摘要
随着芯片功率密度的增大,对多芯片组件(MCM)可靠性设计提出了更高的要求,热阻是MCM可靠性设计的重要参数之一,影响着芯片的结温和运行条件。以MCM发光二极管(LED)为例,对热阻拓扑网络模型和热阻矩阵模型的芯片结温进行了理论分析,研究了在不同芯片功耗、基板导热系数和芯片间距下两种模型与ANSYS仿真结果的误差,并分析了热阻拓扑网络模型误差大的原因。根据芯片热耦合效应和扩散热阻原理,提出了一种耦合扩散热阻的拓扑网络模型,并得出该新型热阻拓扑网络模型的结温表达式与热阻矩阵模型一致,推导出耦合扩散热阻即为芯片与基板热流区域之间的热阻,揭示了二维MCM的热耦合规律,为二维MCM的可靠性设计提供了理论依据。
With the increase of the chip power density,higher requirements are put forward for the reliability design of the multi-chip module(MCM).The thermal resistance is one of the important para-meters of the reliability design of the MCM,which affects the junction temperature and operation condition of the chip.Taking the MCM-LED for example,the chip junction temperature of the thermal resistance topological network model and the thermal resistance matrix model were theoretically analyzed.The errors of the above two models and ANSYS simulation results were analyzed under different chip power consumptions,thermal conductivities of substrate and chip distances.The reasons for the large error of the thermal resistance topological network model were analyzed.A topological network model of the coupling thermal spreading resistance was proposed based on the thermal coupling effect and thermal spreading resistance,which had the same theoretical expression of the junction temperature as the thermal resistance matrix model.It is deduced that the thermal resistance between the chip and the heat flow region of the substrate is the coupling thermal spreading resistance,and the thermal coupling rules of 2D MCM is revealed,which provides theoretical support for the reliability design of 2D MCM.
作者
汪永超
魏昕
章国豪
胡正发
赵忠伟
Wang Yongchao;Wei Xin;Zhang Guohao;Hu Zhengfa;Zhao Zhongwei(School of Electromechanical Engineering,Guangdong University of Technology,Guangzhou 510006,China;Synergy Innovation Institute for Modern Industries of GDUT&Dongyuan,Heyuan 517500,China;Synergy Innovation Institute of GDUT&Heyuan,Heyuan 517025,China)
出处
《半导体技术》
CAS
北大核心
2020年第10期801-808,共8页
Semiconductor Technology
基金
国家自然科学基金资助项目(61574049)
广东省科技计划项目(2018B090902003)
广东省“珠江人才计划”本土创新科研团队项目(2017BT01X168)。