Session: 11-52-01: Computational Thermal/Fluids

Paper Number: 91488

91488 - The Optimal Geometric Arrangement of Heat Pipes in Computer CPU Coolers 

With the advent of powerful computations everywhere, computers nowadays need efficient cooling of their Central Processing Units (CPUs) to operate at a safe temperature. Among the wide variety of CPU coolers available in the market, air-cooled CPU coolers often come with several heat pipes of similar or varying sizes. Heat pipes with a much higher thermal conductivity than other solid materials have been effective heat transfer mediums in CPU coolers for many years. Heat transfer efficiency in a heat-pipe-based CPU cooler varies according to the selection of heat pipe diameter, number of heat pipes, and positioning of heat pipes. This study illustrates an optimal arrangement of heat pipes in a CPU cooler. The cooler model selected in this study comes with a baseplate, eight heat pipes, and many plate fins. One end of heat pipes attached to the base plate collects the heat from the processor and transfers the heat to the other end attached to many plate fins. Eventually, the plate fins transfer the heat to the surrounding air. We investigate a fin and adjacent air gap from the whole CPU cooler as the computational domain. A Grid study on the selected geometry establishes an optimum thermodynamic setting by obtaining the convergent solution while solving the continuity, momentum, and thermal energy equations. The nature of heat transfer in the computational domain is conjugate, while conduction occurs at the plate-fin and convection due to the air with Prandtl number 0.7 as the working fluid. Typical airspeeds for fans employed in the computer CPU cooler suggest the flow regime be laminar. The conjugate heat transfer problem presents two Nusselt number correlations for two heat pipes and brings out that the conjugate heat transfer coefficient remains unaffected by the temperature of the heat pipes. While the correlation itself indicates that the flow Reynolds number affects the correlation, the location of the heat pipe is another subtle and significant factor that is rarely reported numerically in the literature. Depending on the location of heat pipes for a certain Reynolds number, the conjugate Nusselt number ratio for two heat pipes could be as low as 0.65 to a ratio as high as 4.21 relative to the existing cooler model. A high Nusselt number ratio indicates an increase in heat transfer for a heat pipe location relative to another heat pipe location. Ultimately, this investigation predicts an optimal geometric arrangement for the location of the heat pipes. The optimal layout of heat pipes is essential to designing thermally efficient CPU coolers.

Presenting Author: Shuva Das Southern Illinois University Edwardsville

Presenting Author Biography: MSc ME student at the SIUE<br/>Teaching Assistant at the SIUE<br/>BSc ME from Chittagong University of Engineering and Technology

Authors:

Shuva Das Southern Illinois University Edwardsville
Majid Molki Southern Illinois University Edwardsville