SEARCH · Engineering Papers
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.
Four phases of research results are reported: (1) experiments on the compressible turbulent boundary layer flow in a streamwise corner; (2) the two dimensional (2D) interaction of incident shock waves with a compressible turbulent boundary layer; (3) three dimensional (3D) shock/boundary layer interactions; and (4) cooperative experiments at Princeton and numerical computations at NASA-Ames.
The capability of two k-epsilon low-Reynolds number (LRN) turbulence models, those of Jones and Launder (1972) and Lam and Bremhorst (1981), to predict transition in external boundary-layer flows subject to free-stream turbulence is analyzed. Both models correctly predict the basic qualitative aspects of boundary-layer transition with free stream turbulence, but for calculations started at low values of certain defined Reynolds numbers, the transition is generally predicted at unrealistically early locations. Also, the methods predict transition lengths significantly shorter than those found experimentally. An approach to overcoming these deficiencies without abandoning the basic LRN k-epsilon framework is developed. This approach limits the production term in the turbulent kinetic energy equation and is based on a simple stability criterion. It is correlated to the free-stream turbulence value. The modification is shown to improve the qualitative and quantitative characteristics of the transition predictions.
The applicability of and the limits on the applicability of 11 near wall similarity laws characterizing three-dimensional turbulent boundary layer flows were determined. A direct force sensing local wall shear stress meter was used in both pressure-driven and shear-driven three-dimensional turbulent boundary layers, together with extensive mean velocity field and wall pressure field data. This resulted in a relatively large number of graphical comparisons of the predictive ability of 10 of these 11 similarity models relative to measured data over a wide range of flow conditions. Documentation of a complex, separated three-dimensional turbulent flow as a standard test case for evaluating the predictive ability of numerical codes solving such flows is presented.
Prediction formulations for transonic fluctuating pressure environments including protuberance induced turbulent boundary layer flows
Flow-visualization studies, using smoke, were made of the secondary flows in rectangular bends, tandem cascades, and high-turning configurations. The roll-up of the wall boundary layer of a rectangular bend forms a passage vortex near the suction surface similar to that previously observed for cascades. The vortex so formed then shifts out into the main stream. Because of leading-edge effects, the boundary-layer flows in bends were found to be sufficiently different from the flows in blade rows to make direct application of bend results to blade rows inadvisable. Passage vortices are shown, in the tandem-cascade study, to resist turning with the main stream through which they pass and to disturb the flow in subsequent blade rows. This disturbance may explain in part the appreciable size of the losses sometimes attributed to secondary flows in turbomachines despite the fact that the energy involvement in vortex formation is slight. Tip-flow studies of high-turning blades with relative motion between blades and end wall indicated that if the relative sizes of the passage vortex forces, the tip clearance forces, and the blade-scraping effects are properly controlled, it may be possible to improve the blade-tip loading characteristics in turbomachine.
Description of a numerical method that dispenses with the recourse to linearization or 'momentum-integral-correction' schemes in solving nonlinear problems of boundary layer flow in the interaction region of a shock tube or of similar singular parabolic problems of shock-induced unsteady boundary layers. The accuracy of the numerical results obtainable is shown to be satisfactory.
Recent experimental studies suggest that the hairpin vortex plays an important (and perhaps dominant) role in the dynamics of turbulent flows near walls. In this study a numerical procedure is developed to allow the accurate computation of the trajectory of a 3-D vortex having a small core radius. For hairpin vortices which are convected in a shear flow above a wall, the calculated results show that a 2-D vortex containing a small 3-D disturbance distorts into a complex shape with subsidiary hairpin vortices forming outboard of the original hairpin vortex. As the vortex moves above the wall, it induces unsteady motion in the viscous flow near the wall: numerical solutions suggest that the boundary-layer flow near the wall will ultimately erupt in response to the motion of the hairpin vortex and in the process a secondary hairpin vortex will be created. The computer results agree with recent experimental investigations.
Recent experimental studies suggest that the hairpin vortex plays an important (and perhaps dominant) role in the dynamics of turbulent flows near walls. In this study, a numerical procedure is developed to allow the accurate computation of the trajectory of a 3D vortex having a small core radius. For hairpin vortices which are convected in a shear flow above a wall, the calculated results show that a 2D vortex containing a small 3D disturbance distorts into a complex shape with subsidiary hairpin vortices forming outboard of the original hairpin vortex. As the vortex moves above the wall, it induces unsteady motion in the viscous flow near the wall: numerical solutions suggest that the boundary-layer flow near the wall will ultimately erupt in response to the motion of the hairpin vortex and in the process a secondary hairpin vortex will be created. The computer results agree with recent experimental investigations.
Supersonic annular flow passages exist in propulsion applications that include dual combustion ramjet engines where a supersonic annular flow (the outer flow) mixes with a sonic (or supersonic) gas generator flow (the inner flow) in the shock expansion zone downstream of the gas generator nozzle exit. Other engine designs include components in the form of annular ducts whose cross-sectional area varies in the streamwise flow direction. In some of these configurations, it is necessary to support the outer shroud (cowl) by means of struts positioned between the cowl and centerbody. To investigate the distorting influence of these struts on the local flow structure, it is first necessary to ensure that the intrinsic flow without struts is free of wave reflections and the effects of upstream disturbances. It is also necessary to demonstrate that the intrinsic flow exhibits the characteristics of a well-defined turbulent boundary layer flow, so that changes in the local flow structure induced by the presence of struts can be interpreted properly. The purpose of this Note is to demonstrate that a supersonic flow facility that meets these objectives has been developed.
Multiple scaling technique is used to examine the nonparallel instability of supersonic and hypersonic boundary-layer flows to three-dimensional (first mode) and two-dimensional (second mode) disturbances. The method is applied to the flat plate boundary layer for a range of Mach numbers from 0 to 10. Growth rates of disturbances are calculated based on three different criteria: following the maximum of the mass-flow disturbance, using an integral of the disturbance kinetic energy, and using an integral of the square of the mass-flow amplitude. By following the maximum of the mass-flow dusturbance, the calculated nonparallel growth rates are in good quantitative agreement with the experimental results of Kendall (1967) at Mach number 4.5.
Multiple scaling technique is used to examine the nonparallel instability of supersonic and hypersonic boundary-layer flows to three-dimensional (first mode) and two-dimensional (second mode) disturbances. The method is applied to the flat plate boundary layer for a range of Mach numbers from 0 to 10. Growth rates of disturbances are calculated based on three different criteria: following the maximum of the mass-flow disturbance, using an integral of the disturbance kinetic energy, and using the integral of the square of the mass-flow amplitude. By following the maximum of the mass-flow disturbance, the calculated nonparallel growth rates are in good quantitative agreement with the experimental results at Mach number 4.5.
A three-dimensional, laminar boundary-layer method is applied to the incompressible flow over a slender delta wing at incidence. The predictor-corrector finite-difference scheme of Matsuno is used to difference the governing equations. The method has the advantages that no iterations are required to advance the solution and the cross-flow derivatives are formed independent of the cross-field direction. The difference scheme is demonstrated to yield accurate numerical results when compared to the exact solution of the three-dimensional boundary-layer equations for parabolic flow over a moving flat plate. The method is applied to delta wings of various sweep angles at angles of attack up to 20 deg., with the inviscid solution determined using a higher-order, three-dimensional panel method.
The present study is a preliminary investigation into the behavior of the flow within a 28 degree total geometric turning angle hypothetical Mach 10 inlet as calculated with the full three-dimensional Navier-Stokes equations. Comparison between the two-dimensional and three-dimensional solutions have been made. The overall compression is not significantly different between the two-dimensional and center plane three dimensional solutions. Approximately one-half to two-thirds of the inlet flow at the exit of the inlet behave nominally two-dimensionally. On the other hand, flow field non-uniformities in the three-dimensional solution indicate the potential significance of the sidewall boundary layer flows ingested into the inlet. The tailoring of the geometry at the inlet shoulder and on the cowl obtained in the two-dimensional parametric design study have also proved to be effective at controlling the boundary layer behavior in the three-dimensional code. The three-dimensional inlet solution remained started indicating that the two-dimensional design had a sufficient margin to allow for three-dimensional flow field effects. Although confidence is being gained in the use of SCRAM3D (three-dimensional full Navier-Stokes code) as applied to similar flow fields, the actual effects of the three-dimensional flow fields associated with sidewalls and wind tunnel installations can require verification with ground-based experiments.
The first experiment to combine airborne Doppler Lidar and ground-based dual Doppler Radar measurements of wind to detail the lower tropospheric flows in quiescent and stormy weather was conducted in central Oklahoma during four days in June-July 1981. Data from these unique remote sensing instruments, coupled with data from conventional in-situ facilities, i.e., 500-m meteorological tower, rawinsonde, and surface based sensors, were analyzed to enhance understanding of wind, waves and turbulence. The purposes of the study were to: (1) compare winds mapped by ground-based dual Doppler radars, airborne Doppler lidar, and anemometers on a tower; (2) compare measured atmospheric boundary layer flow with flows predicted by theoretical models; (3) investigate the kinematic structure of air mass boundaries that precede the development of severe storms; and (4) study the kinematic structure of thunderstorm phenomena (downdrafts, gust fronts, etc.) that produce wind shear and turbulence hazardous to aircraft operations. The report consists of three parts: Part 1, Intercomparison of Wind Data from Airborne Lidar, Ground-Based Radars and Instrumented 444 m Tower; Part 2, The Structure of the Convective Atmospheric Boundary Layer as Revealed by Lidar and Doppler Radars; and Part 3, Doppler Lidar Observations in Thunderstorm Environments.
An algebraic skin friction law is derived for adiabatic, compressible, equilibrium, turbulent boundary layer flow. An outer solution in terms of the Clauser defect stream function is matched to an inner empirical expression composed of compressible laws of the wall and wake. The modified Crocco temperature-velocity relationship and the Clauser eddy viscousity model are used in the outer solution. The skin friction law pertains for all pressure gradients in the incompressible through supersonic range and for small pressure gradients in the hypersonic range. Excellent comparisons with experiment are obtained in the appropriate parameter ranges. The application to numerical computation is discussed.
Two methods for determining the virtual origin of turbulent boundary layers in hypersonic flow are evaluated. The results of the analyses are restricted to wind-tunnel models having sharp-edged surfaces with zero or small pressure gradients. Virtual origin and skin friction estimates from these two methods are compared with values from a base method for which the virtual origin is calculated from the measured momentum thickness at a station downstream of boundary layer transition.
The inviscid and viscous stability problems are addressed for a boundary layer which can support both Goertler and crossflow vortices. The change in structure of Goertler vortices is found when the parameter representing the degree of three-dimensionality of the basic boundary layer flow under consideration is increased. It is shown that crossflow vortices emerge naturally as this parameter is increased and ultimately become the only possible vortex instability of the flow. It is shown conclusively that at sufficiently large values of the crossflow there are no unstable Goertler vortices present in a boundary layer which, in the zero crossflow case, is centrifugally unstable. The results suggest that in many practical applications Goertler vortices cannot be a cause of transition because they are destroyed by the 3-D nature of the basic state. In swept wing flows the Goertler mechanism is probably not present for typical angles of sweep of about 20 degrees. Some discussion of the receptivity problem for vortex instabilities in weakly 3-D boundary layers is given; it is shown that inviscid modes have a coupling coefficient marginally smaller than those of the fastest growing viscous modes discussed recently by Denier, Hall, and Seddougui (1990). However the fact that the growth rates of the inviscid modes are the largest in most situations means that they are probably the most likely source of transition.
It has long been recognized that flow in the melt can have a profound influence on the dynamics of a solidifying interface and hence the quality of the solid material. In particular, flow affects the heat and mass transfer, and causes spatial and temporal variations in the flow and melt composition. This results in a crystal with nonuniform physical properties. Flow can be generated by buoyancy, expansion or contraction upon phase change, and thermo-soluto capillary effects. In general, these flows can not be avoided and can have an adverse effect on the stability of the crystal structures. This motivates crystal growth experiments in a microgravity environment, where buoyancy-driven convection is significantly suppressed. However, transient accelerations (g-jitter) caused by the acceleration of the spacecraft can affect the melt, while convection generated from the effects other than buoyancy remain important. Rather than bemoan the presence of convection as a source of interfacial instability, Hurle in the 1960s suggested that flow in the melt, either forced or natural convection, might be used to stabilize the interface. Delves considered the imposition of both a parabolic velocity profile and a Blasius boundary layer flow over the interface. He concluded that fast stirring could stabilize the interface to perturbations whose wave vector is in the direction of the fluid velocity. Forth and Wheeler considered the effect of the asymptotic suction boundary layer profile. They showed that the effect of the shear flow was to generate travelling waves parallel to the flow with a speed proportional to the Reynolds number. There have been few quantitative, experimental works reporting on the coupling effect of fluid flow and morphological instabilities. Huang studied plane Couette flow over cells and dendrites. It was found that this flow could greatly enhance the planar stability and even induce the cell-planar transition. A rotating impeller was buried inside the sample cell, driven by an outside rotating magnet, in order to generate the flow. However, it appears that this was not a well-controlled flow and may also have been unsteady. In the present experimental study, we want to study how a forced parallel shear flow in a Hele-Shaw cell interacts with the directionally solidifying crystal interface. The comparison of experimental data show that the parallel shear flow in a Hele-Shaw cell has a strong stabilizing effect on the planar interface by damping the existing initial perturbations. The flow also shows a stabilizing effect on the cellular interface by slightly reducing the exponential growth rate of cells. The left-right symmetry of cells is broken by the flow with cells tilting toward the incoming flow direction. The tilting angle increases with the velocity ratio. The experimental results are explained through the parallel flow effect on lateral solute transport. The phenomenon of cells tilting against the flow is consistent with the numerical result of Dantzig and Chao.