Table Of Content

Analog Circuit Design Volume 2 Immersion in the Black Art of Analog Design Analog Circuit Design Volume 2 Immersion in the Black Art of Analog Design E d i t e d b y Bob Dobkin Jim Williams AMSTERDAM • BOSTON • HEIDELBERG • LONDON • NEW YORK • OXFORD PARIS • SAN DIEGO • SAN FRANCISCO • SINGAPORE • SYDNEY • TOKYO Newnes Newnes is an imprint of Elsevier Newnes is an imprint of Elsevier The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK 225 Wyman Street, Waltham, MA 02451, USA First edition 2013 Copyright © 2013, Linear Technology Corporation. Published by Elsevier Inc. All rights reserved. See separate Publisher’s Note for copyright details of Trade Marks used in this book. The right of Linear Technology Corporation to be identiﬁed as the author of this work has been asserted in ac- cordance with the Copyright, Designs and Patents Act 1988. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher or authors/contributors for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent veriﬁcation of diagnoses and drug dosages should be made. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging-in-Publication Data A catalog record for this book is availabe from the Library of Congress ISBN: 978-0-12-397888-2 For information on all Syngress publications visit our website at http://store.elsevier.com Printed and bound in the United States of America 13 14 15 10 9 8 7 6 5 4 3 2 1 For Jerrold R. Zacharias, who gave me the sun, the moon and the stars. For Siu, who is the sun, the moon and the stars. In memory of Jim Williams, a poet who wrote in electronics. Publisher’s Note This book was compiled from Linear Technology Application Notes, not all of which have made it into Corporation’s original Application Notes. the book. For reference, this conversion table has been These Application Notes have been re-named as included; it shows the book chapter numbers and the origi- chapters for the purpose of this book. However, through- nal Application Note numbers. out the text there is a lot of cross referencing to different CHAPTER NUMBER APPLICATION NOTE CHAPTER NUMBER APPLICATION NOTE 1 2 22 27A 2 104 23 38 3 133 24 79 4 29 25 123 5 44 26 128 6 32 27 131 7 118 28 85 8 65 29 91 9 95 30 129 10 107 31 12 11 15 32 45 12 74 33 52 13 132 34 57 14 4 35 61 15 5 36 67 16 10 37 75 17 13 38 87 18 14 39 98 19 16 40 105 20 18 41 113 21 21 xi SECTION ONE Publisher’s Note Trademarks DirectSense, Dust, Easy Drive, Eterna EZSync, Filter- View, Hot Swap, isoSPI, LDO+, LinearView, LTBiCMOS, LTCMOS, LTPoE++, LTpowerCAD, LTpowerPlan- These trademarks all belong to Linear Technology Corpo- ner, LTpowerPlay, MicropowerSwitcherCAD, Mote-on- ration. They have been listed here to avoid endless rep- Chip, Multimode Dimming, No Latency ∆∑, No Latency etition within the text. Trademark acknowledgment and Delta-Sigma, No R , Operational Filter, PanelProtect, protection applies regardless. Please forgive us if we have SENSE PLLWizard, PowerPath, PowerSOT, PScope, QuikEval, RH missed any. DICE Inside, RH MILDICE Inside, SafeSlot, SmartMesh, Linear Express, Linear Technology, LT, LTC, LTM, SmartMesh IP, SmartStart, SNEAK-A-BIT, SoftSpan, Burst Mode, Dust Networks, FilterCAD, LTspice, OPTI- Stage Shedding, Super Burst, SWITCHER+, ThinSOT, LOOP, Over-The-Top, PolyPhase, SwitcherCAD, Tim- Triple Mode, True Color PWM, UltraFast, Virtual Remote erBlox, µModule and the Linear logo are registered Sense, Virtual Remote Sensing, VLDO and VRS are trade- trademarks of Linear Technology Corporation. Adaptive marks of Linear Technology Corporation. All other trade- Power, Bat-Track, BodeCAD, C-Load, Direct Flux Limit, marks are the property of their respective owners. xxiiii Acknowledgments Analog Circuit Design, Volume 2 emerges a year after the editorial team that ensured that the application notes are irst volume through the efforts of a dedicated team. For clear and consistent—Gary Alexander and Susan Dale. many of us, this is a labor of love, giving further legs to We are indebted to the efforts of the professionals at the timeless application notes of the late Jim Williams and Elsevier/Newnes, including publisher Jonathan Simpson many colleagues at Linear Technology. Thanks irst to the and the production efforts of Pauline Wilkinson and Fiona authors, who do the heavy lifting—in the lab and through Geraghty. Finally, to Bob Dobkin for his insight, time and their insightful writing. Also to the dedicated graphics and belief in this project. John Hamburger Linear Technology Corporation xiii Introduction Why I Write Writing was equally painful. Text low was staccato and disjointed because of the gaps that occurred while I waited In the early 1980s we wanted prospective customers to for results with actual products. I had to keep separate know our name and what we were up to. The real issue notes directing me to uninished text when we inally was inding a way to productively use the seeming dead dropped the products into the breadboards. time before product availability. What readers wanted was The irst article took almost two months, but things a series of credible, full length technical articles in the lan- slowly became easier. Tricks to move along the lab work guage of relevant, working circuits. evolved, and I found ways to write more eficiently, making I moped for weeks over this problem before a possible the manuscripts inherently adaptable to the planned addi- solution became apparent. Instead of waiting for prod- tions and changes. Soon, I was producing an almost-in- ucts, I’d simply go into the lab, develop the applications, ished article every two weeks or so, roaring along, powered and then write the articles. The key to this approach was by adrenaline, solder, pencils, paper, and pizza. to synthesize the expected products using available ICs During the next year, life was a dizzy seven-day-a-week and discretes to build rough equivalents on small plug- blur of breadboards and manuscripts shuttling between in boards. We could develop functional applications and work and my home lab. My diet was a cardiologist’s night- write most of the text. We’d then shelve the manuscript mare. I don’t recall having a meal at home. The refrigera- and breadboards. Later, when products became available, tor was devoid of food but well-provisioned with Polaroid we could put them into the breadboards and implement ilm to feed the oscilloscope camera. All this frenetic bus- the attendant inal changes. Once we had done these tasks, tle boiled off any semblance of a normal social life. At din- we could drop scope photos and speciications into the ner in San Francisco, while nominally listening to my date waiting text, tweak the manuscript, and ship it off. This describe her job intricacies, I silently calculated the opti- approach would speed publication by perhaps a year and mum chapper-channel crossover frequency in a composite synchronize the article’s appearance with product intro- ampliier. The regimen of madness continued for about a duction. year, resulting in 35 full-length feature articles between Initially, the whole scheme appeared absurd and emi- June 1983 and November 1987. nently unworkable, with uncountable technical and I still write, although at a signiicantly less frenetic editorial sinkholes. Getting started was much more dif- pace. Now, when the kids in our lab complain to me about icult than I had imagined. Synthesizing the hardware for writing technical material, I try not to sound like the cur- our unborn ICs proved tricky; my methods, clumsy and mudgeon I am not so slowly becoming. I think that mad stumbling. Breadboarding the applications was laborious tear almost 25 years ago contributes to my current lack of and slow, primarily because I wasn’t sure how accurately empathy. These kids today, with a catalog full of products, I was mimicking the forthcoming IC’s performance. they don’t know what they’ve got. Jim Williams Staff Scientist Linear Technology Corporation Note: This essay originally appeared in EDN magazine. xv Foreword We appreciate the enthusiastic response to the original electronic circuits for testing. Components were screwed book, Analog Circuit Design: A Tutorial Guide to Applica- to the board and wired with Fahnestock clips, such as the tions and Solutions. The acceptance of this book empha- one in this diagram. sizes the need for making good circuit design applications accessible. These writings on applications ill a vacuum, since the majority of application notes and magazine articles do not have suficient depth to teach analog design. At the time I learned analog design, there were not yet any analog ICs. Circuits were all transistorized (maybe some tubes) and the circuit explanations in magazines and books were more complete than many are today. In those days equipment manuals included schematics for repair. I was fortunate to work at a large company with a huge calibration lab for their equipment. I spent many lunch hours perusing the calibration and repair manuals for analog systems. I thank HP, Tektronix and many other companies for their tutorials on analog circuit design that I discov- ered in the calibration lab. It is interesting to note that Jim Williams in his early years at MIT spent much time repairing electronic test equipment that was nonfunctional. Many analog circuits are layout-sensitive and if not Today’s manuals are more sparse. properly laid out, the circuits may not work. I have seen It is relatively hard to ind completed analog designs to many circuits constrained by this problem. study. Books may include analog designs, but they are not Finally, designers need a detailed set of test results for necessarily complete circuits with test results. Likewise the circuit. These show how the circuit works when it’s most application notes are very speciic to the device they working properly and provides a comparison guide for any- are showing and don’t provide a general breadth of infor- one trying to duplicate the circuit. Without all of these, the mation that is useful in the analog world. applications fail as part of a teaching system. I am pleased that this book series is becoming a teach- Analog design is challenging. There are many ways to ing handbook of applications. An application should be get from input to output, and the circuitry in the middle useful, have multiyear life and be easy to duplicate. Any can lead to divergent results. Analog design is like learn- good application note should have a description of the ing a language. When you irst learn a language, you begin application and discussion of where it will be used. It with a vocabulary book and then analyze writings in that should include ancillary information, such as temperature language by looking up words one by one as you encoun- range, power systems, lifetime and other key data usually ter them. Likewise, in analog design you learn the basics of excluded in textbooks that just focus on principles. The the circuit, as well as the function of different devices. You application block diagram needs to be explained so that can write node equations and determine what the circuit is the approach to the solution is understandable. Unless you doing by studying each of the individual circuits. know where you are going, it is dificult to understand how With analog circuit design, you end up using the basic you got there. circuit conigurations you have learned—differential ampli- The circuits for the applications need to be fully devel- iers, transistors, FETs, resistors, and previously studied cir- oped, have part types and construction information. The cuits—to achieve the inal circuit. As with a new language, solutions may include specialized components that require it takes many years to learn to write poetry, and the same the reader to have knowledge of the function and special is true of analog circuit design. properties. The circuit should be shown and the properties Today’s system manuals rarely have any circuit design. and function of each item in the circuit explained with suf- They contain block diagrams and hookup schematics icient detail that readers can extract that information and with thousands of leads to different blocks. So where do use it again. people go for analog circuit design? If you’re just start- The origin of “breadboard” dates back to a time when ing, or if you encounter a problem that’s outside your actual bread cutting boards were used to assemble experience, appropriate references are dificult to ind. xvii