Engineering topics

Sindir, Munir M.

Publications and source records attributed to Sindir, Munir M..

12 records

NASA NTRS1996

Comparative Study of Advanced Turbulence Models for Turbomachinery

A computational study has been undertaken to study the performance of advanced phenomenological turbulence models coded in a modular form to describe incompressible turbulent flow behavior in two dimensional/axisymmetric and three dimensional complex geometry. The models include a variety of two equation models (single and multi-scale k-epsilon models with different near wall treatments) and second moment algebraic and full Reynolds stress closure models. These models were systematically assessed to evaluate their performance in complex flows with rotation, curvature and separation. The models are coded as self contained modules that can be interfaced with a number of flow solvers. These modules are stand alone satellite programs that come with their own formulation, finite-volume discretization scheme, solver and boundary condition implementation. They will take as input (from any generic Navier-Stokes solver) the velocity field, grid (structured H-type grid) and computational domain specification (boundary conditions), and will deliver, depending on the model used, turbulent viscosity, or the components of the Reynolds stress tensor. There are separate 2D/axisymmetric and/or 3D decks for each module considered. The modules are tested using Rocketdyn's proprietary code REACT. The code utilizes an efficient solution procedure to solve Navier-Stokes equations in a non-orthogonal body-fitted coordinate system. The differential equations are discretized over a finite-volume grid using a non-staggered variable arrangement and an efficient solution procedure based on the SIMPLE algorithm for the velocity-pressure coupling is used. The modules developed have been interfaced and tested using finite-volume, pressure-correction CFD solvers which are widely used in the CFD community. Other solvers can also be used to test these modules since they are independently structured with their own discretization scheme and solver methodology. Many of these modules have been independently tested by Professor C.P. Chen and his group at the University of Alabama at Huntsville (UAH) by interfacing them with own flow solver (MAST).

Hadid, Ali H.↗

NASA NTRS1996

Procedure And Criteria For Assessing CFD Codes

Systematic approach to validation helps ensure good designs. Four-phase procedure devised to standardize and systematize assessment of performances of computational fluid dynamics (CFD) software used in designing aircraft, turbines, and other machines interacting with flows. Associated with procedure set of criteria by which one classifies given computer code and measures its ability to predict flows and flow-related quantities accurately.

Lin, Shyi-Jang↗

NASA NTRS1995

Overview of turbulence model development and applications at Rocketdyne

This viewgraph presentation discusses turbulence modeling requirements, development philosophy, and approach; two major areas of concentration (high speed and low speed turbulence modeling); high speed turbulence modeling; compressibility effects; turbulence models adapted to USA code; M = 9.2 flat plate flow; Mach 7.05 flow over axisymmetric flare; Mach 8.6 flow over cold wall edge; low speed turbulence modeling; turbulence models being assessed; turbulence model deck structure and integration with Navier-Stokes solver; nonlinear algebraic-stress model; rotation modified k-epsilon model; and Reynolds stress model.

Hadid, A. H.↗

NASA NTRS1995

Computational System For Rapid CFD Analysis In Engineering

Computational system comprising modular hardware and software sub-systems developed to accelerate and facilitate use of techniques of computational fluid dynamics (CFD) in engineering environment. Addresses integration of all aspects of CFD analysis process, including definition of hardware surfaces, generation of computational grids, CFD flow solution, and postprocessing. Incorporates interfaces for integration of all hardware and software tools needed to perform complete CFD analysis. Includes tools for efficient definition of flow geometry, generation of computational grids, computation of flows on grids, and postprocessing of flow data. System accepts geometric input from any of three basic sources: computer-aided design (CAD), computer-aided engineering (CAE), or definition by user.

Barson, Steven L.↗

NASA NTRS1993

Advanced turbulence models for turbomachinery

Development and assessment of the single-time-scale k-epsilon turbulence model with different near-wall treatments and the multi-scale turbulence model for rotating flows are presented. These turbulence models are coded as self contained module decks that can be interfaced with a number of CFD main flow solvers. For each model, a stand-alone module deck with its own formulation, discretization scheme, solver and boundary condition implementations is presented. These satellite decks will take as input (from a main flow solver) the velocity field, grid, boundary condition specifications and will deliver turbulent quantities as output. These modules were tested as separate entities and, although many logical and programming problems were overcome, only wider use and further testing can render the modules sufficiently 'fool proof'.

Hadid, Ali H.↗

NASA NTRS1992

A numerical study of two-dimensional vortex shedding from rectangular cylinders

An efficient time-marching, non-iterative calculation method is used to analyze time-dependent flows around rectangular cylinders. The turbulent flow in the wake region of a square section cylinder is analyzed using an anisotropic k-epsilon model. Initiation and subsequent development of the vortex shedding phenomenon is naturally captured once a perturbation is introduced in the flow. Transient calculations using standard eddy-viscosity and an anisotropic k-epsilon model averaged over an integral number of cycles to get the fluctuating energy (organized and turbulent) are compared with experimental data. It is shown that the anisotropic k-epsilon model resolves the anisotropy of the Reynolds stresses and gives mean energy distribution closer to the experiment than the standard k-epsilon model.

Hadid, A. H.↗

NASA NTRS1992

A computational design system for rapid CFD analysis

A computation design system (CDS) is described in which these tools are integrated in a modular fashion. This CDS ties together four key areas of computational analysis: description of geometry; grid generation; computational codes; and postprocessing. Integration of improved computational fluid dynamics (CFD) analysis tools through integration with the CDS has made a significant positive impact in the use of CFD for engineering design problems. Complex geometries are now analyzed on a frequent basis and with greater ease.

Ascoli, E. P.↗

NASA NTRS1992

A multi-domain method for subsonic viscous flows

We have developed a Schwarz type domain decomposition method for a pressure base, two- and three-dimensional Navier-Stokes solver. This technique allows one to partition a flow path, which can be characterized by complex geometry and/or complicated flow physics, into smaller sub-domains according to the local geometric simplicity or estimated flow scales. We can, then, sweep the sub-domains in some order and solve the Navier-Stokes equations using as boundary conditions, along the domain interfaces, the Dirichlet conditions which are taken from the most recent update of the solution in the adjacent neighboring domains. With this technique, one can minimize the adverse effects caused by grid skewness and the stiffness problem caused by disparate flow scales. Here, we report the results of a few fundamental flow cases to demonstrate that a judicious use of the multi-domain method can offer a significant convergence acceleration over the traditional one-domain method. This method can be extended to exploit the architecture of a parallel computer to further improve the speed.

Chan, Daniel C.↗

NASA NTRS1992

Some comments on turbulence modeling from an industrial perspective

In propulsion industry, computational techniques (namely computational fluid dynamics) are gradually becoming engineering design and analysis tools, especially in the case of high performance aircraft engines, high speed air breathing propulsion, rocket propulsion, and advanced nuclear propulsion and electric propulsion. The key issues in acceptance of CFD are turn around time, cost, and accuracy. The turn around time of analysis has to fit program schedules. Cost is still the bottom line, the value of CFD to the program should at least balance its cost. CFD can be used if it can be divided into two major parts: numerics and physical models.

Sindir, Munir M.↗

NASA NTRS1988

Positioning Rotors In Turbine Flowmeters

Lengths of wakes roughly proportional to thickness of vanes. Mathematical model simplifies analysis of effects of flow-straightening vanes in turbine flowmeter. Yields numerical solution of differential equations of flow for quick examination of efforts of thicknesses of vanes and rate of flow on extent of wake behind vanes. From examination, minimum distance at which flowmeter rotor placed behind vanes determined.

Lynch, Edward D.↗

NASA NTRS1987

Calculating Leakage Around Turbopump Inducer Shrouds

New mathematical model for leakage flow around shrouded turbopump inducers yields more realistic analyses from which designers determine best geometry for leakage-flow-reinjection ports. Also, designers use calculated velocity profile at inducer leading edge to determine blade-angle distribution of inducer leading edge.

Meng, Sen Yih↗

NASA NTRS1987

Impeller shroud to casing leakage flow simulations in the Space Shuttle Main Engine high pressure fuel pump

Quasi-three-dimensional Navier-Stokes calculations were carried out for the Space Shuttle Main Engine high-pressure fuel pump to simulate the impeller shroud to casing leakage flow. This flow geometry was modeled as an axisymmetric cavity flow with a stationary surface representing the casing, and a rotating surface denoting the impeller. A 63 x 81-node mesh provided sufficient resolution in the regions of greatest flow variations and reduced the effects of numerical diffusion. The turbulence field was closed with the high Reynolds number form of the k-epsilon model supplemented with wall functions in the vicinity of the walls. Finally, a parametric study quantified the effects of through mass flow changes on this leakage flow.

Sindir, Munir M.↗