Effect of Local Hot Spots on the Maximum Dissipation Rates During Flow Boiling in a Microchannel

Abstract

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> One of the most promising technologies to replace air-cooling of micro-processor chips is flow boiling in microchannels. The very high heat flux dissipation from micro-processor chips is highly non-uniform due to the presence of multiple localized hot spots usually related to the localization of bridges and gate oxide shorts. Previous studies focused on the performance of microchannels under uniform heating conditions. Recently, Revellin and Thome <citerefgrp> <citeref refid="ref1"/></citerefgrp> have proposed a new theoretical model to predict the critical heat flux (CHF) in microchannels. This model has been modified here to take into account a non-uniform axial heat flux along a microchannel. The model is used here to perform a local hot spot study to investigate the effects of fluid, saturation temperature, mass flux, microchannel diameter, heated length, size, location and number of hot spots as well as the distance between two consecutive hot spots. Based on the present simulations, to best dissipate a hot spot's heat flux, microchannel heat sinks should follow the following recommendations for a channel of fixed length: determine the optimum channel diameter for the fluid (typically either very small or large is best), utilize as high of mass flux as feasible, and design with as low of saturation temperature as possible. Furthermore, the local hot spot size should be as small as possible, the number of local hot spots as few as possible and the distance between two hot spots as large as possible. Utilizing the present numerical method for individual microchannels arranged in parallel in a multi-microchannel cooling element, it is possible to simulate the entire power dissipation profile of a microprocessor die for local limits of CHF. </para>