Bulletin of the American Physical Society
65th Annual Meeting of the APS Division of Fluid Dynamics
Volume 57, Number 17
Sunday–Tuesday, November 18–20, 2012; San Diego, California
Session L24: Aerodynamics IV |
Hide Abstracts |
Chair: James Hubner, University of Alabama Room: 30E |
Monday, November 19, 2012 3:35PM - 3:48PM |
L24.00001: The Role of Free Stream Turbulence on the Aerodynamic Performance of a Wind Turbine Blade Victor Maldonado, Adrien Thormann, Charles Meneveau, Luciano Castillo In the present research, a 2-D wind turbine blade section based on the S809 airfoil was manufactured and tested at Johns Hopkins University in the Stanley Corrsin wind tunnel facility. A free stream velocity of 10 m/s produced a Reynolds number based on blade chord of 2.08.x10$^{5}$. Free stream turbulence was generated using an active grid placed 5.5 m upstream of the blade which generated a turbulence intensity, $T_u $ of up to 6.1{\%} and an integral length scale, $L_\infty $ of about 0.15 m. The blade was pitched to a range of angles of attack, $\alpha $ from 0 to 18 degrees in order to study the effects of the integral length scales on the aerodynamic characteristics of the wind turbine under fully attached and separated flow conditions. Pressure measurements around the blade and wake velocity deficit measurements utilizing a hot-wire probe were acquired to compute the lift and drag coefficient. Results suggest that turbulence generally increases aerodynamic performance as measured by the lift to drag ratio, $L \mathord{\left/ {\vphantom {L D}} \right. \kern-\nulldelimiterspace} D$ except at 0 degrees angle of attack. A significant enhancement in $L \mathord{\left/ {\vphantom {L D}} \right. \kern-\nulldelimiterspace} D$ results with free stream turbulence at post-stall angles of attack of 16 and 18 degrees, where $L \mathord{\left/ {\vphantom {L D}} \right. \kern-\nulldelimiterspace} D$ increase from 2.49 to 5.43 and from 0.64 to 4.00 respectively. This is a consequence of delaying flow separation with turbulence (which is observed in the suction pressure distribution) which in turn reduces the momentum loss in the wake particularly at 18 degrees angle of attack. [Preview Abstract] |
Monday, November 19, 2012 3:48PM - 4:01PM |
L24.00002: Investigation of the rotational flow effects in a pitching airfoil genetically optimized for vertical axis wind turbines Daniele Ragni, Laura Vitale, Andrea Ianiro, Ben Geurts, Carlos Ferreira In the present study, an airfoil optimized for vertical axis wind turbines applications has been developed with a genetic algorithm, selecting the geometry with maximum (dc$_{l}$/d$\alpha )$/c$_{d}$ among airfoils generated with 16 shape functions. The airfoil, operating in the curved trajectory of a vertical axis wind turbine, is usually optimized adopting conformal mappings in the straight path. Recent experimental results have shown disagreement with this approach, due to the forces determined in the curved flow path. To investigate the effects of flow rotation, an aluminum model (c=0.25m) has been manufactured from the optimized shape and further tested in the LST tunnel of the TUDelft at Reynolds number 10$^{6}$. Planar PIV experiments in combination with the PIV based load determination technique have been performed to simultaneously obtain velocity fields and loads. Results including velocity, pressure distributions, lift and drag are initially discussed in a steady airfoil configuration and compared with numerical results. Successively, the model has been unsteadily pitched using a magnetic linear actuator (up to 3~Hz frequency), with a free stream $V_{\infty}$ = 40 m/s corresponding to Re = 0.7$\times $10$^{6}$. Phase locked PIV vector fields have been acquired and compared to the steadily obtained results. [Preview Abstract] |
Monday, November 19, 2012 4:01PM - 4:14PM |
L24.00003: High resolution velocimetry near the trailing edge of rigid and flexible airfoils undergoing unsteady motion David Olson, Ahmed Naguib, Manoochehr Koochesfahani The advantages of Molecular Tagging Velocimetry (MTV) are exploited in performing highly resolved measurements within the boundary layer, and downstream of the trailing edge of rigid and flexible NACA0012 airfoils undergoing canonical unsteady motions. Experiments are performed over a range of motion and flow parameters in an effort to establish the connections between airfoil motion trajectory, trailing edge flexure, and the time history of vorticity flux at the trailing edge. Specifically, multi-line MTV measurements, which are phase averaged relative to the airfoil motion, are used to examine the formation, evolution and characteristics of the wake vortices near the trailing edge and the concurrent behavior of the boundary layer immediately upstream. Results are used to gain insight into the effect of the trailing edge flexibility on the pattern of vorticity shed from, and the flow details around the trailing edge. [Preview Abstract] |
Monday, November 19, 2012 4:14PM - 4:27PM |
L24.00004: Drag coefficient measurements of spheres with different surface patterns Hendrik Heisselmann, Daniel Strutz, Joachim Peinke, Michael Hoelling Precise drag force measurements of bluff bodies are an under-estimated challenge and in particular drag coefficients of bodies with rough surface structure are not very well documented in literature. In our contribution, we present a new setup for measurements of the acting drag forces on spheres and other bluff bodies. The examined bodies are attached to a slim supporting rod, which is held by thin steel wires in a cubical rigid frame, and the resulting velocity-dependent forces are measured by means of strain gauges. Besides a detailed description of the achieved experimental setup, we will present results from force measurements using smooth spheres and a sphere with a dimpled surface pattern. Measurements were performed for a Reynolds number range of 2,700 up to 230,000 under laminar inflow conditions as well as in turbulent flows generated by classical and fractal grids. An overview of the calculated drag coefficients will be given for different sphere types and for varying turbulence levels. The obtained results will be compared to those documented in literature. [Preview Abstract] |
Monday, November 19, 2012 4:27PM - 4:40PM |
L24.00005: Comparison of Aerodynamic Coefficients from Low Aspect Ratio Membrane Wings and their Time-Averaged Shape Nathan Martin, Andrew Wrist, Zheng Zhang, James Hubner Air flow over flexible membrane wings can induce vibration. The vibrating nature and the time-averaged curvature of a membrane wing may separately contribute to its improved aerodynamic characteristics compared to a flat plate of similar planform. To assess the effect of the time-averaged shape, a comparison of vibrating membrane wings and corresponding time-averaged shape of an aspect ratio two planform was conducted for membranes pre-tensions of 1{\%}, 2{\%}, and 4{\%} and various cell aspect ratios. The membrane displacements were recorded using digital image correlation for each model at 6$^{\circ}$ and 18$^{\circ}$ angles of attack. The displacements were averaged, imported into CAD software, and printed using rapid prototyping equipment. The lift, drag, and pitching moment coefficients were acquired through wind tunnel testing at Reynolds number 50,000. The results indicate that membrane wings generate more drag but are more efficient than their time-averaged shapes due to greater lift. [Preview Abstract] |
Monday, November 19, 2012 4:40PM - 4:53PM |
L24.00006: Coupled Fluid and Structure Measurements over a Low Aspect Ratio Membrane Wing Lawrence Ukeiley, Manuel Arce, Amory Timpe, Zheng Zhang, James Hubner The coupled effect of flow induced membrane deformations and their return influence on the flow are investigated on an aspect ratio 2 thin wing. The wings have multiple cells with a free scalloped trailing edge and are made by adhering heated, thin Silicone membranes to thin rectangular aluminum frames with a rigid leading edge and battens. Time-resolved flow and structure deformations are measured by synchronized acquisition of high-speed two-component particle image velocimetry (PIV) and stereoscopic digital image correlation (DIC) at a chord based Reynolds number of 48,000 and several angles of attack. Flow and structure metrics are compared for membrane with different values of pretension in the rubber. Instantaneous flow fields and mean flow properties are analyzed and compared to a rigid plate of the same dimensions. Specifically, the effects of membrane behavior on flow separation, shear layer size and location, along with vorticity will be analyzed. Power spectral density and correlation techniques are utilized, along with analysis of membrane mean deformation and rms fluctuation behavior to better understand the fluid-structure interactions and how the membranes interact with each other. [Preview Abstract] |
Monday, November 19, 2012 4:53PM - 5:06PM |
L24.00007: Formulation for Time-resolved Aerodynamic Damping in Dynamic Stall Thomas Corke, Patrick Bowles, Dusty Coleman, Flint Thomas A new Hilbert transform formulation of the equation of motion for a pitching airfoil in a uniform stream yields a time resolved aerodynamic damping factor, $\Xi(t)=(\sqrt{(C_{m}^2(t)+\tilde{C}_{m}^2}/\alpha_{max})\sin\psi(t)$, where $C_{m}(t)$ is the instantaneous pitch moment coefficient, and $\tilde{C}_{m}(t)$ is the Hilbert transform of $C_{m}(t)$, $\alpha_{max}$ is the pitching amplitude, and $\psi(t)$ is the time-resolved phase difference between the aerodynamic pitch moment and the instantaneous angle of attack. A $\Xi(t)<0$ indicates unstable pressure loading that can be considered a necessary condition to excite stall flutter in an elastic airfoil. This will be illustrated in experiments with conditions producing ``light" dynamic stall for a range of Mach numbers from 0.3-0.6. These reveal large negative excursions of $\Xi(t)$ during the pitch-up portion of the cycle that correlates with the formation and convection of the dynamic stall vortex. The fact that the cycle-integrated damping coefficient is positive in all these cases underscores how the traditional diagnostic masks much of the physics that underlies the destabilizing effect of the dynamic stall process. This new insight can explain instances of transient limit-cycle growth of helicopter rotor vibrations. [Preview Abstract] |
Monday, November 19, 2012 5:06PM - 5:19PM |
L24.00008: Static versus dynamic stall development Karen Mulleners, Markus Raffel Stall on lifting surfaces is commonly encountered, mostly undesired, and occurs when a critical angle of attack is exceeded. Depending on the unsteady rate of change of the airfoil's angle of attack, static and dynamic stall are distinguished. To design efficient flow control measures, a fundamental understanding of the flow and vortex dynamics during stall development is desirable. Detailed information about the spatial and temporal evolution of the dominant flow features is obtained by time-resolved velocity field measurements combined with an extensive coherent structure analysis. Time-resolved flow field investigations during static and dynamic stall reveal flow reversal near the airfoil's surface at the beginning of stall development. At the interface between the region of reversed and free stream flow, a shear layer develops which plays the key role in the subsequent stall development. During dynamic stall, the shear layer rolls up into a large scale dynamic stall vortex which grows locally and temporally until vortex induced separation occurs. During static stall on the other hand, the shear layer rolls up continuously into large-scale structures that grow spatially. [Preview Abstract] |
Monday, November 19, 2012 5:19PM - 5:32PM |
L24.00009: Large-Scale High-Resolution Cylinder Wake Measurements in a Wind Tunnel using Tomographic PIV with sCMOS Cameras Dirk Michaelis, Andreas Schroeder Tomographic PIV has triggered vivid activity, reflected in a large number of publications, covering both: development of the technique and a wide range of fluid dynamic experiments. Maturing of tomo PIV allows the application in medium to large scale wind tunnels. Limiting factor for wind tunnel application is the small size of the measurement volume, being typically about of 50 x 50 x 15 mm$^{3}$. Aim of this study is the optimization towards large measurement volumes and high spatial resolution performing cylinder wake measurements in a 1 meter wind tunnel. Main limiting factors for the volume size are the laser power and the camera sensitivity. So, a high power laser with 800 mJ per pulse is used together with low noise sCMOS cameras, mounted in forward scattering direction to gain intensity due to the Mie scattering characteristics. A mirror is used to bounce the light back, to have all cameras in forward scattering. Achievable particle density is growing with number of cameras, so eight cameras are used for a high spatial resolution. Optimizations lead to volume size of 230 x 200 x 52 mm$^{3}$ = 2392 cm$^{3}$, more than 60 times larger than previously. 281 x 323 x 68 vectors are calculated with spacing of 0.76 mm. The achieved measurement volume size and spatial resolution is regarded as a major step forward in the application of tomo PIV in wind tunnels. [Preview Abstract] |
Monday, November 19, 2012 5:32PM - 5:45PM |
L24.00010: Transient Response of a Separated Flow over a Two-Dimensional Wing to a Short Duration Pulse David Williams, Thomas Albrecht, Tom Weier, G. Gerbeth A Lorentz force actuator located at the leading edge of a two-dimensional wing at 16 degrees angle of attack was used to introduce short-duration disturbances into a separated flow. The transient response of the separated region at Re = 10,000 was documented using time-resolved PIV measurements. The direction of the Lorentz force was changed between downstream and upstream directed disturbances, and details of the resulting flow field structures and lift measurements were studied. Saturation of the peak lift amplitude occurs as the actuation amplitude is increased from 0.0054 $<$ C$\mu <$ 0.21 percent with the pulse duration fixed at 0.1 convective time. The effect of the pulse duration time on the lift response was examined using a fixed pulse amplitude, which showed that saturation occurred when pulse durations exceed 0.5 convective times. Differences in the coherent structures resulting from the upstream/downstream directed actuation were identified using the FTLE method. The initial development of the disturbed shear layer was strongly dependent on the direction of actuation, but the larger-scale separation did not show much difference. The relaxation of the separated region to the original flow state was essentially independent of the direction of actuation. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700