Analysis of disagreement between numerically predicted and experimental heat transfer data of impinging jet

Abstract

The method of numerical simulation was applied to investigate the effects of jet impinging plate thickness and its thermal conductivity on the local heat flux distribution along the impinging plate. The results show that the two factors have great effects on the heat flux distribution. The non-uniformity of the local heat-flux on the impinging plate surface gets more profound as the plate becomes thicker and thermal conductivity gets larger. When Reynolds number is 5 000, the ratio of nozzle-to-plate spacing to nozzle diameter is 5 and thermal conductivity is 16 W/(m·K), and even for the plate with only 25 µm in thickness, the non-uniformity of the heat flux cannot be neglected. When the plate thickness is 50 µm, only when thermal conductivity is as small as 1 W/(m·K), the heat flux curve can be approximately treated as an iso-heat-flux boundary. In the experimental research, a real non-iso-heat-flux boundary is treated as an iso-heat-flux boundary, which would result in under-estimated Nusselt number value in the stagnation zone and an over-estimated value outside. Such an experimental Nusselt number distribution is taken to evaluate turbulent model, and the conclusion would be drawn that the turbulent model over-predicts the stagnation heat transfer. This is one of the important reasons why many literatures reported that k-e turbulent model dramatically over-predicts the impinging jet heat transfer in the stagnation region.