High-performance sailplane airbrake analysis using CFD

Abstract

Sailplane airbrakes are an essential part of sailplane control and safety. So much so that EASA, (European Aviation Safety Agency) has specific regulations specifically aimed at sailplane airbrake design. For sailplanes to be competitive, the latest in simulation software and techniques are required to design on the cutting edge of what is allowed. This is especially important in the airbrake design as flow over airbrakes is notoriously difficult to predict, leading to larger design margins and loss of performance gains. The aim of this study is to focus on a methodology to achieve accurate, reliable and efficient CFD results, in particular for use during the concept design phase of sailplane aerodynamics. An overall strategy of starting simple and incrementally increasing the complexity was followed to achieve this. The study started by using simple geometries and worked up to more complex geometries, such as the airbrakes. Generally, the methodology was to run mesh refinement studies as well refining meshes in crucial regions of high flow gradients in conjunction with time step and inner iteration independent studies. Once these simulations were completed, validations were done using CFD or wind tunnel data as a reference. This methodology was first applied to unsteady flow over three cylinder cases. Two circular cylinders at Re = 200 and Re = 3.6 × 106 and a square cylinder at Re = 2.2 × 104. Secondly, validation simulations were done on two airfoils using wind tunnel data as a reference. These two airfoils are the FX66-17AII-182 and HPH yn1. Both airfoils are used in sailplane applications, making them suitable for this study. Thirdly the methodology was applied to the two airfoils with airbrakes extended. A good correlation was achieved with the cylinder validation studies and the two FX66 airfoil and airbrake validations. Even though the simulations were only 2D, expected flow characteristics such as vortex shedding and recirculation were simulated and captured. The simulations on the HPH airfoil and airbrake cases did not give such good results and required further investigation. These investigations were done on various other constitutive options and the effects modifying the shear stress limiter and realizability coefficients in the k − ω SST model had on the simulation results. In addition to using more complex turbulence solvers, such as RST and Lag EB k − ϵ, various techniques were studied that were especially used on anisotropic flows, such as recirculation, separation and vortex shedding. The clean HPH airfoil force coefficients correlated well with the wind tunnel results in the linear range (AOA -6° up to 5°), after which the results deviated from that of the wind tunnel. None of these additionally investigated parameters increased the simulation accuracy of the clean airfoil from 8° nor the accuracy of the airfoil and airbrake configuration results.