The current work analyzes the effect of time-step size on the predictive capability and computational cost of the Sliding Mesh (SM) method for modeling flows around the rotating wheels of a mass-production luxury sport utility vehicle (SUV). Two unsteady turbulence models Unsteady Reynolds-Averaged Navier–Stokes (URANS) and Delayed Detached Eddy Simulations (DDES) were tested using time-step sizes ranging from the current recommended time-step size of 1 degree of rotation per time-step (1 D/TS) up to 50 degrees of rotation per time-step (50 D/TS). The flow field predictions compare favorably to the 1 D/TS case for a time-step size as large as 5 D/TS. Using this time-step size leads to a reduction in computational cost of approximately 80% for both unsteady methods. At a time-step of 5 D/TS, the computational cost of the SM method is comparable to the more commonly used Moving Reference Frame (MRF) method. However, drag and flow field predictions by the SM method at this larger time-step compare far more favorably to the 1 D/TS SM case than the standard MRF method. Thus, increasing the time-step size is an effective way to implement the more accurate SM rotation model without increasing the cost over the MRF.
Struk et al. (Fri,) studied this question.