`
`.383m§.§
`
`
`
`_n____&__§§_m3.322m_2_3=_8
`
`WAC 1010
`
`
`
`,.-.......w..¢_«;..«;
`
`
`urmwwgprnaagnvowza-$:vite,g4,W.,,(;+!aapa»pgt4KMaA;iwgai::
`
`
`
`.-'n$~WW«,M_.»,e:a.m-.v»~s‘-;4Ws.srt#a'-at-;
`
`
`
`
`
`
`
`Distributors for North, Central and South America:
`Kluwer Academic Publishers
`101 Philip Drive
`Assinippi Park
`Norwell, Massachusetts 02061 USA
`Telephone (781) 871-6600
`Fax (781) 871-6528
`E-Mail <l<1uwer@wl<ap.com>
`
`Distributors for all other countries:
`Kluwer Academic Publishers Group
`Distribution Centre
`Post Ofiice Box 322
`3300 AH Dordrecht, THE NETHERLANDS
`Telephone 31 78 6392 392
`Fax 31 78 6546 474
`E—Mail services@wkap.nl>
`
`ifi Electronic Services <http://www.wkap.nl>
`
`
`Library of Congress Cataloging-in-Publication
`
`- Erickson, Robert W. (Robert Warren), 1956-
`Fundamentals of power electronics / Robert W. Erickson, Dragan Malc‘simovic.—2“d ed.
`p. cm.
`Includes bibliographical references and index.
`ISBN 0-7923-7270-0 (alk. paper)
`1. Power electronics. 1. Maksimovic, Dragan, 1961- 11. Title.
`
`TK7881.l5 .E75 2000
`621.381-dc21
`
`
`
`00-052569
`
`Copyright 9 2001 by Kluwer Academic Publishers.
`Cover art Copyright 9 1999 by Lucent Technologies Inc. All rights reserved. Used with
`permission.
`'
`
`:w_-Amr-1u,t-irsim.-r«
`
`All rights reserved. No part ofthis publication may be reproduced, stored in a retrieval system or
`transmitted in any form or by any means, mechanical, photo-copying, recording, or otherwise,
`without the prior written permission of the publisher, Kluwer Academic Publishers, 101 Philip
`Drive, Assinippi Park, Norwell, Massachusetts 02061
`
`
`
`Printed on acid-flee paper.
`
`Printed in the United States of America
`
`
`
`
`
`6.2 A Shon List afConver1er5
`
`143
`
`If the converter output voltages V1, V2, and V3 contain the same dc bias, then this dc bias will also appear
`an’
`at the neutral point V". The phase voltages V lg". and V“ are given by
`
`Vun = Vl _ Vn
`Vbrx : V2 _ Vii
`Von = V3 _ Vn
`
`It can be seen that the dc biases cancel out, and do not appear in Va", Vb”, and V6".
`Let us realize converters I, 2, and 3 of Fig. 6.12 using buck converters. Figure 6.l3(a) is then
`obtained. The circuit is re-drawn in Fig. 6.l3(b) for clarity. This converter is known by several names,
`including the voltage-source inverter and the buck—derived three-phase bridge.
`Inverter circuits based on dc—dc converters other than the buck converter can be derived in a
`similar manner. Figure 6.l3(c) contains a three-phase current-fed bridge converter having a boost-type
`voltage conversion ratio, also known as the current—source inverter. Since most inverter applications
`require the capability to reduce the voltage magnitude, a dc—dc buck converter is usually cascaded at the
`dc input port of this inverter. Several other examples of three-phase inverters are given in [S-7], in which
`the converters are capable of both increasing and decreasing the voltage magnitude.
`
`load
`
`6.2
`
`A SHORT LIST OF CONVERTERS
`
`
`
`k converters;
`irrent-source
`
`An infinite number of converters are possible, and hence it is not feasible to list them all. A short list is
`given here.
`Let’s consider first the class of single-input single—output converters, containing a single induc-
`tor. There are a limited number of ways in which the inductor can be connected between the source and
`load. If we assume that the switching period is divided into two subintervals, then the inductor should be
`connected to the source and load in one manner during the first subinterval, and in a different manner
`during the second subinterval. One can examine all of the possible combinations, to derive the complete
`set of converters in this class [8—lO]. By elimination of redundant and degenerate circuits, one finds that
`there are eight converters, listed in Fig. 6.14. How the converters are counted can actually be a matter of
`semantics and personal preference; for example, many people in the field would not consider the nonin-
`verting buck—boost converter as distinct from the inverting buck—boost. Nonetheless, it can be said that a
`converter is defined by the connections between its reactive elements, switches, source, and load; by how
`the switches are realized; and by the numerical range of reactive element values.
`The first four converters of Fig. 6.14, the buck, boost, buck-boost, and the noninverting buck-
`boost, have been previously discussed. These converters produce a unipolar dc output voltage. Vlfith these
`converters, it is possible to increase, decrease, and/or invert a dc voltage.
`the
`Converters 5 and 6 are capable of producing a bipolar output voltage. Converter 5,
`H—bridge, has previously been discussed. Converter 6 is a nonisolated version of a push—pull current-fed
`converter [1 1-15]. This converter can also produce a bipolar output voltage; however, its conversion ratio
`M(D) is a nonlinear function of duty cycle. The number of switch elements can be reduced by using a
`two-winding inductor as shown. The function of the inductor is similar to that of the flyback converter,
`discussed in the next section. When switch 1 is closed the upper winding is used, while when switch 2 is
`closed, current flows through the lower winding. The current flows through only one winding at any
`given instant, and the total ampere-tums of the two windings are a continuous function of time. Advan-
`tages of this converter are its ground-referenced load and its ability to produce a bipolar output voltage
`using only two SPST current-bidirectional switches. The isolated version and its variants have found
`
`
`
`144
`
`Convener Circuits
`
`1. Buck
`
`M(D) = D
`
`5. Bridg.
`
`Fig. 6.14 C(
`
`lyzed. It has
`currents. The
`buck-boost c
`source termir
`inductor com
`also numerot
`transistor anc
`applications 1
`operating poi
`
`0.5
`
`1
`
`D
`
`D D
`
`1
`
`'
`
`0.5
`
`05
`
`[M(D) I
`I
`I
`l
`!
`
`0.5
`
`0
`
`0
`
`MD)
`4
`3
`
`0
`
`°
`
`2
`
`10
`
`0
`-1
`-2
`
`-3
`-4
`
`M(D)
`
`M(D)
`
`4 3 2 1 o
`
`0
`
`0.5
`
`1
`
`D
`
`1
`
`1
`‘
`
`3
`V
`
`2. Boos:
`
`V;
`
`2
`
`1
`
`+
`
`y
`
`-
`
`M(D) = —_1—D
`
`+
`
`V
`
`-
`
`3. Buck—boos1
`I
`
`I
`
`2
`
`M(D) = — T05
`+
`
`V2
`
`V
`
`"
`
`4. Noninverzing buck-boost
`
`M(D) = T95
`
`
`
`Fig. 6.14 Eight members of the basic class of single-input single-output converters containing a single inductor.
`
`application in high-voltage dc power supplies.
`Converters 7 and 8 can be derived as the inverses of converters 5 and 6. These converters are
`capable of interfacing an ac input to a dc output. The ac input current waveform can have arbitrary wave-
`shape and power factor.
`The class of single-input single-output converters containing two inductors is much larger. Sev-
`eral of its members are listed in Fig. 6.15. The Cuk converter has been previously discussed and ana-
`
`
`
`
`
`
`
`6.2 A Short List of Converters
`
`145
`
`5. Bridge
`
`M(D) = 2D -‘ 1
`
`M(D)
`
`6. Watkz'ns~Joh1xs0n
`
`8. Inverse of Watkins—JoImson
`
`M(D) = .,DD_ 1
`
`
`
`Fig. 6.14 Continued
`
`lyzed. It has an inverting buck-boost characteristic, and exhibits nonpulsating input and output terminal
`currents. The SEPIC (single-ended primary inductance converter) [16], and its inverse, have noninverting
`buck-boost characteristics. The Cuk and SEPIC also exhibit the desirable feature that the MOSFET
`source terminal is connected to ground: this simplifies the construction of the gate drive circuitry. Two-
`inductor converters having conversion ratios M(D) that are biquadratic functions of the duty cycle D are
`also numerous. An example is converter 4 of Fig. 6.15 [17]. This converter can be realized using a single
`transistor and three diodes. Its conversion ratio is M(D) = D2. This converter may find use in nonisolated
`applications that require a large step-down of the dc voltage, or in applications having wide variations in
`operating point.
`
`ie inductor.
`
`nveners are
`
`trary wave-
`
`larger. Sev-
`ad and ana-
`
`
`
`146
`
`Convener Cimuits
`
`1. Cuk
`
`v
`
`I
`
`2
`
`+
`
`V
`
`+ v
`
`2. SEPIC
`
`MU’) = ’%
`
`
`
`3. Inverse ofSEPlC
`1
`
`M(D) = T95
`
`vg
`
`2
`
`+
`
`V
`
`M(D) = " T_D“D'
`
`(a)
`
`S
`Fig. 6.16
`taining a mag
`
`cies. Isolatic
`converter ac
`
`improvemen
`operates at t
`W11
`allow better
`rent stresses
`lower cost.
`Mu
`
`ary winding:
`
`Od
`
`uty cycle, 5
`‘ID--il
`mt
`A t
`6.l6(a). A sir
`operation of
`ings and neg
`obey;.th¢.r¢l
`
`In parallel v
`the transfort
`
`Phj
`nect all win.
`core——an in.
`ing inductar
`
`
`
`4. Buck 3
`
`M(D) = D2
`
`
`
`0
`
`0.5
`
`1
`
`D
`
`Fig. 6.15 Several members of the basic class of single—input single-output converters containing two inductors.
`
`6.3
`
`TRANSFORMER ISOLATION
`
`In a large number of applications, it is desired to incorporate a transformer into a switching converter, to
`obtain dc isolation between the converter input and output. For example, in off-line applications (where
`the converter input is connected to the ac utility system), isolation is usually required by regulatory agen-