This manual describes the operation and use of the WIND code, a
computational platform which may be used to numerically solve various
sets of equations governing physical phenomena.
WIND represents a merger of the capabilities of three older CFD codes -
NASTD (the primary flow solver at McDonnell Douglas, now part of Boeing),
NPARC (the original NPARC Alliance flow solver), and NXAIR (an AEDC code
used primarily for store separation problems).
[WIND is a product of the
NPARC Alliance
,
a partnership between the
NASA Glenn Research Center (GRC)
and the
USAF Arnold Engineering
Development Center (AEDC)
dedicated to the establishment of a national,
applications-oriented flow simulation capability.
The Boeing Company
has also been closely associated with the Alliance since its inception, and
represents the interests of the NPARC User's Association.]
Currently, the code supports the solution of the Euler and Navier-Stokes
equations of fluid mechanics, along with supporting equation sets
governing turbulent and chemically reacting flows.
WIND uses multi-zone computational grids, and is capable of computing solutions on a wide variety of grids. Because WIND is written to accommodate arbitrary grid topologies and boundary condition combinations, it may be used to obtain solutions about most of the geometric configurations for which a grid can be generated.
The multi-zone approach to flow solutions makes it possible to decompose virtually any configuration into a number of manageable subregions, or zones. Zonal connectivity information is computed using a pre-processing Grid MANipulation (GMAN) code, and stored in the grid file used by WIND. During the course of a solution, WIND maintains continuity in flow properties across zone boundaries through a process known as zone coupling. [Romer, W. W., and Bush, R. H. (1993) "Boundary Condition Procedures for CFD Analyses of Propulsion Systems - The Multi-Zone Problem," AIAA Paper 93-1971.]
All terms are retained in the governing equations, including secondary flow, reversed flow convection, pressure gradients normal to a wall, streamwise diffusion, and unsteady flow. All heat transfer terms are retained. Several algebraic, one-equation, and two-equation turbulence models are available. Transition may be specified through the use of an external file. Modification of the effective heat transport coefficient due to turbulence is linked to the momentum diffusion coefficient by a turbulent Prandtl number, which is taken to be constant.
The fluid may be treated as a thermally and calorically perfect gas, a thermally perfect gas, equilibrium air, or a mixture undergoing a finite rate chemical reaction. For ideal gas, conventional values are given to the gas constant (R) and the ratio of specific heats (gamma), or they may be specified. Effects of gravity on the fluid (i.e., stratification) are not included.
WIND uses externally generated computational grids. Therefore, all geometric input and capability depend on the grid generator. WIND has no geometric input. All analyses must be preceded by a grid generation run.
The solution is executed iteratively on the computational mesh. The flow equations are evaluated using second-order-accurate finite differences. The partial differential equations are modeled in their conservative form. Explicit terms are computed using either upwind or central differencing, and their order may be controlled through the use of keywords in the input data file. The implicit terms are computed using either an approximately factored or four-stage Runge-Kutta scheme, or they may be disabled altogether. A Global Newton iteration scheme is also available, and may be used for unsteady flows with large time scales or as a convergence acceleration technique for steady flows.
WIND is coded in the Fortran 77, Fortran 90, and C programming languages. The production version of the code is supported on a variety of systems from Silicon Graphics, Hewlett-Packard, Sun, and Cray.