LOW-SPEED SINGLE-ELEMENT AIRFOIL SYNTHESIS

Abstract

Large quantities of experimental data exist on the characteristics of airfoils operating in the Reynolds number range between one and ten million, typical of conventional atmospheric wind tunnel operating conditions. Beyond either end of this range, however, good experienental data becomes scarce. Designers of model airplanes, hang gliders, ultralarge energy efficient transport aircraft, and bio-aerodynamicists attempting to evaluate the performance of natural flying devices, are hard pressed to make the kinds of quality performance/design estimates taken for granted by sailplane and general aviation aerodynamicists. Even within the usual range of wind tunnel Reynolds number, much of the data is for "smooth" models which give little indication of how a section will perform on a wing of practical construction. The purpose of this paper is to demonstrate the use of recently developed airfoil analysis/design computational tools to clarify, enrich and extend the existing experimental data base on 1ow-speed, single element airfoils, and then proceed to a discussion of the problem of tailoring an airfoil for a specific application at its appropriate Reynolds number. This latter problem is approached by use of inverse (or "synthesis") techniques, wherein a desirable set of boundary layer characteristics, performance objectives, and constraints are specified, which then leads to derivation of a corresponding viscous flow pressure distribution. In this plocedure, the airfoil shape required to produce the desired flow characteristics is only extracted towards the end of the design cycle. This synthesis process is contrasted with the traditional "analysis" (either experimental or computational) approach in which an initial profile shape is selected which then yields a pressure distribution and boundary layer characteristic, and finally some perfonnance level. The final configuration which provides the required performance is derived by cut-and-try adjustments to the shape. These two approaches are
shown disgramatically.