DISTRIBUTION A: “Approved for public release: distribution unlimited.” 412TW-PA-14516 1 American Institute of Aeronautics and Astronautics Applicability of Hybrid RANS/LES Models in Predicting Separation Onset of the AVT-183 Diamond Wing Daniel A. Reasor Jr. * Air Force Test Center, 412 th Test Wing, Edwards AFB, CA, U.S.A. Donald J. Malloy † Air Force Test Center, Arnold Engineering Development Complex, Arnold AFB, TN, U.S.A and Derick T. Daniel ‡ Aerospace Testing Alliance, Arnold AFB, TN, U.S.A. This effort contributes to the understanding of hybrid RANS/LES turbulence model behavior for prediction of separation onset via an investigation of the accuracy and predictions of the SARC-DDES variant. The focus is not to interrogate the underlying assumptions of the turbulence model or flow solver employed, but to report the accuracy of flowfield and surface quantity predictions given the choices for grid topology, spatial and temporal resolution, and numerical schemes used. Grid convergence is shown to be difficult to demonstrate for flows of this type. Steady state SARC and unsteady SARC-DDES simulation results demonstrate utility for pre-test predictions, but fail to resolve relevant physics in some instances. Nomenclature c = chord, m C f = skin-friction coefficient C p = pressure coefficient C D = drag coefficient C L = lift coefficient C M = pitching moment coefficient Ma = Mach number Re = Reynolds number St = Strouhal number U ∞ = freestream velocity, m/s α = angle of attack, deg. β = angle of sideslip, deg. Δs = grid spacing, mm Δt = time step, s γ = separation probability I. Introduction difficult phenomenon to predict with modern Reynolds-averaged Navier−Stokes (RANS) and hybrid turbulence models based on a combination of RANS and large-eddy simulation (LES) methods is the incipient separation from a relatively blunt leading edge on a swept wing. Wings with relatively blunt leading edges have a larger radius of curvature (by definition) and generate less adverse pressure gradients as a result. When the curvature is modest, there lacks massive separation and the point or region of separation is difficult to capture accurately. Once separated, the flow consists of small structures and requires a refined grid. An obvious limitation of employing a hybrid RANS/LES turbulence model for this type of separation prediction is that the RANS zone of the simulation will ultimately control the separation point. Furthermore, any prediction insufficiencies or limitations will * Engineer, Engineering Testing Techniques Flight, AIAA Senior Member † Aerodynamics Lead, Analysis and Evaluation Branch, AIAA Associate Fellow ‡ Engineer, Computational Simulation, Modeling & Analysis, AIAA Member A Downloaded by NASA LANGLEY RESEARCH CENTRE on January 12, 2015 | http://arc.aiaa.org | DOI: 10.2514/6.2015-0287 53rd AIAA Aerospace Sciences Meeting 5-9 January 2015, Kissimmee, Florida AIAA 2015-0287 This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States. AIAA SciTech