Hybrid RANS-LES Modeling of the FDA Round Robin Blood Pump

            Blood-contacting medical devices such as blood pumps possess an inherent need for accurate predictions of hemodynamic flow fields. Due to the lack of standardized validation methods for the use of computational fluid dynamics (CFD) simulations in medical devices, the Food and Drug Administration (FDA) developed an idealized centrifugal blood pump and established an inter-laboratory study to gather Particle Image Velocimetry (PIV) data for CFD Validation. This work focuses on benchmarking hybrid RANS-LES models (SBES, IDDES SST) and a standard RANS model (k-ω SST) against PIV results published online for three flow  conditions. Results show that the hybrid SBES and IDDES SST models more accurately predicted velocity fields. The SBES model slightly overpredicted pressure rise of the pump where the k-ω SST possessed a tighter fit to experimental data.