Search

In this paper, different heat sinks are investigated to improve CPU cooling experimentally. Different types of copper heat sinks including a novel "porous heat sink" are proposed and compared to obtain the optimum heat transfer rate. Water and Al2O3/water nanofluids (40 nm) in two different volume fractions (0.1 and 0.2 vol.% from 40-nm nanoparticles) are used as heat transfer fluid to cool the CPU. Different flow rates are also examined in the experimental study. The Nusselt number, wall temperature, thermal resistance and LMTD temperature have been analyzed for a wide range of parameters of the heat sinks. Also, the effect of geometry, heat transfer fluid, input power and flow rate on the heat transfer rate is reported. The result shows that the geometry of the heat sink has an important effect on the thermal performance of the system. The rate of heat transfer in the proposed novel porous heat sink is larger than the other cases. The Nusselt number in the porous heat sink by using nanofluids is 2.2 times larger than the heat sink with an inline pattern. [ABSTRACT FROM AUTHOR]

Abstract: Magnetic convection heat transfer in a two-dimensional square cavity induced by magnetic field gradient is investigated numerically using a semi-implicit finite volume method. The side walls of the cavity are heated with different temperatures, the top and bottom walls are isolated, and a permanent magnet is located near the bottom wall. Thermal buoyancy-induced flow is neglected due to the nongravity condition on the plane of the cavity. Conditions for the different values of non-dimensional variables in a variety of ferrofluid properties and magnetic field parameters are studied. Based on this numerical analysis, a general correlation for the overall Nusselt number on the side walls is introduced for a wide range of effective parameters. Results showed that maximum error produced by use of this correlation is about 6 percent. [ABSTRACT FROM AUTHOR]