Effect of ice on airfoil stall at high Reynolds numbers

Abstract

Introduction I wing design it is very important to determine the maximum lift coefficient as accurately as possible, since this lift coefficient corresponds to the stall speed that is the minimum speed at which controllable flight can be maintained. Any further increase in angle of incidence will increase flow separation on the wing upper surface, resulting in a loss of lift and a large increase in drag. Despite the significant advances in computational fluid dynamics, our ability to predict the maximum lift coefficient and poststall behavior of single airfoils, wings, multielement airfoils, or wings is still not satisfactory. The chief culprit is the lack of an accurate turbulence model to represent flows with extensive separation. Inaccuracies of numerical solutions of the conservation equations at these flow conditions, as well as difficulties in modeling of flow near the trailing edge of an airfoil or wing, add to this dilemma. Recently Cebeci et al. described an interactive boundary-layer method, together with the e approach to the calculation of transition, for predicting stall and poststall behavior of airfoils at low and high Reynolds numbers. The turbulence model was based on the Cebeci-Smith algebraic eddy-viscosity formulation with improvements for strong pressure gradient effects and transitional flows at low Reynolds numbers. Comparison of calculated results for incompressible flows indicated good agreement with experiment for a wide range of Reynolds numbers. Preliminary calculations for low-Mach-number flows with this interactive method that applies compressibility corrections to the panel method also indicated good agreement with data and showed that at a Mach number of 0.30, the compressibility effect on (Q)max was not negligible. The ability of the calculation method of Ref. 1 to predict the effect of ice on airfoil stall at high Reynolds numbers is examined here, and the results are presented in the following section. The interactive boundary-layer method previously developed for iced airfoils and coupled to the LEWICE code for predicting ice shapes' is replaced with the calculation method of Ref. 1. Appropriate changes are made to this method, including roughness effects due to ice, and the results are presented in the following section.