Power-Loss Estimation and Improved Ripple Based Adaptive On-Time Control for DC-DC Buck Converters

Autor: Chia-Hsing Li, 李佳興
Rok vydání: 2014
Druh dokumentu: 學位論文 ; thesis
Popis: 102
This dissertation presents an accurate power-loss estimation method and a ripple based adaptive on-time (RBAOT) control with adjustable virtual-ripple and offset cancellation for buck converters. First, the accurate power-loss estimation method is introduced and analyzed in detail. The switching loss is mainly caused by parasitic capacitance of the metal oxide semiconductor field-effect transistor (MOSFET), and the switching loss on power MOSFETs has been thoroughly researched and analyzed. Besides, considering the thermal characteristics of MOSFET resistance, the dissertation presents iterative method to predict the junction temperature to get the new integration resistance of MOSFET. Recently buck converter control topology uses ripple based adaptive on-time (RBAOT) with virtual-ripple which is a conceptually simple topology and features fast dynamic transient response at lower duty ratio application. This control scheme solves the sub-harmonic oscillation instability when using ceramic capacitors in conventional control scheme. However, this control scheme has some drawbacks, such as poor dynamic transient response at a specified duty ratio range and inherent output voltage offset problem. This dissertation presents a new method which called RBAOT control with adjustable virtual-ripple to improve dynamic transient response, and a new method for output voltage offset cancellation. Theoretical analysis and experimental results are shown and compared to validate the feasibility and reliability of proposed power-loss estimation method. Simulation and experimental comparison results show that the proposed concept of ripple based adaptive constant on-time control with adjustable virtual-ripple and offset cancellation achieved fast dynamic transient responses without output voltage offset.
Databáze: Networked Digital Library of Theses & Dissertations