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.
Undesirable vortexes near floor in small wind tunnels suppressed by simple device that alters flow pattern there. Air is injected along floor and interacts with backflow from wind-tunnel model. Results in smoother, more correct air-flow and to more-reliable wind-tunnel data.
An initial-phase subsonic diffuser has been designed for the turbojet flowpath of the hypersonic x43B flight demonstrator vehicle. The diffuser fit into a proposed mixed-compression supersonic inlet system and featured a cross-sectional shape transitioning flowpath (high aspect ratio rectangular throat-to-circular engine face) and a centerline offset. This subsonic diffuser has been fabricated and tested at the W1B internal flow facility at NASA Glenn Research Center. At an operating throat Mach number of 0.79, baseline Pitot pressure recovery was found to be just under 0.9, and DH distortion intensity was about 0.4 percent. The diffuser internal flow stagnated, but did not separate on the offset surface of this initial-phase subsonic diffuser. Small improvements in recovery (+0.4 percent) and DH distortion (-32 percent) were obtained from using vane vortex generator flow control applied just downstream of the diffuser throat. The optimum vortex generator array patterns produced inflow boundary layer divergence (local downwash) on the offset surface centerline of the diffuser, and an inflow boundary layer convergence (local upwash) on the centerline of the opposite surface.
A new numerical method based on adding a term to the Euler/Navier-Stokes equations has been demonstrated it can effectively treat vortex-dominated flows using low-order numerical schemes and coarse grids. The numerical diffusion introduced by convection schemes can be eliminated by modifying the velocity field as a result of solving the modified flow governing equations. The modification of the velocity field, which conserves the total vorticity, essentially convects the vorticity toward its local extreme to offset the numerical diffusion. This method preserves the vortex structure even when vortices travel on coarse grid region. The method is implemented with two distinctive codes: a Navier-Stokes flow solver based on vorticity-velocity formulation and an Euler code based on velocity-pressure formulation. Problems presented in this paper include airfoil dynamic stall, vortex airfoil interaction, and vortex-fuselage interactions.
The results of an experimental study of spiraling flows in a slightly diverging, circular duct are reported. Seven types of flow disturbances were observed. In addition to the spiral and axisymmetric vortex breakdowns and the double helix mode, four other forms were identified and are reported. The type and axial location of the disturbance depended on the Reynolds and circulation numbers of the flow. Detailed velocity measurements were made by using a laser Doppler anemometer. Measurements made far upstream of any disturbance showed that the introduction of swirl resulted in the formation of a high axial velocity jet centered around the vortex center. A mapping of the velocity field of a so-called axisymmetric breakdown, formed at a Reynolds number of 2560, revealed that the recirculation zone is a two-celled structure, with four stagnation points on the vortex axis marking the axial extremes of the concentric cells. The dominant feature of the flow inside the bubble was the strong, periodic velocity fluctuations. Existing theoretical models do not predict the two-celled structure and the temporal velocity fluctuations that were observed.
The primary project was the numerical simulation, by a finite element/finite difference method, of the viscous flow about an airfoil. The secondary project involved the numerical simulation of the three-dimensional separated and vortex-dominated flow about a hemispherically capped cylinder in the transonic regime. Preliminary calculations were started for the hemisphere-cylinder at 0 and 5 degree angle of attack. The solution of the flow field about airfoils and wings is required to determine the important parameters of lift, moment, and drag. Viscous effects must be accounted for if the drag is to be accurately calculated. At present there are basically two approaches to the numerical simulation of the flow field, the use of fully viscous models and the inviscid/viscous models. The fully viscous models require the solution of an approximation of the Navier-Stokes equations and therefore should simulate most of the physical mechanisms. A fast, accurate, and computationally efficient inviscid flow solver was recently developed by Hartwich. It is thought that Hartwich's program coupled to a fast, accurate, and computationally efficient boundary layer code, will make an excellent tool for airfoil design. The purpose of the primary project was to develop a compressible boundary layer code using the semidiscrete Galerkin finite element method. The numerical scheme employed used the combination of a Dorodnitsyn formulation of the boundary layer equations, with a finite difference/finite element procedure (semidiscrete Galerkin method), in the solution of the compressible two-dimensional boundary layer equations. A laminar compressible boundary layer code was developed and tested for a NACA 0012 airfoil at a Mach number of 0.5, a Reynolds number of 5000, and zero angle of attack. At present the boundary layer program solves up to, but not beyond, separation.
An experimental study has been conducted of the influence of wall to total temperature ratio on the heat transfer to the leeside of a 040A space shuttle configuration. The heat transfer tests were made at a Mach number of 10 and a Reynolds number of one million per foot for angles of attack from 0 deg to 30 deg. Range of wall to total temperature ratio was from 0.16 to 0.43. Where the heat transfer was relatively high and the laminar boundary layer attached, the local heat transfer decreased by about 20 percent as the wall to total temperature ratio was increased from the minimum to the maximum test value. On regions of separated flow and vortex reattachment, very low heating rates were measured at some conditions and indicate significant changes are occurring in the leeside flow field. No single trend of heat transfer variation with wall to total temperature ratio could be observed.
A computer program aimed at the phase separation between gas and liquid at zero gravity, induced by vortex motion, is developed. It utilizes an explicit solution method for a set of equations describing rotating gas-liquid flows. The vortex motion is established by a tangential fluid injection. A Lax-Wendroff two-step (McCormack's) numerical scheme is used. The program can be used to study the fluid dynamical behavior of the rotational two-phase fluids in a cylindrical tank. It provides a quick/easy sensitivity test on various parameters and thus provides the guidance for the design and use of actual physical systems for handling two-phase fluids.
Activities carried out in support of research on flow structure generated by perpendicular blade vortex interaction and implications for helicopter noise prediction are summarized. Progress in the following areas is described: (1) construction of 8 inch-chord NACA 0012 full-span blade; (2) Acquisition of two full-span blades; (3) preparation for hot wire measurements; (4) related work on a modified Betz's theory; and (5) work related to helicopter noise prediction. In addition, a list of publications based on the results of prior experimentation is presented.
Rotating Ekman boundary layer stability and transition for two-dimensional roll vortices superimposed upon basic boundary layer flow
The flow-field characteristics beneath swept and unswept wings as determined by potential-flow theory are compared with the experimentally determined flow fields beneath swept and unswept wing-fuselage combinations. The potential-flow theory utilized considered both spanwise and chordwise distributions of vorticity as well as the wing-thickness effects. The perturbation velocities induced by a unit horseshoe vortex are included in tabular form. The results indicated that significant chordwise flow gradients existed beneath both swept and unswept wings at zero lift and throughout the lift range. The theoretical predictions of the flow-field characteristics were qualitatively correct in all cases considered, although there were indications that the magnitudes of the downwash angles tended to be overpredicted as the tip of the swept wing was approached and that the sidewash angles ahead of the unswept wing were underpredicted. The calculated effects of compressibility indicated that significant increases in the chordwise variation of flow angles and dynamic-pressure ratios should be expected in going from low to high subsonic speeds.
The flow-field characteristics beneath swept and unswept wings as determined by potential-flow theory are compared with the experimentally determined flow fields beneath swept and unswept wing-fuselage combinations. The potential-flow theory utilized considered both spanwise and chordwise distributions of vorticity as well as the wing-thickness effects. The perturbation velocities induced by a unit horseshoe vortex are included in tabular form. The theoretical predictions of the flow-field characteristics were qualitatively correct in all cases considered, although there were indications that the magnitudes of the downwash angles tended to be overpredicted as the tip of the swept wing was approached and that the sidewash angles ahead of the unswept wing were underpredicted. The calculated effects of compressibility indicated that significant increases in the chordwise variation of flow angles and dynamic-pressure ratios should be expected in going from low to high subsonic speeds.
The effects of flight on sound radiated from embedded, uncorrelated ring sources convecting along the midst of the primary and the secondary streams of a coaxial dual flow which emerges from a moving nozzle into the ambience are studied. Cold jets are examined. The problem is posed as a double vortex-sheet flow model which involves deliberate suppression of inherent instabilities of the flow and is formulated, as a linear problem, in terms of the combined contributions of two independent uncorrelated quadrupole-type ring sources, the one convecting in the primary flow representing the sources generated due to the interaction at the primary/secondary interface and the other convecting in the secondary flow representing the sources generated due to the interaction at the secondary/ambient interface. The analysis shows that the effects of flight induce (1) amplication of noise in the forward quadrant, (2) reduction of noise in the aft quadrant and (3) absolutely no impact on radiation of noise at Theta = 90 deg to the jet axis.
The effects of flight on sound radiated from embedded, uncorrelated ring sources convecting along the midst of the primary and the secondary streams of a coaxial dual flow which emerges from a moving nozzle into the ambience are studied. Cold jets are examined. The problem is posed as a double vortex-sheet flow model which involves deliberate suppression of inherent instabilities of the flow and is formulated, as a linear problem, in terms of the combined contributions of two independent uncorrelated quadrupole-type ring sources, the one convecting in the primary flow representing the sources generated due to the interaction at the primary/secondary interface and the other convecting in the secondary flow representing the sources generated due to the interaction at the secondary/ambient interface. The analysis shows that the effects of flight induce: (1) amplication of noise in the forward quadrant, (2) reduction of noise in the aft quadrant and (3) absolutely no impact on radiation of noise at Theta = 90 deg to the jet axis.
Nonrotating strain gauge type vorticity meter for measuring flow field behind delta wings
Results from direct numerical simulations are presented to show that the weakly nonlinear results of Daudpota et al. (1988) are in error with respect to the influence of the Tollmien-Schlichting wave on the Dean vortex. The results of a new weakly nonlinear theory are then presented, and it is shown that the new results are consistent with the direct numerical simulations.
The qualitative responses of a trailing vortex core to change in its convective velocity produced by injection of air into the vortex core and by the flow field of a lifting surface in the path of the core are discussed. Flow pictures of the vortex core and vorticity measurements in the core show that an axial momentum injection of approximately 35% of the total wing drag alters the vortex structure and behavior quite drastically by effectively inducing a more rapid dispersion of the vorticity in the core. The data indicate that the phenomenon is governed by the rate of injection of the momentum rather than the mass flow rate. There also seems to be an optimum rate of injection beyond which increased injection does not bring about equally significant changes in the vortex core. Flow patterns of the region of interaction between the flow field of a lifting surface and the vortex core show two modes of vortex response; (1) the vortex core bends, following the streamline shape until it intercepts the wake of the wing where it is abruptly dispersed, or (2) the vortex core may be sliced into two similar vortices when the vortex hits the leading edge of the wing.
An experimental study was conducted in the Virginia Tech Stability Wind Tunnel to determine surface pressures over a 60 deg sweep delta wing with three vortex flap designs. Extensive pressure data was collected to provide a base data set for comparison with computational design codes and to allow a better understanding of the flow over vortex flaps. The results indicated that vortex flaps can be designed which will contain the leading edge vortex with no spillage onto the wing upper surface. However, the tests also showed that flaps designed without accounting for flap thickness will not be optimum and the result can be oversized flaps, early flap vortex reattachment and a second separation and vortex at the wing/flap hinge line.
A large-scale low-boom inlet concept was tested in the NASA Glenn Research Center 8- x 6- foot Supersonic Wind Tunnel. The purpose of this test was to assess inlet performance, stability and operability at various Mach numbers and angles of attack. During this effort, two models were tested: a dual stream inlet designed to mimic potential aircraft flight hardware integrating a high-flow bypass stream; and a single stream inlet designed to study a configuration with a zero-degree external cowl angle and to permit surface visualization of the vortex generator flow on the internal centerbody surface. During the course of the test, the low-boom inlet concept was demonstrated to have high recovery, excellent buzz margin, and high operability. This paper will provide an overview of the setup, show a brief comparison of the dual stream and single stream inlet results, and examine the dual stream inlet characteristics.