期刊
IEEE TRANSACTIONS ON POWER ELECTRONICS
卷 34, 期 3, 页码 2405-2415出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TPEL.2018.2842029
关键词
High frequency; LLC resonant converter; magnetic shunt; planar transformer; wide input voltage
Achieving high efficiency and high power density is emerging as a goal in many power electronics applications. LLC resonant converter has been proved as an excellent candidate to achieve this goal. To achieve smaller size of passive components, the resonant inductor in the LLC converter is usually integrated into the transformer by utilizing its leakage inductance. However, the leakage inductance of the transformer is usually insufficient and thus the LLC converter has to be operated in a limited frequency range (this limits the input voltage range accordingly), otherwise the power efficiency will drop dramatically. Therefore, a larger resonant inductance in the LLC converter is expected to operate in a wider input voltage range. This paper proposes a new method to create a larger resonant inductance by using a magnetic shunt integrated into planar windings. The accurate leakage inductance modeling, calculation, and optimal design guideline for LLC planar transformer, including optimal magnetic shunt selection and winding layout, are presented. A 280-380 V input and output 48 V-100 W half-bridge LLC resonant converter with 1 MHz resonant frequency is built to verify the design methodology. A comparison is made between two converters with the same parameters, one using magnetic shunt integrated transformer and the others using traditional planar transformer and external inductor. Experimental result shows the proposed converter with magnetic shunt is capable to achieve comparable high efficiency and regulation capability with the other under a wide input voltage, which verifies the optimal design methodology. Above all, this magnetics integration methodology reduces the whole converter's volume and thus increases the power density.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据