Author(s): H. J. Wang; M. J. Davidson
Linked Author(s):
Keywords: Merging jets; Co-flowing ambient; Model predictions; Velocity and concentration distributions
Abstract: A new model for the merging of co-flowing jets is presented. The behavior of the merging flows is governed by the conservation of momentum and tracer fluxes, and the nature of velocity and tracer profiles is determined based on the summation of these fluxes at each location. This model differs from previous attempts, which have employed an additive assumption to model the velocity and tracer profiles. It is shown that the additive assumption is consistent with the summation of momentum and tracer fluxes for the merging of strongly-advected jets, but that the two approaches are inconsistent for the merging of weakly-advected jets. Comparisons are made between the predictions of a model based entirely on the additive assumption and the new model. In addition predictions from the models are compared with available experimental data. The comparisons indicate that, although the additive assumption is inconsistent with the summation of momentum and tracer fluxes for the merging of weakly-advected flows, the errors created by making this assumption are not significant when the predictions are compared with experimental data. To gain a better understanding of the errors generated by the additive assumption in the weakly-advected region, the two models are applied to the merging of jets in a still ambient fluid (the appropriate limiting case). Taking velocity as an example, there are notable differences in the cross-sectional velocity profile predictions from the two models. While both models conserve the overall momentum flux of the flow, the distribution of the momentum flux within the jet cross-section differs. The additive assumption re-distributes the momentum flux in jet cross-section more rapidly and hence merging occurs over a relatively short distance. In addition with the additive assumption model the centerline velocity reduces more rapidly, so that overall momentum flux remains constant. Comparisons of tracer distributions from the two models show similar discrepancies. Although the differences are small, considerable care must be exercised when making use of the additive assumption.
Year: 1999