Methods for improving the efficiency of high-performance audio amplifiers
Bakker, Robert
Bakker, Robert
Loading...
Repository DOI
Publication Date
2022-05-23
Type
Thesis
Downloads
Citation
Abstract
High-performance class-D audio amplifiers have seen application in many modern consumer electronic devices including smartphones, wireless speakers and headphones, laptops, TVs, etc. While the efficiency and size of modern class-D amplifiers are far superior to their analogue counterparts, the ever-increasing demands for smaller devices and lower power consumption motivate the drive to optimize the amplifier designs further. Furthermore, statistical analysis of real-world audio signals such as music shows that the actual time-averaged efficiency of a class-D amplifier is often many times lower than its advertised peak efficiency. As such, the work in this thesis aims to both miniaturize as well as improve the efficiency of high-performance class-D amplifiers. Firstly, the class-D amplifier design itself is examined. The results of this analysis show that the output low-pass filter inductor is responsible for a significant proportion of the overall power loss, as well as often being the largest component of the amplifier. As such, a systematic approach to designing and optimising the inductor design is presented to both reduce the overall power loss, as well as minimize its footprint. Secondly, it was identified during the analysis of the inductor performance that the class-D amplifier can benefit significantly from the use of envelope tracking (ET) to reduce the overall power losses. As such, an investigation into the benefits of ET on the class-D amplifier, as well as its associated power supply is discussed. Two ET demonstrator systems are designed and analysed, including a battery-operated step-up converter, and a mains-operated step-down converter, respectively. Measurement results from both indicate a significant increase in overall time-averaged efficiency of up to 48% in certain use cases, particularly when used with high crest factor signals such as music.
Funder
Publisher
NUI Galway