Follow us on

HiFUN: High Resolution Flow Solver on Unstructured meshes

The code HiFUN the primary product of SandI, is a general purpose flow solver employing unstructured data based algorithms. HiFUN is fine tuned to solve typical aerospace applications and certain flow problems encountered in automotive industries. The code has been extensively used for solving a number of problems, over a wide range of Mach numbers, ranging from airship aerodynamics to aerodynamics of hypersonic vehicles.

The solver allows the use of four basic mesh elements, namely, hex, tet, prism and pyramid, in combination. The adaptive capability of the solver, allowing for hanging nodes, results in arbitrary polyhedral volumes.

Higher order spatial accuracy is achieved by using a linear reconstruction procedure based on either method of least squares or Green-Gauss procedure. Use of Venkatakrishnan limiter ensures monotonicity of the solution for high speed flows.

The solver supports a number of numerical flux formulas with a provision for the user to include his favorite flux formula.

A positive viscous discretization ensures robustness of the flow solver even on grids with highly skewed cells. This is indeed an unique feature of this solver.

The flow solver is equipped with an efficient algorithm to handle non conformal grids at an interface between two grid blocks.

It is possible to simulate translating and rotating walls in the flow solver.

The flow solver is equipped with an efficient algorithm to handle multiple rotating frames of reference in a given problem.

The flow solver is capable of simulating flows through porous media.

The solver supports Spalart-Allmaras and k- Omega turbulence models. The model equations are solved in a decoupled manner. A robust discretization used for the model equations ensure high levels of convergence even for the turbulence quantities.

For turbulent computations, standard equilibrium wall function gets automatically activated if grid resolution near the wall is not adequate to resolve viscous sub–layer.

A matrix-free implicit procedure ensures rapid convergence, both for steady and unsteady computations (in dual time mode).

A unique four layered approach to data handling on each of the processors ensures there is no degeneration in the performance of the parallel code as compared to a serial code, while at the same time achieving a linear speed up even for large number of processors. The MPI is used for message passing across the processors.

Formal second order time accuracy is achieved both on stationary and moving grids using a dual time stepping procedure. On the moving grids the solver is GCL compliant.

A hybrid adaptive strategy employing sensors based on both residual error estimator (referred to as R-Parameter) and error indicator is made use of. The cell division is isotropic.

This unique feature of HiFUN is typically suited for serial applications, where by the problem size a given machine can handle is substantially enhanced.

For more details, please contact:,