An experimental investigation of turbulent boundary layer relaminarization is underway at the Hessert Center for Aerospace Research at the University of Notre Dame. A turbulent boundary layer was exposed to a large favorable pressure gradient for several values of Launder's relaminarization parameter. It was observed that many of the turbulent boundary layer parameters behaved as if the boundary layer relaminarized, despite values of the relaminarization parameter below which relaminarization would be expected. Preliminary results lead one to question the validity of the relaminarization parameter as the proper criterion for the onset of relaminarization. Further, evidence suggests the turbulent burst-sweep event may be a direct indicator of relaminarization and preliminary experimental data supports this claim.

A low-speed wind tnnnel investigation was conducted to investigate the aerodynamics of a forebody undergoing pitsch oscillations. Two different 3.5 tangent ogive forebodies (axisymmetric and chined cross-section) were pitched sinusoidally through an angle of attack range of O-90” and O-45” at a Reynolds number of 76,300. Force balance data was acquired to resolve the aerodynamic loads for static and dynamic testing. The data indicates that during the dynamic motion the developed aerodynamic loads are a function of the forebody cross-section, the upstroke/downstroke portion of the cycle, and the amplitude of the motion. NOMENCLATURE AOA angle of attack CN Normal Force Coefficient = FN/(qS) CY Side Force Coefficient = Fy/(qS) D forebody base diameter of tangent ogive DFT discrete Fourier transform f sinusoidal pitching frequency FN Normal Force FY Side Force k reduced frequency=2n;fL/U, +Permanently assigned to AF’RLIVAAA, WPAFB, OH; Member AIAA *Professor & Chair, Dept of Aerospace and Mechanical Engineering, University of Notre Dame; Fellow, AIAA This paper is declared a work ilf the U.S. Government and is not subject to copyright protection in the-united States. 1 L forebody length 9 freestream dynamic pressure phi model roll angle Re Reynolds number = pU,D& S forebody base area = nD214 U, freestream velocity X axial distance from model apex INTRODUCTION The study of an axisymmetric body in a crossflow has long been of interest to fluid dynamicists due to the nature of the separated leeward flow. Of Q particular interest to aerodynamicists is the fact that as a pointed body of revolution is inclined to the flow, asymmetric forces develop with increasing angle of attack (AOA). The development of the asymmetric forces is usually detrimental to the stability of a missile or aircraft that possesses an axisymmetric-type forebody configuration. l-4 Current military desires are pushing the performance envelopes of contemporary aircraft/missile designs to be super-maneuverable fly faster and turn quicker and tighter, or to improve point-and-shoot capabilities. In order for current and future air vehicles to efficiently fly with increasing performance, they must also be able to fly at greater angles of attack. Flight at higher angles of attack requires an understanding of the asymmetric vertical flow patterns emanating from the vehicles’ forebody configuration. In addition, forebody vortices that interact with LEX/strake and wing flows can cause premature “breakdown” of (c)l999 American Institute of Aeronautics & Astronautics . l these flows resulting in limited overall performance of the vehicles. As a result, extensive research into the flowfields and aerodynamic loads of forebody configurations (at high angles-of-attack) has been conducted.5.6 This research has also included techniques to control the forebody vertical flowfield. Forebody Flow The flow arou-nd a slender forebody is extremely complex as illustrated in Figure 1. At low angles of attack the flow separates around the forebody on the leeward side to form two symmetric vortices. As the angle of attack is increased further the forebody vortices become asymmetric. The asymmetric forebody vortices create a side force and yawing moment on the body. The magnitude of the yawing moment created by the asymmetry can be quite large. Figure 2 is a curve showing the typical yaw control capability due to the rudder for a high performance fighter aircraft. As the angle of attack is increased the rudder becomes less effective. On the other hand, the yaw moment due to the asymmetric forebody vortices increases with increasing angle of attack. Control of the forebody vortices can be used to provide adequate lateral/directional control for high angle of attack flight,! An overview of forebody vortex control can be found in Malcolm.7 Successful control of forebody vortices has been proven through pneumatic and mechanical techniques. Jet nozzles, when located in specific axial and circumferential locations and pointed properly, can control the trajectory of the vortices by manipulating the location of flow separation.*-” Investigations by Roes” havedetermined that synthetic-jet microblowing (zero-net-mass-flux) can manipulate high AOA forebody vortices as effectively as steady jets of equivalent mean mass flow rate. Flight test results on an F-18 have proven successful control authority can be achieved by actuating forebody mechanical flippers. In addition, it has been shown that the natural asymmetry can be manipulated by altering the apex region of the forebody geometry with surface perturbations.‘2 Adequately controlling the forebody flowfield at the high AOA ranges will aid the stability and maneuverability of the air vehicle, and lends itself to potentially reducing the radar signature by reducing the size of other empennage control surfaces. The emphasis on LO features has already driven the aircraft designs to posses chined forebody cross-sections. This chined feature fixes the circumferential location of flow separation on the forebody. An aerodynamic characteristic of the chined configuration is that it generates stronger leeward vertical flow structures resulting in greater aerodynamic lifting forces.13 The stronger flowfield also makes it more difficult to control. In order to gain a thorough understanding of the forebody flow physics and its implications to an air vehicle as it maneuvers at high AOA, it must be studied dynamically in addition to statically. Smith and Nunn14 . investigated an axisymmetric body undergoing uniform pitch-up rates from O-90”. They noted an increment of normal force developed due to the pitching motion and is closely coupled with the Reynolds number. However, for slower pitch rates, the normal force development is only slightly different than the static AOA case. Mange and Bragg15’16 have investigated a cliined forebody configuration as it pitches sinusoidally through moderately high AOA. The normal force developed sooner in the upstroke and unloaded quicker in the downstroke than in the static case for all AOA. By visualizing the displacement of the leeward vertical flow structures, the primary core was noticed to be “tighter” and closer to the body during the upstroke phase compared to the static case. It was also determined that the windward surface pressure distribution was just as important to the leeward surface pressure in determining the integrated aerodynamic~loads. As it is understood that the magnitude of control power from forebody control techniques varies with angle of attack, it is also realized that the effectiveness will vary when the forebody geometry is in motion. The findings of Smith and Nunn14 and Mange and Bragg15*” highlight the fact that the forebody aerodynamics are influenced by the dynamic motion of the body in both magnitude and

A fundamental investigation of the confluent boundary layer formed by the interaction of the slat wake with the main element boundary layer of a multi-element airfoil was reported in Thomas, Liu and Nelson. One unanticipated result of that study was the observation of significant unsteady high lift flow behavior for certain slat positions. These unsteady motions can have a strong effect on the aerodynamic performance of the high lift system. This appears related to the observation that the unsteady flow caused more aggressive mixing between the slat wake and main wing boundary layer. This results in a decrease in high lift performance and an increase in airframe noise. This unsteady aspect of the high lift flow field forms the focus of this paper. Experimental evidence of unsteady flow field behavior is presented first. These results suggest that the unsteady flow behavior is a manifestation of unsteady vortex shedding and separated shear layer reattachment associated with the slat cove separation region. A discussion of the wavelet analysis technique is presented and this is shown to be a powerful tool in the characterization of transient flow phenomena. Some representative wavelet transform results obtained in the slat wake are presented and discussed. This work represents part of an ongoing study and the paper concludes by describing our current research effort to understand the unsteady aspects of high-lift flows.

This progress report is a series of overviews outlining experiments on the flow physics of confluent boundary layers for high-lift systems. The research objectives include establishing the role of confluent boundary layer flow physics in high-lift production; contrasting confluent boundary layer structures for optimum and non-optimum C(sub L) cases; forming a high quality, detailed archival data base for CFD/modelling; and examining the role of relaminarization and streamline curvature. Goals of this research include completing LDV study of an optimum C(sub L) case; performing detailed LDV confluent boundary layer surveys for multiple non-optimum C(sub L) cases; obtaining skin friction distributions for both optimum and non-optimum C(sub L) cases for scaling purposes; data analysis and inner and outer variable scaling; setting-up and performing relaminarization experiments; and a final report establishing the role of leading edge confluent boundary layer flow physics on high-lift performance.

The methodology outlined here was shown to be an effective means by which detailed unsteady surface pressure measurements may be obtained on a wing undergoing a self-induced oscillation in a wind tunnel study. Previous methods have relied on forcing the wing through a sinusoidal motion at a fixed amplitude and frequency. The advantage of the technique outlined here is that pressure measurements are obtained for the exact motion time history, including all transient motions. In fact, almost any self-induced or forced motion may be implemented by the system. The method also has the advantage of a significant reduction in the cost of pressure transducers due to the repeatability of the unsteady flowfield.

An experimental investigation on the flowfield surrounding a delta wing undergoing a transient pitching motion was conducted. The leading-edge vortices were marked with smoke, and the model and vortex motions were recorded using high-speed motion picture photography. These film records were then analyzed to yield information on the vortex location and trajectory as functions of both angle of attack and time. Hysteresis effects on the vortex dynamics were apparent for both the pitch-up and pitch-down motions. The amount of lag from the static position was seen to be a function of the combined effects of nondimensional pitch rate and initial breakdown location. Reynolds number effects were inconsistent. Upon completion of the wing motion, the vortex break-down typically would not have reached the static position, but would continue moving relative to the wing surface.

A study of the dynamic behavior of the leading-edge vortices on a delta wing undergoing oscillatory pitching motions is presented, A sharp-edged, flat-plate, delta wing having a sweep angle of 70 deg was used in this investigation. The wing was sinusoidally pitched about its one-half chord position at reduced frequencies ranging from /r = 2ir/c/i/ = 0.05 to 0.30 at root chord Reynolds numbers between 9xl04 and 3.5 x 10s, for angle-ofattack ranges of a = 29 to 39 deg and a = 0 to 45 deg. During these dynamic motions, visualization of the leadingedge vortices was obtained by injecting TiCl4 through ports located near the model apex. The location of vortex breakdown was recorded using high-speed motion-picture photography. The motion-picture records were analyzed to determine the vortex trajectory and breakdown position as a function of angle of attack. When the wing was sinusoidally pitched, hysteresis was observed in the location of the breakdown position. This hysteresis increased with reduced frequency. The velocity of breakdown propagation along the wing and the phase-lag between model motion and breakdown location were also determined. Detailed information was also obtained on the oscillation of breakdown position in both static and dynamic cases.

This paper discusses the use of a seven-hole probe on measurements of leading edge vortices of highly sweep delta wing planforms. Intrusive probe data taken with the pressure probe were compared with non-intrusive measurements made with laser Doppler anemometry system. In addition to probe size, the natural position of breakdown and the sweep angle of the wing are also factors in determining sensitivity of the flow to probe interference. At low angles of attack vortex breakdown does not occur in the vincinity of the model and the seven hole probe was found to yield reasonably accurate measurements. When the angle of attack of the model was increased so that vortex breakdown was near the trailing edge, introducing the probe over the wing would cause the breakdown position to move ahead of the probe. However, when breakdown naturally occurred ahead of the mid-chord of the wing the vortices were found to be less sensitive to a probe placed behind the breakdown point. Vortex breakdown on a lower swept wing is found to be more sensitive to interference. Near the breakdown region, seven hole probe measurement is less accurate due to a combination of probe interference and flow reversal.

The study of the near wake on a cable yawed with respect to a uniform flow was performed using five rigid cable models and a cylinder in a low speed wind tunnel. The models were tested at a Reynolds number of 6,000 based on diameter. Smoke-wire flow visualization was used to qualitatively and quantitatively evaluate the near wake flows. Hot wire measurements of the wake frequency spectrum were obtained. Results show that cable stranding affects shedding characteristics and near wake structure.

An experimental investigation of vortex breakdown on delta wings at high angles of attack is presented. Smoke flow visualization and the laser light sheet technique were used to obtain cross-sectional views of the leading-edge vortices as they break down for a series of flat-plate delta wings having sweep angles of 70, 75, 80, and 85 deg. At low tunnel speeds (as low as 3 m/s), details of the flow that are usually imperceptible or blurred at higher speeds can be clearly seen. A combination of lateral and longitudinal cross-sectional views provides information on the three-dimensional nature of the vortex structure before, during, and after breakdown. Whereas details of the flow are identified in still photographs, the dynamic characteristics of the breakdown process have been recorded using high-speed movies. Velocity measurements have been obtained using a laser Doppler anemometer with the 70 deg delta wing at 30 deg angle of attack. The measurements show that, when breakdown occurs, the core flow is transformed from a jet-like to a wake-like flow.