散热片
电子设备冷却
热阻
材料科学
微通道
热的
热管
传热
机械工程
机械
热力学
物理
工程类
纳米技术
作者
khavar fazeli,Kambiz Vafai
标识
DOI:10.1016/j.ijheatmasstransfer.2023.124842
摘要
In this work, innovative heat sink designs consisting of Microchannel Heat Sinks (MCHS) and Flat Plate Micro Heat Pipes (FPM-HP) are investigated to improve thermal management in electronic components. The three-dimensional finite volume method is used to analyze the fluid flow and heat transfer in the heat sinks. The validity of the numerical models is confirmed by comparison with pertinent experimental data. The thermal performances of the new heat sink designs are compared to that of double-layer microchannel heat sinks with parallel flow (DLPF MCHS) and counter flow (DLCF MCHS) configurations, as two of the more efficient heat sink designs in the industry. The investigated configuration of Silicon-Water Counter Dual MCHS with rectangular cross-section improves the temperature uniformity by 43–50 % and 26–31 % compared to DLPF and DLCF MCHSs, respectively, and the triangular cross-section configuration improves it by 38–47 % and 21–26 %, respectively. The thermal resistance of heat sinks is reduced by 10–15 % compared to DLPF MCHS and by 2–8 % compared to DLCF MCHS. Additionally, innovative hybrid heat sink designs combined with MCHS and FPM-HP are proposed to optimize hotspot management in electronics cooling applications that involve local high heat generation. The hybrid DLPF MCHS & FPM-HP reduces the maximum temperature and surface temperature gradient of the heat sink by 13–23 % and 20–49 %, while these are reduced by 16–40 % and 27–64 % for the hybrid DLCF MCHS & FPM-HP compared with the corresponding copper base (no heat pipe) designs. The hybrid Counter Dual Rectangular MCHS & FPM-HP reduces the maximum temperature and surface temperature gradient of the heat sink by 15–24 % and 29–65 % compared with the DLCF MCHS with copper base (no heat pipe) design, while these are reduced by 10–26 % and 24–63 % for the Counter Dual Triangular MCHS, respectively.
科研通智能强力驱动
Strongly Powered by AbleSci AI