A sensitivity analysis of the transitional flow over a NACA662 - 415 foil and the elliptical Arndt wing is carried out. The SST turbulence model is complemented with the ? - Re ~?t transition model to determine the effect of varying turbulence intensity and eddy-viscosity ratio on the integral quantities and transition locations. Local grid refinement at the transition location is used to improve convergence. The skin friction drag coefficient is found to be more sensitive to the inflow conditions for 5? angle of attack compared to 9?. The movement of the transition location on the suction side is found to be responsible for this. The transition model captures a laminar separation bubble at the pressure side for both angles of attack, causing the lift coefficient to drop slightly. 3D calculations for the Arndt wing with the same foil section show that applying a transition model can decrease the boundary layer thickness by a factor of three, which is expected to influence the viscous core radius and consequently the minimum pressure in the tip vortex.@en

This study evaluates the influence of transition and turbulence modeling on the prediction of wetted and cavitating tip vortices for an elliptical wing, while investigating the numerical errors. Transition modeling increases the quality of numerical predictions since the assumption of a fully turbulent boundary layer, commonly found in literature, contributes to underprediction of the tip vortex cavity size. By applying the local correlation-based transition model (LCTM) and controlling the boundary layer thickness using different turbulent inflow conditions, the pressure in the vortex was found to reduce by 20% for an Angle of Attack (AoA) of 5 deg. The consequent increase in cavity size was found to be of a similar order of magnitude. At 9 deg AoA, transition always occurs just downstream of the leading edge, making this AoA more suitable to investigate the effect of different turbulence modeling approaches. Azimuthal and axial velocity fields are validated against stereographic-particle image velocimetry (S-PIV) measurements. The time-averaged velocity profiles predicted by delayed detached-eddy simulation (DDES) and improved delayed detached-eddy simulation (IDDES) are close to the experiments; however, no velocity fluctuations and vortex dynamics are observed around the vortex. A comparison of wetted and cavitating simulations shows that the cavity leads to a change in the balance between production and destruction of turbulence kinetic energy, which reduces the turbulent diffusion in and around the cavity compared to wetted flow conditions. Consequently, the vapor flow exhibits the characteristics of a potential flow. Whether this is physically plausible remains to be investigated. @en