Book Description

Power conversion is a necessity in almost all modern electric systems and machines: energy must be regulated and delivered in the intended manner if a system is to perform well, or at all. Power converters, the electronic circuits used to control this energy flow, have been a subject of intense study and rapid development in recent years and are widely acknowledged to be a fundamental enabler for modern day human societal capabilities. Many market sectors have strongly advocated for further development of energy conversion systems with improved efficiency and power density as these traits often directly dictate practical viability. While advancements in semiconductor device physics have yielded improved parts for use inconverter solutions, it is becoming apparent that there is additional massive potential and merit in revisiting fundamental converter topologies and circuit techniques. To date, power converters that use capacitors as their primary energy transfer elements (termed "switchedcapacitor" power converters) are far less ubiquitous than their switched-inductor counterparts, and seemingly for good reason: characteristics such as poor output regulation and intrinsic transient inrush currents that lead to inefficiency have largely prevented switched-capacitor topologies from gaining practical consideration in general power converter markets. Solutions to these negative attributes are strongly desired as capacitors can offer energy densities up to three orders of magnitude greater than inductors, with these energy transfer elements typically consuming the majority of a power converter's weight/volume. Recent work has demonstrated significant potential for hybrid switched-capacitor-inductor converter techniques: here, small inductive element(s) are used to eliminate the conventional drawbacks of a converter which is predominantly capacitor based. The hybridized approach helps unlock the full potential of capacitor-based converters and has been demonstrated to offer compelling results at the cost of added complexity. This work offers an exploration into a collection of state-of-the-art power converter techniques and topological methods, primarily within the field of hybridized switched-capacitor-inductor converters. The first two chapters give a background on fundamental considerations such as conventional loss mechanisms and the slow-switching-limit (SSL), as well as several established loss mitigation techniques. An integrated converter system and its associated functional blocks is then discussed in Chapters 3 and 4, exemplifying a hybridized two-stage converter and illustrating the implementation of several loss mitigation methods and practical circuit techniques. Next, several hybridized variations of the Dickson topology are discussed: this family of DC-DC converters is well suited for non-isolated large voltage conversion ratios. A number of these variants are proposed here for the first time, illustrating significant potential for further converter development. The steady-state bias points, resonant switching frequency, duty cycle and voltage ripple as a function of load are calculated for several example converters, including the non-trivial case of a converter undergoing split-phase operation and whose operating points exhibit a strong load dependence. To facilitate comparative analysis between topologies, a mathematical method is presented that characterizes the total energy density utilization of fly capacitors throughout a converter, accounting for large voltage ripple and iii highly nonlinear reverse-bias transitions. This analysis assists with optimal topology selection as energy density utilization directly dictates the required capacitor volume at a specified power level and switching frequency. An expanded family of fly capacitor networks is then introduced in Chapter 6; here it is shown that there are a large number of unexplored yet practical fly capacitor configurations that are eligible for use in hybridized converters. It is calculated that a 6-7 % reduction in capacitor volume can be achieved relative to conventional Dickson fly capacitor networks, while preserving the desirable characteristic of equal voltage ripple on its branches. N-phase and split-phase switching methods and their respective trade-offs are then discussed in detail, offering control techniques that allow a departure from conventional two-phase operation while retaining high-efficiency zero-voltage and zero-current switching (ZVS/ZCS) conditions. A Cockcroft-Walton prototype demonstrates both methods implemented on the same piece of hardware, significantly improving the efficiency range with respect to load and resulting in a state-of-the-art power density of 483.3 kW/liter (7, 920W/inch3). Next, a method termed "resonant charge redistribution" (RCR) is proposed that greatly reduces output capacitance (C[subscript OSS]) related switching losses in all switches of a complex switched-capacitor network. Despite little effort being put towards optimization, a prototype using RCR measures a 61 % reduction in total losses at light load for a near negligible 0.74 % increase in total solution volume. Lastly, resonant gate drive techniques are discussed. Here, within a proposed resonant gate-driver topology, a capacitive decoupling technique is demonstrated that allows power to be delivered to a "flying" high-side N-channel device which commutes between two variable voltages. The implemented prototype achieves up to a 72 % reduction in gating loss when switching over 20 MHz and with rise/fall times ≤ 7 ns. Combining several of the novel techniques described herein can result in near complete mitigation of all primary switching loss mechanisms observed throughout the complex structure of a switched-capacitor converter network. This relatively new field of hybridized converter design has already yielded converters with record-breaking performance, as is demonstrated here. With contemporary techniques, including those described in this work, the field of power electronics is on the cusp of seeing widespread dramatic improvements in energy handling capability, power density, specific power and efficiency at reduced cost, with huge potential for growth and improved energy consumption in both developed and emerging markets.