# Publications

Listed below are the papers published by Dr. Ray Ridley in the field of power electronics. After the title and reference for the paper is the abstract, taken directly from the official proceedings. Following this, *in italics*, is a commentary on the paper and its contributions and important results.

The papers are in approximate order of their relative importance to power systems designers.

Ridley, R. B., *A New Continuous-Time Model for Current-Mode Control *

IEEE Transactions on Power Electronics, April, 1991, pp. 271-280.

(Special issue on Modeling for Power Electronic Circuits and Systems.)

The accuracy of sampled-data modeling is combined with the simplicity of pole-zero representation to give a new current-mode control model, accurate to half the switching frequency. All of the small signal characteristics of current-mode control are predicted, including high-frequency subharmonic oscillation which can occur even at duty cycles of less than 0.5. The best representation for the control-to-output transfer function is shown to be third-order. Model predictions are confirmed with measurements on a buck converter.

*The first model which simultaneously allows prediction of the current-loop instability, transition from voltage-mode to current-mode control, and a simple pole-zero representation. The paper shows why you have to use a three-pole representation of a current-mode system to get results which closely model those seen in the lab. *

Ridley, R. B., ** A New Continuous-Time Model for Current-Mode Control with Constant On-Time, Constant Off-Time, and Discontinuous Conduction Mode**; IEEE Power Electronics Specialists Conference Record, San Antonio, Texas, June 1990, pp. 382-389.

A new small-signal model for current-mode control of PWM converters is extended for constant on-time control, constant off-time control, and discontinuous conduction mode. Constant on- or off-time converters in CCM have two complex right-half-plane zeros in the current loop feedback, but reduced modulator gain eliminates current-loop instability. The modulator gain exhibits frequency-dependent phase which introduces an extra gain term in the circuit model. The new model for DCM operation does not have RHP zeros in the current feedback loop, but the buck converter exhibits an instability for higher conversion ratios.

*This paper completes the set of models for current-mode control. Some interesting results were obtained for modulator gain with the variable frequency control schemes. The buck converter instability is a little-known phenomenon that can cause problems for modern power supplies converting 5 V down to 3.3 V, for example.*

Ridley, R. B., ** Secondary LC-Filter Analysis and Design Techniques for Current-Mode Controlled Converters**; IEEE Transactions on Power Electronics, Vol. 3, No. 4, October 1988, 499-507.

Small-signal characteristics of current-mode-controlled PWM converters with a second-stage LC filter are analyzed. A secondary filter can be designed to provide good attenuation of the switching ripple while maintaining adequate stability margins with capacitive loading. Design guidelines for the filter are given.

*This paper shows you how to properly design and model a second-stage filter. This is a very good and practical guide on the right way to reduce the noise on the output of your power supply without introducing instability. The results are contrary to the design approach taken by many engineers.*

Ridley, R. B., S. Kern, B. Fuld, ** Analysis and Design of a Wide Input Range Power Factor Correction Circuit for Three-Phase Applications**, IEEE Applied Power Electronics Conference Proceedings, San Diego, 1993, pp. 299-305.

A combined buck and boost topology power factor correction circuit which can operate with input voltages from 150 - 540 VAC is presented. The design and analysis of the operation and dominate losses of this circuit are given. The features of the topology are compared with those of a standard boost power factor correction circuit.

*This topology is at the heart of one of the most advanced and reliable power systems in use today in the mainframe computer industry. The power cord for the system can literally be plugged into either a low-line 208 VAC or a high-line 480 VAC feed, either single phase or three phase, with no tap switches to be made. The system can also survive an outage on one of three input lines for extended periods of time, without any of the power stages needing to be oversized. A very important technology for the 2-20 kW power range in systems that need an isolated output bus. *

Ridley, R. B.,** New Simulation Techniques for PWM Converters**, IEEE Applied Power Electronics Conference Proceedings, San Diego, 1993, pp. 517-523.

New simulation techniques are presented which take advantage of the special structure of PWM converter power stages and their compensation circuits. These techniques provide reduction of system order, and allow for the fastest possible simulation without any iteration algorithms. This is achieved while retaining accurate large-signal simulation with details of cycle-by-cycle waveforms.

*This paper gives some of the theory behind the special and unique simulator used in POWER 4-5-6, demonstrating why this is the fastest method of simulation that you can use and still get very accurate results.*

Vlatkovic, Vlatko, Juan A. Sabate, Raymond B. Ridley, and Fred C. Lee, ** Small Signal Analysis of the Phase-Shifted PWM Converter**, IEEE Transactions on Power Electronics, Vol. 7, No. 1, January, 1992, pp. 128-135.

The specific circuit effects in the phase-shifted pulse-width-modulated (PS-PWM) converter and their impact on the converter dynamics are analyzed. The small-signal model is derived incorporating the effects of phase-shift control and the utilization of transformer leakage inductance and power FET junction capacitances to achieve zero-voltage resonant switching. The paper explains the differences in the dynamic characteristics for the PS-PWM converter and its PWM counterpart. Model predictions are confirmed by experimental measurements.

*The phase-shifted full bridge is a circuit which has substantial performance advantages over its PWM counterparts. If you are going to use a full bridge, you should look very hard at implementing the phase-shifted version, especially since Unitrode now has a controller for driving this topology. This converter also forms a part of the system mentioned above.*

Ridley, R., M. Reynell, and S. Kern, ** Thermal Considerations for Distributed Power Converter Systems**, IEEE Applied Power Electronics Conference Proceedings, San Diego, 1993, pp. 615, 866-872.

Modern high-density DC-DC power supplies present a significant challenge to the system designer in cooling the package. In many applications, cooling air is restricted and large heatsinks are needed to prevent excessive temperature rise. This paper presents thermal design examples for typical dc-dc systems, and uses a three-dimensional thermal modeling tool (FLOTHERM) to predict temperature rise.

*This paper is high in the list to stress the importance of proper thermal design. Even in a simple system with just one heat-dissipating element, surprising results can be obtained, and some basic thermal modeling can predict these results before hardware is built.*

Hsiao, C. J., R. B. Ridley, H. Naitoh, and F. C.. Lee, ** Circuit-Oriented Discrete-Time Modeling and Simulation for Switching Converters**, IEEE Power Electronics Specialists Conference Record, Blacksburg, Virginia, June 1987, pp. 167-176.

A generalized discrete-time modeling and simulation program, applicable to any PWM, resonant or quasi-resonant converter, has been developed. From a circuit description, this program automatically generates state-space equations corresponding to each switching interval and performs time-domain simulations by

using state-transition equations with a fast-convergence algorithm for topological change.

*The program this paper described, COSMIR, was one of several computer programs developed around this time to address the specific problems of PWM converter simulation. It ran very fast, but the code was prone to logic errors on occasion. Also, the circuit description for anything but a simple PWM converter was rather cumbersome. The biggest lesson learned from this was not to try and write a special purpose program as a stand alone application. In today's computer software market, the resources needed to create and maintain such a complex program are formidable. This lesson has been applied in developing POWER 4-5-6.*

Ridley, R. B., and F. C. Lee, ** Practical Nonlinear Design Optimization Tool for Power Converter Components**, IEEE Power Electronics Specialist Conference Record, 1987, pp. 314-323.

A computer-aided design tool for power converter components is described. This tool allows a designer with a minimum of programming and optimization experience to interface with nonlinear optimization routines. Realistic design values and available vendor components can be incorporated in a design without using an extensive data base program structure.

*This paper used a nonlinear optimization tool to design a power supply from the basic design equations and specifications. It worked OK, but the optimization tool was very cumbersome and not recommended for this task. POWER 4-5-6 takes this work to a more sophisticated level, incorporating many more design rules and substituting experience for an optimization engine.*

Ridley, R. B., C. Zhou, and F. C. Lee, ** Practical Nonlinear Design Optimization for Power Converter Components**, IEEE Transactions on Power Electronics, Vol. 5, No. 1, January 1990, pp. 29-39.

A computer-aided design approach for power converter components is described. A designer with a minimum of programming and optimization experience can interface with nonlinear optimization routines to rapidly perform design trade-offs which would be impossible by hand. A power converter design using MOSFET and bipolar junction transistor (BJT) switches is shown to illustrate the power of optimization routines in power electronics. Realistic design values and available vendor components can be incorporated in a design without using an extensive data base program structure. A practical example is given with experimental data to verify the accuracy and usefulness of optimization software.

*Same topic as the above paper, with some more details added for this IEEE Power Electronics Transactions version.*

Kelkar, S. S., R. B. Ridley, C.J. Hsiao, R. Ramkumar, and F. C. Lee, ** A Computer-Aided Design and Simulation Tool for Switching Converters**, IEEE Power Electronics Specialists Conference Record, Blacksburg, Virginia, June 1987, pp. 3-12.

A fully automated computer-aided design approach is presented which results in an optimal power stage and control circuit design. The design meets all dc, small signal and large signal closed-loop performance specifications. The proposed approach is very efficient; it can help in reducing component and manufacturing costs as well as design time and is successfully demonstrated on a multiple-output flyback converter breadboard.

*Gives a practical demonstration of the use of the tools described in the previous three papers. Is the result truly "optimal"? Depends on what you use to measure this. The more reliable and faster processes used in Power 4-5-6 improve greatly on these methods. *

Sable, D. M., R. B. Ridley, and B. H. Cho, ** Comparison of Performance of Single-Loop and Current-Injection Control for PWM Converters which Operate in Both Continuous and Discontinuous Modes of Operation**, IEEE Power Electronics Specialists Conference Record, San Antonio, Texas, June 1990, pp. 74-79.

A comparison of single-loop and current-mode controlled power converters operating in the continuous and discontinuous mode is performed using the PWM switch model and a new, continuous time model of current-injection-control (CIC). The theoretical and experimental results show a significant performance improvement that can be realized when CIC is employed in converters which operate over a wide load range.

*Another good practical paper. Bottom line: if you build a converter which will operate in both CCM and DCM, you should use current-mode control if you expect to get good transient performance. There is no reason why a converter should NOT operate in both regions, though you will find many designers who go to great pains to avoid this.*

*The lessons of this paper are being forgotten in the late 90's as many semiconductor companies have started making voltage-mode only controllers again. *

Sable, Dan M., Raymond B. Ridley, and Bo H. Cho, ** Comparison of Performance of Single-Loop and Current-Injection Control for PWM Converters that Operate in Both Continuous and Discontinuous Modes of Operation**, IEEE Transactions on Power Electronics, Vol. 7, No. 1, January, 1992, pp. 136-142.

A analysis of current-injection-controlled (CIC) power converters operating in both the continuous and discontinuous modes is performed using the PWM switch model and a new, continuous-time model of current-injection control. The stability, output impedance, audio susceptibility, and transient response are compared with single-loop control. The control of an example buck converter is designed with CIC and single-loop control. It is shown how single-loop controlled power converters exhibit a large change in the dynamic performance when crossing the boundary between continuous mode and discontinuous mode. This is especially true for the output impedance and transient response. The dynamic performance of current-injection-controlled power converters remains relatively fixed when crossing this boundary. A significant performance improvement that can be realized when CIC is employed in converters that operate over a wide load range.

*IEEE Transactions version of the previous paper. This is the preferred version.*

Vorperian V., and R. B. Ridley, ** A Simple Scheme for Unity Power Factor Rectification for High-Frequency AC Buses**, IEEE Transactions on Power Electronics</I>, Vol. 5, No. 1, January 1990.

A simple scheme is proposed for off-line unity power factor rectifications for high frequency ac buses (20 kHz). In this scheme, a bandpass filter of the series resonant type centered at the line frequency is inserted between the line and the full-wave rectified load. The Q = Zo/Rl formed by the load and the characteristic impedance of the tank circuit determine the power factor, the boundary between the continuous and discontinuous conduction modes, the peak stresses and the transient response of the rectifier. It is shown that for Q < 2 the line current is nearly sinusoidal with less than five percent third harmonic distortion while the power factor is essentially unity. A increase in the value of Q causes an increase in the peak voltages of the tank circuit and a slower transient response of the rectifier circuit. The dc small-signal and transient analyses of the rectifier circuit are determined and the results are in good agreement with simulation and experimental results.

*Note: this paper is not an endorsement of high-frequency ac versus dc power distribution, a topic of very heated debate for the space station in the late 1980s. However, if you are planning on using a high-frequency ac bus for power distribution, you should at least make sure that the power system operates with low harmonics. This paper presented a simple scheme with a series-resonant LC notch filter to greatly improve the harmonics and power factor of the system.*

*This system is practical, it really does work quite well in the right application. A recent research project applied the techniques here to a 20 kHz, 10 kW power system which needed to transmit power over a 750 foot long, 1/4 inch diameter cable.*

*Some of the analysis presented in this paper is not new, it was originally done by Steve Freeland at Caltech. The small-signal modeling is original, and has been verified experimentally.*

Ridley, Raymond B., Wojciech A. Tabisz, Fred C. Lee, and Vatche Vorperian, ** Multi-Loop Control for Quasi-Resonant Converters**, IEEE Transactions on Power Electronics, Vol. 6, No. 1, January 1991, pp.28-38.

A new, multi-loop control scheme for quasi-resonant converters is described. Similar to current-mode control for PWM converters, this control offers excellent transient response and replaces the voltage-controlled oscillator with a simple comparator. A signal proportional to the output-inductor current is compared with an error voltage signal to modulate the switching frequency. The control can be applied to either zero-voltage-switched or zero-current-switched quasi-resonant converters. Computer simulation is used to demonstrate the effectiveness of the control method applied to a zero-current-switched buck quasi-resonant converter. Experimental results are presented for zero-current flyback and zero-voltage buck quasi-resonant converters, operating up to 7 MHz.

*Some interesting experiments in pushing the loop gain of a converter to its limits. Using some wide-bandwidth components, and a passive current-mode sensor, crossover frequencies in excess of 100 kHz were obtained. There are some techniques used in this paper that are very relevant and applicable to PWM converters operating at higher frequencies. The paper is not an endorsement of zero-voltage quasi-resonant operation, but if you are going to use this technology, this is a good way to control it. *

Ridley, R. B., A. Lotfi, V. Vorperian and F. C.. Lee, ** Design and Control of a Full-Wave, Quasi-Resonant Flyback Converter**, IEEE Applied Power Electronics Conference Proceedings<, New Orleans, 1988, pp. 41-49.

Design considerations for a quasi-resonant converter for a typical distributed power system application are described. Different topologies are compared in terms of the peak stresses on the power switch. The design of a full-wave, zero-current-switched flyback quasi-resonant converter with a novel, multi-loop control is described in detail. Circuit waveforms, small-signal and large signal measurements are compared with theoretical predictions.

*If you are going to build a full-wave, quasi-resonant flyback converter, this is a good way to control it. Whether or not you should use this topology is another matter-- generally it is not recommended, though some very special applications may be able to make use of it. A simple PWM circuit at significantly lower frequency will often outperform this technology.*

Ridley, R. B., W. A. Tabisz, and F. C. Lee, ** Multi-Loop Control for High-Frequency Quasi-Resonant Converters**, Intertec Communications Inc., High Frequency Power Conversion International, 1988, pp. 381-389.

Limitations of multi-loop control for a 1MHz flyback quasi-resonant converter are discussed. A new control circuit is presented which allows operation up to 10MHz. Circuit waveforms and small-signal characteristics for a 7MHz zero-voltage-switched quasi-resonant converter are presented to demonstrate the effectiveness of the new control circuit.

*Same comment as for the earlier paper: the control works well, but this is not necessarily an endorsement for any kind of resonant converter for most applications. *

Sable D. M., and R. B. Ridley, ** A High Frequency Multi-Module Spacecraft Boost Regulator**, International Telecommunications Energy Conference, 1988.

Choi, B., B. H. Cho, R. B. Ridley, and F. C.. Lee, ** Control Strategy for Multi-Module Parallel Converter System**, IEEE Power Electronics Specialists Conference Record, San Antonio, Texas, June 1990, pp. 225-234.

The control strategy for a multi-module converter system for high-current low-voltage applications is investigated. The system consists of several converter modules in parallel to effectively deliver a high-current output. A multi-stage output filter is employed to efficiently attenuate the ripple and high-frequency noise. In addition to output voltage and inductor current feedback, a feedback from the intermediate filter stage is employed to optimize the transient response in the event of failure of a converter module, and also to improve the other closed-loop performances of the system. Based on the small-signal analysis, a systematic control-loop design procedure for optimal performance of the system is presented.

*A complex power system used in high-power mainframe computers. The simulation program Saber proved to be the most useful in analyzing this system. A paper for the control design fanatic who likes to think in the Laplace domain.*

Zhou, C., R. B. Ridley, and F. C.. Lee, ** Design and Analysis of a Hysteretic Boost Power Factor Correction Circuit**, IEEE Power Electronics Specialists Conference Record, San Antonio, Texas, June 1990, pp. 800-807.

The design of an active unity power factor correction circuit with variable-hysteresis control for off-line switching power supplies is described. Design equations relating the boost inductor current ripple to the boost inductor selection and circuit performances are developed and are verified with measurements. A computer-aided design program is developed to select the optimal circuit components. Design guidelines for the low-frequency feedback network are presented using the switch model for the power factor correction circuit. Small-signal transfer functions for open and closed-loop responses are derived.

*Hysteretic current-mode was very popular at this time for PFC applications. Since the introduction of the Unitrode Average current-mode chip, this popularity has faded. In general, the hysteretic control is more problematic to implement, and more noise sensitive.*

Sable, D. M., B. H. Cho and R. B. Ridley, ** Elimination of the Positive Zero in Fixed Frequency Boost and Flyback Converters**, IEEE Applied Power Electronics Conference Proceedings, Los Angeles, California, March 1990, pp.

205-211.

It is shown how a fixed-frequency, leading-edge modulated PWM can eliminate the undesirable positive zero in practical boost and flyback converters. This allows a substantial improvement in the closed-loop characteristics. Several techniques are employed to predict this result. The design procedure for elimination of the positive zero is presented. Experimental verification is provided.

*Note: this is an analysis of an existing satellite power system which had succeeded in eliminating the RHP zero by clever and intuitive design. It is not recommended that you build your boost or flyback converters this way, but you may run into this phenomenon at some time. Interesting paper f you are into control theory of converters. It's not practical in most cases since it requires the use of high esr caps which leads to elevated loss and noise.*

Tang, W., F. C. Lee, R. B. Ridley, I. Cohen, ** Charge Control: Modeling, Analysis and Design**, IEEE Power Electronics Specialists Conference Record, Spain, 1992, pp. 503-511.

A new power converter control method, charge control, is studied. A complete small-signal analysis is performed for the control scheme. Subharmonic oscillation similar to that of CIC control is found and the relationship between the subharmonic oscillation and line/load condition of charge control is defined. Based on analysis, the design guidelines which guarantee the stability of the control system under given line and load ranges are proposed. The small-signal model was confirmed experimentally.

Tang, W., F. C. Lee, and R. B. Ridley, ** Small-signal Modeling of Average Current-Mode Control**, IEEE Applied Power Electronics Conference Proceedings, Boston, 1992, pp. 747-755.

A recently proposed average current-mode control is analyzed. A complete small-signal model for the control scheme is developed. The model is accurate up to half the switching frequency. By closing the current loop, a flat control-to-inductor current transfer function, up to half the switching frequency, can be achieved. This control scheme enables the converter to behave as an ideal current source. The subharmonic oscillation, as frequently reported in peak current-mode control, also exists in this control. This subharmonic oscillation can be eliminated by choosing a proper gain of the compensation network in the current loop. Model predications are confirmed experimentally.

*Paper shows the analysis of the average current-mode control, widespread in the PFC field. It shows that this control scheme can also have oscillations in the current loop.*

Choi, Byungcho, Bo H. Cho, Fred C. Lee, and Raymond B. Ridley, ** The Stacked Power System: A New Power Conditioning Architecture for Mainframe Computer Systems**, IEEE Transactions on Power Electronics, Vol. 9, No. 6, November, 1994, pp. 616-623.

The stacked power system combines series and parallel connections of multimodule power supplies to produce several low voltage outputs at high current levels. The system could generate ultra low voltage outputs with higher efficiencies than conventional centralized or distributed approaches, using standard single-output converters. This paper presents the architecture of the stacked power system and establishes a design procedure for the system.

Chin, Shaoan A., John Tero, Milan M. Jovanovic, Raymond B. Ridley, and Fred C. Lee, ** A New IC Controller for Resonant-Mode Power Supplies**, IEEE Applied Power Electronics Conference Proceedings, Los Angeles, California, March 1990, pp. 459-466.

The NE5580 is an all-purpose control IC providing control functions for zero-current-switching (ZCS) or zero-voltage-switching (ZVS) resonant or quasi-resonant power supplies. It features: frequency-modulated constant on-time or constant off-time control with 10 MHz voltage-controlled oscillator; 10 MHz error amplifier; dual 1A peak totem-pole output drivers with no cross conduction current; and other functions as will be described.

*The controller worked well for a wide range of resonant converters. Unfortunately, Signetics never solved the quiescent dissipation and output driver shoot-through problems, and the chip was never commercially successful.*

Chin, S. A., J. Tero, M. M. Jovanovic, R. B. Ridley, and F. C. Lee, ** A 10 MHz Universal Control IC for Resonant-Mode Power Supplies**, High-Frequency Power Conversion Conference, Santa Clara, California, May 1990.

Ridley, R. B., B. H. Cho and F. C.. Lee, ** Analysis and Interpretation of Loop Gains of Multi-Loop-Controlled Switching Regulators**, IEEE Transactions on Power Electronics, Vol. 3, No. 4, October 1988, pp. 489-498.

Two different loop gains of power supplies with current-mode control are considered. The relationships between the loop gains and closed-loop characteristics are derived. Both loop gains can be used to aid design. A common problem with the implementation of current-mode control is discussed.

*This is at the bottom of the list for a good reason. An interesting paper, it seems to make a lot of sense. Unfortunately, for current-mode control, most of it is wrong, due to an early error in modulator gain! See the later paper on a new small-signal model for current mode control for proper analysis of the current-mode system. Don't read this paper unless you are particularly curious. The best lesson contained in this paper: don't believe everything that you read, publication is not necessarily an indication of accuracy.*

Post Script: The same error in modulator gain has appeared again in yet another publication on current-mode control.