SEARCH · Search NASA
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Influence of suction on shock wave turbulent boundary layer interactions for two dimensional and axially symmetric flows
An upstream-influence scaling law, previously formulated through analysis of Mach 3 data, has been extended to Mach numbers from 2.5 through 4. For adiabatic, equilibrium, turbulent boundary layers, there is no Mach number effect on the constants in the Reynolds number parameters of this law. In addition, based on local similarity, a new Mach number parameter, namely, the Mach number component of the incoming stream normal to the farfield upstream influence, is proposed. Scaling by either the incoming Mach number normal to the inviscid shock or by the incoming Mach number normal to the farfield upstream influence is equivalent to scaling by the hypersonic similarity parameter.
Interaction between a shock wave and an unseparated turbulent boundary layer is considered. The method of matched asymptotic expansions is used, with solutions valid in the double limit as Reynolds number tends to infinity and Mach number tends to unity. The shock is weak enough that interaction effects can be considered as perturbations to the undisturbed flow; the case considered is that where the sonic line is near the outer edge of the boundary layer. It is shown that, with order estimates for Reynolds stress perturbations, the induced wall pressure distribution can be calculated using only the two outer interaction regions, independent of a specific closure condition and that this solution is in fact a turbulent free interaction solution. A detailed analysis of the inner regions, for which an eddy viscosity model for the Reynolds shear stress is used, provides a description of the variations in velocity, temperature and density near and at the wall.
Tabulated data from a series of experimental studies of the interaction of a shock wave with a turbulent boundary layer in axisymmetric flow configurations is presented. The studies were conducted at the walls of circular wind tunnels and on the cylindrical centerbody of an annular flow channel. Detailed pitot pressure profiles and wall static pressure profiles upstream of, within and downstream of the interaction region are given. Results are presented for flows at nominal freestream Mach Numbers of 2, 3 and 4. For studies at the tunnel sidewalls, the shock waves were produced by conical shock generators mounted on the centerline of the wind tunnel at zero angle of attack. The annular ring generator was used to produce the shock wave at the centerbody of the annular flow channel. The effects of boundary layer bleed were examined in the investigation. Both bleed rate and bleed location were studied. Most of the bleed studies were conducted with bleed holes drilled normal to the wall surface but the effects of slot suction were also examined. A summary of the principal results and conclusions is given.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
The accuracy of numerical computations of the three-dimensional sharp fin flow field using the Reynolds-averaged Navier-Stokes equations is examined. Computed flows are compared with a set of benchmark experiments at Mach 3 for two different Reynolds numbers. A detailed comparison is performed between experimental data and separate computational results for the three-dimensional sharp fin for a 10 deg fin angle in order to examine the accuracy of two different turbulence models. The computed surface pressure for the fin at a 20 deg angle is compared with recent experimental data. The computed yaw angle profiles are shown to be sensitive to the turbulence model. In the immediate vicinity of the surface, the calculated results using the Baldwin-Lomax (1978) model are in better agreement with the data than the Jones-Launder (1972) model. Within the inner portion of the boundary layer, both models tend to underpredict the yaw angle.
Explore the source record for details and available documents.
A numerical investigation of transonic turbulent flows separated by curvature and shock wave - boundary layer interaction is presented. The free stream Mach numbers considered are 0.4, 0.5, 0.6, 0.7, 0.8, 0.825, 0.85, 0.875, 0.90, and 0.925. In the numerical method, the conservation of mass equation is replaced by a pressure correction equation for compressible flows and thus incremental pressure is solved for instead of density. The turbulence is described by a multiple-time-scale turbulence model supplemented with a near-wall turbulence model. The present numerical results show that there exists a reversed flow region at all free stream Mach numbers considered whereas various k-epsilon turbulence models fail to predict such a reversed flow region at low free stream Mach numbers. The numerical results also show that the size of the reversed flow region grows extensively due to the shock wave - turbulent boundary layer interaction as the free stream Mach number is increased. These numerical results show that the turbulence model can resolve the turbulence field subjected to extra strains caused by the curvature and the shock wave - turbulent boundary layer interaction and that the numerical method yields a significantly accurate solution for the complex compressible turbulent flow.
An experiment is described that tests and guides computations of a shock-wave turbulent boundary-layer interaction flow over a 20-deg compression corner at Mach 2.85. Numerical solutions of the time-averaged Navier-Stokes equations for the entire flow field, employing various turbulence models, are compared with the data. Each model is critically evaluated by comparisons with the details of the experimental data. Experimental results for the extent of upstream pressure influence and separation location are compared with numerical predictions for a wide range of Reynolds numbers and shock-wave strengths.
Interaction regions created by the impingement of full span, externally generated, shock waves on a nozzle wall boundary layer were investigated. Incident shock strength was varied to produce unseparated, incipient, and fully separated flow fields. Significant departures from two-dimensionality were observed over the entire range of shock strengths tested and were identified with sidewall and corner boundary layer effects. However, comparisons of present centerline results with published two-dimensional data, obtained under similar test conditions and geometrical constraints, showed excellent agreement (e.g., incipient separation pressure levels, wall pressure distributions, free-interaction, and scale of the interaction region). This raises some question concerning the degree of two-dimensionality achieved in these previous investigations.
A finite difference solution to the problem of the interaction between an impinging oblique shock wave and the laminar boundary layer on a flat plate is presented. The boundary layer equations coupled with the Prandtl-Meyer relation for the external flow are used to calculate the flow field. A method for the calculation of the separated flow region is presented and discussed. Comparisons between this theory and the experimental results of other investigators show fairly good agreement. Results are presented for the case of a cooled wall with an oncoming flow at Mach number 2.0 without and with suction. The results show that a small amount of suction greatly reduces the extent of the separated region in the vicinity of the shock impingement location.
A combined experimental and numerical study has been conducted on the hypersonic shock-wave turbulent-boundary layer interaction at Mach 8.2 generated by a single fin of angles alpha = 10 and 15 deg. Three models are considered: (1) the 3D compressible Reynolds-averaged Navier-Stokes (RANS) equations using the k-epsilon turbulence model, (2) the 3D RANS using the Rodi turbulence model, and (3) the conical RANS using the Baldwin-Lomax algebraic turbulence model. The computations are compared with various experimental data. The computations using models (1) and (2) show quantitatively very similar results and very good agreement with experimental data for surface pressure and skin friction. Comparison with boundary layer profiles of pitot pressure and yaw angle are also generally good, but the peak surface heat transfer is overestimated by up to 48 percent. The effect of the laminar boundary layer on the fin is restricted to the immediate vicinity of the fin surface. Conical calculations using model (3) show substantially poorer agreement with experiment.
This paper describes a combined experimental and theoretical study of three-dimensional swept shock wave-turbulent boundary layer interactions at Mach 4 generated by a sharp fin of angles alpha equals 16 and 20 degrees. The theoretical model is the three-dimensional compressible Reynolds-averaged Navier-Stokes equations with turbulence incorporated through the algebraic eddy viscosity model of Baldwin and Lomax. Previous computations have been performed by Horstman using the Baldwin-Lomax, Cebeci-Smith and Jones Launder models. Computed results for the surface pressure, skin friction and streamline angles are compared with experiment and previous numerical results. The present results display good agreement with experimental data for surface pressure and surface flow direction. All turbulence models fail to accurately predict the peak skin friction. The computed flowfields are in agreement with many of the features of the quasi-conical flowfield model of Settles.
A review is given of computations for a series of nominally 2-D laminar viscous-inviscid interactions. Comparisons were made with detailed experimental shock tunnel results. The shock wave boundary layer interactions considered were induced by a compression ramp in one case and by an externally generated incident shock in the second case. In general, good agreement was reached between the grid refined calculations and experiment for the incipient and small separation conditions. For the highly separated flow, 3-D calculations which included the finite span effects of the experiment were required in order to obtain agreement with the data.
Method for predicting interaction produced by externally generated, oblique shock wave impinging on laminar or turbulent boundary layers
A thin film heat transfer gauge is applied to the measurement of heat transfer coefficients in the interaction regions of incident shock waves and fully developed turbulent boundary layers. It was developed to measure heat flux with high spatial resolution and fast response for wind tunnels with long flow duration. To measure the heat transfer coefficients in the interaction region in detail, experiments were performed under the conditions of Mach number = 4, total pressure = 1.2 MPa, 0.59 to approximately 0.65. Reynolds number = 1.3 to approximately 1.5 x 10 to the 7th power and incident shock angles from 17.8 to 22.8 degrees. The results show that the heat transfer coefficient changes complicatedly in the interaction region. At the beginning the interaction region, the heat transfer coefficient decreases at first, reaches its minimum value at the point where the pressure begins to increase, and then increases sharply. When the boundary layer begins to separate, even a small separation bubble causes significant changes in the heat transfer coefficient, while the pressure does not show any changes which suggests that the boundary layer begins to separate.