Hotspot thermal management in microchannel heat sinks with vortex generators

2020 
Abstract Vortex generators (VGs) have been widely applied in the microchannel heat sinks to improve the thermal-hydraulic performance in multicore processors. However, most studies focus on global heat transfer enhancement of microchannel, while the hotspot temperature is often neglected. In this paper, a numerical study of hotspot-targeted thermal management using VGs is performed in a rectangular microchannel with a heat flux of 400 W/cm2 and 50 W/cm2 respectively at the hotspot and background region. With the maximum thermal resistance, pumping power and the overall performance factor as the key metrics, the performances of VGs in four configurations (co-flow-up, co-flow-down, counter-flow-up and counter-flow-down) are compared with a smooth microchannel without VG. Deionized water with temperature-dependent thermo-physical properties is used as the coolant with the Re in the range of 66–330. The results indicate that the adoption of VGs can significantly improve the cooling effect over the hotspot with a lower pressure loss penalty comparing with the smooth microchannel. In addition, the VGs in a “co-flow” configuration lead to the fluid circulation effect and transport the coolant in the core region of the microchannel to the sidewalls. It is also noted that comparing with the smooth microchannel, more than 78.8% pumping power can be saved by using the “co-flow-down” configuration to achieve the same cooling effect to maintain the average temperature at 304.7 K over the hotspot. Geometric parameter analysis reveals that the increase in the height and attack angle of VGs results in more pressure loss, whereas the enhanced heat transfer is insignificant when the VG's height ratio exceeds 0.2 or the attack angle surpasses 45°. Taking full account of the heat transfer performance and pressure loss, VGs with a height ratio of 0.2 and attack angle of 45° are recommended for practical applications.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    28
    References
    2
    Citations
    NaN
    KQI
    []