Search NASA⌕ Search

Engineering topics

Benson, T. J.

Publications and source records attributed to Benson, T. J..

An Integration of the Turbojet and Single-Throat Ramjet

A turbine-engine-based hybrid propulsion system is described. Turbojet engines are integrated with a single-throat ramjet so as to minimize variable geometry and eliminate redundant propulsion components. The result is a simple, lightweight system that is operable from takeoff to high Mach numbers. Non-afterburning turbojets are mounted within the ramjet duct. They exhaust through a converging-diverging (C-D) nozzle into a common ramjet burner section. At low speed the ejector effect of the C-D nozzle aerodynamically isolates the relatively high pressure turbojet exhaust stream from the ramjet duct. As the Mach number increases, and the turbojet pressure ratio diminishes, the system is biased naturally toward ramjet operation. The common ramjet burner is fueled with hydrogen and thermally choked, thus avoiding the weight and complexity of a variable geometry, split-flow exhaust system. The mixed-compression supersonic inlet and subsonic diffuser are also common to both the turbojet and ramjet cycles. As the compressor face total temperature limit is approached, a two-position flap within the inlet is actuated, which closes off the turbojet inlet and provides increased internal contraction for ramjet operation. Similar actuation of the turbojet C-D nozzle flap completes the enclosure of the turbojet. Performance of the hybrid system is compared herein to that of the discrete turbojet and ramjet engines from takeoff to Mach 6. The specific impulse of the hybrid system falls below that of the non-integrated turbojet and ramjet because of ejector and Rayleigh losses. Unlike the discrete turbojet or ramjet however, the hybrid system produces thrust over the entire Mach number range. An alternate mode of operation for takeoff and low speed is also described. In this mode the C-D nozzle flap is deflected to a third position, which closes off the ramjet duct and eliminates the ejector total pressure loss.

Trefny, C. J.↗

Calculation of a circular jet in crossflow with a multiple-time-scale turbulence model

Numerical calculation of a three dimensional turbulent flow of a jet in a crossflow using a multiple time scale turbulence model is presented. The turbulence in the forward region of the jet is in a stronger inequilibrium state than that in the wake region of the jet, while the turbulence level in the wake region is higher than that in the front region. The calculated flow and the concentration fields are in very good agreement with the measured data, and it indicated that the turbulent transport of mass, concentration, and momentum is strongly governed by the inequilibrium turbulence. The capability of the multiple time scale turbulence model to resolve the inequilibrium turbulence field is also discussed.

Kim, S.-W.↗

AGARD WG13 aerodynamics of high speed air intakes: Assessment of CFD results

A brief review of the work accomplished by the numerical subgroup of AGARD Working Group 13 on the aerodynamics of high speed air intakes is presented. This work comprised the selection of test cases for which experimental data were available. The test cases were chosen to range in complexity from normal-shock/boundary-layer interaction to full forebody-inlet combinations. Computations for these test cases were solicited from a large number of organizations and individual researchers within the NATO countries. The computation methods reached from Euler solvers (with and without boundary layer corrections) to full Reynolds averaged Navier-Stokes codes. The group compared these results with the test data available for each test case. A short overview of the CFD methods employed, a description of the test cases selected, and some of the comparisons between CFD solutions and test data are presented. The conclusions and recommendations drawn from this assessment are given.

Bissinger, N. C.↗

Comparison of the SMAC, PISO, and iterative time-advancing schemes for unsteady flows

Calculations of unsteady flows using a simplified marker and cell (SMAC), a pressure implicit splitting of operators (PISO), and an iterative time advancing scheme (ITA) are presented. A partial differential equation for incremental pressure is used in each time advancing scheme. Example flows considered are a polar cavity flow starting from rest and self-sustained oscillating flows over a circular and a square cylinder. For a large time step size, the SMAC and ITA are more strongly convergent and yield more accurate results than PISO. The SMAC is the most efficient computationally. For a small time step size, the three time advancing schemes yield equally accurate Strouhal numbers. The capability of each time advancing scheme to accurately resolve unsteady flows is attributed to the use of new pressure correction algorithm that can strongly enforce the conservation of mass. The numerical results show that the low frequency of the vortex shedding is caused by the growth time of each vortex shed into the wake region.

Kim, S.-W.↗

Fluid-flow of a row of jets in crossflow - A numerical study

A detailed computer-visualized flow field of a row of jets in a confined crossflow is presented. The Reynolds averaged Navier-Stokes equations are solved using a finite volume method that incorporates a partial differential equation for incremental pressure to obtain a divergence-free flow field. The turbulence is described by a multiple-time-scale turbulence model. The computational domain includes the upstream region of the circular jet so that the interaction between the jet and the crossflow is simulated accurately. It is shown that the row of jets in the crossflow is characterized by a highly complex flow field that includes a horse-shoe vortex and two helical vortices whose secondary velocity components are co-rotating in space. It is also shown that the horse-shoe vortex is a ring of reversed flows located along the circumference of the jet exit.

Kim, S.-W.↗

Calculation of a circular jet in crossflow with a multiple-time-scale turbulence model

Numerical calculation of a three dimensional turbulent flow of a jet in a crossflow using a multiple time scale turbulence model is presented. The turbulence in the forward region of the jet is in a stronger inequilibrium state than that in the wake region of the jet, while the turbulence level in the wake region is higher than that in the front region. The calculated flow and the concentration fields are in very good agreement with the measured data, and it indicated that the turbulent transport of mass, concentration, and momentum is strongly governed by the inequilibrium turbulence. The capability of the multiple time scale turbulence model to resolve the inequilibrium turbulence field is also discussed.

Kim, S.-W.↗

Comparison of 3-D viscous flow computations of Mach 5 inlet with experimental data

A time marching 3-D full Navier-Stokes code, called PARC3D, is validated for an experimental Mach 5 inlet configuration using the data obtained in the 10 x 10 ft supersonic wind tunnel at the NASA Lewis Research Center. For the first time, a solution is obtained for this configuration with the actual geometry, the tunnel conditions, and all the bleed zones modeled in the computation. Pitot pressure profiles and static pressures at various locations in the inlet are compared with the corresponding experimental data. The effect of bleed zones, located in different places on the inlet walls, in eliminating the low energy vortical flow generated from the 3-D shock-boundary layer interaction is simulated very well even though some approximations are used in applying the bleed boundary conditions and in the turbulence model. A further detailed study of the effect of individual bleed ports is needed to understand fully the actual mechanism of efficiently eliminating the vortical flow from the inlet. A better turbulence model would help to improve the accuracy even further in predicting the corner flow boundary layer profiles.

Reddy, D. R.↗

Comparison of 3-D viscous flow computations of Mach 5 inlet with experimental data

The accuracy of the time-marching three-dimensional Navier-Stokes code PARC3D (Cooper et al., 1987) is evaluated using experimental data on a Mach 5 mixed-compression inlet configuration, obtained in the 10 x 10-ft working section of the supersonic wind tunnel at NASA Lewis. The basic principles of the PARC3D computations (in which the inlet and tunnel geometry and the bleed conditions are fully modeled) and the experimental setup are described, and the results are presented in extensive graphs and characterized in detail. PARC3D is shown to give reasonably accurate predictions of the bleed-zone elimination of low-energy vortical flow due to shock/boundary-layer interaction; various possible improvements are briefly considered.

Reddy, D. R.↗

Three dimensional viscous analysis of a hypersonic inlet

The flow fields in supersonic/hypersonic inlets are currently being studied at NASA Lewis Research Center using 2- and 3-d full Navier-Stokes and parabolized Navier-Stokes solvers. These tools have been used to analyze the flow through the McDonnell Douglas Option 2 inlet which has been tested at Calspan in support of the National Aerospace Plane Program. Comparisons between the computational and experimental results are presented. These comparisons lead to better overall understanding of the complex flows present in this class of inlets. The aspects of the flow field emphasized in this work are the 3-D effects, the transition from laminar to turbulent flow, and the strong nonuniformities generated within the inlet.

Reddy, D. R.↗

Real gas effects on the numerical simulation of a hypersonic inlet

The present use of a parabolized Navier-Stokes solver to accurately simulate the flowfield in a supersonic inlet yields good agreement between numerical analysis and experiment for a Mach 7.4 inlet under cruise conditions, with an internal compression ratio of 8. The significance of real gas effects on the performance calculation of a hypersonic inlet is demonstrated, with small changes in the ratio of specific heats resulting in a substantial change in the calculated pitot pressure ratio.

Ng, W.-F.↗

Validation of a three-dimensional viscous analysis of axisymmetric supersonic inlet flow fields

A three-dimensional viscous marching analysis for supersonic inlets was developed. To verify this analysis several benchmark axisymmetric test configurations were studied and are compared to experimental data. Detailed two-dimensional results for shock-boundary layer interactions are presented for flows with and without boundary layer bleed. Three dimensional calculations of a cone at angle of attack and a full inlet at attack are also discussed and evaluated. Results of the calculations demonstrate the code's ability to predict complex flow fields and establish guidelines for future calculations using similar codes.

Benson, T. J.↗

Numerical solution to the glancing sidewall oblique shock wave/turbulent boundary layer interaction in three dimension

A supersonic three-dimensional viscous forward-marching computer design code called PEPSIS is used to obtain a numerical solution of the three-dimensional problem of the interaction of a glancing sidewall oblique shock wave and a turbulent boundary layer. Very good results are obtained for a test case that was run to investigate the use of the wall-function boundary-condition approximation for a highly complex three-dimensional shock-boundary layer interaction. Two additional test cases (coarse mesh and medium mesh) are run to examine the question of near-wall resolution when no-slip boundary conditions are applied. A comparison with experimental data shows that the PEPSIS code gives excellent results in general and is practical for three-dimensional supersonic inlet calculations.

Anderson, B. H.↗

Numerical solution to the glancing sidewall oblique shock wave/turbulent boundary layer interaction in three-dimension

A supersonic three-dimensional viscous forward-marching computer design code called PEPSIS is used to obtain a numerical solution of the three-dimensional problem of the interaction of a glancing sidewall oblique shock wave and a turbulent boundary layer. Very good results are obtained for a test case that was run to investigate the use of the wall-function boundary-condition approximation for a highly complex three-dimensional shock-boundary layer interaction. Two additional test cases (coarse mesh and medium mesh) are run to examine the question of near-wall resolution when no-slip boundary conditions are applied. A comparison with experimental data shows that the PEPSIS code gives excellent results in general and is practical for three-dimensional supersonic inlet calculations.

Anderson, B. H.↗