Showing papers on "Freestream published in 1985"

Experimental study of disturbances produced in a pre-transitional laminar boundary layer by weak freestream turbulence

Abstract: The high-amplitude, low-frequency, disturbances generated in a Blasius boundary layer by grid-produced freestream turbulence were studied by means of hot-wire anemometry and by flow visualization. Emphasis was placed on determining the cause and nature of these, and on their effect upon the mean flow profile. It was found that the motion in the layer was intrinsically three-dimensional, according to which the fronts, of Tollmien-Schlichting waves became warped. Evidence of streamwise vorticity was noted. At Reynolds numbers near 1.6 million, the maximum attained, sharp velocity gradients appeared locally. With increase of grid size, there appeared turbulent spots which were wider than the coherence scale of the background motion.

Unsteadiness of the shock wave structure in attached and separated compression ramp flows

Abstract: Wall pressure fluctuations have been measured upstream of the corner-line in several two dimensional, adiabatic, compression ramp flows. The nominal freestream Mach number was 3 and the Reynolds number, based on boundary layer thickness, was 1.4 million. The measurements show that the shockwave structure is unsteady in both separated and attached flows, resulting in a region in which the wall pressure signal is intermittent. Statistical properties of this intermittent region, and of the separated flow, are presented and correlated with results from other studies.

An Experimental Investigation of an Airfoil Undergoing Large-Amplitude Pitching Motions

Abstract: FLOW visualization and near-surface hot-wire experiments were performed in the U.S.A.F Academy Aeronautics Laboratory subsonic wind tunnel on an airfoil undergoing large-amplitude pitching motions about its quarter chord. The experiments were conducted using a NACA 0015 airfoil at an airfoil Reynolds number of 45,000. Two cases are presented in which the angular pitching rate a is maintained constant during the motion. These two cases represent two different nondimensional pitching rates a+, where ot+ is equal to 6; nondimensionalized by the chord c and the freestream velocity U^ (a + ^ac/U^). Data for the two cases where values of a+ are equal to 0.2 and 0.6 show the dramatic effect of pitch rate on flow structure. Largescale vortical structures are seen in both cases at high angles of attack but appear much later and are of a different form for the case with the larger a+ value. These structures are very energetic, producing reverse flow velocities near the airfoil surface of 1.0-2.1 times the freestream velocity.

Flowfield scaling in sharp fin-induced shock wave/turbulent boundary-layer interaction

Abstract: This paper presents the results of an experimental investigation of the three-dimensional interaction of a swept planar shock wave with a turbulent boundary layer. The shock wave was generated by a sharp, unswept fin mounted normal to a flat test surface. On the two test surfaces used, the incoming boundary layers varied in thickness by 3:1. In both cases, the freestream Mach number was nominally 3, the freestream Reynolds number 6.3 X 10 m~ *, and the wall temperature close to adiabatic. Detailed yaw angle and pitot pressure surveys in the two cases reveal a similar flowfield structure that can be correlated using a simple scaling technique.

Response of a Supersonic Inlet to Downstream Perturbations

Abstract: Experimental results are reported for flows in a ramp-type, external compression inlet with a large-aspectratio, rectangular cross section. The inlet was operated at a freestream Mach number of 1.84 with mechanically generated downstream perturbations. High-speed schlieren and time-dependent pressure measurements were employed extensively. In supercritical operation, pressure fluctuations throughout the inlet caused by the excitation varied linearly with the fluctuations at the exit station, even for large exit station amplitudes. In subcritical operation (buzz), the excitation interacted nonlinearly with the naturally present, highly periodic oscillations by either modifying the natural frequency, if the excitation was near a natural harmonic, or by having the excitation modulate the naturally occurring oscillation. In addition, the conditions at the two criticality boundaries were determined as a function of excitation amplitude and frequency.

Asymmetrical wake generated by a spinning cylinder

Abstract: An extended set of measurements has been carried out on a developing turbulent wake behind a spinning circular cylinder immersed in a uniform stream. Various moments of the velocity signals have revealed the effect of the rotation upon characteristic parameters of the flowfield. For small values of rotation, where the circumferential velocity is less than the freestream velocity, a quasiconventional turbulent wake behavior obtains. At high values of rotation, the magnitudes of all dynamic turbulent fluctuations decrease substantially due to the suppression of the Kantian vortex street. As well, it was observed that a region of negative production was generated in the developed flow for those cases where the circumferential velocities were comparable to the freestream velocity. This aspect is explored in considerable detail.

Experimental study of flows in a two-dimensional inlet model

Abstract: Experimental results are reported for flows in a ramp-type, external-compression inlet with a rectangular, large-aspect-ratio cross section, at a freestream Mach number of 1.84. Variation of the exit throttle area created a wide range of operating conditions, from highly supercritical to fully subsonic internal flows with a detached shock ahead of the ramp lip. Both time-mean and dynamic aspects of the flows were investigated. At high terminal-shock Mach numbers (1.5-2.2), three different, massively separated, subsonic flow patterns were found, which depended on throttle setting and the history of prior flow conditions. A specific pressure ratio could be associated with at least two such patterns. At low pressure ratios, the terminal shock was weak and strongly influenced by the ramp/cowl configuration. The presence of leading edges is believed to have been closely involved in the large-amplitude, periodic oscillations (buzz) observed at the low end of the pressure-ratio range.

Modification of vortex interactions in a reattaching separated flow

Abstract: Recent experimental observations have shown that large-scale organized vortices are produced in reattaching separated flows. Interactions between these vortices are important in the development of these flows downstream. Experimental studies from a downstream-fac ing step flow are presented to demonstrate that substantial changes in a reattaching flow can be produced by controlled forcing techniques. The forcing apparently works by affecting the vortex merging process in a fashion similar to that observed in forced mixinglayer experiments. The separated mean flow spreading rate could be increased most effectively by forcing at a nondimensional frequency (based on step height and freestream velocity) between 0.2 and 0.4. This result was found to be relatively independent of step Reynolds numbers over the range (26,000-76,000) studied. A significant decrease in the reattachment length accompanied the increased growth of the separated shear layer. Considerable changes in the turbulence energy and the Reynolds stress levels were also observed for the forced flows.

Space–time results for a separating turbulent boundary layer using a rapidly scanning laser anemometer

Abstract: A rapidly scanning one-velocity-component directionally sensitive fringe-type laser-Doppler anemometer which scans the measurement volume perpendicular to the optical axis of the transmitting optics was used to investigate the flow structure of the steady freestream separated turbulent boundary layer of Simpson, Chew & Shivaprasad (1981a). Space–time correlations were obtained for the first time in a separated turbulent boundary layer and showed that the integral lengthscale Ly for the large eddies grows in size towards detachment, although the ratio of this lengthscale to the boundary-layer thickness remains constant. Results also indicate local dependence of the backflow on the middle and outer regions of the boundary layer at a given instant in time.

Heat transfer and fluid mechanics measurements in transitional boundary layers on convex-curved surfaces.

Abstract: The test section of the present experiment to ascertain the effects of convex curvature and freestream turbulence on boundary layer momentum and heat transfer during natural transition provided a two-dimensional boundary layer flow on a uniformly heated curved surface, with bending to various curvature radii, R. Attention is given to results for the cases of R = infinity, 180 cm, and 90 cm, each with two freestream turbulence intensity levels. While the mild convex curvature of R = 180 cm delays transition, further bending to R = 90 cm leads to no signifucant further delay of transition. Cases with both curvature and higher freestream disturbance effects exhibit the latter's pronounced dominance. These data are pertinent to the development of transition prediction models for gas turbine blade design.

Conical Similarity of Shock/Boundar y-Layer Interactions Generated by Swept and Unswept Fins

Abstract: A parametric experimental investigation has been made of the class of three-dimensi onal shock wave/turbulent boundary-layer interactions generated by swept and unswept leading-edge fins. The fin sweepback angles were 0-65 deg at 5, 9, and 15 deg angles of attack. Two equilibrium two-dimensional turbulent boundary layers with a freestream Mach number of 2.95 and a Reynolds number of 6.3 x 107/m were used as incoming flow conditions. AH of the resulting interactions were found to possess conical symmetry of the surface flow patterns and pressures outside of an initial inception zone. Further, these interactions were found to obey a simple conical similarity rule based on inviscid shock wave strength irrespective of fin sweepback or angle of attack. This is one of the first demonstrations of similarity among three-dimensional interactions produced by geometrically dissimilar shock generators. Nomenclature a = exponent in Reynolds number scaling law h = fin height, cm LI = length along inviscid shock wave trace on test surface for the inception of conical flow, cm Ls = length along inviscid shock wave trace from fin leading edge, cm LuN = upstream influence length normal to inviscid shock wave trace on test surface, cm MO, = freestream Mach number MN =M00sinj80, component of freestream Mach number normal to inviscid shock wave trace on test surface p = surface static pressure, N/m2 Pec = static pressure of the incoming freestream flow, N/m2 Red = Reynolds number based on boundary-layer thickness

Ram constriction vane diffuser for jet engine

Abstract: A ram constriction vane diffuser adapted as an air intake of an air plenum-engine pod wherein vertical multiple vanes are fitted on the vertically inclined air intake opening and include fixed vanes positioned adjacent the centerline portion of the air plenums which define an envelope for the airflow including nozzles locate on both sides of a jet engine, a plurality of deflectable vanes positioned in an equally spaced relationship in the air intake openings on both sides of fixed vanes, and wherein each deflectable vane has a leading edge and a drivable trailing section pivotally mounted on the rigid leading section of the deflectable vane and with the drivable trailing sections of each vane and with an actuator for adjusting the angle of the deflectable vanes relative to the freestream air flow for enabling the drivable trailing section of each vane to be turned away from a parallel position to the centerline of an engine suction diffuser, which is the original position thereof for speed flight, to reduce the area of the air intake openings to produce oblique ram-airstream adjacent the engine suction diffuser intake and with each ram-airstream flowing in a tangentially constricted path at critical pressure which forms a convergent-divergent freestream throat and a ram pressure stream in the air plenum on each side of the engine pod resulting in the velocity of the airstream decreasing with an increase in the static pressure thereof and a gain in the air plenums producing a ram constriction pressure airflow to air-inducing nozzles is shown.

Freestream turbulence effects on wake properties of a flat plate at an incidence

Abstract: The effect of freestream turbulence on the characteristics of a turbulent wake developed from the trailing edge of a flat plate at angles of attack of 3, 0, and 6 deg is discussed. Correlations based on analytical and experimental considerations are developed to predict the mean characteristics of a turbulent wake at an incidence in the freestream turbulence environment. The analytical approach is based on the Navier-Stokes meanmomentum and continuity equation. The experimentation was carried out in a subsonic wind tunnel using crosswire anemometry. Experimental data of the flat plate at three incidences and three turbulence levels (0.40, 5.23, and 7.23%) from self-similarity considerations are presented to predict the semiempirical correlations for the wake velocity defect, length scale, wake velocity profile, shape factor, and eddy viscosity. The effect of pressure gradient on various parameters is also discussed. It is concluded that the wake velocity defect and shape factor decrease while the length scale and eddy viscosity increase when there is an increase in freestream turbulence.

Transverse Jet Break-up and Atomization with Rapid Vaporization along the Trajectory.

Abstract: Experiments for transverse injection of chilled Freon-12 into the Virginia Tech 23 x 23 cm blowdown wind tunnel were run at a freestream Mach number of 0.44 and freestream stagnation pressure and temperature of 2.5 atm and 298 K, respectively. The spray plume was documented with photographs and droplet measurements. The results showed a clear picture of the mechanisms of jet decomposition in the presence of rapid vaporization. Immediately after injection, a vapor cloud was formed in the jet plume, which then dissipated downsteam leaving droplets on the order of 8-10 microns in diameter. This represented a substantial reduction compared to baseline tests run at the same conditions with water, which had little vaporization. A simulation approach to studying hot-flow subsonic cross-stream fuel-injection problems in a less complex and costly cold-flow facility is proposed. The simulation parameters were developed and refined with the aid of a numerical solution for the simpler case of a rapidly evaporating laminar jet in a coaxial airstream. The experimental case was transformed (through two new similarity parameters involving injection and freestream properties) to a simulated case of a typical ramjet-combustion-chamber fuel-injection problem where ambient-temperature fuel (kerosene) is injected into a hot airstream. 18 references.

On the dynamics scaling of forced unsteady separated flows

Abstract: The active control of separated flows in turbulent boundary layers was achieved experimentally using the periodic oscillations of a spoiler-like flap inside the separated zone. Results were obtained for three heights of the separation generator, using several freestream velocities and numerous forcing frequencies. The dynamical scaling analysis of reattachment control as a function of reduced frequency revealed the existence of at least two distinct mechanisms controlling flow reattachment. At low reduced frequencies (k < 0.04) , the primary mechanism corresponds to the momentum exchange induced by the modulation of the separated shear layer, and leads to the periodic shedding of the Separation bubble. This mechanism scales with the characteristic height of the separation region. For larger values of the reduced frequency parameter, the dominating mechanism is the formation and shedding of energetic vortices caused by the oscillation of the flap. This mechanism scales with the reduced frequency based an the flap height.

Experimental aerodynamic coefficients on a Shuttle-like vehicle at Mach 6 and 10 and comparison to prediction

Abstract: Aerodynamic coefficients measured for an Orbiter-like configuration are presented for a range of angle of attack from 15 deg to 45 deg at Mach 6 and 10. The low-gamma aspect of a real gas on aerodynamic characteristics and shock detachment distance were simulated by testing this configuration in Mach 6 air (freestream gamma equal to 1.4) and Mach 6 CF4 (freestream gamma equal to 1.17). The effects of Mach number, Reynolds number, and gamma on these measurements are examined and comparisons made to an inviscid flowfield computer code known as HALIS. Pitching-moment coefficients measured in CF4 revealed a nose-up increment when compared with measurements in perfect air, indicative of real-gas effects. In general, the HALIS code accurately predicted the measured aerodynamic coefficients in air and CF4.

Aerodynamic Characteristics of Noncircular Bodies in Flat Spin and Coning Motions

Abstract: Experimental results for bodies of square cross-section coning at angles up to 90 deg show that the measured side force characteristics are extremely nonlinear, exhibiting both discontinuities and hysteresis effects. The present paper analyzes these results to determine to what extent the nonsteady wall boundary condition has influenced the flow separation and associated lateral characteristics of the coning body. It is shown that the socalled moving wall effects have a dominant influence on the flow separation and can explain the unusual side force characteristics measured in the experiments. Nomenclature c = cross-sectional chord (diameter for a circular cross section) h = cross-sectional maximum height L1 = sectional lift; coefficient, cl=LI/(pw U2^ /2)c M = freestream Mach number TV = coning rate N1 = sectional normal force coefficient, cn=N1/(PooUl/2)c p = roll rate P = static pressure; coefficient, Cp = (p—p(X)/(p(X>t/i/2) r = corner radius Re = Reynolds number, ^U^c/v^ U = velocity

A consistent spatial differencing scheme for the transonic full-potential equation in three dimensions

Abstract: A full-potential steady transonic wing flow solver has been modified so that freestream density and residual are captured in regions of constant velocity. This numerically precise freestream consistency is obtained by slightly altering the differencing scheme without affecting the implicit solution algorithm. The changes chiefly affect the fifteen metrics per grid point, which are computed once and stored. With this new method, the outer boundary condition is captured accurately, and the smoothness of the solution is especially improved near regions of grid discontinuity.

Shuttle vertical fin flowfield by the direct simulation Monte Carlo method

Abstract: The flow properties in a model flowfield, simulating the shuttle vertical fin, determined using the Direct Simulation Monte Carlo method. The case analyzed corresponds to an orbit height of 225 km with the freestream velocity vector orthogonal to the fin surface. Contour plots of the flowfield distributions of density, temperature, velocity and flow angle are presented. The results also include mean molecular collision frequency (which reaches 1/60 sec near the surface), collision frequency density (approaches 7 x 10 to the 18/cu m sec at the surface) and the mean free path (19 m at the surface).

An experimental investigation of an airfoil undergoing large amplitude pitching motions

Abstract: FLOW visualization and near-surface hot-wire experiments were performed in the U.S.A.F Academy Aeronautics Laboratory subsonic wind tunnel on an airfoil undergoing large-amplitude pitching motions about its quarter chord. The experiments were conducted using a NACA 0015 airfoil at an airfoil Reynolds number of 45,000. Two cases are presented in which the angular pitching rate a is maintained constant during the motion. These two cases represent two different nondimensional pitching rates a+, where ot+ is equal to 6; nondimensionalized by the chord c and the freestream velocity U^ (a + ^ac/U^). Data for the two cases where values of a+ are equal to 0.2 and 0.6 show the dramatic effect of pitch rate on flow structure. Largescale vortical structures are seen in both cases at high angles of attack but appear much later and are of a different form for the case with the larger a+ value. These structures are very energetic, producing reverse flow velocities near the airfoil surface of 1.0-2.1 times the freestream velocity.

Ignition of a fuel spray by a hot surface

Abstract: An experimental study of the ignition of Jet-A fuel sprays by an isothermal hot surface was conducted in a vertical axisymmetric duct. The ranges of flow conditions under which ignition was investigated were: 1) freestream velocity, 1-5 m/s; 2) boundary-laye r momentum thickness, 3-20 mm; 3) freestream air temperature, 40-250°C; 4) fuel concentration, ignitability limits; and 5) droplet size (SMD), 20-200 /im. In addition to measurements of the wall temperature necessary for ignition under the above conditions, local measurements of velocity, "turbulence" intensity, fuel concentration, and the fraction of fuel vaporized were measured in the boundary layer at surface temperatures just below that required for ignition. The results exhibited vapor ignition trends for most of the flow conditions, with some exceptions where single-droplet ignition appeared to be present. The experimental data are compared with existing vapor ignition theory.

Numerical simulation of viscous supersonic flow over a generic fighter configuration

Abstract: A procedure is presented, as well as some results, to calculate the flow over a generic fighter configuration. A parabolized marching Navier-Stokes code is used to obtain the solution over a wing-canopy body. The flow conditions simulate supersonic cruise with a freestream Mach number of 2.169 and angles of attack of 4 and 10 deg. The body surface is considered to be an adiabatic wall and the flow is assumed to be turbulent for the given Reynolds number.

Knudsen layer characteristics for a highly cooled blunt body in hypersonic rarefied flow

Abstract: R are presented summarizing the first detailed Knudsen-layer characteristics at \00/RN = 0.15 (freestream mean free path normalized by the spherical nose radius, RN) for a cold wall (TW/T^ =2) at Mach 26 using a Monte Carlo (MC) code developed for a virtual memory minicomputer. Flow profiles are in reasonable agreement with previous Monte Carlo results outside the Knudsen layer, but differ appreciably from contiriuum predictions with boundary conditions simulating rarefild flow effects of shock slip and body slip. MC results for slip velocity and temperature jump at the wall are considerably lower than predicted by slip models used in the continuum codes. Breakdown of the continuum description for heat transfer and shear stress in the Knudsen layer is documented at the rarefied flow conditions corresponding to a Reynolds number Re^ =65, based on RN.

Effects of non-uniform velocity profiles on dual jets in a crossflow

Abstract: The interaction between engine exhaust jets and the freestream affects the aerodynamic and stability characteristics of VTOL aircraft during the transition from hover to forward flight. This interaction is often modeled as a simple uniform jet issuing from a flat plate into a subsonic crossflow. The distribution of pressures induced by the jet on the surface of the plate can be used to predict the lift loss and pitching moment for a full-scale aircraft. The uniform jet model has limitations because an actual turbofan engine generates a rather nonuniform exit velocity profile. The purpose of this work is to study the effect of a nonuniform velocity profile on the surface pressure distribution. The mutual interaction of dual jets also is investigated in side-by-side and tandem configurations. Detailed pressure distributions are presented for two jet-to-freestream velocity ratios of 2.2 and 4.0. One important finding is that a nonuniform jet with a high velocity periphery and a low velocity core has a higher effective velocity ratio than a uniform jet with the same mass flow.

Heat transfer and pressure drop in a tube bank inclined with respect to the flow

Abstract: This research is intended to lend understanding and to quantify the heat-transfer and fluid-flow characteristics for yawed tube banks in both staggered and in-line arrays. The investigated range of yaw angle was from 90 (crossflow) to 45/sup 0/, while the freestream Reynolds number (based on the tube diameter) ranged between 7000 and 45,000. The transverse and longitudinal center-to-center distances between the tubes were S/sub T//D = S/sub L//D = 2, respectively. The heat-transfer experiments were carried out on a row-by-row basis. Pressure drop measurements were made not only upstream and downstream of the tube bank but also within it. The patterns of fluid flow adjacent to the tubes were visualized using the oil-lampblack technique. A detailed study was carried out to determine the heat-transfer characteristics of a yawed single cylinder. The yaw angle range was between 90 and 30/sup 0/, and flow visualization was also performed. The pressure measurements showed that the overall dimensionless pressure drop for the staggered array is higher than that for the in-line array for a given Reynolds number or yaw. The flow-visualization patterns showed that the boundary layer separation depends on the yaw angle. For the single cylinder, the Nusselt number varied with the yawmore » angle in an undulating manner and did not correlate with the Independence Principle.« less

The Effects of Freestream Turbulence on the Profile Boundary Layer and Losses in a Compressor Cascade

Abstract: The turbulent profile boundary layer on a one-foot chord compressor cascade blade has been measured with varying levels of freestream turbulence. Increased levels of freestream turbulence were found to increase the fullness of the velocity profiles, with a consequent decrease in displacement thickness and an increase in the skin friction coefficient. A small increase in freestream turbulence causes the cascade total-pressure loss to increase initially, while at the higher turbulence levels boundary layer separation was delayed, resulting in a decrease in the total-pressure loss and deviation angle.Copyright © 1984 by ASME