Effects of Solver Parameters and Boundary Conditions on RANS CFD Flow Predictions over a Gen-6 NASCAR Racecar 2022-01-0372
Racecar aerodynamic development demands rapid and incremental development cycles using extensive and well-correlated simulation data. The successful implementation of such a process is a major performance differentiator between race teams. Reynolds Averaged Navier-Stokes (RANS) simulations are an industry-wide tool of choice for their relatively quick turn-around times and cost-effectiveness. A limitation of RANS simulation is an inability to fully resolve flow separation and wake structures of the racecar geometry thereby reducing the accuracy of simulation and the confidence in incremental development work. However, race organizers of both Formula1 and NASCAR are placing increasing limits on aerodynamic development such as number of runs in a wind tunnel and CPU hours for CFD simulation. This prevents widespread use of LES or DES methodologies that require 5-10 times more computational resources. The confidence in a RANS simulation must increase to meet the development limitations while continuing to provide the necessary aerodynamic performance advantage. The current study aims to develop a framework that increases the confidence in RANS simulation using the popular SST k-ω turbulence model by evaluating the effects of solver parameters, closure coefficients and boundary conditions. For this a full-scale detailed model of a Gen-6 NASCAR was investigated using simulations run in the commercial CFD code Star-CCM+ (version 2020.2.1). The RANS simulations were validated against moving-ground, open-jet wind tunnel (Windshear) data using two ride-heights and two crosswind angles. The influence of realizability, compressibility, and simulation setup were analyzed. The proposed framework has shown good correlation in aerodynamic coefficients with lift and drag predictions within 2% of wind tunnel data thus providing high confidence in incremental aerodynamic development.