeling the effects of smaller sub-grid-scale eddies. In comparison to RANS-based models, LES can provide instantaneous information about the flow and can be more accurate for flows with complex, anisotropic turbulence; but requires significantly more computational resources.
M
Mesh
The discrete spatial framework that divides the computational domain into a finite number of cells. FLOW-3D uses a structured rectangular mesh, in which cells are organized in regular rows and columns. Mesh resolution directly affects simulation accuracy, with finer cells resolving smaller flow features at greater computational cost.
Mesh convergence study
A systematic process of progressively refining the computational mesh and repeating the simulation to verify that results no longer change significantly with further refinement. A mesh-converged solution indicates that discretization errors are acceptably small and results are reliable.
N
Navier-Stokes equations
The fundamental partial differential equations governing the motion of viscous fluids, expressing the conservation of momentum. Combined with the continuity equation and, where relevant, the energy equation, the Navier-Stokes equations form the mathematical foundation of all CFD simulation.
Numerical diffusion
Artificial smearing of sharp gradients introduced by truncation errors in the discretization scheme. Numerical diffusion can reduce accuracy near interfaces and sharp flow features. FLOW-3D's TruVOF method is specifically designed to minimize numerical diffusion at free surfaces.
P
Pressure-velocity coupling
The numerical procedure used to enforce consistency between the pressure and velocity fields in an incompressible flow solver. Because pressure does not have its own transport equation in incompressible flow, iterative methods are required to ensure that the computed velocity field satisfies continuity at every time step.
R
Reynolds number
A dimensionless number expressing the ratio of inertial forces to viscous forces in a flow. Low Reynolds numbers indicate laminar flow; high values indicate turbulent flow. Reynolds number is a fundamental parameter for characterizing flow regime and for scaling experimental results to simulation conditions.
Reynolds-averaged Navier-Stokes (RANS)
An approach to turbulence modeling in which the Navier-Stokes equations are averaged, decomposing flow variables into mean and fluctuating components. RANS models, including k-epsilon and RNG k-epsilon, solve for the mean flow and model the effect of turbulence through an additional turbulent viscosity term.
S
Sharp interface
A free-surface representation in which the transition between fluid and void is confined to a single cell face rather than smeared over multiple cells. Sharp interface methods preserve physical accuracy at the fluid boundary, particularly for wave dynamics, jetting, and droplet formation.
Steady state
A flow condition in which all quantities at every point in the domain remain constant over time. Many engineering flows are analyzed at steady state to reduce computational cost. Where flow is inherently unsteady (such as in filling, sloshing, or turbulent recirculation) a transient simulation is required.
Surface tension
The cohesive intermolecular force at the interface between a liquid and a gas that acts to minimize surface area. Surface tension drives capillary rise, droplet formation, and interface stability, and is particularly significant in small-scale flows where it dominates over gravitational and inertial forces.
T
Transient simulation
A simulation that resolves how flow conditions evolve over time, rather than converging to a single steady-state solution. FLOW-3D is inherently a transient solver, tracking free-surface motion, filling, and other time-dependent phenomena from start to finish.
TruVOF
Flow Science's proprietary implementation of the volume-of-fluid method for tracking free surfaces. TruVOF maintains a sharp fluid interface confined to a single cell face, preventing the numerical diffusion that degrades accuracy in standard VOF implementations. This precision is central to FLOW-3D's ability to accurately simulate wave dynamics, jetting, splashing, and free-surface filling.
Turbulence
A chaotic, unsteady flow regime characterized by eddies, vortices, and rapid fluctuations in velocity and pressure. Turbulence occurs at high Reynolds numbers when inertial forces overwhelm viscous damping. Because resolving all turbulent scales directly is computationally prohibitive for most engineering flows, turbulence models are used to represent its effects on the mean flow.
Turbulent kinetic energy
A measure of the energy contained in turbulent velocity fluctuations, typically denoted k. It is one of the two transported quantities in the standard k-epsilon and RNG k-epsilon turbulence models, and is used to compute the turbulent viscosity that augments molecular viscosity in the momentum equations.
V
Validation
The process of demonstrating that a simulation model accurately represents real physical behavior by comparing results against experimental measurements or physical observations. Validation establishes credibility for a simulation's intended use and is distinct from verification.
Verification
The process of confirming that a simulation correctly solves the governing equations as intended, typically through comparison with analytical solutions or established benchmarks. A verified simulation may still require validation against experimental data to confirm physical accuracy.
Viscosity
A measure of a fluid's resistance to deformation under shear stress. Dynamic viscosity quantifies the absolute resistance to flow; kinematic viscosity is dynamic viscosity divided by fluid density. Viscosity governs boundary layer thickness, flow regime, and the balance between viscous and inertial forces expressed by the Reynolds number.
Volume of fluid (VOF)
A numerical method for tracking free surfaces by computing the fraction of each mesh cell occupied by a given fluid. A cell value of 1 indicates fully fluid, 0 indicates void, and intermediate values mark the interface region. FLOW-3D's TruVOF method extends the standard VOF approach to maintain a sharp, single-cell-face interface.
Vorticity
A vector quantity describing the local rotation rate of the fluid, defined as the curl of the velocity field. Vorticity is used in post-processing to identify vortex structures, recirculation zones, and turbulent flow features that may not be apparent from velocity or pressure fields alone.
W
Wall functions
Empirical relationships used to approximate the flow behavior in the near-wall boundary layer without resolving it with a fine mesh. Wall functions reduce mesh requirements near solid surfaces by bridging between the wall and the fully turbulent outer flow, and are used in conjunction with RANS turbulence models in FLOW-3D. Also known as wall models.
Wall shear stress
The tangential force per unit area exerted by the fluid on a solid surface, arising from the velocity gradient in the boundary layer. Wall shear stress is relevant for erosion prediction, drag calculations, and no-slip boundary condition modeling.