Introduction

It is well known that small modifications to the trailing edges of airfoils can have a significant influence on the Kutta condition and therefore on the global circulation and the mean lift. The most popular version among these TEDs (Trailing Edge Devices) was first introduced by Dan Gurney in the seventies in order to enhance the aerodynamic downforce of his racing car. He assembled the car rear wing with a small flap mounted at the trailing edge of the pressure side which was later called ``Gurney-flap''. It delivered stronger downforce and hence allowed increased speed on track bends. Due to its considerable effect compared to the small size combined with low weight and cost, this passive flow control tool can be found on almost any modern racing car.

The first theoretical investigations were published by Liebeck who introduced the concept of trailing edge devices to aircraft aerodynamics [1]. In recent years the effect of Gurney-flaps on airfoil flows have been studied experimentally and numerically [2,3]. Investigations with multi-element configurations have demonstrated that Gurney-flaps can also be combined with standard high lift devices for further increase of effectiveness [4,5].

The gain in lift obtained is unfortunately coupled to an increase in drag [6]. Experiments by Bechert et al. [7] indicate that this extra drag is mainly due to the flow instabilities in the wake of the Gurney-flap. Such unsteady flow structures behind bluff bodies could also be responsible for additional noise. Although the effect of TEDs and diverging trailing edges on the mean flow properties has been investigated in great detail (see e.g. Sauvage [8]), less attention has been paid to the unsteady behavior.

Bechert et al. showed that controlling the wake flow holds the potential to improvements of the overall flow characteristics. Several modifications of standard Gurney-flaps were suggested that contribute to a reduction of drag. Most promising is one version similar to the wing of a dragonfly [7].

Another way to stabilize the unsteady flow in the wake may be to introduce a splitter plate into the flow field behind the Gurney-flap. The effect of such a device is comparable to that of a splitter plate in the wake of a cylinder. Here the size of the shed vortices as well as the amplitude of lift oscillations could be damped [9].

In the present study, the unsteady flow structures in the wake of an HQ17 airfoil with Gurney-flap is investigated by numerical simulations. This airfoil is a laminar profile and has a characteristic bluff trailing edge with a height of $h/c=0.33\%$. It is equipped with flaps of $h/c=0.5\%c$ up to $h/c=2.0\%c$ flap height. Computations include URANS simulations based on the LLR $k$-$\omega$ turbulence model by Rung [10] and Detached Eddy Simulations.

These two different turbulence modeling approaches are compared with respect to their applicability to the requirements of unsteady turbulent flows in order to combine their particular advantages and to improve the capabilities for accurate prediction of high lift flows.

The detailed mechanisms which give rise to the observed drag penalty of Gurney-flaps are investigated, and different methods targeting its reduction are compared. The goal is to identify improved solutions by analysis of different modifications of Gurney-flaps and their effect on the flow structures as well as on the mean flow. Two- as well as three-dimensional simulations were performed to examine the differences between both modeling practices. Results are compared to experiments by Bechert et al. [7]