ISSCC 2016 / SESSION 8 / LOW-POWER DIGITAL CIRCUITS / 8 . 5 8 . 5 A 60 %-Efficiency 20 nW-500 μ W Tri-Output Fully Integrated Power Management Unit with Environmental Adaptation and Load-Proportional Biasing for IoT Systems

Wanyeong Jung,Junhua Gu, Paul D. Myers, Minseob Shim, Seokhyeon Jeong,Kaiyuan Yang, Myungjoon Choi, ZhiYoong Foo, Suyoung Bang,Sechang Oh,Dennis Sylvester,David Blaauw

semanticscholar(2018)

引用 0|浏览3
暂无评分
摘要
As Internet-of-Things (IoT) systems proliferate, there is a greater demand for small and efficient power management units. Fully integrated switched-capacitor (SC) DC-DC converters are promising candidates due to their small form factor and low quiescent power, aided by dynamic activity scaling [1-3]. However, they offer a limited number of conversion ratios, making them challenging to use in actual systems since they often require multiple output voltages (to reduce power consumption) and use various input power sources (to maximize flexibility). In addition, maintaining both high efficiency and fast load response is difficult at low output current levels, which is critical for IoT devices as they often target low standby power to preserve battery charge. This paper presents a fully integrated power management system that converts an input voltage within a 0.9-to-4V range to 3 fixed output voltages: 0.6V, 1.2V, and 3.3V. A 7-stage binaryreconfigurable DC-DC converter [1-2] enables the wide input voltage range. Three-way dynamic frequency control maintains converter operation at nearoptimum conversion efficiency under widely varying load conditions from 5nW to 500μW. A proposed load-proportional bias scheme helps maintain high efficiency at low output power, fast response time at high output power and retains stability across the entire operating range. Analog drop detectors improve load response time even at low output power, allowing the converter to avoid the need for external sleep/wakeup control signals. Within a range of 1-to-4V input voltage and 20nW-500μW output power, the converter shows >60% conversion efficiency, while maintaining responsiveness to a 100× sudden current increase.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要