CFD modelling and simulation on a lambda wing at subsonic speed

This paper presents and discusses the results of a study at subsonic airspeed of the aerodynamic characteristics of the Swept Wing Flow Test (SWIFT) lambda wing configuration, which was undertaken as part of the NATO AVT-298 Task Group activity on “Reynolds Number Scaling Effects on Swept Wing Flows”. While the task group studied the aerodynamics of this unconventional wing shape across the subsonic and transonic Mach number, and a wide range of Reynolds numbers via cryogenic testing in the NASA NTF wind tunnel, this paper focuses only on the Mach 0.2 conditions at a Reynolds number, based on mean chord, of 2.5 million, for which the model was tested at the ARA Transonic Wind Tunnel in the UK. Various fidelity CFD methods were employed for comparison with experimental data, over a pitch sweep from -4 to 20 degrees angle of attack, including the Viscous Full Potential (VFP) method, RANS, Unsteady RANS and Delayed Detached Eddy Simulation (DDES). The results for this case, highlight the complex 3D stall, initiating inboard, associated with this class of swept wing, which is very different from that seen on conventional swept, tapered wings typically seen on civil transport aircraft, which tends to initiate towards the tip. While the results show that, of the RANS turbulence models tested, the k-omega SST turbulence model most effectively predicted the experimental data, but none of the linear eddy viscosity models could resolve the benign stall characteristics captured in the experiment. Only the DDES method was found to effectively predict the post stall characteristics to some degree of accuracy. The VFP method generated results in a fraction of the time (seconds compared with hours), required for higher fidelity CFD solution, and was found to provide data with equivalent accuracy to RANS based methods for pre-stall conditions.