High-performance energy-saving miniature loop heat pipe for cooling compact power semiconductors

High-performance energy-saving miniature loop heat pipe for cooling compact power semiconductors
复制标题

DOI:
10.1016/j.enconman.2021.114081
复制
发表时间:
2021-05
影响因子:
10.4
通讯作者:
N. Watanabe;T. Mizutani;H. Nagano
N. Watanabe;T. Mizutani;H. Nagano
中科院分区:
工程技术1区
文献类型:
--
作者:
N. Watanabe;T. Mizutani;H. Nagano

文献摘要

被引文献

相似文献

首次开发了一种用于冷却高热流密度(≥30 W/cm 2)和高允许工作温度(≤250 °C)的紧凑型宽带隙功率半导体的微型回路热管(mLHP)。由于mLHP由蒸发器(15 × 16 × 3 mm)、冷凝器(12 × 12.5 × 2 mm)和传输管(100 mm)组成,因此可以安装在相当于游戏智能手机大小的小型产品上。为了实现高的热通量性能,mLHP使用Shirasu多孔玻璃作为芯,使用乙醇作为工作流体。此外,对于冷凝器,使用薄铜板(60 × 60 × 0.5 mm)、薄铝板(35 × 20 × 3 mm)或小型冷却风扇(17 × 17 × 3 mm)来增强散热。当冷凝器使用薄铝冷板散热时,mLHP在水平姿态下实现了15 W(60 W/cm 2)的最大传热性能。在最大热负荷下,15 W时的热源温度、蒸发器温度和系统热阻分别为249.8 °C、142.2 °C和0.72 °C/W。此外,蒸发器处的蒸发传热系数约为24,000 W/m2/K。除冷凝器部分外,所建立的稳态数值模型与实验结果基本一致,相对误差为7%。
A miniature loop heat pipe (mLHP) is developed to cool compact wide bandgap power semiconductors with a high heat flux (≥30 W/cm2) and high allowable operating temperature (≤250 °C) for the first time. Since the mLHP consists of an evaporator (15 × 16 × 3 mm), a condenser (12 × 12.5 × 2 mm), and transport tubes (100 mm), it can be mounted on a small product equivalent to the size of a gaming smartphone. To achieve high heat flux performance, the mLHP uses Shirasu porous glass for the wick and ethanol for the working fluid. Moreover, for the condenser, heat dissipation is enhanced using a thin copper plate (60 × 60 × 0.5 mm), a thin aluminum cold plate (35 × 20 × 3 mm), or a small cooling fan (17 × 17 × 3 mm). When the condenser dissipates heat using the thin aluminum cold plate, the mLHP achieves a maximum heat-transport performance of 15 W (60 W/cm2) in the horizontal attitude. At the maximum heat load, the heat-source temperature, evaporator temperature, and system thermal resistance at 15 W are 249.8 °C, 142.2 °C, and 0.72 °C/W, respectively. Additionally, the heat transfer coefficient of evaporation at the evaporator is approximately 24,000 W/m2/K. The proposed steady-state numerical model is consistent with the experimental results, excluding the condenser part, with a relative error of 7%.