Effects of Inflow Conditions and Subgrid Model on LES for Turbulent Jets

Paper i proceeding, 2005

The turbulent mixing process prescribing the spreading rate of the jet and the length of the potential core region is influenced by a number of factors. Using large-eddy simulation (LES), the four factors that are believed to be the most important in this respect are: subgrid-scale properties, the accuracy of the numerical scheme, the entrainment boundary conditions, and the inflow conditions. In a previously performed study of a subsonic (Mach 0.75) jet, the turbulence mixing was found to be too efficient and hence the length of the potential core region was underpredicted. In that study indications were found of that the overpredicted mixing was due to the inflow conditions. For a model nozzle, capturing the initial turbulent shear flow might not be of that great importance for accurate prediction of radiated sound since most of these effects will appear in the high-frequency range. When dealing with real engine geometries, however, it becomes quite important. Moreover, methods for industrial use have to cope with complex geometries and high temperature and velocity ratios making the ability to capture the initial flow physics even more important. In the present work LES has been used for the same Mach 0.75 jet. The acoustic field is extracted to the far field using Kirchhoff surface integration. The effects of inflow conditions, Reynolds number, and subgrid-scale model on flowfield and acoustic signature are investigated. The Favre-filtered Navier-Stokes equations were solved using a finite-volume method solver with a low-dissipation third-order upwind scheme for the convective fluxes, a second-order centered difference approach for the viscous fluxes and a three-stage second-order Runge-Kutta technique in time. The computational domain was discretized using a block-structured boundary-fitted mesh with approximately 3,000,000 cells. The calculations were performed on a parallel computer, using message-passing interface (MPI). A compressible form of Smagorinsky's subgrid-scale model was used to compute the subgrid-scale stresses. Absorbing boundary conditions based on characteristic variables were adopted for all free boundaries.