Steady Reynolds-averaged Navier--Stokes (RANS) simulations were performed to examine the ability of four turbulence models—realizable k–ε (k–ε), shear-stress transport (SST), Reynolds stress model with linear pressure strain (RSM-LPS), and stress-omega RSM (RSM-τω)—to predict the turbulent flow and heat transfer in a trapezoidal U-duct with and without a staggered array of pin fins. Results generated for the heat-transfer coefficient (HTC) were compared with experimental measurements. For the smooth U-duct, the maximum relative error in the averaged HTC in the up-leg is 2.5% for k–ε, SST, and RSM-τω and 9% for RSM-LPS. In the turn region, the maximum is 50% for k–ε and RSM-LPS, 14.5% for RSM-τω, and 29% for SST. In the down-leg, SST gave the best predictions and RSM-τω being a close second with maximum relative error less than 10%. The ability to predict the separated flow downstream of the turn dominated the performance of the models. For the U-duct with pin fins, SST and RSM-τω predicted the best, and k–ε predicted the least accurate HTCs. For k–ε, the maximum relative error is about 25%, whereas it is 15% for the SST and RSM-τω, and they occur in the turn. In the turn region, the staggered array of pin fins was found to behave like guide vanes in turning the flow. The pin fins also reduced the size of the separated region just after the turn.
Read full abstract