Abstract:
Numerical simulations using the Reynolds-Averaged Navier-Stokes (RANS) equations were conducted to study the development and turbulent decay of the tip vortices in the wake produced by a hovering rotor blade. An extensive series of calculations was performed with the OVERFLOW 2.2l code, the results being compared to detailed dual-plane Particle Image Velocimetry (PIV) measurements of a turbulent tip vortex trailed from a single-bladed rotor. The work investigated both the required mesh resolution and most suitable turbulence closure models with corrections for rotation and streamline curvature by assessing their predictions of the tip vortex properties and the overall physical nature of the rotor wake. It was found that even when using a higher-order spatial discretization scheme, a minimum off-body grid spacing equal to 0.625% of the chord length was required to accurately predict the core size, peak swirl velocity and strength of the tip vortex. The rotational/curvature corrections applied to the Spalart-Allmaras turbulence model better preserved the vortex characteristics to later wake ages than the same corrections applied to the k-w SST model. In both cases, the correction proposed by Spalart and Shur outperformed the simplified correction proposed by Dacles-Mariani et al., with the latter providing little impact on the k-w SST model.