Computation of turbulent axisymmetric and nonaxisymmetric jet flows using the K-epsilon model

It is known that the standard K-e model does not provide an accurate prediction of the mean flow of turbulent jets. This is so even when the Pope and Sarkar correction terms are included. It is suggested that the K-e model, together with the Pope and Sarkar terms for nonplanar and high convective Mach number flow corrections, does contain the essential ingredients of turbulence physics for adequate jet mean flow prediction. The problem lies in the standard coefficients that were calibrated by using boundary-layer and low Mach number plane mixing layer data. By replacing these coefficients by a new set of empirical coefficients, it is demonstrated that the model can offer good predictions of axisymmetric, rectangular, and elliptic jet mean flows over the Mach number range of 0.4-2.0 and jet total temperature to ambient temperature ratio of 1.0-4.0. The present result conveys the message that it is possible that there is no universally applicable turbulence model. The reason is that although the characteristics and dynamics of fine-scale turbulence may be the same for all turbulent flows, the large turbulence structures, having dimensions comparable to the local length scale of the flow, are significantly influenced by local boundary conditions and geometry. Thus overall turbulence dynamics are somewhat problem type dependent.