Aerodynamics of wings at low Reynolds numbers: boundary layer separation and reattachment

Unrestricted Due to advances in electronics technology, it is now possible to build small scale flying and swimming vehicles. These vehicles will have size and velocity scales similar to small birds and fish, and their characteristic Reynolds number will be between 104 and 10^5. Currently, these flying and swimming vehicles do not perform well, and very little research has been done to characterize them, or to explain why they perform so poorly. This dissertation documents three basic investigations into the performance of small scale lifting surfaces, with Reynolds numbers near 10^4.; Part I. Low Reynolds number aerodynamics. Three airfoil shapes were studied at Reynolds numbers of 1 and 2 x 10^4: a flat plate airfoil, a circular arc cambered airfoil, and the Eppler 387 airfoil. Lift and drag force measurements were made on both 2D and 3D conditions, with the 3D wings having an aspect ratio of 6, and the 2D condition being approximated by placing end plates at the wing tips.; Comparisons to the limited number of previous measurements show adequate agreement. Previous studies have been inconclusive on whether lifting line theory can be applied to this range of Re, but this study shows that lifting line theory can be applied when there are no sudden changes in the slope of the force curves. This is highly dependent on the airfoil shape of the wing, and explains why previous studies have been inconclusive.; Part II. The laminar separation bubble. The Eppler 387 airfoil was studied at two higher Reynolds numbers: 3 and 6 x 10^4. Previous studies at a Reynolds number of 6 x 10^4 had shown this airfoil experiences a drag increase at moderate lift, and a subsequent drag decrease at high lift. Previous studies suggested that the drag increase is caused by a laminar separation bubble, but the experiments used to show this were conducted at higher Reynolds numbers and extrapolated down.; Force measurements were combined with flow field measurements at Reynolds numbers 3 and 6 x 10^4 to determine whether the drag increase is really caused by the formation of a laminar separation bubble. The results clearly indicate that the reverse is true, and that the subsequent drag decrease is caused by the laminar separation bubble.; Part III. The leading edge vortex. Four wings with different sweep angles were studied at Reynolds number 5 x 10^4: sweep angles of 0deg, 20deg, 40deg, and 60deg. The wings had a simple cambered plate airfoil similar to the cambered airfoil of part I above. Each wing was built to have the same aspect ratio, wing area, and streamwise airfoil shape. Previous studies on bird wings speculate that simply sweeping the wings can cause a leading edge vortex to form, which could cause substantial improvements in performance. However, these studies were not well controlled, and were conducted from a biological perspective.; Qualitative and quantitative flow field measurements were combined with force measurements to conduct a well controlled engineering experiment on the formation and effect of a leading edge vortex on simple swept wings. A stable vortex was found to form over the 60deg swept wing at one particular angle of attack, but it was not similar to the traditional notion of a leading edge vortex. The vortex has a small radius, and extends over little of the span. Force measurements indicate that the vortex has no significant impact on the forces measured. Thus, simply sweeping a wing is not sufficient to form a significant leading edge vortex, and other effects must be considered.