Gradient–transport models for the turbulent scalar fluxes have proved to be deficient in a number of important practical flows, especially in those where the effects of body forces and extra rates of strain are significant. There are numerous examples in this respect, most notably in the prediction of the heat transfer rates in flows where the effects of stabilizing streamline curvature are sufficiently strong to completely extinguish turbulent mixing. The cause of the problem appears to be in the omission from these models of an explicit dependence of these fluxes on the mean strain field — a dependence that is required by the exact equations that govern the evolution of the turbulent scalar fluxes. The purpose of this paper is to determine whether the incorporation of such dependence in an algebraic model for the turbulent scalar fluxes is on its own sufficient to capture the main effects of streamline curvature. The model used here is an explicit, non–linear model that was developed in collaboration with the late Professor Speziale (Younis, Speziale & Clark, 1996). It is first validated in this paper by comparisons with data from one– and two–dimiensional heated free shear flows and is then used to predict the main features of a heated homogeneous shear layer in local equilibrium. Comparisons made with experimental data and with other models demonstrate the validity of the present approach.

This content is only available via PDF.
You do not currently have access to this content.