Experimental investigations on the thermal superposition effect of multiple hotspots for embedded microfluidic cooling

With the development of high-performance multi-core microprocessors, the increase in hotspots density poses a severe challenge to chip thermal management. Embedded cooling can improve the cooling performance at the chip level. However, the cooling efficiency will be deteriorated by the interaction of hotspots. This paper proposes the concept of thermal superposition effect based on the relationship between the cooling efficiency of two hotspots and their relative positions. The effective cooling area and superposition area describe the thermal characteristics and interaction of hotspots, and the superposition factor beta evaluates the intensity of interaction. By optimizing the superposition factor, the average temperature reduces 22%. Moreover, the factor can predict the average temperature of hotspots within the error under14%. Therefore, the thermal superposition effect and factor beta is an effective analysis tool in multi-hotspot cooling, simplifying the analysis process and making it possible to predict and optimize the interaction by experiment and finite element simulation.