Kuo-Hsiang Chien

Two Si-based micro pulsating heat pipes (µPHPs) charged using HFE-7100 were either horizontally or vertically oriented and were tested using several heating powers. The width of each channel was 0.8 mm in one µPHP containing uniform channels, and the channel width was 1.0 mm or 0.6 mm in the other µPHP, which did not contain uniform channels. The depth of each channel was 0.25 mm. The overall size of each µPHP was 60 × 10 × 1.25 mm. Visual observation and temperature measurement of the µPHPs under various conditions were performed and the results were analyzed. The results indicated that when the µPHPs were operated horizontally at a heating power ranging from 1 to 7 W, the pulsating two-phase flow in the channels of the µPHPs could not begin, except when the µPHP containing nonuniform channels was tested at a heating power of 7 W. With a heating power less than 5 W, the frequency of the sine-like oscillating displacement of the vapor slug increased and the displacement of the vapor slug reduced in either vertically oriented μPHP, as the heating power increased With a heating power higher than 5 W, periodic “start-stop” behaviors were observed in the vertical μPHP containing nonuniform channels.

This study presents an approximation for determining an optimized thickness of a concentric heated rectangular plate and derives an analytical solution for spreading resistance of a spreader having orthotropic conductivities. The solution for the orthotropic plate is obtained by separation of variables, and the optimized thickness is determined by taking the derivative of the thermal resistance with respect to the spreader thickness. According to the calculated results, an enhanced in-plane spreading effect can reduce the spreading resistance. The spreading resistance dominates the overall resistance of thin plates, whereas the one-dimensional conduction resistance becomes important for thick plates. However, the predicted optimized thickness from the approximation shows a disparity from the analytical results, while the aspect ratio between a spreader and heat source is less than 0.2. Even so, the thermal resistance corresponding to the predicted thickness is still in good agreement with the analytical solution. The proposed approximation will be useful for practical thermal design of heat sinks by predetermining the spreader thickness.

This study examines the spreading ability of rectangular plates numerically, analytically, and experimentally. The effect of aspect ratio, defined as an equivalent radius of a heater divided by that of a spreader plate, is investigated. The numerical results show a very good agreement with the analytical solutions. From the calculated results, the spreading resistance of the conduction plates with a small aspect ratio is higher than the one-dimensional conduction resistance. Calculated results also show that the spreading ability of a metal plate would be affected slightly by the external convective heat-transfer coefficient when the ratio of the longitudinal heat convection to the lateral heat spreading is less than 0.1. In addition to the numerical analysis, experimental comparisons between copper∕aluminum plates and a vapor chamber having the same thickness have been conducted. The experimental results show that the thermal resistance of the metal plates is independent of input power whereas that of the vapor chamber shows a noticeable drop with increased power. For the influence of concentrated heat source, the surface temperature distributions for the metal plates become concentrated with a reduced aspect ratio. However, the variations of the aspect ratio and the input power would yield minor effects to the surface temperature distribution of the vapor chamber. As compared with the conduction plates, the vapor chamber would offer a lower temperature rise and a more uniform temperature distribution. Thus, the vapor chamber provides a better choice as a heat spreader for concentrated heat sources.