University of Wollongong
Browse

Design and analysis of the IGBT heat dissipation structure based on computational continuum mechanics

Download (1.26 MB)
journal contribution
posted on 2024-11-15, 22:42 authored by Xin Lin, Huawei Wu, Zhen Liu, Baosheng Ying, Congjin Ye, Yuanjin Zhang, Zhixiong Li
© 2020 by the authors. With the trend of high integration and high power of insulated gate bipolar transistor (IGBT) components, strict requirements have been placed on the heat dissipation capabilities of the IGBT devices. On the basis of traditional rectangular fins, this paper developed two new types of heat-dissipating fins to meet the high requirements of heat dissipation for the IGBT devices. One is the rectangular radiator with a groove length of 2.5 mm and a width of 0.85 mm, the other is the arc radiator with the angle of 125 arc angle, 0.8 mm arc height, and 1.4 mm circle radius. After theoretically calculating the IGBT junction temperature, numerical simulations have been implemented to verify the theoretical result. The commercial CFD software, STAR-CCM+, was employed to simulate the heat dissipation characteristics of the IGBT module under different wind speeds, power, and fin structures. By analyzing the temperature field and vector field of the IGBT module, the analysis results demonstrate that the error between the simulation result and the theoretical calculation is within 5%, which proves the feasibility of the newly designed heat-dissipating fins. When the wind speed is 12.5 m/s, the power is 110 W, the fin height is 31.2 mm, and the fin thickness is 2.3 mm, the rectangular radiator can achieve the best heat dissipation performance.

History

Citation

Lin, X., Wu, H., Liu, Z., Ying, B., Ye, C., Zhang, Y. & Li, Z. (2020). Design and analysis of the IGBT heat dissipation structure based on computational continuum mechanics. Entropy, 22 (8), 816-1-816-12.

Language

English

RIS ID

145043

Usage metrics

    Categories

    Keywords

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC