Journal of Emerging Trends in Engineering and Applied Sciences: Research Areas

Article Title: Prediction and Optimization of Design Parameters of Microelectronic Heat Sinks
by Mathias Ekpu, Raj Bhatti, Michael I. Okereke, Sabuj Mallik, Kenny C. Otiaba

Abstract:
This research presents an approach for predicting and optimizing the thermal resistance and mass of microelectronics heat sinks parameters. The Minitab software used the Taguchi design of experiments method to generate L27 Orthogonal array combinations for the heat sink geometry. The combinations consist of 6 factors (fin height, base thickness, fin thickness, length, number of fins, and width of heat sink) and 3 levels (low, medium, high). Ansys finite element software was used in the thermal analysis of the heat sink to determine the thermal resistance while the mass was calculated. The thermal resistance and mass responses were used to develop a multiple linear regression analysis model for predicting both thermal resistance and mass of heat sink. The developed regression model showed over 90% accuracy in predicting the thermal resistance and mass of heat sinks. In addition, the optimized heat sink geometry showed a 50.75% reduction in mass and 5.26% reduction in thermal resistance. This model could be used as an alternative to simulation software and could save cost and time in selecting an optimal heat sink for microelectronic applications.

Keywords: heat sink, multiple linear regression analysis, Taguchi design of experiment, mass, thermal resistance.

Download full paper