ABSTRACT -- CTS-9304485 PI -- Ta-Shen Chen The proposed research is a coordinated experimental and numerical study on buoyancy-dominated, turbulent separated flows over backward-facing and forward-facing step geometries that are oriented in vertical and horizontal positions. It is aimed at providing quantitative results for the momentum and thermal transport in the separates and reattached flow regions under the effect of strong buoyancy forces and low flow velocities. Emphasis will be placed on the experimental study to examine the turbulent transport characteristics (including momentum and heat transfer) in natural convection and in buoyancy-dominated mixed convection adjacent to these geometries. The existing turbulence models, such as the k-E models for high and low Reynolds numbers, the algebraic stress model, and the RNG k-E model have been applied with success to forced convection, but they have not been fully tested and validated for buoyancy- dominated or buoyancy-affected complex flows involving flow separation and reattachment. This is because at the present time there is a lack of experimental measurements of buoyancy-dominated turbulent separated flows, even for such simple geometry as a backward-facing or forward-facing step. Such measurements are needed in order to test, validate, and/or improve the existing turbulence models for applications in strong turbulent buoyant flows involving flow separation. Simple geometries of backward-facing and forward-facing steps are chosen for the proposed research in order to establish a benchmark of experimental and numerical data for these geometries in buoyancy-dominated turbulent mixed convection flows, similar to what has been done for the same geometries in laminar mixed convection flows by the authors. Measurements of the mean and turbulent quantities of the momentum and thermal transport under buoyancy-dominated flow conditions for these benchmark geometries will be carried out and these measured data will be used in the numer ical phase of the proposed study, with the hope of validating or optimizing, and maybe, if needed, modifying the existing turbulence models. These measurements will also be of great value for future studies by others in modeling turbulent transport in buoyancy-dominated separated flows. Results from the proposed research will be correlated to provide the building blocks that are needed in dealing with turbulent transport phenomena in complex, buoyancy-dominated separated flows that are prevalent in many engineering devices.
