Computational Fluid Dynamics Enhancements for Scramjet Flow Simulations

Combusting flows in general, and scramjet flows in particular, strain the performance of existing CFD solvers to their limit, the weakest link being the turbulence closure. Prediction of high-speed combusting flows is particularly sensitive to the level of accuracy in computing mass transport. Solvers based on eddy viscosity models relate turbulent mass flux to species gradients through an exchange mechanism involving eddy viscosity and a so-called turbulent Schmidt number (Sct). A similar approach is used for energy transport, the equivalent parameter being the turbulent Prandtl number (Prt). Both numbers are treated as constants throughout the flow-field. However, experimental evidence suggests that these parameters vary significantly, thus models that treat them as constants cannot predict the flow details with sufficient accuracy. The proposed approach introduces a method for computing variable Sct and Prt numbers with the potential to significantly improve scramjet flow predictive capability.