Experiments of flow boiling heat transfer coefficient of FC-72 were carried out over simulated silicon chip of 10×10×0.5 mm3 for electronic cooling. Four kinds of micro-pin-fins with the dimensions of 30×60, 30×120, 50×60, 50×120 μm2 (thickness, t × height, h) respectively, were fabricated on the chip surfaces by the dry etching technique to enhance boiling heat transfer. A smooth chip was also tested for comparison. The experiments were conducted at three different fluid velocities (0.5, 1 and 2m/s) and three different liquid subcoolings (15, 25 and 35K). All micro-pin-finned surfaces show a considerable heat transfer enhancement compared to the smooth surface. Both the forced convection and nucleate boiling heat transfer contribute to the total heat transfer performance. The contribution of each factor to the total heat transfer has been clearly presented in the flow boiling heat transfer coefficient curves. In a lower heat flux region, the heat transfer coefficient increases greatly with increasing fluid velocity, but increases slightly with increasing heat flux, indicating that the single-phase forced convection dominates the heat transfer process. With further increasing heat flux to the onset of nucleate boiling, the heat transfer coefficient increases remarkably. For a given liquid subcooling, the curves of flow boiling heat transfer coefficient at fluid velocities of 0.5 and 1 m/s almost follow one line for each surface, showing insensitivity of nucleate boiling heat transfer to fluid velocity. However, at the largest fluid velocity of 2 m/s, the slope of the flow boiling heat transfer coefficient curves for micro-pin-finned surfaces becomes smaller, indicating that the forced convection also plays an important role besides the nucleate boiling heat transfer. The curves of the flow boiling heat transfer coefficient can be used to determine the boiling incipience at different fluid velocities, which provides a basis for the suitable fluid velocity selection in designing highly efficient cooling scheme for electronic devices.
Optimization of R245fa Flow Boiling Heat Transfer Prediction inside Horizontal Smooth Tubes Based on the GRNN Neural Network
