A Comparative Study of Turbulence Models in Axisymmetric Nozzle Flow

Two turbulence models have been studied to determine which of the models should be used in further Computational Fluid Dynamics (CFD) research. A zero-equation turbulence model, Baldwin-Lomax (B-L), is easy to use, requires no history of the flow, and requires little in the way of additional computations or additional computer memory space [1]. A two-equation k-ε model, Yang-Shih (Y-S), is more difficult to implement, does require flow history, and requires many more computations and much more computer space; however, it is potentially more accurate than the B-L model [2]. Using both Navier-Stokes (NS) and Parabolized Navier-Stokes (PNS) solvers, the two models and their codes were validated against the testbed of the Wright Laboratory (WL) Mach 12 wind tunnel nozzle. Nozzle flow solution wall data and exit data were compared with experimentally obtained nozzle data, with the surprising result that the B-L model provided the more accurate solution, making it the chosen turbulence model for further nozzle research.