Orthogonal study and analysis of variance on a thermal management system for high-power LED package

Determination of influences of various factors is important for improvement and optimization of a thermal management system for microelectronics. In this study, a series of orthogonal experiments and analysis of variance were performed on an LED cooling system which integrates thermoelectric cooler with microchannel heat sink using nanofluid as coolant. The significance of four factors, including nanofluid concentration, ambient temperature, nanofluid temperature, and power of the thermoelectric cooler, were evaluated. Substrate temperature was measured and temperature distribution of the LED was estimated. Results indicate that nanofluid demonstrated better cooling capacity compared with water. Results of difference analysis and analysis of variance indicate that all the four factors showed significant effects on the LED temperature, while the dominant factor was nanofluid concentration. The optimum nanofluid concentration was obtained. The variation of the nanofluid effect on the cooling performance with ambient temperature and nanofluid temperature were observed. The influence of dispersion stability of the nanofluid on the cooling performance was discussed.