## Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

Hands-On Electronics

Packed full of real circuits to build and test, Hands-On Electronics is a unique introduction

to analog and digital electronics theory and practice. Ideal both as a college textbook and

for self-study, the friendly style, clear illustrations and construction details included in the

book encourage rapid and effective learning of analog and digital circuit design theory.

All the major topics for a typical one-semester course are covered, including RC circuits,

diodes, transistors, op amps, oscillators, digital logic, counters, D/A converters and more.

There are also chapters explaining how to use the equipment needed for the examples

(oscilloscope, multimeter and breadboard), together with pinout diagrams for all the key

components referred to in the book.

Hands-On Electronics

A One-Semester Course for

Class Instruction or Self-Study

Daniel M. Kaplan

and

Christopher G. White

Illinois Institute of Technology

caxniioci uxiviisiry iiiss

Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press

The Edinburgh Building, Cambridge cn: :iu, United Kingdom

First published in print format

isnx-:+ ,¬ï-c-s::-ï:s+o-,

isnx-:+ ,¬ï-c-s::-ï,+s:-c

isnx-:+ ,¬ï-c-s::-c¬ooï-ï

© Cambridge University Press 2003

2003

Information on this title: www.cambridge.org/9780521815369

This book is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place

without the written permission of Cambridge University Press.

isnx-:c c-s::-c¬ooï-:

isnx-:c c-s::-ï:s+o-+

isnx-:c c-s::-ï,+s:-ï

Cambridge University Press has no responsibility for the persistence or accuracy of

uiis for external or third-party internet websites referred to in this book, and does not

guarantee that any content on such websites is, or will remain, accurate or appropriate.

Published in the United States of America by Cambridge University Press, New York

www.cambridge.org

hardback

paperback

paperback

eBook (EBL)

eBook (EBL)

hardback

Contents

List of ﬁgures page xi

List of tables xv

About the authors xvi

To the Reader xvii

Acknowledgments xviii

Introduction xix

1 Equipment familiarization: multimeter, breadboard,

and oscilloscope 1

1.1 Multimeter 1

1.2 Breadboard 2

1.2.1 Measuring voltage 4

1.2.2 Measuring current; resistance and Ohm’s law 5

1.2.3 Measuring resistance 8

1.3 Oscilloscope 8

1.3.1 Probes and probe test 10

1.3.2 Display 11

1.3.3 Vertical controls 11

1.3.4 Horizontal sweep 12

1.3.5 Triggering 12

1.3.6 Additional features 13

2 RC circuits 15

2.1 Review of capacitors 15

2.1.1 Use of capacitors; review of AC circuits 17

2.1.2 Types and values of capacitors 19

v

vi Contents

2.2 Review of current, voltage, and power 20

2.2.1 Destructive demonstration of resistor power rating 21

2.3 Potentiometer as voltage divider 22

2.3.1 DC voltage divider 23

2.3.2 AC voltage divider 23

2.4 RC circuit 24

2.5 RC circuit as integrator 24

2.6 Low-pass ﬁlter 25

2.7 RC circuit as differentiator 27

2.8 High-pass ﬁlter 28

2.9 Summary of high- and low-pass ﬁlters 28

3 Diodes 31

3.1 Semiconductor basics 31

3.2 Types of diodes 35

3.3 Rectiﬁcation 36

3.4 Diode action – a more sophisticated view 37

3.5 Measuring the diode characteristic 38

3.6 Exploring rectiﬁcation 40

3.7 Input and output impedance 45

4 Bipolar transistors 47

4.1 Bipolar-junction-transistor basics 47

4.1.1 Basic deﬁnitions 50

4.1.2 Simplest way to analyze transistor circuits 51

4.1.3 Ebers–Moll transistor model 52

4.2 Experiments 54

4.2.1 Checking transistors with a meter 54

4.2.2 Emitter follower 55

4.2.3 Common-emitter ampliﬁer 57

4.2.4 Collector as current source 59

4.2.5 Transistor switch 60

4.3 Additional exercises 61

4.3.1 Darlington connection 61

vii Contents

4.3.2 Push–pull driver 62

4.3.3 Common-base ampliﬁer 63

5 Transistors II: FETs 65

5.1 Field-effect transistors 65

5.1.1 FET characteristics 66

5.1.2 Modeling FET action 68

5.2 Exercises 69

5.2.1 FET characteristics 69

5.2.2 FET current source 70

5.2.3 Source follower 71

5.2.4 JFET ampliﬁer 73

6 Transistors III: differential amplifier 75

6.1 Differential ampliﬁer 75

6.1.1 Operating principle 76

6.1.2 Expected differential gain 76

6.1.3 Measuring the differential gain 77

6.1.4 Input offset voltage 78

6.1.5 Common-mode gain 78

6.2 Op amps and their building blocks 79

6.2.1 Current mirror 79

6.2.2 Differential ampliﬁer with current-source loads 80

6.2.3 Improved current mirror 82

6.2.4 Wilson current mirror 82

7 Introduction to operational amplifiers 85

7.1 The 741 operational ampliﬁer 85

7.1.1 741 pinout and power connections 86

7.1.2 An ideal op amp 87

7.1.3 Gain of inverting and noninverting ampliﬁers 88

7.1.4 Op amp ‘golden rules’ 90

7.1.5 The nonideal op amp 90

viii Contents

7.2 Experiments 91

7.2.1 Testing open-loop gain 91

7.2.2 Inverting ampliﬁer 92

7.2.3 Noninverting ampliﬁer 93

7.2.4 Voltage follower 94

7.2.5 Difference ampliﬁer 95

7.3 Additional experiments 97

7.3.1 Current source 97

7.3.2 Noninverting summing amp with difference ampliﬁer 98

8 More op amp applications 101

8.1 Op amp signal processing 101

8.1.1 Differentiator 102

8.1.2 Integrator 103

8.1.3 Logarithmic and exponential ampliﬁers 105

8.2 Experiments 106

8.2.1 Differential and integral ampliﬁers 106

8.2.2 Logarithmic and exponential ampliﬁers 108

8.2.3 Op amp active rectiﬁer 108

8.2.4 Op amp with push–pull power driver 109

8.3 Additional exercises 111

9 Comparators and oscillators 113

9.1 Experiments 113

9.1.1 Op amp as comparator 113

9.1.2 Unintentional feedback: oscillation 115

9.1.3 Intentional positive feedback: Schmitt trigger 116

9.1.4 RC relaxation oscillator 117

9.1.5 555 timer IC 118

9.2 Additional experiments 121

9.2.1 Alarm! 121

9.2.2 Sine/cosine oscillator 122

9.2.3 Active bandpass ﬁlter 123

ix Contents

10 Combinational logic 125

10.1 Digital logic basics 125

10.1.1 Logic levels 126

10.1.2 Logic families and history 127

10.1.3 Logic gates 129

10.1.4 Summary of Boolean algebra 130

10.2 CMOS and TTL compared 131

10.2.1 Diode logic 131

10.2.2 Transistor–transistor logic (TTL) 132

10.2.3 Complementary MOSFET logic (CMOS) 133

10.2.4 Powering TTL and TTL-compatible integrated

circuits 136

10.3 Experiments 137

10.3.1 LED logic indicators and level switches 137

10.3.2 MOSFETs 138

10.3.3 CMOS NAND gate 140

10.3.4 Using NANDs to implement other logic functions 140

10.3.5 TTL quad XOR gate 141

10.4 Additional exercises 142

10.4.1 7485 4-bit magnitude comparator 142

11 Flip-flops: saving a logic state 143

11.1 General comments 144

11.1.1 Schematics 144

11.1.2 Breadboard layout 144

11.1.3 Synchronous logic 144

11.1.4 Timing diagrams 144

11.2 Flip-ﬂop basics 145

11.2.1 Simple RS latch 145

11.2.2 D-type ﬂip-ﬂop 147

11.3 JK ﬂip-ﬂop 148

11.4 Tri-state outputs 149

x Contents

11.5 Flip-ﬂop applications 151

11.5.1 Divide-by-four from JK ﬂip-ﬂops 151

11.5.2 Contact bounce 152

11.5.3 Electronic coin toss 153

12 Monostables, counters, multiplexers, and RAM 155

12.1 Multivibrators 156

12.2 Counters 156

12.3 Experiments 157

12.3.1 Bi-quinary ripple counter 157

12.3.2 Monostable multivibrator 159

12.3.3 Multiplexer and ﬁnite-state machine 162

12.3.4 RAM 162

13 Digital↔analog conversion 167

13.1 A simple D/A converter fabricated from familiar chips 168

13.2 Tracking ADC 170

13.3 080x ADC and DAC chips 171

13.3.1 Successive-approximation ADC 171

13.4 Additional exercises 177

13.4.1 Digital recording 177

13.4.2 Successive-approximation ADC built from

components 178

Further reading 183

Appendix A Equipment and supplies 185

Appendix B Common abbreviations and circuit symbols 188

Appendix C RC circuits: frequency-domain analysis 191

Appendix D Pinouts 194

Glossary of basic electrical and electronic terms 197

Index 199

Figures

1.1 Illustration showing many of the basic features of the

PB-503 powered Protoboard. page 3

1.2 Measuring voltage. 5

1.3 Measuring current. 5

1.4 Illustration of the Tektronix TDS 210 digital oscilloscope. 9

2.1 Representation of an arbitrary, periodic waveform. 18

2.2 Circuit demonstrating destructive power loading. 21

2.3 Three schematics representing a resistive voltage divider. 22

2.4 The voltage-divider concept for RC circuits. 24

2.5 High-pass ﬁlter or voltage differentiator. 27

2.6 Relationships among input voltages and capacitor and

resistor voltages for high- and low-pass RC ﬁlters. 29

3.1 Representation of a junction between P-type and N-type

semiconductor material. 33

3.2 Diode circuit symbol and biasing. 33

3.3 Typical current–voltage characteristics for germanium

and silicon diodes. 34

3.4 Representation of physical diodes and symbols used in

circuit diagrams. 35

3.5 Measuring the forward characteristic of a diode. 39

3.6 Power transformer supplies V

out

≈ 25 V r.m.s. 41

3.7 Power transformer with half-wave rectiﬁcation. 42

3.8 Half-wave rectiﬁer with ﬁlter capacitor. 42

3.9 An example of how to insert a diode bridge into a breadboard. 43

3.10 Full-wave rectiﬁcation using diode bridge. 44

3.11 Full-wave rectiﬁcation with ﬁlter capacitor. 45

3.12 Complete rectiﬁer circuit. 46

4.1 Construction and circuit symbols and biasing examples for

NPN and PNP junction transistors. 48

xi

xii List of figures

4.2 Schematic representation of how an NPN transistor operates. 49

4.3 Characteristic curves for an NPN bipolar transistor. 51

4.4 Transistor as back-to-back diodes; TO-92 pinout. 55

4.5 Emitter follower. 55

4.6 Emitter follower with optional load circuit for measurement

of Z

out

. 56

4.7 Common-emitter ampliﬁer. 57

4.8 Transistor current source. 59

4.9 Transistor switch. 60

4.10 Darlington pair. 62

4.11 Driving loudspeaker with push–pull buffer. 63

4.12 Common-base ampliﬁer. 64

5.1 Construction and circuit symbols of JFETs. 66

5.2 Schematic representation of JFET operation. 67

5.3 Idealized common-source characteristic curves for a JFET. 67

5.4 Circuit for measuring the common-source characteristic curves. 70

5.5 Self-biasing JFET current source. 71

5.6 Source follower. 72

5.7 Source follower with current-source load. 73

5.8 JFET ampliﬁer. 73

6.1 Differential ampliﬁer and function generator with

100-to-1 attenuator. 76

6.2 Current sink for differential ampliﬁer. 79

6.3 Current mirror. 80

6.4 Differential ampliﬁer with current-mirror load. 81

6.5 Differential ampliﬁer with Wilson-current-mirror load. 82

7.1 Diagram of 8-pin DIP 741 package showing ‘pinout’. 86

7.2 Op amp inverting-ampliﬁer circuit. 88

7.3 Op amp noninverting-ampliﬁer circuit. 89

7.4 Open-loop op amp test circuit. 91

7.5 Circuit for demonstrating summing junction. 93

7.6 Op amp voltage follower and voltage follower as the input

stage to an inverting-op-amp circuit. 95

7.7 Difference ampliﬁer. 96

7.8 Op amp current source. 98

7.9 Fancy summing circuit. 99

8.1 Generalized op amp inverting-ampliﬁer circuit. 102

xiii List of figures

8.2 Basic op amp differentiator. 102

8.3 Improved op amp differentiator. 103

8.4 Basic op amp integrator. 104

8.5 Improved op amp integrator. 104

8.6 Op amp logarithmic ampliﬁer. 105

8.7 Op amp exponential ampliﬁer. 105

8.8 Simple and improved versions of an op amp half-wave rectiﬁer. 109

8.9 Op amp follower with push–pull output-buffer power driver

with two feedback arrangements. 110

8.10 Block diagram showing how to build an ‘exponentiator’. 111

9.1 Poor comparator and 311 comparator. 114

9.2 311 comparator with 10 k series input resistor. 115

9.3 Schmitt trigger using 311 comparator. 116

9.4 RC relaxation oscillator using comparator. 118

9.5 Block diagram for the 555 timer IC. 119

9.6 555 timer IC used as an oscillator and as a one-shot or timer. 120

9.7 555 timer conﬁgured as an alarm. 121

9.8 Sine/cosine oscillator. 122

9.9 Active bandpass ﬁlter. 123

10.1 Logic levels for various 7400-family lines. 127

10.2 Labeling of 7400-series chips. 129

10.3 Standard logic gates with truth tables. 130

10.4 De Morgan’s theorems expressed symbolically. 131

10.5 Two-input diode gate. 132

10.6 Diode–transistor NAND gate using 2N3904s. 132

10.7 Schematic representation of an ‘enhancement-mode’

N-channel MOSFET. 134

10.8 Schematic representations of a CMOS inverter constructed

using one N-channel and one P-channel MOSFET. 135

10.9 Schematic representation of a CMOS NAND gate with LED

logic-level indicator. 136

10.10 Logic-level switch using either an SPST or SPDT switch and a

pull-up resistor. 137

10.11 Circuits for measuring the channel resistance as a function of

gate voltage. 138

11.1 Timing diagram with timing deﬁnitions for a rising-edge-triggered

ﬂip-ﬂop. 145

xiv List of figures

11.2 Simple RS latch made of two-input NANDs with state table. 146

11.3 7474 D-type ﬂip-ﬂop with state table. 147

11.4 Sample timing diagram for a (positive-edge-triggered) 7474 D-type

ﬂip-ﬂop. 147

11.5 Pinout of the 74112 JK ﬂip-ﬂop. 149

11.6 Pinout and power connections for the 74373 and input and output

connections for testing the tri-state output. 150

11.7 Divide-by-four ripple counter. 151

11.8 Synchronous divide-by-four counter. 152

11.9 Looking at contact bounce by driving a divide-by-four counter

from a switch. 153

12.1 Pinout of 7490 decade counter. 157

12.2 Pinout of TIL311 hex display. 158

12.3 Timing diagram for a gated clock signal. 160

12.4 Pinout of ’121 and ’123 one-shots with external RC timing network. 160

12.5 Substandard outputs resulting from gating clock signals. 161

12.6 Pinout of 74150 16-to-1 multiplexer. 163

12.7 Pinout of 7489 16×4 RAM. 163

13.1 Simple D/A converter and output waveform resulting from input

counting sequence. 168

13.2 Simple A/D converter. 171

13.3 Pinout for ADC080x series of A/D converters and the on-chip

self-clocking conﬁguration. 172

13.4 Pinout for DAC080x series of D/A chips. 175

13.5 Method for producing a DC-shifted waveform. 176

13.6 Control logic for 8-bit successive-approximation ADC. 179

13.7 8-bit successive-approximation ADC. 180

C.1 Series RC circuit. 193

C.2 Right triangle to illustrate Eq. C.17. 193

Tables

1.1 Digital multimeter inputs. page 2

1.2 Color code for nonprecision resistors. 7

2.1 Some typical dielectric materials used in capacitors. 16

3.1 A sample of commercially available diodes. 36

4.1 A sample of commercially available bipolar transistors. 50

10.1 Common families within the 7400 series. 128

xv

About the authors

Dr Daniel M. Kaplan received his Ph.D. in Physics in 1979 from the State

University of New York at Stony Brook. His thesis experiment discovered

the b quark, and he has devoted much of his career to experimentation

at the Fermi National Accelerator Laboratory on properties of particles

containing heavy quarks. He has taught electronics laboratory courses for

non-electrical-engineering majors over a ﬁfteen-year period at Northern

Illinois University and at Illinois Institute of Technology, where he is cur-

rently Professor of Physics and Director of the Center for Accelerator

and Particle Physics. He also serves as Principal Investigator of the Illinois

Consortiumfor Accelerator Research. He has been interested in electronics

since high school, during the junior year of which he designed a computer

based on DTL integrated circuits. Over more than twenty-ﬁve years in

experimental particle physics he has often been responsible for much of

his experiments’ custom-built electronic equipment. He is the author or

co-author of over 150 scientiﬁc papers and one encyclopedia article, and

co-editor of three books on heavy-quark physics and related ﬁelds.

Dr Christopher G. White is Assistant Professor of Physics at Illinois

Institute of Technology. He received his Ph.D. in Physics from the

University of Minnesota in 1990. He has authored or co-authored over

100 scientiﬁc articles in the ﬁeld of high-energy particle physics, and his

current research interests involve neutrinos and hyperons. Dr White is an

enthusiastic and dedicated teacher who enjoys helping students to over-

come their fear of electronics and to gain both conﬁdence and competence.

xvi

To the Reader

Some of you may be encountering electronic circuits and instruments for

the ﬁrst time. Others may have ‘played around’ with such stuff if, for

example, you were ever bitten by the ‘ham radio’ bug. In either case, this

sequence of laboratory experiments has been designed to introduce you to

the fundamentals of modern analog and digital electronics.

We use electronic equipment all the time in our work and recreation.

Scientists and engineers need to know a bit of electronics, for example to

modify or repair some piece of equipment, or to interface two pieces of

equipment that may not have been designed for that purpose. To that end,

our goal is that by the end of the book, you will be able to design and build

any little analog or digital circuit you may ﬁnd useful, or at least understand

it well enough to have an intelligent conversation about the problem with

an electrical engineer. A basic knowledge of electronics will also help you

to understand and appreciate the quirks and limitations of instruments you

will be using in research, testing, development, or process-control settings.

We expect few of you to have much familiarity with such physical theo-

ries as electromagnetismor quantummechanics, so the thrust of this course

will be from phenomena and instruments toward theory, not the other way

round. If your curiosity is aroused concerning theoretical explanations, so

much the better, but unfamiliarity with physical theory should not prevent

you from building or using electronic circuits and instruments.

xvii

Acknowledgments

We are grateful to Profs Carlo Segre and Tim Morrison for their contri-

butions and assistance, and especially to the IIT students without whom

this book would never have been possible. Finally, we thank our wives and

children for their support and patience. It is to them that we dedicate this

book.

xviii

Introduction

This book started life as the laboratory manual for the course Physics 300,

‘Instrumentation Laboratory’, offered every semester at Illinois Institute of

Technology to a mix consisting mostly of physics, mechanical engineering,

and aeronautical engineering majors. Each experiment can be completed

in about four hours (with one or two additional hours of preparation).

This book differs from existing books of its type in that it is faster paced

and goes into a bit less depth, in order to accommodate the needs of a one-

semester course covering the elements of both analog and digital electron-

ics. In curricula that normally include one year of laboratory instruction in

electronics, it may be suitable for the ﬁrst part of a two-semester sequence,

with the second part devoted to computers and computer interfacing – this

scheme has the virtue of separating the text for the more rapidly changing

computer material from the more stable analog and digital parts.

The book is also suitable for self-study by a person who has access to

the necessary equipment and wants a hands-on introduction to the subject.

We feel strongly, and experience at IIT has borne out, that to someone who

will be working with electronic instrumentation, a hands-on education in

the techniques of electronics is much more valuable than a blackboard-

and-lecture approach. Certainly it is a better learning process than simply

reading a book and working through problems.

The appendices suggest sources for equipment and supplies, provide

tables of abbreviations and symbols, and list recommendations for fur-

ther reading, which includes chapter-by-chapter correspondences to some

popular electronics texts written at similar or somewhat deeper levels to

ours: the two slim volumes by Dennis Barnaal, Analog Electronics for

Scientiﬁc Application and Digital Electronics for Scientiﬁc Application

(reissued by Waveland Press, 1989); Horowitz and Hill’s comprehensive

The Art of Electronics (Cambridge University Press, 1989); Diefenderfer

and Holton’s Principles of Electronic Instrumentation (Saunders, 1994);

xix

xx Introduction

and Simpson’s Introductory Electronics for Scientists and Engineers (2nd

edition, Prentice-Hall, 1987). There is also a glossary of terms and pinout

diagrams for transistors and ICs used within. The reader is presumed to

be familiar with the rudiments of differential and integral calculus, as well

as with elementary college physics (including electricity, magnetism, and

direct- and alternating-current circuits, although these topics are reviewed

in the text).

The order we have chosen for our subject matter begins with the basics –

resistors, Ohm’s law, simple AC circuits – then proceeds towards greater

complexity by introducing nonlinear devices (diodes), then active devices

(bipolar and ﬁeld-effect transistors). We have chosen to discuss transistors

before devices made from them (operational ampliﬁers, comparators, dig-

ital circuitry) so that the student can understand not only how things work

but also why.

There are other texts that put integrated circuits, with their greater ease

of use, before discrete devices; or digital circuits, with their simpler rules,

before the complexities of analog devices. We have tried these approaches

on occasion in our teaching and found them wanting. Only by considering

ﬁrst the discrete devices from which integrated circuits are made can the

student understand and appreciate the remarkable properties that make

ICs so versatile and powerful. A course based on this book thus builds

to a pinnacle of intellectual challenge towards the middle, with the three

transistor chapters. After the hard uphill slog, it’s smooth sailing fromthere

(hold onto your seatbelts!).

The book includes step-by-step instructions and explanations for the

following experiments:

1. Multimeter, breadboard, and oscilloscope;

2. RC circuits;

3. Diodes and power supplies;

4. Transistors I;

5. Transistors II: FETs;

6. Transistors III: differential ampliﬁer;

7. Introduction to operational ampliﬁers;

8. More op-amp applications;

9. Comparators and oscillators;

10. Combinational logic;

11. Flip-ﬂops: saving a logic state;

xxi Introduction

12. Monostables, counters, multiplexers, and RAM;

13. Digital↔analog conversion.

These thirteen experiments ﬁt comfortably within a sixteen-week

semester. If you or your instructor prefers, one or two experiments may

easily be omitted to leave a couple of weeks at the semester’s end for inde-

pendent student projects. To this end, Chapter 6, ‘Transistors III’, has been

designed so that no subsequent experiment depends on it; obviously this is

also the case for Chapter 13, ‘Digital↔analog conversion’, which has no

subsequent experiment.

As you work through the exercises, you will ﬁnd focus questions and

detailed instructions indicated by the symbol ‘

**’. Key concepts for each
**

exercise will be denoted by the symbol ‘

r

’. Finally, the standard system of

units for electronics is the MKS system. Although you may occasionally

run across other unit systems, we adhere strictly to the MKS standard.

1

Equipment familiarization: multimeter,

breadboard, and oscilloscope

In this chapter you will become acquainted with the ‘workhorses’ of elec-

tronics testing and prototyping: multimeters, breadboards, and oscillo-

scopes. You will ﬁnd these to be indispensable aids both in learning about

and in doing electronics.

Apparatus required

One dual-trace oscilloscope, one powered breadboard, one digital multi-

meter, two 10X attenuating scope probes, red and black banana leads, two

alligator clips.

1.1 Multimeter

You are probably already familiar with multimeters. They allow measure-

ment of voltage, current, and resistance. Just as with wristwatches and

clocks, in recent years digital meters (commonly abbreviated to DMM for

digital multimeter or DVM for digital voltmeter) have superseded the ana-

log meters that were used for the ﬁrst century and a half or so of electrical

work. The multimeters we use have various input jacks that accept ‘banana’

plugs, and you can connect the meter to the circuit under test using two

banana-plug leads. The input jacks are described in Table 1.1. Depending

on how you conﬁgure the meter and its leads, it displays

r

the voltage difference between the two leads,

r

the current ﬂowing through the meter from one lead to the other, or

r

the resistance connected between the leads.

Multimeters usually have a selector knob that allows you to select what is

to be measured and to set the full-scale range of the display to handle inputs

of various size. Note: to obtain the highest measurement precision, set the

knob to the lowest setting for which the input does not cause overﬂow.

1

2 Hands-on electronics

Table 1.1. Digital multimeter inputs.

Input jack Purpose Limits

a

COM reference point used for all measurements

V input for voltage or resistance measurements 1000 V DC/750 V AC

mA input for current measurements (low scale) 200 mA

10 A input for current measurements (high scale) 10 A

a

For the BK Model 2703B multimeters used in the authors’ labs.

To avoid damaging the meter, be sure to read the safety warnings in its

data sheet or instruction booklet.

1.2 Breadboard

‘Breadboard’ may seem a peculiar term! Its origins go back to the days

when electronics hobbyists built their circuits on wooden boards. The

breadboards we use represent a great step forward in convenience, since

they include not only sockets for plugging in components and connecting

themtogether, but also power supplies, a function generator, switches, logic

displays, etc.

The exercises that follow were designed using the Global Specialties

PB-503 Protoboard. If you do not have access to a PB-503, any suitable

breadboardwill do, providedyouhave a functiongenerator andtwovariable

power supplies. Additional components that you will need along the way

(that are built into the PB-503) include a 1 k and a 10 k potentiometer,

a small 8 speaker, two debounced push-button switches, several LED

logic indicators, and several on–off switches.

Fig. 1.1 displays many of the basic features of the PB-503. (For simplic-

ity, some PB-503 features that will be used in experiments in later chapters

have been omitted.) While the following description is speciﬁc to the

PB-503, many other breadboards share some, if not all, of these features.

The description will thus be of some use for users of other breadboard

models as well.

The breadboard’s sockets contain spring contacts: if a bare wire is pushed

into a socket, the contacts press against it, making an electrical connec-

tion. The PB-503’s sockets are designed for a maximum wire thickness of

22 AWG (‘American Wire Gauge’) – anything thicker (i.e., with smaller

3 1 Equipment familiarization

−15 V +15 V

+5 V

Horizontal Row

(Group of 5)

Function

Generator

Analog

Digital

Frequency

Slider

Amplitude

Slider

10 k pot 1 k pot

8

Speaker

Ω

Voltage Adjustment Knobs

+15 −15

SPDT Switches

Logic Switch Bank

Push Button

De-Bounced

Switches

Vertical Column

(Group of 25)

Fig. 1.1. Illustration showing many of the basic features of the PB-503 powered

Protoboard, with internal connections shown for clarity. Note that each vertical column is

broken into halves with no built-in connection between the top and bottom.

AWG number) may damage the socket so that it no longer works reliably

for thin wires. The PB-503 sockets are internally connected in groups of

ﬁve (horizontal rows) or twenty ﬁve (vertical columns; see Fig. 1.1).

Each power supply connects to a ‘banana’ jack and also to a row of

sockets running along the top edge of the unit. The three supplies, +5 V

(red jack), +15 V (yellow jack), and −15 V (blue jack), have a common

4 Hands-on electronics

‘ground’ connection (black jack). The +15 V and −15 V supplies are

actually adjustable, using the knobs provided, from less than 5 volts to

greater than 15 volts.

1.2.1 Measuring voltage

Voltage is always referenced to something, usually a local ground. For the

following exercises you will measure voltage with respect to the breadboard

ground, which is also the common ground for the three power supplies. To

measure a voltage, you will ﬁrst connect the ‘common’ jack of the meter to

the breadboard common (i.e., breadboard ground). Next you will connect

the meter’s ‘voltage’ jack to the point of interest. The meter will then tell

you the voltage with respect to ground at this one point.

When connecting things, it’s always a good idea to use color coding to

help keep track of which lead is connected to what. Use a black banana-

plug lead to connect the ‘common’ input of the meter to the ‘ground’ jack

of the breadboard (black banana jack labeled with a ‘ ’ or ‘ ’ symbol).

Use a red banana-plug lead with the ‘V’ input of the meter.

Since the DMM is battery powered, it is said to ‘ﬂoat’ with respect to

ground (i.e., within reason,

1

one may connect the DMM’s common jack to

any arbitrary voltage with respect to the breadboard ground). It is therefore

possible to measure the voltage drop across any circuit element by simply

connecting the DMM directly across that element (see Fig. 1.2).

Warning: This is not true for most AC-powered meters and oscilloscopes.

**To practice measuring voltages, measure and record the voltage between
**

each power supply jack and ground. In each case set the meter’s range

for the highest precision (i.e., one setting above overﬂow).

**Adjust the +15 V and −15 V supplies over their full range and record
**

the minimum and maximum voltage for each. Carefully set the +15 V

supply to a voltage half-way between its minimum and maximum for

use in the next part.

1

If you wonder what we mean by ‘within reason’, ask yourself what bad thing would happen if

you connected the DMM common to, say, twenty million volts – if you’re interested, see e.g.

H. C. Ohanian, Physics, 2nd edition, vol. 2, ‘Interlude VI’ (Norton, New York, 1988), esp.

pp. VI–8 for more information on this.

5 1 Equipment familiarization

DMM Power Supply

+

_

Power

Supply

Multimeter

V

+

VΩ mA COM A

Ground

(Common)

(b) (a)

1 F µ

Fig. 1.2. Measuring voltage. (a) An arbitrary circuit diagram is shown as an illustration of

how to use a voltmeter. Note that the meter measures the voltage drop across both the

resistor and capacitor (which have identical voltage drops since they are connected in

parallel). (b) A drawing of the same circuit showing how the leads for a DMM should be

connected when measuring voltage. Notice how the meter is connected in parallel with

the resistor.

DMM

Power Supply

+

_

Power

Supply

Multimeter

Potentiometer

Slider

(Center Tap)

Ground

(Common)

0.011 A

VΩ mA COM A

A

+

(b) (a)

Fig. 1.3. Measuring current. (a) Schematic diagram of series circuit consisting of power

supply, 10 k potentiometer, and multimeter. (Note that the center tap of the potentiometer

is left unconnected in this exercise – accidentally connecting it to power or ground could

lead to excessive current ﬂow and burn out the pot.) (b) A drawing of the same circuit

showing how the DMM leads should be conﬁgured to measure current. Note that the

meter is connected in series with the resistor.

1.2.2 Measuring current; resistance and Ohm’s law

Current is measured by connecting a current meter (an ammeter, or a DMM

in its ‘current’ mode) in series with the circuit element through which the

current ﬂows (see Fig. 1.3). Note carefully the differences between Fig. 1.2

and Fig. 1.3.

6 Hands-on electronics

Recall that Ohm’s law relates current I , voltage V, and resistance R

according to

V = IR. (1.1)

This is not a universal law of electrical conduction so much as a statement

that there exist certain materials for which current is linearly proportional

to voltage.

2

Materials with such a linear relationship are used to fabricate

‘resistors’: objects with a known and stable resistance. Usually they are

little cylinders of carbon, carbon ﬁlm, metal ﬁlm, or wound-up wire, en-

cased in an insulating coating, with wire leads sticking out the ends. Often

the resistance is indicated by means of colored stripes according to the

resistor color code (Table 1.2). Resistors come in various sizes accord-

ing to their power rating. The common sizes are

1

8

W,

1

4

W,

1

2

W, 1 W,

and 2 W.

You can easily verify this linear relationship between voltage and current

using the ﬁxed 10 k (10 000 ohm) resistance provided between the two

ends of one of the breadboard’s ‘potentiometers’. Apotentiometer is a type

of resistor that has an adjustable ‘center tap’ or ‘slider’, allowing electrical

connections to be made not only at the two ends, but also at an adjustable

point along the resistive material.

The ‘10 k pot’ (as it is called for short) is located near the bottom edge

of the breadboard, and can be adjusted by means of a large black knob.

3

Inside the breadboard’s case, the ends of the pot (as well as the center tap)

connect to sockets as labeled on the breadboard’s front panel. By pushing

wires into the sockets you can make a series circuit (Fig. 1.3) consisting of

an adjustable power supply, the 10 k pot, and the multimeter (conﬁgured

to measure current). You can attach alligator clips to the meter leads to

connect them to the wires. But, before doing so, be sure to observe the

following warnings:

r

First, turn off the breadboard power to avoid burning anything out if you

happen to make a mistake in hooking up the circuit.

r

Be careful to keep any exposed bits of metal from touching each other

and making a ‘short circuit’! Note that most of the exposed metal on

2

Of course, the existence of other materials (namely semiconductors) for which the I –V relationship

is nonlinear makes electronics much more interesting and underlies the transformation of daily life

brought about by electronics during the twentieth century.

3

If you don’t have a PB-503 breadboard, ﬁnd a 10 k pot on your breadboard if it has one; otherwise

you will have to purchase a separate 10 k pot.

7 1 Equipment familiarization

Table 1.2. Color code for nonprecision resistors (5, 10, or 20% tolerance).

The resistance in ohms is the sum of the values in columns 1 and

2, multiplied by the value in column 3, plus or minus the tolerance

in column 4. For example, the color code for a 1 k resistor would

be ‘brown--black--red’, for 51 ‘green--brown--black’, for 330

‘orange--orange--brown’, etc.

Stripe: 1 2 3 4 (tolerance)

Black 0 0 10

0

Brown 10 1 10

1

Red 20 2 10

2

Orange 30 3 10

3

Yellow 40 4 10

4

Green 50 5 10

5

Blue 60 6 10

6

Violet 70 7 10

7

Gray 80 8 10

8

White 90 9 10

9

Gold 5%

Silver 10%

None 20%

Stripe 1

Stripe 2

Stripe 3

Tolerance Stripe

the breadboard (screw heads for example) has a low-resistance path to

ground.

r

If you accidentally connect power or ground to the potentiometer’s center

tap, you can easily burn out the pot, rendering it useless! If in doubt, have

someone check your circuit before turning on the power.

**Use Ohm’s law to predict the current that will ﬂow around the circuit
**

if you use the power supply that you set to its midpoint in the previous

exercise. What current should ﬂow if the supply is set to its minimum

voltage? What is the current if the supply is set to its maximum voltage?

8 Hands-on electronics

**Now turn on the breadboard power, measure the currents for these three
**

voltages, and compare with your predictions. Make a graph of voltage vs.

current from these measurements. Is the relationship linear? How close

is the slope of voltage vs. current to 10 k?

1.2.3 Measuring resistance

Now turn off the breadboard power and disconnect your series circuit. In

this and the following part, the pot should connect only to the meter.

**Set the meter for resistance and measure and record the resistance
**

between the two ends of your 10 k pot. Due to manufacturing tolerances,

you will probably ﬁnd that it is not exactly 10 k. By what percentage

does it differ from the nominal 10 k value? Does the measured value

agree more closely with the slope you previously measured than with

the nominal value? Explain.

**Now connect the meter between the center tap and one end of the pot.
**

What resistance do you observe? What happens to the resistance as you

turn the potentiometer’s knob?

**Leavingthe knobinone place, measure andrecordthe resistance between
**

the center tap and each end. Do the two measurements add up to the total

you measured above? They should – explain why.

1.3 Oscilloscope

With its many switches and knobs, a modern oscilloscope can easily in-

timidate the faint of heart, yet the scope is an essential tool for electronics

troubleshooting and you must become familiar with it. Accordingly, the

rest of this laboratory session will be devoted to becoming acquainted with

such an instrument and seeing some of the things it can do.

The oscilloscope we use is the Tektronix TDS210 (illustrated in Fig. 1.4).

If you don’t have a TDS210, any dual-trace oscilloscope, analog or digital,

can be used for these labs as long as the bandwidth is high enough – ideally,

30 MHz or higher. While the description belowmay not correspond exactly

to your scope, with careful study of its manual you should be able to ﬁgure

out how to use your scope to carry out these exercises.

The TDS210 is not entirely as it appears. In the past you may have

used an oscilloscope that displayed voltage as a function of time on a

9 1 Equipment familiarization

TDS 210

60 MHz

1 GS/s

MEASURE

CURSOR

CH 1 CH 2

TRIGGER MENU

VERTICAL

CONTROLS

VOLTS/DIV

SEC/DIV

HORIZONTAL

CONTROLS

TRIGGER LEVEL

MENU

OPTION

BUTTONS

AUTOSET

CALIBRATION

CONTACT POINT

Trigger Info

Horizontal and Vertical Info

Menu

Options

MENU

Trigger Level

Menu

Options

Menu

Options

Menu

Options

Menu

Options

Volts per

Division

Seconds per

Division

Fig. 1.4. Illustration of the Tektronix TDS210 digital oscilloscope. The basic features to

be used in this tutorial are marked. Note and remember the location of the ‘autoset’

button – when all else fails, try autoset!

cathode-ray tube (CRT). While the TDS210 can performa similar function,

it does not contain a CRT (part of the reason it is so light and compact).

Until the 1990s, most oscilloscopes were purely ‘analog’ devices: an

input voltage passed through an ampliﬁer and was applied to the deﬂection

plates of a CRT to control the position of the electron beam. The position

of the beam was thus a direct analog of the input voltage. In the past few

years, analog scopes have been largely superseded by digital devices such

as the TDS210 (although low-end analog scopes are still in common use

for TV repair, etc.).

Adigital scope operates on the same principle as a digital music recorder.

In a digital scope, the input signal is sampled, digitized, and stored in

memory. The digitized signal can then be displayed on a computer screen.

One of your ﬁrst objectives will be to set up the scope to do some of

the things for which you may already have used simpler scopes. After

that, you can learn about multiple traces and triggering. In order to have

something to look at on the scope, you can use your breadboard’s built-in

function generator, a device capable of producing square waves, sinusoidal

waves, and triangular waves of adjustable amplitude and frequency. But

start by using the built-in ‘calibrator’ signal provided by the scope on a

metal contact labeled ‘probe comp’ (or something similar), often located

near the lower right-hand corner of the display screen.

10 Hands-on electronics

Note that a leg folds down from the bottom of the scope near the front

face. This adjusts the viewing angle for greater comfort when you are seated

at a workbench, so we recommend that you use it.

1.3.1 Probes and probe test

Oscilloscopes come with probes: cables that have a coaxial connector (sim-

ilar to that used for cable TV) on one end, for connecting to the scope, and

a special tip on the other, for connecting to any desired point in the circuit

to be tested. To increase the scope’s input impedance and affect the cir-

cuit under test as little as possible, we generally use a ‘10X’ attenuating

probe, which has circuitry inside that divides the signal voltage by ten.

Some scopes sense the nature of the probe and automatically correct for

this factor of ten; others (such as the TDS210) need to be told by the user

what attenuation setting is in use.

As mentioned above, your scope should also have a built-in ‘calibrator’

circuit that puts out a standard square wave you can use to test the probe

(see Fig. 1.4). The probe’s coaxial connector slips over the ‘ch 1’ or ‘ch 2’

input jack and turns clockwise to lock into place. The probe tip has a spring-

loaded sheath that slides back, allowing you to grab the calibrator-signal

contact with a metal hook or ‘grabber’.

An attenuating scope probe can distort a signal. The manufacturer there-

fore provides a ‘compensation adjustment’ screw, which needs to be tuned

for minimum distortion. The screw is usually located on the assembly that

connects the probe to the scope, or, occasionally, on the tip assembly.

**Display the calibrator square-wave signal on the scope. If the signal looks
**

distorted (i.e., not square), carefully adjust the probe compensation using

a small screwdriver. (If you have trouble achieving a stable display, try

‘autoset’.)

**Check your other probe. Make sure that both probes work, are prop-
**

erly compensated, and have equal calibrations. Sketch the observed

waveform.

(Consult your oscilloscope user manual for more information about car-

rying out a probe test.)

Note that each probe also has an alligator clip (sometimes referred to

as the ‘reference lead’ or ‘ground clip’). This connects to the shield of the

coaxial cable. It is useful for reducing noise when looking at high-frequency

11 1 Equipment familiarization

(time intervals of order nanoseconds) or low-voltage signals. Since it is

connected directly to the scope’s case, which is grounded via the third prong

of the AC power plug, it must never be allowed to touch a point in a circuit

other than ground! Otherwise you will create a short circuit by connecting

multiple points to ground, which could damage circuit components.

This is no trouble if you are measuring a voltage with respect to ground.

But if you want to measure a voltage drop between two points in a circuit,

neither of which is at ground, ﬁrst observe one point (with the probe) and

then the other. The difference between the two measurements is the voltage

across the element. During this process, the reference lead should remain

ﬁrmly attached to ground and should not be moved! (Alternatively, you

can use two probes and conﬁgure the scope to subtract one input from the

other.)

Warning: A short circuit will occur if the probe’s reference lead is connected anywhere other

than ground.

1.3.2 Display

Your oscilloscope user’s manual will explain the information displayed

on the scope’s screen. Record the various settings: timebase calibration,

vertical scale factors, etc.

**Explain brieﬂy the various pieces of information displayed around the
**

edges of the screen.

The following exercises will give you practice in understanding the vari-

ous settings. For each, you should study the description in your oscilloscope

user’s manual. The description belowis speciﬁc to the TDS210; if you have

a different model, your manual will explain the corresponding settings for

your scope.

1.3.3 Vertical controls

There is a set of ‘vertical’ controls for each channel (see Fig. 1.4). These

adjust the sensitivity (volts per vertical division on the screen) and offset

(the vertical position on the screen that corresponds to zero volts). The

‘ch 1’ and ‘ch 2’ menu buttons can be used to turn the display of each

12 Hands-on electronics

channel on or off; they also select which control settings are programmed

by the push-buttons just to the right of the screen.

**Display a waveform from the calibrator on channel 1. What happens
**

when you adjust the position knob? The volts/div knob?

1.3.4 Horizontal sweep

To the right of the vertical controls are the horizontal controls (see Fig. 1.4).

Normally, the scope displays voltage on the vertical axis and time on the

horizontal axis. The sec/div knob sets the sensitivity of the horizontal axis,

i.e. the interval of time per horizontal division on the screen. The position

knob moves the image horizontally on the screen.

**How many periods of the square wave are you displaying on the screen?
**

Howmanydivisions are there per period? What time interval corresponds

to a horizontal division? Explain how these observations are consistent

with the known period of the calibrator signal.

**Adjust the sec/div knob to display a larger number of periods. Now
**

what is the time per division? How many divisions are there per period?

1.3.5 Triggering

Triggering is probably the most complicated function performed by the

scope. To create a stable image of a repetitive waveform, the scope must

‘trigger’ its display at a particular voltage, known as the trigger ‘threshold’.

The display is synchronized whenever the input signal crosses that volt-

age, so that many images of the signal occurring one after another can be

superimposed in the same place on the screen. The level knob sets the

threshold voltage for triggering.

You can select whether triggering occurs when the threshold voltage is

crossed frombelow(‘rising-edge’ triggering) or fromabove (‘falling-edge’

triggering) using the trigger menu (or, for some scope models, using trigger

control knobs and switches). You can also select the signal source for the

triggering circuitry to be channel 1, channel 2, an external trigger signal,

or the 120 V AC power line, and control various other triggering features

as well.

Since setting up the trigger can be tricky, the TDS210 provides an

automatic setup feature (via the autoset button) which can lock in on

13 1 Equipment familiarization

almost any repetitive signal presented at the input and adjust the voltage

sensitivity and offset, the time sensitivity, and the triggering to produce a

stable display.

**After getting a stable display of the calibrator signal, adjust the level
**

knob in each direction until the scope just barely stops triggering. What

is the range of trigger level that gives stable triggering on the calibra-

tor signal? How does it compare with the amplitude of the calibrator

waveform? Does this make sense? Explain.

Next connect the scope probe to the breadboard’s function generator –

you can do this by inserting a wire into the appropriate breadboard socket

and grabbing the other end of the wire with the scope probe’s grabber. The

function generator’s amplitude and frequency are adjusted by means of

sliders and slide switches.

**Look at each of the waveforms available from the function generator:
**

square, sine, and triangle. Try out the frequency and voltage controls

and explain how they work. Adjust the function generator’s frequency

to about 1 kHz.

**Display both scope channels, with one channel looking at the output of
**

the function generator and the other looking at the scope’s calibrator

signal. Make sure the vertical sensitivity and offset are adjusted for each

channel so that the signal trace is visible.

What do you see on the screen if you trigger on channel 1? On channel 2?

**What do you see if neither channel causes triggering (for example, if the
**

trigger threshold is set too high or too low)?

**How does this depend on whether you select ‘normal’ or ‘auto’ trigger
**

mode? Why? (If you ﬁnd this confusing, be sure to ask for help, or study

the oscilloscope manual more carefully.)

1.3.6 Additional features

The TDS210 has many more features than the ones we’ve described so far.

Particularly useful are the digital measurement features. Push the measure

buttontoprogramthese. Youcanuse themtomeasure the amplitude, period,

and frequency of a signal. The scope does not measure amplitude directly.

How then can you derive the amplitude from something the scope does

measure?

**Using the measurement features, determine the amplitude, frequency,
**

and period of a waveform of your choice from the function generator.

14 Hands-on electronics

You can also use the on-screen cursors to make measurements.

**Use the cursors to measure the half-period of the signal you just
**

measured.

**Explain how you made these measurements and what your results were.
**

(Afeature that comes in particularly useful on occasion is signal averaging;

this is programmed via the acquire button and allows noise, which tends

to be random in time, to be suppressed relative to signal, which is usually

periodic.)

2

RC circuits

Capacitors are not useful in DC circuits since they contain insulating gaps,

whichare opencircuits for DC. However, for voltages that change withtime,

a simple series circuit with a capacitor and a resistor can output the time

derivative or integral of an input signal, or can ﬁlter out low-frequency or

high-frequency components of a signal. But before plunging into the world

of time-varying voltage and current (i.e., alternating-current circuits), we

explore the voltage-divider idea using direct current, since it gives us a

simple way to understand circuits containing more than one component in

series. Then we apply it to the analysis of RC circuits as ﬁlters. Note that

the series RC circuit can be analyzed in two different ways:

r

via the exponential charging/discharging equation, and

r

as an AC voltage divider.

Both approaches are valid – in fact, they are mathematically equivalent –

but the ﬁrst is more useful when using capacitors as integrators or differ-

entiators, whereas the second is more useful when analyzing low-pass and

high-pass ﬁlters. The ﬁrst is referred to as the time-domain approach, since

it considers the voltage across the capacitor as a function of time, and the

second as the frequency-domain approach, since it focuses on the ﬁlter

attenuation vs. frequency.

Apparatus required

Oscilloscope, digital multimeter, breadboard, 68 and 10 k resistors,

0.01 F ceramic capacitor.

2.1 Review of capacitors

As you may recall froman introductory physics course, a capacitor consists

of two parallel conductors separated by an insulating gap. The capacitance,

15

16 Hands-on electronics

Table 2.1. Some typical dielectric materials used in capacitors.

Material Dielectric constant (κ)

Vacuum 1.0

Air (at STP) 1.00054

Paper 3.5

Mica 5.4

Ceramic ≈100

C, is proportional to the area of the conductors, A, and inversely propor-

tional to their separation, s, multiplied by the dielectric constant, κ, of the

insulating material:

C = κ

0

A/s,

where, in the MKS system of units, A is in meters squared, s in meters, and

C in farads, abbreviated F (1 farad ≡1 coulomb per volt). (The constant of

proportionality is the so-called permittivity of free space and has the value

0

= 8.854 ×10

−12

F/m).

The farad is an impractically large unit: for a conductor area of 1 cm

2

and separation of 1 mm, with dielectric constant of order 1, the capaci-

tance is ∼ picofarads. To achieve the substantially larger capacitances (of

order microfarads) often found in electronic circuits, manufacturers wind

ribbon-shaped capacitors up into small cylinders and use insulators of high

dielectric strength, such as ceramics or (in the so-called electrolytic ca-

pacitors) special dielectric pastes, that chemically form an extremely thin

insulating layer when a voltage is applied. Table 2.1 gives dielectric con-

stants for some typical dielectrics used in capacitors.

Capacitors thus come in a variety of types, categorized according to

the type of dielectric used, which determines how much capacitance can

be squeezed into a small volume. Electrolytic and tantalum capacitors are

polarized, which means that they have a positive end and a negative end,

and the applied voltage should be more positive at the positive end than at

the negative end – if you reverse-voltage a polarized capacitor it can burn

out, or even explode! Paper, mica, and ceramic capacitors are unpolarized

and can be hooked up in either direction. The large dielectric constants of

the polarized dielectrics permit high capacitance values – up to millifarads

in a several-cubic-centimeter can.

17 2 RC circuits

2.1.1 Use of capacitors; review of AC circuits

The fundamental rule governing the behavior of capacitors is

Q = CV, (2.1)

where Q is the charge stored on the capacitor at a given time, V is the

voltage across the capacitor at that time, and C is the capacitance. Current

can ﬂow into or out of a capacitor, but only to the extent that the charge

on the capacitor is changing. In other words, the current into or out of a

capacitor is equal to the time derivative of the charge stored on it. You can

see the resemblance between Eq. 2.1 and Ohm’s law (Eq. 1.1). The key

difference is that, for a resistor, it is the time derivative of the charge that

is proportional to voltage, whereas for a capacitor it is the charge itself.

Capacitors are thus useful only in circuits in which voltages or currents

are changing in time, namely AC circuits. We will consider the response

of circuits to periodic waveforms; these can be characterized by their

frequency, f , and period, T, which of course are related by

T = 1/f,

as well as their angular frequency ω = 2π f . (If f is expressed in hertz ≡

cycles per second, then T is in seconds and ω is in radians per second.) A

periodic waveform is also characterized by its amplitude, which, assuming

the wave is pure AC(i.e. symmetric withrespect toground), is the maximum

voltage that it reaches.

There are an inﬁnite variety of AC waveforms, but to understand how

capacitors are used, it is sufﬁcient to focus on two: square waves and

sine waves. You have already encountered a square wave in the previous

chapter – the scope’s calibrator signal. A square wave of amplitude V

0

is a signal that oscillates back and forth between two voltage levels, one

at +V

0

and one at −V

0

, spending 50% of the time at each level. Note that

the peak-to-peak voltage is twice the amplitude:

V

p–p

= 2A.

A sine wave is a particularly important case because, by Fourier de-

composition, any periodic waveform can be represented as a sum of sine

waves of various amplitudes and frequencies. Most of AC circuit analysis

therefore concerns itself with the response of circuits to sine waves.

18 Hands-on electronics

V

p--p

V

0

V

0

T

Fig. 2.1. Representation of an arbitrary, periodic waveform with period T, amplitude V

0

and peak-to-peak voltage V

p–p

.

A sine wave can be represented mathematically by

V(t ) = V

0

sin(ωt +φ). (2.2)

If t is in seconds, this describes a voltage with amplitude V

0

changing sinu-

soidally in time at the rate of ω radians per second. The phase φ determines

the voltage at t = 0:

V(0) = V

0

sin φ. (2.3)

Now suppose such a voltage signal is applied to a capacitor; for sim-

plicity, we choose φ = 0, i.e. at t = 0 the voltage is zero. Since we are

assuming there is no resistance in the circuit, there is no possibility of a

voltage drop anywhere in the circuit.

1

Thus, at any moment in time, the

voltage across the capacitor must equal the voltage out of the signal source.

To ﬁnd the resulting current we can differentiate Eq. 2.1:

I =

dQ

dt

(2.4)

= C

dV

dt

(2.5)

= C

d

dt

(V

0

sin ωt ) (2.6)

= ωCV

0

cos ωt (2.7)

= ωCV

0

sin (ωt +90

◦

). (2.8)

We see that the current is also a sine wave, but shifted in time with a phase

shift of 90

◦

, i.e. the current leads the voltage by 90

◦

.

We can write an ‘Ohm’s-law equivalent’ for a capacitor, as long as it is

understood that we are talking about sinusoidal waveforms only:

V

0

= I

0

X

C

, (2.9)

1

Of course, this is only an idealized approximation, since in any real circuit there is at least the

resistance of the wires, and in practice any signal source has some internal resistance.

19 2 RC circuits

where I

0

is the amplitude of the current sine wave and X

C

≡ 1/ωC is the

capacitive reactance of the capacitor. The reactance is thus the effective

resistance of the capacitor. Note that it is frequency-dependent, in keeping

with our intuition that for DC a capacitor should look like an open circuit

(inﬁnite resistance), while at high frequency it should approach a short

circuit (zero resistance).

For completeness, we mention here the inductive reactance X

L

≡ ωL, where L is the

inductance of an inductor. Inductors are coils of wire and satisfy the equation

V = L

dI

dt

. (2.10)

Just as capacitors often employ a dielectric, inductors are often wound on a ferrite core

to increase their inductance. Note that the inductor equation relates the voltage across

an inductor to the derivative of the current through it, while the capacitor equation

(Eq. 2.1) relates the voltage across a capacitor to the integral of the current. Thus,

where the current through a capacitor leads the voltage across it, the current through

an inductor lags the voltage across it by 90

◦

. With respect to its function in a circuit,

an inductor can thus be thought of as the opposite of a capacitor. Whereas capacitors are

relatively small, light, cheap, and have negligible resistance, inductors tend to be large,

heavy, expensive, and have appreciable resistance. Nevertheless, they ﬁnd important

use in ﬁltering applications, e.g. bandpass ﬁlters, crossover circuits for hi-ﬁ speakers,

radio-frequency circuits, and so forth. In the interests of time we omit inductor exercises

from our course, but if you understand capacitors you will have very little difﬁculty in

applying inductors.

2.1.2 Types and values of capacitors

For some reason the various manufacturers’ conventions for marking ca-

pacitors are particularly confusing – probably it has to do with the fact that

many small-value capacitors are physically too small to permit much print-

ing on them. Some common sense is required. Keep in mind that 1 farad

is a huge unit! Most capacitors are in the picofarad and microfarad ranges,

and these are the two commonly used units. A physically large capacitor

that says ‘10M’ on it is usually 10 microfarads, not 10 millifarads (for some

reason, most manufacturers don’t want to print Greek letters, so they use

‘M’ instead of ‘’). A 10 millifarad cap would be labeled ‘10000M’. A

small capacitor that says just ‘10’ on it is 10 picofarads. The other important

number is the maximum operating voltage, which is usually printed on the

capacitor if there is room.

20 Hands-on electronics

Some small capacitors are labeled like resistors, either with a color code

or with numbers that mean the same thing. The ﬁrst digit of this capacitance

code is the tens, the second is the ones, the third is the power of 10, and

the units are picofarads. This is sometimes ambiguous – for example, a

capacitor that says ‘470’ could be 470 pF or 47 ×10

0

= 47 pF! Usually

the clue is the presence of a letter, following the capacitance code, that

indicates the tolerance – J for ±5%, K for ±10%, M for ±20%, etc. – so

that ‘470 K’ means 47 pF ±10%, whereas just 470 means 470 pF! Note

that there is no ambiguity if it says 471 – since normal capacitors are not

manufactured with enough precision to warrant a third signiﬁcant digit,

the ‘1’ must be the power of ten. When in doubt, you can always check it

out by putting it in an RC circuit with a known R value and measuring the

time constant (see below), or by plugging it into a capacitance meter, if you

have one.

2.2 Review of current, voltage, and power

Before we get started on RCcircuits, let us brieﬂy reviewpower dissipation

and component ratings – you need to understand these to avoid damaging

components.

Voltage is related to potential-energy difference. The voltage drop across

any circuit element is directly proportional to the change in energy of a

charge as it traverses the circuit element. Speciﬁcally,

1 volt = 1 joule/coulomb.

The potential energy (with respect to some reference point) is equal to the

voltage multiplied by the charge.

Current refers to the motion of charges. The current through a given

surface (e.g. the cross-section of a wire) is deﬁned as the net charge passing

through that surface per unit time. The unit for current is the ampere:

1 ampere = 1 coulomb/second.

The product of voltage and current has units of joules/second, otherwise

known as watts.

If the voltage drop across a circuit element equals the change in poten-

tial energy per unit charge, and the current equals the amount of charge

21 2 RC circuits

moving through the element per unit time, then their product equals the

power released within the device! The power dissipated within any device is

given by

P = IV. (2.11)

For resistive elements (or when an effective resistance can be deﬁned),

Eq. 2.11 can be combined with Ohm’s law to give:

P = IV = I

2

R = V

2

/R. (2.12)

Resistors, diodes, transistors, relays, integrated-circuit chips, etc., are

rated (in part) by their maximum allowed power. Exceeding these ratings

can have disastrous effects on your circuit, and may even cause a ﬁre! To

illustrate this point, our ﬁrst exercise will deliberately lead to the destruction

of a carbon-ﬁlm resistor.

2.2.1 Destructive demonstration of resistor power rating

Caution: In the following exercise, care must be taken to prevent burns.

The resistor in the following exercise will become very hot and may even

catch ﬁre (brieﬂy). Keep the body of the resistor well above the breadboard.

Do not touch the resistor with your ﬁngers. Remove the destroyed resistor

using pliers or a similar tool.

Be sure that the power is turned off, and construct the circuit shown in

Fig. 2.2 using a

1

4

watt carbon-ﬁlm resistor.

**Turn on the power and observe the effect on the resistor. Be sure to
**

turn off the power as soon as the resistor begins to smoke. Record your

observations and comments.

+

68 Ω

1/4 watt

Resistor

15 V

Fig. 2.2. This circuit can be used to demonstrate destructive power loading. Note that the

resistor will heat up rapidly.

22 Hands-on electronics

**Calculate the power that was dissipated by the resistor before it burned
**

out. What is the minimum resistor value that can be safely used in this

circuit? (Assume that only

1

4

watt resistors are available.)

**Calculate the current that ﬂowed through the resistor (before it burned
**

out).

Note that even though the voltage was low and the current was well

under 1 A, damage was nevertheless done! Because your body’s resistance

is large, lowvoltages can’t give you a shock, but in the wrong circumstances

they can still cause trouble. The key to safe work in electronics is always

to estimate power dissipations in components before turning on the power,

and to make sure you are not exceeding the ratings.

2.3 Potentiometer as voltage divider

The voltage-divider idea is very useful in analyzing almost all circuits,

so you will need to become thoroughly familiar with it. A resistive voltage

divider is simply two resistors in series (see Fig. 2.3). A voltage differ-

ence, V

in

, is applied across the two, and a smaller voltage, V

out

, results

at the junction between them. A potentiometer can be used as a vari-

able voltage divider, and you will now try this out using the breadboard’s

10 k pot.

Warning: You can easily burn out the pot in this exercise if you are not careful!

1. Turn off the power before hooking up the circuit!

2. If you accidentally connect the pot’s slider to ground while one end is

connected to the supply voltage (or vice versa) you can easily burn out

~

V V

R

2

R

1

V

R

2

R

1

V

~

potentiometer

V

Fig. 2.3. Three schematics representing a resistive voltage divider. In all cases you can

show using Ohm’s law that V

out

= V

in

R

2

/(R

1

+ R

2

). Note that the far right

representation is implemented using a potentiometer. In this case, the output voltage is

variable and ranges between ground and V

in

(depending on the position of the slider).

23 2 RC circuits

the pot – brieﬂy explain why this is true. (Hint: how much power can be

dissipated in the pot in such a situation?)

3. If you connect the multimeter on a current or resistance setting between

the slider and some other point in the circuit while the circuit is powered,

you can easily burn out the pot, since on these settings a meter can act

as a low impedance (short circuit).

2.3.1 DC voltage divider

First, use the ohmmeter setting of your multimeter to measure the resistance

between the slider and each end of the 10 k pot.

2

Turn the knob to set the

slider exactly half-way between the ends, using the meter to tell you when

you get there.

Next, use the voltmeter setting to adjust the variable positive power

supply to +10 V. Then, turn off the power, connect one end of the pot to

ground, and connect the other end to +10 V. Connect the meter (on the volt-

age setting) betweenthe slider andground. Double-checkyour connections.

**Without moving the slider or changing the supply voltage, turn on the
**

power, and measure the voltage between the slider and ground.

**What are the values of R
**

1

and R

2

? Using the voltage-divider equation

V

out

= V

in

R

2

R

1

+ R

2

, (2.13)

explain why the predicted output voltage is +5 V. How close to this

prediction is your measured voltage? What is the percentage error?

2.3.2 AC voltage divider

Now verify that a resistive voltage divider works the same way for AC as

for DC. Apply a sinusoidal signal from the function generator to the pot in

place of the +10 V DC.

**Lookat the functiongenerator’s output signal withthe scope andmeasure
**

its peak-to-peak voltage, amplitude, and r.m.s. voltage. (The scope’s

measure menu is useful here.) Look at the signal at the pot’s center tap –

what are the peak-to-peak, amplitude, and r.m.s. values there? Howdoes

the pot’s voltage-division ratio, R

2

/(R

1

+ R

2

), compare for DCand AC?

2

The 10 k pot is located near the center of the bottom edge of the PB-503 breadboard and is adjusted

by means of a large black knob. If you don’t have a PB-503, ﬁnd a 10 k pot on your breadboard if it

has one; otherwise you will have to use a separate 10 k pot.

24 Hands-on electronics

V

R

C

V V

in

C

R

~

Fig. 2.4. The voltage-divider concept works perfectly well for RC circuits. This circuit is

also known as a low-pass ﬁlter, or as a voltage integrator.

2.4 RC circuit

Now hook up a 10 k resistor and a 0.01 F ceramic capacitor in series.

Ground one end of the capacitor, connect the other end to the resistor, and

connect the other end of the resistor to a 500 Hz square wave from the

function generator (see Fig. 2.4). Display the input signal (output of the

function generator) on channel 1 of the scope, and the output signal (at

the junction of the resistor and capacitor) on channel 2.

What are the amplitudes of the input and output waveforms?

**Sketch the output waveform. You may recall from your general physics
**

that it obeys the equation

V(t ) = V

0

e

−t /RC

, (2.14)

where t = 0 corresponds to a rising or falling edge of the square wave.

**Show that the time to fall to 37% of the peak value (i.e. V(t ) = 0.37 V
**

0

)

is the ‘time constant’, RC, and determine RC using the scope – you

should ﬁnd your oscilloscope’s ‘cursor’ feature useful here. (Be sure to

set the time and voltage scales sufﬁciently sensitive to yield an accurate

measurement – what settings should you use, and why?)

**Based on the nominal component values, what do you predict for RC?
**

Is your measurement consistent with this prediction? What are the

tolerances of the components you are using – does this explain any

discrepancy?

2.5 RC circuit as integrator

Nowswitch the function-generator frequency to 50 kHz. Observe what hap-

pens to the output waveform’s shape and amplitude. This can be explained

25 2 RC circuits

quantitatively from Eq. 2.1

V

out

(t ) =

Q(t )

C

(2.15)

=

1

C

t

0

I (t )dt (2.16)

=

1

C

t

0

V

in

− V

out

R

dt (2.17)

≈

1

RC

t

0

V

in

dt, (2.18)

where V

out

is the voltage across the capacitor, and the approximation is

valid as long as V

out

V

in

.

**What does Eq. 2.18 predict when V
**

in

is a constant, as it is during half of

each period of the square wave? Carefully measure and sketch the output

waveform. Compare your observations with your expectations based on

Eq. 2.18 and explain your results.

**What output amplitude would you expect at 25 kHz? Change the input
**

frequency and see if your prediction is correct.

**At approximately what frequency will Eq. 2.18 cease to predict the
**

output waveform accurately? Change the input frequency and test your

prediction.

2.6 Low-pass filter

Now switch from a 50 kHz square wave to a 50 kHz sine wave. Since

Eq. 2.18 should still apply, the output waveform should be the integral of

a sine wave, i.e. a cosine wave.

**What does this imply about the phase shift between input and output?
**

Measure the phase shift: 360

◦

multiplied by the time t between the zero

crossing of the input signal and the zero crossing of the output signal,

divided by the period, or

φ = 360

◦

t

T

. (2.19)

(The cursors are useful here.)

**Is the measured phase shift consistent with your prediction? Does the
**

voltage across the capacitor lag or lead the current through it? Explain.

26 Hands-on electronics

The other way to analyze this circuit is as an AC voltage divider, using

Eq. 2.13 with R

2

replaced by the reactance of the capacitor, X

C

, and R

1

+

R

2

replaced by the total impedance, Z, of the resistor and capacitor in

series. Since the voltage across the capacitor is 90

◦

out of phase with the

current through the resistor, these add in Pythagorean fashion:

V

out

= V

in

X

C

Z

(2.20)

= V

in

X

C

R

2

+ X

2

C

(2.21)

= V

in

1

1 +(ωRC)

2

. (2.22)

We see that the attenuation (V

out

/V

in

) depends on the frequency.

**What attenuation do you observe at 50 kHz? Calculate what you expect,
**

and compare.

At the breakpoint or half-power frequency f

0

, ωRC = 1, and thus the

attenuation is 1/

√

2 = 0.707. This is also referred to as the −3 dB point,

since it is the frequency at which the output voltage is attenuated by 3 dB.

The breakpoint is a convenient way to parametrize simple ﬁlters. (The

decibel is a logarithmic measure of the ratio of two signals:

number of dB = 20 log

A

2

A

1

,

where A

1

is the amplitude of the ﬁrst signal and A

2

is the amplitude of

the second; in this case A

1

= V

in

is the amplitude of the output signal and

A

2

= V

out

is the amplitude of the input signal.)

**By varying f until the output amplitude is 70.7% of the input ampli-
**

tude, measure f

0

. Calculate what you expect, and compare with your

measurement.

**What are the attenuation and phase shift at low frequency, say 50 Hz?
**

Compare with the predictions of Eq. 2.22. Compare the phase shifts with

φ = arctan

R

X

C

. (2.23)

The phase shift at low frequency is easy to understand: in the limit of DC

the capacitor must act like an open circuit, i.e. inﬁnite impedance, and thus

does not affect the output signal. Conversely, in the high-frequency limit

the capacitor must look like a short circuit to ground, so the output signal

goes to zero and the phase shift becomes dominated by the capacitor.

27 2 RC circuits

2.7 RC circuit as differentiator

Nowinterchange the capacitor and resistor so that the input signal is applied

at the capacitor (see Fig. 2.5). Drive the circuit with a 50 Hz square wave.

**What waveform do you see at the output? What are the input and output
**

amplitudes?

You can think of the shape of the output in terms of the exponential RC

charging/discharging curve, with f 1/RC, or you can think of it as an

approximation to the derivative of the input signal. Mathematically, the

derivative of an ideal square wave would be inﬁnite at the voltage steps and

zero in between, but of course an electrical signal can never be inﬁnite! In

this circuit the voltage spikes are limited in size to twice the input amplitude.

Using Eq. 2.16,

V

out

= IR (2.24)

= R

dQ

dt

(2.25)

= RC

d(V

in

− V

out

)

dt

(2.26)

≈ RC

dV

in

dt

, (2.27)

where the approximation is again valid when V

out

V

in

. So, indeed, the

circuit puts out an approximation to the time derivative of the input signal.

You can see why the approximation of Eq. 2.27 breaks down in the case

of a square wave, since at the rising and falling edges of the square wave

V

out

> V

in

.

**What does Eq. 2.27 imply if the input is a triangle wave? Try it out and
**

compare quantitatively with what you expect.

**What does Eq. 2.27 imply if the input is a sine wave? Try it out and com-
**

pare quantitatively with what you expect. Sketch the output waveform.

~

V V

V

R

C

R

C

Fig. 2.5. High-pass ﬁlter or voltage differentiator.

28 Hands-on electronics

If you are surprised at all the wiggles on the output signal, you can

verify that they are real (as opposed to noise) using the signal-averaging

feature of the scope’s acquire menu. You’ve discovered a poorly kept

secret of function-generator design! The sine waveform is rather difﬁcult

to generate, and most function generators actually use an approximation to

it that is piecewise-linear around the peaks and valleys. The derivative of a

piecewise-linear function is a series of steps and plateaus.

2.8 High-pass filter

**What attenuation and phase shift do you observe with a 50 Hz sine wave
**

as input?

What about with a 50 kHz sine wave?

Why do these phase shifts make sense?

**Should the breakpoint frequency be any different in this conﬁguration
**

than in the low-pass ﬁlter? Check it and make sure. Compare your mea-

surements with

V

out

= V

in

R

Z

(2.28)

= V

in

R

R

2

+ X

2

C

(2.29)

= V

in

ωRC

1 +(ωRC)

2

. (2.30)

**Show that well below the breakpoint frequency, Eq. 2.30 predicts that
**

the output amplitude should increase linearly with frequency. Take a few

measurements to demonstrate that this prediction is correct.

2.9 Summary of high- and low-pass filters

For reference, here once again are the key equations describing high-pass

and low-pass RC-ﬁlter operation in the frequency domain.

High-pass:

V

out

V

in

=

ωRC

1 +(ωRC)

2

, (2.31)

φ = arctan

1

ωRC

. (2.32)

29 2 RC circuits

Low-pass:

V

out

V

in

=

1

1 +(ωRC)

2

, (2.33)

φ = arctan ωRC. (2.34)

Note, as expected, that at high frequency, the voltage-division ratio goes

to unity in the ﬁrst case and to zero in the second case, while the phase

shift goes to zero in the ﬁrst case and to 90

◦

in the second. Also, in both

cases, the breakpoint frequency is the point at which the two terms in the

Fig. 2.6. Right triangles depicting the relationships among input voltages (always

represented by the hypoteneuse of the triangle) and capacitor and resistor voltages for (a)

high-pass and (b) low-pass RC ﬁlters. In each case, the center diagram shows the isosceles

triangle representing the case f = f

0

; the triangles on the left are for a frequency well

below f

0

; and those on the right for a frequency well above f

0

.

30 Hands-on electronics

square-root are equal:

ω

0

RC = 1 (2.35)

f

0

=

ω

0

2π

(2.36)

=

1

2π RC

. (2.37)

These relationships are all illustrated in Fig. 2.6, which shows, for low-

and high-pass ﬁlters, how the phase and magnitudes of the voltages across

the resistor and capacitor are related, for a ﬁxed input voltage, in three

frequency regimes: f = f

0

and frequencies that are well below and well

above f

0

. See Appendix C for further discussion of frequency-domain

analysis of RC circuits.

3

Diodes

In this chapter we will explore semiconductor diodes and some circuits

using them. We’ve seen that resistors have a simple linear relationship

between the voltage across themand the current through them(Ohm’s law).

On the other hand,

r

diodes have an exponential relationship between current and voltage.

Mathematically this may seem much more complicated than Ohm’s law,

but we think you’ll agree that the idea as just stated is simple enough – it just

takes some getting used to! As we’ll see, an important consequence of the

exponential characteristic is that diodes conduct much more readily in one

direction than in the other. This makes them ideally suited for rectiﬁcation:

the conversion of AC into DC.

Apparatus required

Breadboard, oscilloscope, one or two multimeters, one 1N914 (or similar)

silicon signal diode, one 1N4001 (or similar) 1 Asilicon rectiﬁer diode, one

100 and one 10 k

1

4

W resistor, one 1 k 2 W resistor, power transformer

with 12.6 V r.m.s. output on each side of the center tap, one diode bridge

element, one 100 F electrolytic capacitor, and one 1000 F electrolytic

capacitor.

3.1 Semiconductor basics

Current will ﬂow through a material provided that there are charge carriers

free to move and an electric ﬁeld to move them. Conductors (such as

copper) have lots of charge carriers (electrons) ready to move in response

to the slightest electric ﬁeld. Insulators (such as diamond) possess very

few free charge carriers – all the electrons are tightly bound to the crystal

lattice, so that, even in the presence of a strong electric ﬁeld, no current

31

32 Hands-on electronics

ﬂows. Semiconductors (such as silicon and germanium) are somewhere in

between. The conductivity of a semiconductor can be enhanced through

doping, the deliberate inclusion of impurities within the semiconductor

lattice.

Silicon, for example, has four valence electrons, which are used to make

covalent bonds with neighboring silicon atoms. Phosphorus has ﬁve va-

lence electrons, and boron has three. In a silicon crystal, if a silicon atom

is replaced with a ‘donor’ material, such as phosphorus, an extra valence

electron becomes available that is loosely bound to the lattice. If an

‘acceptor’ material such as boron is substituted for silicon, a ‘hole’ appears

in the electron structure of the lattice. Doping silicon with a donor material

creates an ‘N-type’ semiconductor, whereas doping with an acceptor cre-

ates a ‘P-type’ semiconductor. (Note that, despite their names, these doped

semiconductors are electrically neutral: the ‘extra’ electrons in N-type ma-

terial are compensated for by the additional protons in the atomic nuclei

of the donors, while the ‘missing’ electrons in P-type are compensated for

by the missing protons in the acceptor nuclei.)

The ‘extra’ electrons within N-type material can move under the inﬂu-

ence of an electric ﬁeld. Thus, the dominant charge carriers are electrons;

i.e., N-type material has negative charge carriers.

For P-type material, electrons fromneighboring atoms can jump into the

holes, moving the holes from one place to another. The holes can migrate

in the direction of an electric ﬁeld. The charge motion is thus due to the

motion of the holes, i.e., P-type material has positive charge carriers.

If a junction between P-type and N-type semiconductor material is cre-

ated within a single crystal, in such a way that the crystalline structure is

preserved across the junction, the result is a junction diode. Electrons from

the N-region migrate across the junction into the P-region, ﬁlling holes

as they go. This creates a net charge build-up around the junction (see

Fig. 3.1) – positive in the N-region and negative in the P-region – leading

to an internal electric ﬁeld as shown. Once the holes are ﬁlled, the junction

region becomes devoid of charge carriers and thus acts as an insulator,

preventing further current ﬂow.

If an external ﬁeld is applied in the same direction as the internal ﬁeld,

the ‘depletion region’ (region around the junction devoid of charge carriers)

increases in size, so current does not ﬂow. On the other hand, if an external

ﬁeld is applied opposite to the internal ﬁeld, free charge carriers ﬂowtoward

the junction. Electrons ﬂowinto the N-type material fromthe metal contact

33 3 Diodes

N-type P-type

+

_

Internal Electric Field

Hole

Free Electron

Silicon, Germanium

Boron, Aluminum

Arsenic, Phosphorus

Fig. 3.1. Representation of a junction between P-type and N-type semiconductor material.

Free electrons from the N-region will migrate into the P-region, combining with holes.

N-type P-type

+

+

+

+

+

+

+

_

_

_

_

_

_

_

V

R

I

s

I

s

Electric Field

N-type P-type

V R

I

I

Electric Field

Free Electron

Hole

Depletion Region

Forward-Biased Diode Circuit

Reverse-Biased Diode Circuit

+

+

Fig. 3.2. Diode circuit symbol and biasing.

(see Fig. 3.2). New holes are created within the P-material as electrons

jump from the semiconductor to the metal contacts. At the junction, the

holes fromthe P-type material meet electrons fromthe N-type material and

combine. A PN junction thus allows current to ﬂow easily in one direction

but blocks current ﬂow in the reverse direction.

For such a diode the current I ﬂowing through the device is given

approximately by

I = I

s

(e

eV/nkT

−1) (3.1)

where I

s

(sometimes called I

R

or I

0

) is the ‘reverse saturation current’, e is

the electron charge, V is the voltage across the junction, n is an empirical

constant between 1 and 2, k is Boltzmann’s constant, and T is the junction

temperature in kelvin. For simplicity, we’ll assume for now that n = 1.

This dependence of current on voltage is illustrated in Fig. 3.3.

34 Hands-on electronics

V

diode

(Volts)

−0.2

0.2 0.4 0.6 0.8

2

4

6

8

( A) µ

(mA)

Forward

Current

Reverse

Current

−0.2 −20

Ge

Si

−0.4 −18

Reverse

Breakdown

Fig. 3.3. Typical current–voltage characteristics for germanium and silicon diodes; note

that the current scales in the forward and reverse directions differ by a factor of 10 000,

and that the voltage scale changes at large reverse voltage. If a large enough reverse

voltage is applied, the junction breaks down and allows a large reverse current to ﬂow (the

‘Zener effect’).

When the P-type material is at a more positive voltage than the N-type

material, the diode is said to be ‘forward-biased’; this corresponds to V > 0

in Fig. 3.3. When the P-type material is more negative than the N-type

material, the diode is said to be ‘reverse-biased’; this corresponds to V < 0

in Fig. 3.3.

Some useful approximations

In the forward-biased case, when V is greater than ≈100 mV or so, the ‘1’

in Eq. 3.1 becomes negligible compared with the exponential term, and

I ≈ I

s

e

eV/kT

. (3.2)

When the diode is reverse-biased and |V| is greater than ≈100 mV or

so, the exponential term is negligible, and the reverse current is almost

constant, with

I ≈ −I

s

. (3.3)

35 3 Diodes

3.2 Types of diodes

In addition to standard junction diodes, light-emitting diodes (LEDs),

Schottky diodes, and Zener diodes are also common. LEDs are junction

diodes typically made fromgalliumarsenide phosphide (GaAsP). They act

very much like silicon junction diodes except that they emit light when

conducting forward current and have forward voltage drops about twice as

large as silicon diodes. Infrared, red, orange, yellow, green, and blue LEDs

are commercially available.

Zener diodes are manufactured with controlled reverse-breakdown prop-

erties. Their forward characteristics are similar to those of junction diodes;

however, Zener diodes are used in reverse-biased mode. While reverse

breakdown typically destroys a standard junction diode, Zener diodes are

designedtooperate at andaroundtheir reverse-breakdown(‘Zener’) voltage.

The Zener voltage is determined during the manufacturing process by ad-

justing the semiconductor doping. Typical Zener voltages range from 3.3

to 75 volts.

Schottky diodes are manufactured by bonding a metal conductor to an

N-type semiconductor. Electrons fromthe N-type material migrate into the

metal. This migration creates a potential barrier across the boundary, which

then behaves in a similar fashion to a PN junction. In general, Schottky

diodes are used in applications requiring high speed and low capacitance.

Physically, most diodes look like a little cylinder with wires sticking

out the two ends (Fig. 3.4). To distinguish the ends from each other, the

manufacturer often prints the diode circuit symbol on the diode body.

Alternatively, sometimes a ring is marked around the body close to the

‘cathode’ (the N-type end), to distinguish it from the ‘anode’ (the P-type

end). Diodes are manufactured with speciﬁed values for maximumcurrent,

forward voltage drop, leakage current (reverse saturation current), reverse-

breakdown voltage, and switching speed (time required for the diode to

Fig. 3.4. Representation of physical diodes along with the symbols used in circuit

diagrams.

36 Hands-on electronics

Table 3.1. A diverse selection of diodes is commercially available, of which a tiny sampling is

given here.

Diodes are commonly rated by their switching speed, maximum power dissipation, maxi-

mumforward current, maximumforward voltage at a speciﬁed forward current, and reverse-

breakdown voltage. The junction capacitance is sometimes listed as well.

P

max

I

F

max

V

F

max

@ I

F

V

BR(R)

C

Diode Type (W) (A) (V @ A) (V) (pF)

1N914 small signal 0.5 0.3 1.0 @ 0.01 75 4.0

1N4001 rectiﬁer 1.0 1.1 @ 1.0 50 8.0

1N4004 rectiﬁer 1.0 1.1 @ 1.0 400 8.0

1N5402 rectiﬁer 3.0 1.1 @ 3.0 200 40

FR601 fast rectiﬁer 6.0 1.3 @ 6.0 50 200

MBD301 Schottky 0.28 0.1 0.6 @ 0.01 30 1.0

1N4733A Zener 1.0 1.2 @ 0.2 5.1

switch fromforward to reverse bias or vice versa). Afewexamples are given

in Table 3.1.

3.3 Rectification

A rectiﬁer is a device that converts AC to DC by blocking the ﬂow of

current in one direction. Rectiﬁcation used to be almost the exclusive

province of vacuum tubes, with the exception of the ‘detector crystals’

(naturally occurring semiconductor diodes) used in crystal sets in the early

days of radio. Nowadays, semiconductor diodes are universally used for the

purpose.

An ideal rectiﬁer would offer zero resistance when forward-biased, i.e.,

the voltage across it would be zero, independent of the amount of forward

current ﬂow. It would offer inﬁnite resistance when reverse-biased, i.e., no

current would ﬂow, regardless of the size of the applied reverse voltage.

You can see from Fig. 3.3 that real diodes typically approximate the ideal

reasonably well, with only a few hundred millivolts of forward voltage

over a wide range of forward current, and with reverse current measured in

nanoamps, even for volts of reverse voltage (as long as the reverse voltage

is kept small enough to avoid breakdown). For the typical range of currents

encountered in most electronic circuits (a few to hundreds of milliamps), a

37 3 Diodes

handy approximation is that a forward-biased germanium diode has about

a 300 mV voltage drop across it, while a forward-biased silicon diode has

about a 600 mV drop.

3.4 Diode action -- a more sophisticated view

If we want, we can think of a diode as a resistor whose resistance depends

on the current ﬂowing through it, i.e., the resistance is dynamic rather than

having a constant (or static) value. While the static resistance of a device is

the voltage across the device divided by the current through it (R = V/I ,

Ohm’s law), the dynamic resistance is the slope dV/dI of the V–I curve at

any point. You can see that, for a resistor, the static and dynamic resistances

are the same, but for a device whose V–I curve is nonlinear, they become

different. Note in particular that a nonlinear device does not have a static

resistance, since V/I is not constant.

We can ﬁnd the dynamic resistance of a diode by differentiating Eq. 3.1:

dI

dV

=

e

kT

I

s

e

eV/kT

; (3.4)

thus, the dynamic resistance is

dV

dI

=

kT/e

I

s

e

eV/kT

(3.5)

≈

kT/e

I

, (3.6)

where the approximation is valid for forward-biasing such that the ‘1’ in

Eq. 3.1 is negligible. Note that at T = 300 K (room temperature),

e

kT

=

1.6 ×10

−19

C

(1.38 ×10

−23

J/K) ×300 K

= 39 V

−1

≈

1

25 mV

, (3.7)

so the dynamic resistance of a forward-biased diode can be simply approx-

imated by

dV

dI

≈

25 mV

I

. (3.8)

These results – the exponential dependence of current on voltage and

the consequent dynamic-resistance formula – are important to remember,

38 Hands-on electronics

as they also characterize the behavior of transistors. (This model of diode

behavior neglects some features that matter in practice. For example, the

semiconductor has some ‘ohmic’, or static, resistance that is in series with

the junction resistance just described, so in practice the dynamic resistance

is somewhat larger than that given by Eq. 3.8.)

3.5 Measuring the diode characteristic

As before, be careful not to burn out your breadboard’s 1 k pot: hook up

the following circuits with the power off, and double-check each circuit

before powering it up.

**For the circuit shown in Fig. 3.5, estimate the maximum current in your
**

circuit and the maximum power dissipation in the 100 resistor – is the

resistor’s

1

4

Wrating safe for its worst-case power dissipation? Caution:

These small diodes are easily damaged by overcurrent. To be on the safe

side, do not let the forward current exceed 50 mA.

**With a 1N914 (or similar) silicon diode, forward-biased as shown in
**

Fig. 3.5, increase the voltage across the diode starting from 0 V in steps

of 100 mV and record the diode current in each case.

For the portion of the characteristic curve close to the origin, the mi-

croamp range of the multimeter will be required to measure the forward

current. If you don’t have a second meter available, you can use your os-

cilloscope to measure the diode voltage.

**Plot your results for I vs. V on a linear scale – do they seemqualitatively
**

consistent with the functional shape of Eq. 3.1?

When unity is negligible in comparison with the exponential term of

Eq. 3.1, Eq. 3.2 may be re-expressed as

ln I ≈ ln I

s

+

e

kT

V. (3.9)

**Plot ln I vs. V. Is ln I approximately linear in V? How does its slope
**

compare with e/kT? What value for n (from Eq. 3.1) is implied by your

measured slope?

**Nowreverse the diode – for a reverse voltage of 5 Vwhat reverse current
**

do you observe? Is it consistent with the range of I

s

expected for a silicon

diode?

39 3 Diodes

0.64 V

+5 V

V

+

VΩ mA COM A

Ground

(Common)

mA

100 Ω

1 k pot

+5 V

+

100 Ω

1 k pot

1.2 mA

VΩ mA COM A

Z

in

R

in

Ammeter

DMM or

Oscilloscope

(b)

(a)

fuse

Fig. 3.5. (a) Measuring the forward characteristic of a diode. (b) When used to measure

current, the DMM is equivalent to an ideal ammeter in series with a small input

impedance. Most ammeters will also have a series fuse to protect the meter. When

measuring voltage, the DMM or oscilloscope looks like an ideal voltmeter in parallel

with a large input impedance.

If the reverse current seems to be much bigger than you expect, consider

that you have a voltage-measuring device (a scope or voltmeter) in parallel

with the diode (Fig. 3.5(b)).

**Disconnect the scope or meter – nowhowmuch current do you observe?
**

Reconnect it and disconnect the diode – how much current ﬂows with

the scope or meter alone? What do you infer to be the input resistance of

the scope or meter? Explain by applying Ohm’s law to relate the voltage

being measured to the current you observe.

Keep this experience in mind – it is often necessary to consider the effect

of your measuring device on the circuit being studied. Can you see why

an ideal voltmeter would have inﬁnite resistance, while an ideal ammeter

would have zero resistance?

40 Hands-on electronics

3.6 Exploring rectification

Next we take up the basic principles of rectiﬁcation. Almost all electronic

equipment requires power from a steady voltage source, i.e., a DC power

supply. For portable equipment, the power is supplied by batteries. How-

ever, the most convenient power source is the 120 V60 Hz ACline.

1

(120 V

is in fact the r.m.s. value of the sinusoidal voltage, the amplitude being

V

0

=

√

2 ×120 V = 170 V, (3.10)

and the peak-to-peak voltage is, of course, twice this, or V

p−p

= 340 V,

as you can easily verify from the deﬁnition of the root-mean-square by

integrating over the sine wave.)

Within most electronic equipment using the AC line, there is a power

transformer that steps down the 120 V AC to a more convenient voltage,

a rectiﬁer that converts the alternating voltage from the transformer to a

DCvoltage, and a regulator that maintains the output voltage at the desired

level.

Caution: In using a power transformer, bear in mind that an especially large transient current

sometimes flows when the line cord is first plugged in.

You will probably blow fewer fuses if you leave the power transformer

plugged in at all times. Attach banana-plug leads to the transformer’s sec-

ondary only after you are sure your circuit will not damage any of the

equipment. Do not permit powered lines to dangle loosely; when reconﬁg-

uring your circuit, it is safest to disconnect the leads at the transformer, not

at the breadboard.

**Set up the circuit shown in Fig. 3.6(a) using a 10 k resistor as the load,
**

R

L

. Observe the sinusoidal voltage waveform across R

L

. Measure the

amplitude V

0

and the r.m.s. voltage. Check the relation

√

2V

rms

= V

0

. (3.11)

You will probably ﬁnd V

rms

> 25 V. Since the windings of the transformer

have some ohmic resistance, the transformer’s output voltage depends on

1

In North America, the supply voltage froma standard wall socket is 120 V, and the supply frequency

is 60 Hz; the discussion is equally valid for other values, which may be substituted according to

your local supply voltage and frequency.

41 3 Diodes

Center

Tap

V

out

R

L

Transformer

V

rms

V

p--p

V

0

V

0

fuse

120

VAC

(a)

(b)

Fig. 3.6. (a) Power transformer supplies V

out

≈ 25 V r.m.s.; (b) waveform produced by

the circuit in (a).

the current drawn, and its 25.2 V r.m.s. nominal output voltage is for sub-

stantially higher current than is drawn by the 10 k load.

**Add a 1N4001 diode to give the half-wave rectiﬁer of Fig. 3.7(a) with
**

R

L

= 10 k.

**Observe and record the voltage waveform. Measure the amplitude V
**

0

using the oscilloscope (due to the rectiﬁcation it is now equal to the

peak-to-peak voltage).

**Compare the amplitude of the half-rectiﬁedwaveformwiththe amplitude
**

of the unrectiﬁed waveform measured above. By how much has the

amplitude decreased? Is this the amount you expect? Explain.

**Measure the average voltage V
**

av

across the load with a DC voltmeter.

Check that for a half-wave rectiﬁer

V

av

=

V

0

π

. (3.12)

**Add a ﬁlter capacitor in parallel with the load as shown in Fig. 3.8(a).
**

Caution: The 100 F electrolytic capacitor is polarized -- be careful not to hook it up

backwards! The negative terminal should be labeled with a ‘−’ sign.

42 Hands-on electronics

V

out

R

L

fuse

120

VAC

Transformer

V

av

V

p--p

V

0

Rectifier Diode

(a)

(b)

Fig. 3.7. (a) Power transformer with half-wave rectiﬁcation; (b) waveform produced by

circuit shown in (a).

V

out

R

L

Transformer

Rectifier Diode

+ 100 F µ

1

60

s

fuse

120

VAC

(a)

(b)

Ripple Voltage

1

60

s

(c)

Ripple Voltage

Fig. 3.8. (a) Half-wave rectiﬁer with ﬁlter capacitor; (b) waveform produced by circuit

shown in (a); (c) simple approximation to waveform produced by circuit shown in (a).

43 3 Diodes

**What is the voltage rating of your capacitor? Make sure it is sufﬁcient
**

for the voltage that will be applied!

**Observe and record the output-voltage waveform across the load resis-
**

tance R

L

and measure the peak-to-peak ‘ripple voltage’ (i.e., the amount

by which the output voltage is varying; see Fig. 3.8(b)).

Here is that rare situation – measuring accurately a small AC signal on

top of a large DC offset – in which you should use the AC-coupling feature

of the scope’s vertical menu. If you are troubled with noise, you may want

to trigger the scope on ‘line’ and employ signal averaging.

A simpliﬁed analysis approach for predicting the expected output wave-

form, illustrated in Fig. 3.8(c), is to assume that the capacitor charges up to

the peak voltage instantaneously and discharges at a uniform rate dQ/dt ,

equal to the average load current. The average current through the load can

be determined using the known resistance R

L

and the average DC voltage

across it as measured with a voltmeter. Using these assumptions, and the

fundamental capacitor equation Q = CV,

**Calculate the output-voltage droop in each cycle, and compare with the
**

peak-to-peak ripple voltage as measured. Is the observed percentage

discrepancy within the tolerances of your components? Explain.

**Replace the 100 F electrolytic capacitor with a 1000 F capacitor. Do
**

you expect the ripple voltage to increase or decrease? Explain.

**Measure the ripple voltage and compare with your expectations.
**

Full-wave rectiﬁcation should decrease the ripple by about a factor of

two. This can be accomplished using two diodes and the transformer’s

center tap or by using a diode bridge. Diode-bridge rectiﬁers are available

as a single encapsulated unit, making the bridge-rectiﬁer approach partic-

ularly convenient. These bridges have four diodes within. The terminals

are labeled: ‘∼’ marks the two terminals that should be connected to the

transformer secondary, and ‘+’ and ‘−’ denote the positive and negative

outputs (see Fig. 3.9).

~

~ +

_

Diode Bridge

Fig. 3.9. An example of how to insert a diode bridge into a breadboard.

44 Hands-on electronics

V

out

R

L

Transformer

Diode Bridge

+

_

~ ~

fuse

120

VAC

Fig. 3.10. Full-wave rectiﬁcation using a diode bridge.

Caution: A defective bridge element can blow the power transformer fuse -- check it be-

fore placing it in service. It should show essentially infinite resistance between the terminals

marked ‘∼’. When using an ohmmeter to check the resistance, remember to measure it for

both orientations of the terminals -- since you are dealing with diodes, the resistance could

be different in each direction.

**Set up the bridge rectiﬁer circuit of Fig. 3.10 with R
**

L

= 10 k. (Insert

the rectiﬁer package straddling the central groove of a breadboard socket

unit, with the long dimension of the package running along the groove, as

shown in Fig. 3.9.) Four rectiﬁer diodes can be used if a bridge rectiﬁer

is not available. As before, observe and record the voltage waveform

across R

L

.

**Add a ﬁlter capacitor to the full-wave rectiﬁer as shown in Fig. 3.11(a) –
**

again, be careful not to connect the capacitor backwards. This form

of power supply is very common. Measure the average output voltage

across the 10 k load. Record the peak-to-peak ripple voltage.

**Repeat these measurements for R
**

L

= 1 k. Caution: What power rating

must the 1 k resistor have, and why? Use socket adapters to handle the

fat leads of the 2 W resistor.

**Repeat the ripple voltage calculations for these two values of R
**

L

, keeping

in mind that the ﬁlter-capacitor discharge time is now one-half of the

AC cycle.

To make this circuit into a ‘complete’ power supply, one would want to

regulate the output, that is, employ feedback to make the output voltage

and ripple less dependent on the load resistance. This could be done using

Zener diodes, but a more effective technique is a transistorized regulating

45 3 Diodes

V

out

R

L

Transformer

Bridge Rectifier

+ 1000 F µ

1

120

s

+

_

~ ~

Ripple Voltage

fuse

120

VAC

(a)

(b)

Fig. 3.11. (a) Full-wave rectiﬁcation with ﬁlter capacitor; (b) waveform produced by

circuit shown in (a).

circuit. Integrated three-terminal voltage regulators (such as the 7800 and

7900 series) have made this particularly simple.

3.7 Input and output impedance

Input and output impedance are key ideas that are used all the time in

analyzing circuits. You’ve already encountered the input impedance of the

scope or voltmeter in section 3.5.

A good way to think about the effect of an instrument on the circuit

to which it is connected is via the instrument’s Th´ evenin equivalent. The

Th´ evenin equivalent of the multimeter (when set to measure voltage) is a

large resistor in parallel with an ideal voltmeter (Fig. 3.5). In practice, an

input also has some small capacitance and inductance and hence is more

completely characterized by its impedance vs. frequency, which takes into

account both the resistance and the capacitive and inductive reactances.

Outputs can also be characterized by their impedance (see Fig. 3.12).

You’ve already taken data that determine the output impedance of your

ﬁltered full-wave rectiﬁer circuit.

From

Z

out

= −V

out

/I

out

, (3.13)

46 Hands-on electronics

V

out

R

L

Z

out

Complete Rectifier Circuit

I

V

0

Ideal

Voltage

Source

Output

Impedance

+

Fig. 3.12. A rectiﬁer circuit can be modeled (as a Th´ evenin equivalent) using an ideal

voltage source in series with an output impedance. The output impedance is measured by

observing the output voltage as a function of the output current: Z

out

= −V

out

/I

out

.

and your data on V

out

vs. R

L

, compute the circuit’s output impedance,

Z

out

, in ohms.

**As another example, determine the output impedance of your bread-
**

board’s function generator, by measuring its sine-wave output amplitude,

ﬁrst with no load, and then with a load resistance of 1 k to ground.

If the output impedance has negligible frequency dependence, it can be

approximated as a pure resistance, in which case the function generator’s

Th´ evenin-equivalent circuit consists of an ideal AC voltage source (one

having zero internal resistance) in series with a single resistor.

**Check the function generator’s Z
**

out

both at low and high frequencies

(say 50 Hz and 50 kHz) – do you observe any appreciable frequency

dependence?

**Sketch schematic diagrams (with component values labeled) of the
**

Th´ evenin-equivalent circuits of your voltmeter, full-wave-rectiﬁedpower

supply, and function generator.

4

Bipolar transistors

Invented in 1947, transistors (and integrated circuits made fromthem) have

been the basis for the explosive proliferation of electronic devices that

revolutionized so much of life in the latter half of the twentieth century.

Although discrete (i.e. individually packaged) transistors are now used

mainly in special situations (e.g. where high power or speed is required),

since transistors form the basis of a large class of integrated circuits, an

understanding of how they work remains valuable. This will be the subject

of the next few chapters. This chapter will introduce you to some basic

bipolar-junction-transistor circuits.

Apparatus required

Breadboard, oscilloscope, two multimeters, 2N3904 and 2N3906 transis-

tors, red light-emitting diode (LED), 1N914 (or similar) silicon signal

diode, two 330 , two 10 k, and one each of 100 , 1 k, 3.3 k, 22 k,

and 100 k

1

4

W resistors, 1 F capacitor.

4.1 Bipolar-junction-transistor basics

Why and how transistors work is a bit subtle and can easily confuse the

beginning student, but it is something you must master. Study the following

description carefully, and compare it with the descriptions in other books.

You may also want to re-read both our description and others after you’ve

had some experience building and analyzing transistor circuits. (If you want

more of the background detail on semiconductor physics, good places to

look are Simpson’s Introductory Electronics for Scientists and Engineers,

or any textbook on modern physics.)

A bipolar junction transistor consists of two PN junctions sandwiched

very close together within a single crystal of semiconductor (Fig. 4.1(a)).

47

48 Hands-on electronics

P P

P N N

N

E E

E E

B

B

B

B

C C

C C

R

E

R

C

R

E

R

C

(a)

(b)

Fig. 4.1. (a) Construction and (b) circuit symbols and biasing examples for NPN and PNP

junction transistors.

The region common to the two junctions, called the base, may be of either

N-type or P-type material. This thin region is surrounded by material of the

opposite type, in regions known as the emitter and collector. Wire leads

are attached to the three regions.

The circuit symbols for NPN and PNP junction transistors are shown in

Fig. 4.1(b). Note that, in the circuit symbol, the arrow on the emitter lead

points in the direction of positive current ﬂow. You can tell whether a tran-

sistor in a schematic diagram is PNP or NPN by the direction of the arrow.

The simplest way to think of transistor action is as current ampliﬁca-

tion: a small current ﬂowing into the base controls a large current ﬂowing

into the collector. Both the base and collector currents ﬂow out from the

emitter. (This description assumes an NPN transistor. For PNP, the current

directions are opposite: a small current ﬂowing out from the base controls

a large current ﬂowing out from the collector, with both currents ﬂowing in

through the emitter.) The ratio of collector current to base current is called

β (or h

fe

) and is typically in the range 20 to 300.

More precisely, however, a transistor is a voltage-controlled current

source: small changes in the base voltage cause large changes in collec-

tor current. Such a device, in which an input voltage controls an output

current, is called a transconductance ampliﬁer. The transconductance (g

m

)

for a given device is deﬁned as the change in output current per change in

control voltage and has units of (ohm)

−1

(otherwise known as a mho).

To understand the operation of an NPNtransistor in more detail, it is con-

venient to consider the ﬂow of electrons, since electrons are the ‘majority

carriers’ in the N-type regions. Note that the ﬂow of electrons is of course

opposite in direction to the ﬂowof conventional positive current. As above,

49 4 Bipolar transistors

(lightly doped) (heavily doped)

Base

N-type P-type N-type

Electric Field Electric Field

Emitter Collector

V

CB

V

BE

Depletion

Region

I

C

I

E

I

B

Free Electron

Hole

+ +

Fig. 4.2. Schematic representation of how an NPN transistor operates. External bias

voltages create an electric ﬁeld, which pulls electrons (emitted into the base by the

emitter) across the base and into the collector. This results (seemingly paradoxically) in a

large ﬂow of electrons through the (reverse-biased) base–collector junction, a current that

is easily controlled by small changes in base voltage. The large hollow arrow represents

the ﬂow of electrons from the emitter to the collector.

the following description applies also to PNP transistors, but with the

current carriers and directions reversed.

In normal transistor operation, the base–emitter diode is forward-biased

and the base–collector diode is reverse-biased (Fig. 4.2). The depletion

region between the base and the collector extends essentially throughout

the thin base region (creating an electric ﬁeld as shown in Fig. 4.2) and

blocks the ﬂow of majority current carriers – holes ﬂowing from base to

collector and electrons ﬂowing through the collector to the base. At the

same time, the emitter lead injects electrons into the emitter, which ﬂow

across the (forward-biased) base–emitter junction. While (as just stated)

the base–collector bias inhibits the ﬂow of holes from the base into the

collector, the electrons with which the base is now ﬁlled are drawn by the

electric ﬁeld through the junction and into the collector. They do this even

though the base–collector junction is reverse-biased. This is the essence

of transistor action. Essentially, the construction of the transistor results

in large numbers of the ‘wrong’ current carrier entering the base and then

continuing ‘downhill’ into the collector.

Typically, ≈99%of electrons entering the base fromthe emitter continue

into the collector, and only ≈1% emerge as base current. The base current

results from the small fraction of electrons entering the base that combine

50 Hands-on electronics

Table 4.1. A diverse selection of bipolar transistors is commercially available, of which a small

sampling is given here.

Transistors are commonly rated by their speed (e.g. toggle frequency f

T

), voltage capability,

maximum current, typical h

fe

, and power capability .

I

C

max

f

T

P

max

Cost

Part # Type Application (A) h

fe

(MHz) (W) ($US)

2N3904 NPN gen’l purpose 0.20 100–300 300 0.625 0.06

2N3906 PNP gen’l purpose 0.20 100–300 250 0.625 0.06

2N5089 NPN gen’l purpose 0.05 450–1800 50 0.625 0.14

2N2369 NPN switching 0.2 40–120 500 0.4 1.40

2N5415 PNP power 1.0 30–150 10 0.99

TIP102 NPN pwr Darlington 8.0 1000–2000 80 0.69

MJ10005 NPN pwr Darlington 20 50–600 175 6.90

with holes. It is desirable to have as large a ratio of collector current to base

current as possible. To achieve this, in addition to being as thin as possible,

the base is made of lightly doped material.

Since the base–emitter junction obeys the exponential diode law for cur-

rent as a function of voltage, small changes in the base–emitter voltage dif-

ference have a large effect on the collector current. Thus, a transistor can be

used as an ampliﬁer: a device in which a small signal controls a large signal.

It is worth keeping in mind that β is not well controlled in the transistor

manufacturing process. Although β is approximately constant for a given

transistor, it varies from transistor to transistor even if they are of the

same type. (For examples of the range of variation, see Table 4.1.) In

addition, β depends signiﬁcantly on temperature, collector current, and

collector-to-emitter voltage. Since β can vary over a substantial range, it

is good practice to design transistor circuits in such a way that their proper

functioning does not depend strongly on its exact value. Because β is so

variable, the current-ampliﬁer picture of transistor action is less useful than

the transconductance-ampliﬁer picture.

4.1.1 Basic definitions

To discuss transistor action quantitatively, we need to deﬁne three voltage

differences, three currents, and the relationships among them:

r

V

BE

≡ V

B

− V

E

= potential of base relative to emitter,

r

V

CB

≡ V

C

− V

B

= potential of collector relative to base,

51 4 Bipolar transistors

r

V

CE

≡ V

C

− V

E

= potential of collector relative to emitter,

r

associated identity (from Kirchhoff’s voltage law):

V

CE

= V

BE

+ V

CB

. (4.1)

For an NPN transistor in its normal operating mode, all the above potential

differences are positive.

r

I

C

= current ﬂowing into collector,

r

I

E

= current ﬂowing out of emitter,

r

I

B

= current ﬂowing into base,

r

associated identity (from Kirchhoff’s current law):

I

E

= I

C

+ I

B

. (4.2)

For an NPN transistor in its normal operating mode, all the above currents

are positive.

The relationships between these voltages and currents are usually ex-

pressed in terms of characteristic curves. Fig. 4.3 displays sets of represen-

tative curves for an arbitrary bipolar transistor. Study their shapes carefully

and refer to them as you perform the following exercises.

4.1.2 Simplest way to analyze transistor circuits

In the following section we consider a mathematical model (the Ebers–

Moll model) that gives a good approximation to transistor performance.

However, you don’t need the model to understand transistor operation in

most circuits. We will see from the Ebers–Moll model that, if a transistor

V

CB

Linear

Region I

C

V

CE

Linear

Region

Breakdown

Breakdown

Saturation

Region

0

5 10 15 20

(Volts) (Volts)

0

5 10 15 20

10

20

30

40

(mA)

I

B

= 30 A µ

I

E

= 3 mA

I

B

= 250 A µ

I

B

= 150 A µ

I

B

= 90 A µ

I

E

= 25 mA

I

E

= 15 mA

I

E

= 9 mA

I

B

= 350 A µ

I

E

= 35 mA

10

20

30

40

I

C

(mA)

Fig. 4.3. Characteristic curves for an NPN bipolar transistor.

52 Hands-on electronics

is on and conducting milliamperes of collector current, its base–emitter

voltage difference V

BE

is approximately constant at about 700 mV. Fur-

thermore, the base is a point of high impedance, whereas the emitter is

a point of low impedance. Thus, a changing voltage at the base causes

matching voltage changes at the emitter. In other words, the base voltage

controls the emitter voltage:

r

the emitter voltage follows the base.

Since β is large,

r

the collector current nearly equals the emitter current.

Since the collector is of even higher impedance than the base,

r

the collector assumes any voltage required by Ohm’s law as applied to

the rest of the circuit.

These three rules are all you need to know in most situations.

(Suggestion: You may want to skip the next section for now and use

these rules to analyze the circuits of Figs. 4.5, 4.7, and 4.8, described in

sections 4.2.2, 4.2.3, and 4.2.4. Then come back and study the next section.)

4.1.3 Ebers–Moll transistor model

To see why the simple picture just described is valid, we next consider the

Ebers–Moll model, which is based on the physical description in section 4.1

and provides a reasonably good description of transistor action. In this

model, the amount of collector current that ﬂows is determined by the

amount of forward-bias that is applied to the base–emitter diode. We thus

have

I

C

= I

s

e

eV

BE

/kT

−1

. (4.3)

As with Eq. 3.1, I

s

is the reverse saturation current. (In practice, the expo-

nential usually dominates and the ‘1’ can be neglected.) Also, as mentioned

above, the base current is related to the collector current by

I

B

=

I

C

β

. (4.4)

Dynamic resistance of emitter

Since the emitter current equals the collector current to a good approxima-

tion, Eq. 4.3 also detemines the emitter current. We can thus use it to derive

a useful expression for the dynamic resistance, r

e

, of the emitter. Recalling

the deﬁnition of dynamic resistance, r

e

is the partial derivative of emitter

53 4 Bipolar transistors

voltage with respect to emitter current:

r

e

≡

∂V

E

∂ I

E

. (4.5)

The dynamic resistance tells us how, for ﬁxed base voltage, the emitter

voltage would change in response to a change in emitter current – for

example, how the emitter voltage would differ if the emitter were driving

a small load resistance as opposed to a large one. To determine r

e

, we

differentiate Eq. 4.3 at ﬁxed base voltage, giving

r

e

=

kT

e

1

I

C

. (4.6)

(In addition there is the ohmic resistance of the emitter, which is typically

a few ohms.)

Dynamic resistance of base

On the other hand, if we ﬁx the emitter voltage, we can ﬁnd from Eqs. 4.3

and 4.4 the dynamic resistance r

BE

to the emitter as seen at the base:

r

BE

=

∂V

B

∂ I

B

= β

kT

e

1

I

C

. (4.7)

This tells us how the base loads the circuit that is driving it. We see that

the base–emitter junction appears to have a low resistance when viewed

from the emitter end, but appears to have a higher resistance (by a factor

β) when viewed from the base end.

Some useful approximations

Since at room temperature, e/kT = 39 V

−1

, in practice a reasonable ap-

proximation is

r

e

=

25 mV

I

C

, (4.8)

r

BE

= β

25 mV

I

C

. (4.9)

Since the emitter acts as a low impedance (only a few ohms for typical

collector-current values), its voltage hardly depends on the current ﬂowing

through it. But the base acts as a high impedance, so it is easy to apply a sig-

nal voltage to the base. This comes about because the base current is smaller

than the emitter current by the factor β, which is of order 100. Although

r

BE

is only a few hundred ohms, the factor β applies also to any resistor

54 Hands-on electronics

R

E

that is in series with the emitter, i.e. the apparent resistance seen at

the base is β(r

e

+ R

E

) = r

BE

+βR

E

, which is typically tens to hundreds

of kilohms.

We saw in the case of the silicon diode that a crude approximation in

which the forward diode drop is taken to be approximately constant at

600 mV is adequate for most applications. As mentioned above, in many

practical transistor applications (including the circuits you will build in

this chapter) a simple approximation is sufﬁcient: treat the base–emitter

voltage difference V

BE

as constant at about 700 mV and r

e

as constant at

a few ohms. This reﬂects the fact that the order of magnitude for I

C

in a

typical small-signal-transistor circuit is several milliamperes. Often r

e

is

much smaller than R

E

and can be neglected.

4.2 Experiments

4.2.1 Checking transistors with a meter

Since a transistor is constructed as a pair of back-to-back PN junctions,

a quick way to test a transistor is to verify its junction resistances in the

forward- and reverse-biased directions. Often this can be done using an

ohmmeter, which sends current through the device under test and measures

the resulting voltage. However, our digital multimeters are not designed for

this kind of measurement; instead, they provide a diode test function that

measures the forward voltage corresponding to a forward current of about

600 A. To test a diode or transistor junction, connect it between the ‘V’

and ‘common’ jacks with clip leads and set the meter’s selector knob to

the position marked with the diode symbol. The diode is forward-biased if

its anode is connected to ‘V’ and its cathode to ‘common’.

**Test the base–collector and base–emitter junctions of a 2N3904 (NPN)
**

and 2N3906 (PNP), and record your readings in both the forward- and

reverse-biased directions.

To tell which pin of the transistor is which, refer to Fig. 4.4, which

shows the pinout for the TO-92 plastic case in which these transistors are

packaged. If your transistors are good, each junction should show about

700 mV in the forward direction, and an out-of-range indication in the

reverse direction.

55 4 Bipolar transistors

2

N

E

3

9

0

4

B

C

2

N

E

3

9

0

6

B

C

TO-92

case

E E

B B

C C

PNP NPN

(a) (b)

Fig. 4.4. Transistor as back-to-back diodes; TO-92 pinout.

V

in

330

3.3 k

Vout

+15 V

2N3904

−15 V

R

B

R

E

V

in

10 k

R

B

Z

in

Emitter

Follower

(a) (b)

Fig. 4.5. (a) Emitter follower. (b) Emitter-follower model used for input-impedance

measurements. The value for Z

in

is found using the voltage-divider equation.

4.2.2 Emitter follower

This simple transistor circuit (shown in Fig. 4.5(a)) is often used to ‘buffer’

an AC signal, as well as to change its DC voltage level by V

BE

. (It is also a

close relative of the common-emitter ampliﬁer that we will study next.) It is

called an emitter follower because the voltage at the emitter follows voltage

changes at the base, with V

BE

almost constant at about 0.7 V. Although it

has voltage gain (V

out

/V

in

) of about unity (actually slightly less because

of the logarithmic dependence of V

BE

on I

C

implied by Eq. 4.3), it is still

an ampliﬁer: it has high input impedance, and low output impedance, and

can thus provide current gain (i.e., output current > input current), which

is what is meant by buffering.

1

Start by grounding V

in

. You can determine I

B

and I

C

(within the uncer-

tainties due to the resistor tolerances) by measuring the voltage drops

1

Since power is the product of voltage and current, power ampliﬁcation can occur through an increase

in either quantity.

56 Hands-on electronics

across R

B

and R

E

. Derive an approximate β value for your transistor by

computing the ratio of collector current to base current.

Nowdrive V

in

with a sine wave fromthe function generator and compare

the input and output signals. What are the input and output amplitudes?

(If they seem to differ signiﬁcantly, make sure both scope-probe com-

pensations are properly adjusted.) Measure the DC voltage shift V

BE

between the base and emitter.

The 330 base resistor is in series with the input resistance r

BE

+βR

E

,

so it has little effect on the signal. It is there to prevent parasitic oscillation,

which might otherwise occur due to the inductance of the wire jumpers

and the ‘parasitic’ capacitive coupling between the emitter and the base.

(‘Parasitic’ refers tocapacitance that is there inevitably, due tothe proximity

of the leads to each other, rather than by design.) If you’re curious about

this, try omitting the base resistor and see what happens. Not all 2N3904s

are guaranteed to oscillate in this circuit; whether yours does could also

depend on details of the wiring arrangement.

**Measure the input impedance. To do so, replace the base resistor with
**

10 k and measure the small decrease in amplitude fromone side of R

B

to

the other (Fig. 4.5(b)). Explain using the voltage-divider idea how this

measures the input impedance. (When you’re done with this exercise,

restore the 330 base resistor for use in the following parts.)

**Measure the output impedance. To do so, add a blocking capacitor and
**

load resistor as shown in Fig. 4.6(a) and infer the output impedance from

the small decrease in output amplitude (see Fig. 4.6(b)).

The blocking capacitor allows the 330 load to affect the ACsignal voltage

without changing the DC biasing of the transistor.

Vin

1.0 F µ

330

3.3 k

Vout

+15 V

2N3904

330

−15 V

Vout

330

~

Emitter

Follower

Z

out

R

L

R

L

(a) (b)

Fig. 4.6. (a) Emitter follower with optional load circuit for measurement of Z

out

.

(b) Emitter follower modeled as an ideal voltage source in series with an output

impedance.

57 4 Bipolar transistors

**Explain what would happen to the DC bias voltage of the emitter if the
**

capacitor were omitted.

When using the blocking capacitor, be sure to use a small signal am-

plitude (about 1 V) so as not to apply too large a reverse voltage to the

capacitor, and use a high enough frequency so that the capacitor causes

negligible attenuation – about 10 kHz. Note that polarized capacitors can

be safely reverse-voltaged by a volt or two r.m.s., but typically not by more

than 15% of their voltage rating.

**How do your measured impedances compare with what you expect?
**

The input impedance should equal β times the emitter resistor, and the

output impedance should equal the dynamic resistance of the emitter (as

described in section 4.1.3).

4.2.3 Common-emitter amplifier

The common-emitter ampliﬁer is a very common transistor ampliﬁer con-

ﬁguration, but that is not how it gets its name – the name reﬂects the idea

that the emitter is in common between the input circuit and the output

circuit. Construct the common-emitter ampliﬁer shown in Fig. 4.7.

**Predict and measure the quiescent DC voltages (i.e. the voltages when
**

no input signal is present) at the base, emitter, and collector.

The predictions are easy:

1. Apply Ohm’s law to the base-bias resistive voltage divider to determine

the quiescent base voltage: V

B

= V

CC

R

2

/(R

1

+ R

2

). This is an approx-

imation since it neglects the base current, but, as you’ll see, given the

large value of β, the base current is small enough that the approximation

is a good one.

Vin

1 F µ

Vout

2N3904

R

1

= 100 k

R

C

V

CC

= +15 V

R

2

R

E

R

1

R

2

= 10 k

R

C

= 10 k

R

E

= 1 k

Fig. 4.7. Common-emitter ampliﬁer.

58 Hands-on electronics

2. Then apply Ohm’s law to the emitter resistor to determine the emitter

current, taking into account the transistor’s expected V

BE

drop: I

E

=

(V

B

− V

BE

)/R

E

.

3. ThenapplyOhm’s lawtothe collector resistor todetermine the quiescent

collector voltage: V

out

= V

CC

− I

C

R

C

. You know the collector current

well enough since it equals the emitter current to a good approximation.

**Compared to your measurements, by what percentages are your voltage
**

predictions wrong? Is this as expected given the resistor tolerances and

uncertainties in β and V

BE

?

**Using the measured voltages, predict the collector, emitter, and base
**

currents.

**Calculate the change inquiescent base voltage if youtake the base current
**

into account. Assume that the base current ﬂows through the Th´ evenin

equivalent of the base-bias voltage divider (i.e., the input impedance of

the base is in parallel with R

2

).

You can understand how the circuit ampliﬁes by applying Ohm’s law

to the emitter and collector resistors. Since the emitter follows the base, a

voltage change at the base causes a larger voltage change at the collector:

V

out

= −I

C

· R

C

= −I

E

· R

C

I

E

= V

E

/R

E

= V

B

/R

E

= V

in

/R

E

, (4.10)

Therefore,

V

out

= −V

in

R

C

/R

E

. (4.11)

**Measure the voltage gain (V
**

out

/V

in

) and compare with what you

expect.

So as not to exceed the available output-voltage range of the circuit, be

careful to keep the input amplitude less than about 700 mV. (You can

check what happens to the output waveform as you exceed this amplitude,

but be sure not to exceed the 1 V reverse-voltage capability of the input

capacitor.) Also, use a high enough frequency that the input high-pass ﬁlter

does not attenuate the signal too much – it is a little tricky to estimate the

breakpoint frequency of the ﬁlter because three resistances in parallel need

to be taken into account: those of the base-bias voltage divider as well as

the input impedance of the base.

Is the ampliﬁer inverting?

**Look at the signal at the emitter and explain what you see.
**

59 4 Bipolar transistors

+5

330

1 k

+15

2N3904

mA

Fig. 4.8. Transistor current source.

**Try a triangle-wave input – can you see any distortion in the output
**

waveform? There should be some due to the variation of r

e

with col-

lector current, but the effect is small since r

e

is in series with the 1 k

emitter resistor. How big should the effect be according to the Ebers–

Moll model?

In the grounded-emitter ampliﬁer, i.e., for R

E

= 0, the voltage gain is

greater, but so is the distortion, since r

e

alone appears between the emitter

and ground.

4.2.4 Collector as current source

Since the base–collector junction is reverse-biased, the collector should

act as a very high impedance. This means that the collector is a good

approximation to a current source: a device that outputs a constant current,

independent of its voltage. You can verify this using the circuit of Fig. 4.8,

using a second meter or the scope to make various voltage measurements.

Start with the load resistance (10 k pot – as always, be careful to connect

it properly so as not to burn it out!) set to 0 .

**How much collector current should ﬂow? Is this conﬁrmed by your
**

measurement of the emitter voltage?

**Slowly increase the load resistance until the output current starts to
**

decrease. At the pot setting where the current source starts to fail (output

current starts to vary rapidly with load resistance), measure the collector

voltage. What is the compliance of your current source (the range of

output voltage over which the current is approximately constant)?

**Where the current source starts to fail, how does the collector voltage
**

compare with the base voltage? You should be observing transistor sat-

uration: when |V

CE

| < |V

BE

|, the current-source behavior stops since

60 Hands-on electronics

ground

or +5

330

10 k

+5

2N3904

Red

LED

Fig. 4.9. Transistor switch.

the collector-base junction becomes forward-biased and its impedance

decreases.

Another way of thinking about the same phenomenon is that when the

transistor saturates, β decreases sharply, thus the base current increases as

the base ‘steals’ current from the collector.

**By measuring the voltage drop across the base resistor, take a few mea-
**

surements of β as you turn up the load resistance and the transistor

becomes more and more saturated. Make a graph of β vs. V

CE

.

4.2.5 Transistor switch

Transistor saturation is put to good use in the saturated switch. Construct

the circuit of Fig. 4.9. You should see the light-emitting diode (LED) turn

on when you connect the input to +5 V and off when you leave it open or

connect it to ground. (If you are unsure which terminal of the LED is the

anode and which is the cathode, you can check it with a meter, or feel free

to try it both ways. Often the cathode is indicated by a ﬂat spot on the rim

of the plastic that encapsulates the diode, and some manufacturers make

the anode lead slightly longer than the cathode lead.)

Note that the transistor inthis common-emitter connectionis an‘inverter’,

in the sense that a high voltage level at its input (the open end of the series

base resistor) causes a low voltage level at its output (the collector). This

circuit is called a saturated switch since the transistor goes into saturation

(|V

CE

| < |V

BE

|) when turned on. It has the virtue of dissipating very lit-

tle power, since, when the transistor is on, the voltage across it is small,

61 4 Bipolar transistors

whereas, when the transistor is off, the current through it is essentially

zero.

**When the transistor is on, what are the base and collector currents? (You
**

don’t need to bother with an ammeter for this measurement – just look

at the voltage drops across the base and collector resistors.)

**What is your transistor’s saturation voltage V
**

CE(sat)

? What is the ‘on

voltage’ across the LED?

**Approximately what minimum value of β must the transistor have to be
**

sure of saturating when +5 V is applied at the input?

**Drive the switchfromthe ‘TTL’ output of the functiongenerator (‘digital’
**

square-wave with a low voltage level near zero and a high level near

+5 V) at 100 kHz and use the dual-trace oscilloscope to measure the

turn-on and turn-off delays in nanoseconds. (Trigger the scope with the

function generator while looking at both the function generator signal

and the collector voltage.)

The relatively slowturn-on and turn-off delays of the saturated switch are

due to the charge stored in the base when the transistor saturates. It takes

time for the transistor to switch states since this saturation charge must

ﬂow in or out of the base through the input resistor. High-speed switching

transistors (such as the 2N2369) are manufactured to minimize this effect

and can operate at frequencies as high as 1400 MHz.

4.3 Additional exercises

The following optional exercises offer additional practice.

4.3.1 Darlington connection

To provide high input impedance and reduce the input base current, one

can cascade two transistors in series, i.e. ‘buffer’ the input with an emitter-

follower stage. This Darlington transistor pair acts like a single transistor

whose current gain is the product of the two β’s and whose V

BE

drop is the

sum of the two V

BE

’s. Build the circuit of Fig. 4.10.

**With the input grounded, what quiescent currents do you observe through
**

R

B

and R

E

? What does this imply for the combined β value of the

Darlington pair?

**Now apply an input signal – what do you see at the output? What is the
**

DC voltage drop from input to output?

62 Hands-on electronics

V

in

10 k

3.3 k

Vout

+15

Q = Q =

1 2

2N3904

−15

Q

1

Q

2

R

B

R

E

Fig. 4.10. Darlington pair.

**The input impedance should be so big that you can’t measure any de-
**

crease insignal amplitude across the 10kresistor –checkthis assumption.

What minimum value does this imply for the input impedance? What

input impedance do you expect, and why?

Darlington pairs are available encapsulated in three-lead packages, for

example the 2N6426 with combined β value of about 100 000. The Darling-

ton connection is particularly useful for power transistors, to compensate

for their low β (β ≈ 20 is not uncommon). For example, the TIP110 50 W

power Darlington has a minimum combined β value of 500.

4.3.2 Push–pull driver

To provide low output impedance, a push–pull buffer stage is often used.

This consists of two emitter followers, one PNP, and one NPN, arranged so

that the PNP conducts during one half of the output period (when V

out

< 0)

and the NPN during during the other half (V

out

> 0).

**First, drive your breadboard’s speaker (nominal impedance = 8 ) from
**

the function generator. With the amplitude set to maximum and the

frequency at 1 kHz, measure the function-generator amplitude both with

and without the speaker connected. What is the attenuation due to the 8

load? Is it consistent with the measurement of the function generator’s

output impedance you made in section 3.7?

**Since the speaker impedance may depend on frequency, repeat the mea-
**

surement at 10 kHz and compare.

**Go back to 1 kHz and add a push–pull buffer as shown in Fig. 4.11. (Be
**

sure you have adjusted the supply voltages to not more than ±5 V so

as not to overheat the transistors – they are rated for 200 mA collector

current and 625 mW power dissipation.) You should see a larger output

amplitude and hear a louder tone from the speaker.

63 4 Bipolar transistors

V

in

+5

2N3904

2N3906

−5

Speaker

Fig. 4.11. Driving loudspeaker with push–pull buffer.

The circuit may oscillate because of inadvertent positive feedback from

output to input; if it does, try (1) rearranging your circuit so that the wire

jumpers are as short as possible, (2) adding a 330 resistor in series with

the input, and if that’s not enough to stabilize it, (3) adding a few-hundred-

picofarad cap to ground at the output.

**The output waveform will display ‘crossover distortion’ – what does it
**

look like and why does it occur? (Hint: for a transistor to be on, there

must be about 700 mV between base and emitter – is there a time during

the cycle when neither transistor is on?) By how much does it reduce

the output amplitude, and why? What is the minimum input amplitude

required for an audible output? Explain.

For high-power applications, a power-Darlingtonpush–pull stage is often

used. Two methods can be used to alleviate the crossover distortion:

r

shifting the bias points of the two transistors apart to minimize the portion

of the cycle when neither transistor is on;

r

using feedback to apply a signal to the bases that compensates for the

distortion.

We will explore these techniques in Chapter 8.

4.3.3 Common-base amplifier

Build the common-base ampliﬁer of Fig. 4.12.

**Predict and measure the transistor’s quiescent currents and bias voltages.
**

The diode at the base should bias the emitter approximately at ground.

Since the diode and base–emitter voltage drops are unlikely to be exactly

the same, the input will have a small DC offset – how big is it?

64 Hands-on electronics

V

in

22 k

V

out

+15

2N3904

10 k 10 k

100

−15

Fig. 4.12. Common-base ampliﬁer.

**Connect a small sine-wave input and determine the voltage gain. Note
**

that, in contrast to the common-emitter ampliﬁer, the input and output

currents are almost equal, and the ampliﬁer is noninverting.

**What do you predict for the input and output impedances? You can
**

measure the input impedance easily using the 400 output impedance

of the function generator: howmuch smaller does the function-generator

output become when you connect it to the ampliﬁer input? What input

impedance does this imply for the ampliﬁer? Explain.

5

Transistors II: FETs

In this chapter we introduce the ﬁeld-effect transistor (FET). A majority of

today’s integrated circuits are built using FETs of one type or another. FET

operation is easier to explain than that of bipolar transistors; however, due

to the variability of FET parameters, many people ﬁnd FETs more difﬁcult

to use. As with bipolar technologies, it is essential that you master the

basics of FET operation, and you will ﬁnd that knowledge useful later on.

Apparatus required

Breadboard, oscilloscope, multimeter, two 2N5485 JFETs, one 1N4733

Zener diode, two 1 k, one 3.3 k, two 10 k, one 100 k, and one 1 M

1

4

W re-

sistors, 0.1 Fceramic capacitor, 1.0 Fand 100 Felectrolytic capacitors.

5.1 Field-effect transistors

Like bipolar junction transistors, ﬁeld-effect transistors (FETs) are three-

terminal semiconductor devices capable of power gain. Qualitatively, they

operate much like junction transistors, but they have much higher input

impedance and lower transconductance and voltage gain. Also, they have

a larger variation in their ‘V

BE

’ equivalent (called V

GS

) than bipolar tran-

sistors. They come in a confusing variety of types, but we will concentrate

for today on junction FETs (JFETs).

Fundamentally, there are two types of FETs: junction FETs and metal-

oxide-semiconductor FETs (MOSFETs). In both types, a conducting chan-

nel betweenthe drainandsource terminals is controlledbya voltage applied

to the gate terminal. The channel can be made of either N-type or P-type

material (Fig. 5.1). N-channel is more common since the conductivity of

N-type semiconductor (in which electrons carry the current) is higher than

that of P-type (in which holes do).

65

66 Hands-on electronics

P

N N

S

G

S

P

G

N-Channel

N

P P

D

G

S

N

G

P-Channel

G

D

D

S

G

D

N-Channel

JFET

P-Channel

JFET

Fig. 5.1. Construction and circuit symbols of JFETs (note that other variants of these

symbols are also used).

The gate region of a JFET consists of material of opposite type to that

of the channel; thus, the gate and channel form a diode. In our preferred

symbol for JFETs (Fig. 5.1), to distinguish the source from the drain,

the gate terminal is drawn at the source end, even though the channel is

physically spread out between the source and drain. As for any diode, the

arrowon the gate symbol indicates the direction of forward-bias. However,

JFETs are normally used with the gate–channel diode reverse-biased.

In a JFET, the channel conducts unless it is turned off by an applied

reverse-bias voltage between the gate and the channel. As the reverse-bias

is increased, more and more current carriers are repelled out of the channel

until it is ‘pinched off’ and the drain–source current drops to zero (see

Fig. 5.2). The voltage at which this occurs is called V

P

or V

GS(off)

.

Note that the drain, gate and source of an FET play similar roles to the

collector, gate, and emitter (respectively) of a bipolar transistor. Unlike the

bipolar case, the source and drain are roughly interchangeable and it is

possible for an FET to be used ‘backwards’.

5.1.1 FET characteristics

The simplest way to think of FET action is as a voltage-controlled current

source, i.e. the drain current I

D

is approximately constant for a given gate–

source voltage V

GS

, depending only slightly on the voltage V

DS

between the

drain and source. Since the gate–channel diode is normally reverse-biased,

the gate current is extremely tiny (typically ∼nanoamperes), so that for all

67 5 Transistors II: FETs

Drain Source

V

V

V

Drain Source

V

V

V

+

+

+

+

+

+

(a)

(b)

GS

GS

GS

GS

DS

DS

Gate

Gate

Gate

Gate

Fig. 5.2. Schematic representation of JFET operation: (a) gate–channel diode slightly

reverse-biased; (b) gate–channel diode highly reverse-biased (V

GS

≥ V

P

so that channel is

pinched off).

V = 0 V

V = −3 V

V = −2 V

V = −1 V

I

V

I

Saturation Region

Linear

Region

5 10 15 20

(Volts)

0

(mA)

5

10

15

20

GS

GS

GS

GS

DSS

DS

D

Fig. 5.3. Idealized common-source characteristic curves for a JFET.

practical purposes the drain and source currents are equal (I

D

= I

S

). The

voltage-controlled current-source behavior occurs as long as the drain–

source voltage V

DS

is sufﬁciently high. This is called the saturation region

of the FET characteristic (see Fig. 5.3).

68 Hands-on electronics

For V

DS

smaller than a volt or two, a JFET behaves like a voltage-

controlled resistor rather than a current source, i.e. the slope of the I –V

characteristic is controlled by the gate–source voltage. This is the linear

region of the FET characteristic, useful for automatic gain control (AGC)

and modulation applications.

Don’t confuse FET saturation with bipolar-transistor saturation – they

are entirely different phenomena! For example, recall that bipolar-transistor

saturation occurs at small V

CE

, whereas FETsaturation occurs at large V

DS

.

5.1.2 Modeling FET action

Recall that for bipolar transistors the collector current varies exponentially

with the base–emitter voltage. For FETs operated in the saturation region,

the relationship is quadratic:

I

D

= I

DSS

1 −

V

GS

V

P

2

, (5.1)

where V

P

is the pinch-off voltage and I

DSS

is the saturation drain current

for V

GS

= 0 (i.e. gate shorted to source). Thus, the transconductance is

proportional to

√

I

D

:

g

m

= I

D

/V

GS

. (5.2)

As in the case of bipolar transistors, this is only a model and should

not be expected to be exact. Like the parameter β for bipolar transistors,

I

DSS

and V

P

are temperature-dependent and vary substantially even among

devices of the same type, so good designs minimize the dependence of

circuit performance on them.

Since the transconductance of a bipolar transistor increases linearly with

I

C

((g

m

)

bipolar

= 1/r

e

), but that of an FET only as the square root of I

D

,

bipolar transistors typically have higher transconductance than FETs for

a given current, and thus can give higher gain in ampliﬁcation appli-

cations. This has led to the common practice of combining FETs with

bipolar transistors in analog integrated circuits to exploit the advantages

of both, e.g. the CA3140 MOSFET-input op amp with its teraohm input

impedance.

69 5 Transistors II: FETs

5.2 Exercises

5.2.1 FET characteristics

As shown in Fig. 5.1, the gate–source and gate–drain connections are PN

junctions.

**Use the diode-test feature of the multimeter and verify this picture us-
**

ing a 2N5485 JFET (the pinout for the 2N5485 is shown in Fig. 5.4).

Note: the pinout of the 2N5485 does not correspond to that of the

2N3904.

Unlike an NPN transistor, whose emitter and collector are distinct N-type

regions separated by the P-type base, the drain and source of an N-channel

JFET occupy opposite ends of a single N-type region, connected via the

channel.

**Use an ohmmeter to show that the drain and source are connected. If
**

the meter reading ﬂuctuates, try connecting the gate and source together

using the breadboard and a small piece of wire. Explain why this will

stabilize your measurement. What resistance do you measure?

Next verify Eq. 5.1. To measure I

DSS

and V

P

, set the drain voltage to 10 V,

ground the source, and apply a variable negative voltage to the gate. This

can be accomplished using either the 1 k or 10 k potentiometer as a voltage

divider (ground one end of the pot, set the other end to −5 V, and connect

the slider to the gate).

Measure I

DSS

. Adjust the pot until the gate is at ground while leaving

the drain voltage at 10 V. According to Eq. 5.1, the drain current now

equals (by deﬁnition) I

DSS

.

Measure V

P

. Using an ammeter to measure the drain current, adjust the

gate voltage until the drain current drops to zero. The pinch-off voltage

will be equal to this gate voltage. Why is this true?

**Verify Eq. 5.1. With the drain voltage set to 10 V, adjust V
**

GS

while

measuring the drain current. Plot I

D

versus V

GS

.

The common-source characteristic curves for the 2N5485 can be mea-

sured using the circuit shown in Fig. 5.4. These curves illustrate the basic

dependences among I

D

, V

DS

, and V

GS

, and will be useful while performing

the remaining exercises.

Since JFETs are notoriously variable, more so than bipolar transistors,

try to use the same JFET for all exercises. Use one pot to adjust the gate

70 Hands-on electronics

2

N

5

4

8

5

D

J

F

E

T

S

G

A

0–15 V

−5 to 0 V

D

S

G

(a) (b)

Fig. 5.4. Circuit for measuring the common-source characteristic curves.

voltage and the other to adjust the drain voltage. Use the scope probes to

measure the drain and gate voltages while using a meter to measure the

drain current.

**Using the pot, adjust V
**

GS

to be 0.5 V more positive than V

P

(keep

in mind that V

P

is negative!). Slowly increase the drain voltage from

zero to 15 V while measuring the drain current. Record and plot your

measurements.

For V

DS

less than a few volts, the current should increase linearly with

drain voltage. This is the ‘linear region’, in which the JFETacts as a voltage-

controlled variable resistor. As you further increase V

DS

, the current should

then ‘saturate’ at an approximately constant value.

JFET saturation occurs because the increasing drain voltage creates an

increasing depletion region between the gate and drain. Since V

GS

> V

P

,

the channel will never be pinched off completely, with an equilibrium (of

sorts) created. The size of the depletion region (and thus the resistance of

the channel) increases approximately linearly with drain–source voltage

difference, resulting in approximately constant current.

**Repeat your measurement procedure for several V
**

GS

values between V

P

and zero. Plot the data on a single graph and clearly label each curve,

indicating the linear region, the saturation region, and I

DSS

.

5.2.2 FET current source

Adding a resistor improves the JFET’s current-source performance com-

pared with that of the bare JFET.

**Hook up the circuit shown in Fig. 5.5 and measure the dependence of the
**

drain current on the drain–source voltage as you adjust the pot; record

and plot your measurements.

What is V

GS

for each data point?

71 5 Transistors II: FETs

10 k

pot

1 k

+15

A

2N5485

open

R

S

Fig. 5.5. Self-biasing JFET current source.

You can compute V

GS

from I

D

and the known resistance of R

S

. You should

see the drain current start to vary substantially as you make the transition

from the saturation region to the linear region.

What is the compliance?

Within the saturation region, how constant is the current?

**Calculate the approximate output impedance in the saturation region
**

(see Eq. 3.13).

**Compare the performance of this current source with that of the bare
**

JFET and of the bipolar current source that you built in section 4.2.4.

Even though V

GS

is not exactly constant as V

DS

is varied, this circuit

actually works better (has larger output impedance) than one in which V

GS

is held constant. This is because negative feedback is at work. For example,

suppose I

D

increases; then, so does the drop across R

S

, increasing the mag-

nitude of V

GS

and moving the FET closer to pinch-off, thus decreasing I

D

.

5.2.3 Source follower

As with the emitter follower from section 4.2.2, a JFET source follower

provides power ampliﬁcation via current buffering, even though the voltage

gain is less than or equal to unity. Build the circuit of Fig. 5.6, and try it

out with a 1 kHz sine wave of small amplitude.

The operation of this circuit is reasonably straightforward. Given the

nanoampere gate current, the 1 M resistor is small enough to bias the gate

very near ground. Quiescently, the FETis thus in the saturation region, with

I

D

determined by I

DSS

and V

P

according to Eq. 5.1. If the input voltage

increases, V

GS

moves closer to zero, the channel opens, and I

D

increases.

72 Hands-on electronics

V

in

0.1 µF

1 MΩ 1 k

V

out

+15

2N5485

R

S

Fig. 5.6. Source follower.

Thus, V

out

follows the source. When V

in

decreases, the channel closes and

V

out

drops.

What is the DC offset at the output?

**Measure the voltage gain.
**

You should see that the voltage gain is less than unity, since the dynamic

resistance of the source (=1/g

m

) forms a voltage divider with R

S

. (This

effect was also present for the bipolar transistor, but was much smaller

due to the bipolar transistor’s larger value of g

m

.) Draw a diagram of this

voltage divider.

**Fromyour observed attenuation, derive a value for g
**

m

and compare with

that of a bipolar transistor at the same current.

You can improve the source follower by providing it with much higher

load resistance. Since an ideal current source would have inﬁnite resistance,

a current-source load is often used; it can be constructed by adding another

FET, as in the clever circuit of Fig. 5.7. (Since we are using N-channel

JFETs, it is actually a current sink.) Try it out.

**Measure the voltage gain and the DC offset from input to output.
**

Note that if the two 2N5485s approximately match in their character-

istics, not only is the voltage gain unity, but the DC offset is small: the

constant current due to Q

2

creates a constant voltage drop across R

1

. Fur-

thermore, since the gate of Q

2

is at the same voltage as the bottom of R

2

,

to the extent that the two FETs (and the two resistors) match, this should

also be true for Q

1

and R

1

. Thus the output voltage must follow the source

voltage of Q

1

.

Explain the operation of this circuit in your own words.

What are I

D

, V

GS1

, and V

GS2

?

73 5 Transistors II: FETs

Vin

0.1 F µ

1 MΩ 1 k

Vout

+15

2N5485

1 k

2N5485

−15

Q

1

Q

2

R

1

R

2

Fig. 5.7. Source follower with current-source load.

Vin

0.1 F µ

1.0 F µ

3.3 k

1 MΩ

1 k

Vout

+15

+

R

S

R

D

Fig. 5.8. JFET ampliﬁer.

The offset can be much improved by using a matched FET pair, e.g. the

2N3958 dual JFET. Such a circuit is often used in the input stage of an

oscilloscope.

5.2.4 JFET amplifier

Construct the ampliﬁer shown in Fig. 5.8. The principle at work is essen-

tially the same as for the common-emitter ampliﬁer: a varying input voltage

controls a varying current, which the drain resistor ‘converts’ to an output

voltage.

**Using the common-source curves that you measured in section 5.2.1,
**

predict the quiescent gate–source bias voltage and output voltage for this

74 Hands-on electronics

ampliﬁer. Power the ampliﬁer and compare the measured values with

your predictions.

How much power is dissipated by the FET?

What is the input impedance?

**Is this an inverting or noninverting amp? Explain why.
**

The voltage gain (A) is deﬁned as the ratio of the output amplitude to the

input amplitude. If the source voltage remains ﬁxed, then

V

G

= V

GS

.

As discussed previously,

g

m

= I

D

/V

GS

,

and since

V

D

= I

D

· R

D

,

A = g

m

· R

D

.

**What is the predicted voltage gain for this ampliﬁer? Using a 1 kHz
**

small-amplitude sine-wave input, measure the voltage gain and compare

with the expected gain.

The source capacitor is used to ‘ﬁx’ the source voltage even as the drain

current ﬂuctuates due to the AC input. (This trick can also be used to in-

crease the voltage gain of the bipolar-transistor common-emitter ampliﬁer.)

This implies that the gain will be frequency-dependent.

**Switch the input to a triangle wave and adjust the frequency widely. Ex-
**

plain what you see. Replace the source capacitor with a 100 Fcapacitor.

How does this change things?

**Try several different 2N5485s and record the voltage gain and quiescent
**

drain current and output voltage. How reproducible are the results?

**Comment on the design and operation of simple transistor and JFET
**

circuits. For example, when would you choose a bipolar transistor over

a JFET or vice versa? Feel free to include any general comments you

have on the experience you’ve gained from the last few chapters.

6

Transistors III: differential amplifier

Inthis chapter we will studythe transistor differential-ampliﬁer circuit. This

is a very important transistor circuit, as it is the basis of the operational

ampliﬁer (or op amp), one of the most useful devices for analog signal

processing. Probably the most surprising thing about op amps is their very

large voltage gain, usually exceeding 100 000. This chapter will give you a

clearer idea how such performance is achieved. We will also look at some

other circuits that serve as building blocks for op amps.

Apparatus required

Breadboard, oscilloscope, multimeter, three 2N3904 and three 2N3906

transistors, one 5.1 V Zener diode, three 100 , ﬁve 10 k, two 22 k one

each of 560 , 2.2 k, and 100 k

1

4

W resistors.

6.1 Differential amplifier

A differential ampliﬁer is an ampliﬁer for differential input signals, i.e.

it ampliﬁes the voltage difference of its two inputs. This is useful in two

important ways:

1. A differential ampliﬁer can be used to amplify a differential signal (the

voltage difference between the two inputs) while suppressing any noise

that is common to the two inputs.

2. As we will see in future chapters, differential ampliﬁers make it easy to

build circuits that use negative feedback.

Don’t confuse the differential ampliﬁer with the differentiator: although

the names sound similar, the two circuits perform entirely different

operations.

**What does each do?
**

75

76 Hands-on electronics

Vin(−)

10 k

Vout (+)

+15

2N3904

−15

Q

1

Q

2

R

1

R

E

10 k

10 k

100 100

R

E

R

L

R

L

Vin(+)

Vout (−)

(a)

~

(b)

Function

generator

100

10 k

Vin(+)

Fig. 6.1. (a) Differential ampliﬁer; (b) function generator with 100-to-1 attenuator.

6.1.1 Operating principle

Fig. 6.1(a) shows a differential ampliﬁer made of two NPN transistors.

You can think of it as a ‘current divider’. Quiescently, the current through

R

1

is shared equally between Q

1

and Q

2

. If a differential input signal is

applied to the bases of Q

1

and Q

2

, some collector current shifts from one

transistor to the other. This change, I , causes a positive voltage change at

one collector and a negative voltage change at the other, i.e. a differential

output signal, as explained in more detail below.

6.1.2 Expected differential gain

The differential voltage gain can be understood as follows. Suppose we

apply a differential signal consisting of equal and opposite changes in the

base voltages of Q

1

and Q

2

:

V

in+

= −V

in−

≡ V. (6.1)

Since the emitter voltages follow the base voltages, a similar voltage

change occurs between the two emitters. This causes a current I to ﬂow

across the two emitter resistors. Since these are in series with the dynamic

emitter resistances of the two transistors (each of which has an approxi-

mate value r

e

= 1/g

m

= 25 mV/I

C

according to the Ebers–Moll model),

we have

I = V/(R

E

+r

e

). (6.2)

77 6 Transistors III: differential amplifier

Since a purely differential signal raises the voltage at one emitter by

the same amount that it lowers the voltage at the other emitter, it does

not change the voltage at the top of R

1

; thus, the current through R

1

is

constant. This means that the current due to the differential signal must add

to the collector current of one transistor and subtract from that of the other

transistor, so the differential voltage gain should be R

L

/(R

E

+r

e

). One-

half of the ampliﬁed differential signal appears at each collector. In other

words, a small differential input voltage applied to the two bases causes a

large differential output signal at the two collectors.

6.1.3 Measuring the differential gain

Wire up the circuit, check it against your schematic (Fig. 6.1), and try it out.

Because it has a large gain, you will need to use a small enough amplitude

out of the function generator so that the ampliﬁed signal is not ‘clipped’

(cut off at the top or bottom). However, when the function generator is

set to a small amplitude, it puts out a rather noisy signal with a relatively

large DCoffset. To avoid these problems, run the function-generator output

through a 100-to-1 attenuator (voltage divider) made from a 10 k resistor

in series with a 100 resistor to ground (Fig. 6.1(b)).

**Measure the attenuation (voltage-division ratio) by setting the function
**

generator for a large amplitude and measuring the signal amplitudes

before and after the voltage divider. Compare your observed attenuation

with the theoretical value.

Now that you know the attenuation, you can display the function-

generator output on the scope and calculate from it the size of the actual

input signal to the ampliﬁer.

Connect the output of your attenuator to the base of Q

1

, and observe the

ampliﬁer outputs at the collectors of Q

1

and Q

2

.

**Measure the differential voltage gain A
**

diff

=(V

out+

− V

out−

)/(V

in+

−

V

in−

) for a fewdifferent input amplitudes and frequencies; compare with

what you expect.

The ampliﬁer circuit clips its output when all of the available current has

been switched to one transistor or the other; this determines maximum and

minimum voltages, beyond which the output cannot go.

**Try it and see. At what output voltages does clipping set in? Compare
**

with what you expect – you can estimate the amount of current that is

available to either transistor by measuring the voltage drop across R

1

.

78 Hands-on electronics

6.1.4 Input offset voltage

If you ground both inputs, what are the output voltages? For an ideal differ-

ential ampliﬁer, they should be equal, but you will probably ﬁnd that, due

to small mismatches between the two collector resistors, the two emitter re-

sistors, and the two transistors, they are not. To obtain exactly equal outputs

you would have to input a small voltage difference, called the input offset

voltage, which you can estimate as the output-voltage difference divided

by the voltage gain.

**How big an input offset voltage do you obtain this way?
**

6.1.5 Common-mode gain

It is desirable for a differential ampliﬁer to be insensitive to common-

mode input, i.e. identical signals applied to both inputs. This feature (called

common-mode rejection) is useful when sensing a small signal in the pres-

ence of noise, since often the noise is in common on both inputs and can

be subtracted away by a differential ampliﬁer.

Test the common-mode rejection of your differential ampliﬁer:

**Connect both inputs to the same sine wave from the function genera-
**

tor (i.e. V

in+

= V

in−

). What do you observe at the outputs? You should

see almost identical signals on both outputs. What common-mode gain

(A

CM

+ = V

out

+/V

in

+ and A

CM

− = V

out

−/V

in

−) do you ob-

serve for each output, and why?

Common-mode rejection is usually speciﬁed in terms of the common-

mode rejection ratio expressed in decibels:

CMRR = 20 log

10

A

diff

A

CM

. (6.3)

**What value of CMRR do you observe?
**

Although the common-mode gain of this circuit is small, it can still be

a nuisance in practice. It could be reduced by increasing the size of R

1

,

but that would reduce the amount of current ﬂowing through Q

1

and Q

2

,

increasing r

e

and reducing the differential gain. A better solution is to

replace R

1

with a current source (shown in Fig. 6.2).

1

Try this.

**Measure how much current your current sink sinks, calculate what you
**

expect, and compare.

1

Strictly speaking, it is a current ‘sink’ since positive current ﬂows into the collector as indicated.

79 6 Transistors III: differential amplifier

560

2.2 k

2N3904

−15

~ 2 mA

1N4733

5.1 V

Fig. 6.2. Current sink for differential ampliﬁer.

**What output signals do you observe now for a common-mode input?
**

What is the common-mode gain now?

It should be very much smaller than previously – if the output signal is so

small that you have trouble measuring it, you can at least set an upper limit

on it and on the common-mode gain.

2

Save your three-transistor differential ampliﬁer for use below.

6.2 Op amps and their building blocks

An operational ampliﬁer is a differential ampliﬁer with a single-ended

output and as high a differential gain as possible (typically >10

5

). Op

amps are manufactured as integrated circuits. They are typically used with

DCcoupling and with negative feedback fromoutput to input. Their internal

design includes level-shifting circuitry so that the single output is at ≈0 V

if the two input voltages are equal.

6.2.1 Current mirror

To achieve high gain, in op amps the emitter resistors are typically omitted,

and the collector resistors are replaced by current sources. Acurrent mirror

is a convenient conﬁguration for this purpose.

Build the PNP current mirror of Fig. 6.3. The current out of Q

4

‘pro-

grams’ an approximately equal current out of Q

5

as follows: since Q

4

’s

collector is connected to its base, it is held at a V

BE

drop below the positive

supply. This determines the current out of Q

4

by Ohm’s law applied to R.

2

To set an upper limit, assume that the output signal equals the precision of your measurement.

80 Hands-on electronics

10 k

2N3906

+15

R

Q

4

Q

5

I

out

I

prog

load

10 k pot

1 k

Fig. 6.3. Current mirror.

Since the bases are connected together, both transistors have the same value

of V

BE

, and thus their collector currents will match if they have matching

values of I

s

and β and are at the same temperature.

In practice, there will always be a slight current mismatch since the

programming current includes the base currents of Q

4

and Q

5

and the

output current does not. Also, since β increases with V

CE

, and V

BE

at a

given collector current depends slightly on V

CE

(called the ‘Early effect’),

the current mismatch will depend on the output voltage.

You can explore this using a variable load, as shown in Fig. 6.3. Monitor

the collector voltage as you adjust the load resistance.

**How does the output current vary with collector voltage? What is the
**

approximate dynamic resistance of the output? Is this a better or a worse

current source than the ones you built in previous chapters?

6.2.2 Differential amplifier with current-source loads

Various tricks can be used to improve the performance of the current mirror,

as we will see below. But ﬁrst, hook up your simple current mirror to your

three-transistor differential ampliﬁer, replacing the 10 k collector resistors,

and remove the 100 emitter resistors (see Fig. 6.4). Connect a scope

probe to the collector of Q

2

. If the inputs balance, the currents through Q

1

and Q

2

will be equal, and so will the currents through Q

4

and Q

5

. But if

a differential signal is present at the input, the current mismatch between

Q

2

and Q

5

must ﬂow through the 10 M scope-probe input impedance. You

should then see an enormous differential gain! If Q

5

were an ideal current

source, the gain would be 10 M/2r

e

.

**Explain the last statement.
**

81 6 Transistors III: differential amplifier

V

in(−)

V

out

2N3904

Q

1

Q

2

V

in(+)

560

2.2 k

−15

1N4733

5.1 V

2N3906

+15

Q

4

Q

3

Q

5

1 k pot

offset

adjust

+15 −15

22 k 22 k

Fig. 6.4. Differential ampliﬁer with current-mirror load.

In practice, the collector of Q

5

will have an output impedance less than

10 M, and so the gain will be lower.

You may ﬁnd it desirable to increase the attenuation of your input voltage

divider from 100 to 1000 here. Also, to avoid clipping of the output signal,

you will want to arrange for the quiescent collector voltage of Q

2

to be

around 7 V – you can adjust it by hooking up the base of Q

2

not to ground,

but to the wiper of the 1 k pot, with one end of the pot connected through

a resistor to +15 V and the other end connected through a second resistor

to −15 V, as shown in Fig. 6.4. Due to the high gain of this ampliﬁer, the

output voltage will be sensitive to small variations at either input. If you

observe a DC output signal near either ground or +15 V, it is likely that

you need to ﬁne-tune the offset voltage using the 1 k pot. Also note that a

gain value approaching 1000 is not unreasonable.

**What gain do you observe? Compare with the gain you would expect if
**

Q

2

’s effective load were 10 M. What effective load resistance do you

infer? Is this consistent with the dynamic resistance of the current-mirror

output you observed in section 6.2.1?

**The sensitivity of this circuit is easily demonstrated. Try warming either
**

Q

4

or Q

5

by gently squeezing the transistor between your ﬁnger and

thumb. Observe how the output changes and suggest an explanation for

your observation. Try doing the same for the other transistor.

82 Hands-on electronics

6.2.3 Improved current mirror

You can improve the output impedance of your current sources by adding a

small resistor in series between the positive supply and the emitter of each

PNP transistor.

**Try this with two 100 resistors and see by howmuch the gain increases.
**

What effective load resistance do you infer now?

6.2.4 Wilson current mirror

You can do even better by converting the simple current mirror to a Wilson

current mirror. Do this by adding a third PNPtransistor as shown in Fig. 6.5.

This clever circuit beats the Early effect by ﬁxing Q

5

’s V

CE

– at the same

time, it also symmetrizes the base-current mismatch of the simple current

mirror. Explain.

**Try this to see by how much the gain increases. Note that you will need
**

to re-tune the collector voltage of Q

2

by adjusting the offset voltage.

What effective load resistance do you infer now?

V

out

Q

1

Q

2

V

in(+)

560

2.2 k

−15

1N4733

5.1 V

+15

Q

4

Q

3

Q

5

Q

6

V

BE

V

BE

2 mA

V

in(−)

1 k pot

offset

adjust

+15 −15

22 k 22 k

Fig. 6.5. Differential ampliﬁer with Wilson-current-mirror load.

83 6 Transistors III: differential amplifier

IC op amps have even higher gain than this, of course, as well as higher

input impedance. Higher input impedance can be achieved by using

Darlington transistor pairs in place of the input transistors, or by using

FET inputs instead of bipolar transistors. The gain can be increased fur-

ther by adding a second stage of ampliﬁcation after the differential pair.

To achieve low output impedance, the output is usually buffered with an

additional transistor stage.

7

Introduction to operational amplifiers

An operational ampliﬁer is a high-gain DC-coupled ampliﬁer with differ-

ential inputs and single-ended output. Op amps were originally developed

as vacuum-tube circuits to be used for analog computation. Nowadays they

are packaged as integrated circuits (ICs). Such devices can closely approxi-

mate the behavior of an ideal ampliﬁer, and their use avoids the necessity of

coping with the messy internal details of ampliﬁer circuitry. Thus, an IC

op amp is often the device of choice in scientiﬁc instrumentation. In this

chapter we will introduce the op amp and its most common applications.

Apparatus required

Breadboard, oscilloscope, multimeter, two 741 op amps, one further 741

(optional), one 100 , three 10 k, two 100 k, one 1 M

1

4

W resistors, and

four more 10 k resistors (optional).

7.1 The 741 operational amplifier

The IC we shall be using is a general purpose op amp designated by the

number 741. The 741 is a very popular and successful design, useful for sig-

nals fromDCto beyond audio frequency (though in recent years FET-input

op amps such as the LF411 have been gaining on the 741 in popularity).

It is available from most manufacturers of linear integrated circuits (chips

that produce an output proportional to their inputs, as opposed to digital

ICs, whose outputs have typically only two states).

Each manufacturer has a different system of nomenclature for ICs, e.g.

National Semiconductor calls the 741 an LM741, Fairchild a A741, etc.,

but the 741s made by different manufacturers are all electrically compat-

ible. To add complication, the 741 is available in various package styles

and is rated for use in various temperature ranges. The one we use is the

85

86 Hands-on electronics

5

6

7

8

4

3

2

1 offset null

inverting

input

noninverting

input

V−

V+

output

offset null

no connection

−

+

741 op amp

(top view)

Fig. 7.1. Diagram of eight-pin DIP 741 package showing ‘pinout’. Often, in addition to

(or instead of) the notch at the ‘pin 1’ end of the package, there is a dot next to pin 1.

741C(commercial temperature range, 0 to 70

◦

C) in the eight-pin mini-DIP

(dual in-line pin) plastic package. This package is convenient because the

two rows of pins easily straddle the groove running down the center line

of a breadboard socket block.

Youcanﬁndthe manufacturer’s data sheet for the 741onthe webor injust

about any electronic textbook or linear-IC data book. As shown on the data

sheet, the 741 actually has on its single silicon crystal a multiple-stage two-

input DC ampliﬁer consisting of twenty transistors, eleven resistors, and

one capacitor.

1

Its output voltage is proportional to the voltage difference

between the two inputs, i.e. it is a ‘differential’ ampliﬁer. A differential

ampliﬁer is convenient when one wants to study the difference between

two almost identical signals, as well as in many other applications, as we

shall see. (Of course, this is not to be confused with a differentiator! Recall

that a differentiator outputs the time derivative of its input signal, not at all

what a differential ampliﬁer does.)

7.1.1 741 pinout and power connections

Fig. 7.1 shows the layout of the eight-pin DIP package. The package is a

rectangular piece of black plastic containing the silicon chip itself as well

as the ﬁne gold wires that connect the chip to the contact pins. Look at the

top face of the package and orient yourself as to which pin is pin 1 – looking

at the top, you see the pins bending away from you and the pin numbers

increasing in the counterclockwise direction. The end at which pin 1 is

located is indicated by a special mark – depending on the manufacturer,

not necessarily the mark shown in Fig. 7.1. Pins 1–4 are located along one

long edge of the package, while pins 5–8 are on the opposite edge.

1

Nowadays, of course, this is nothing – for example, the Pentium chip had 3.3 million transistors.

87 7 Introduction to op amps

The 741C is rated for maximum supply voltages of ±18 V, and the

recommended range is ±(≤15) V. To be on the safe side, before you begin

to build your circuit, turn on the breadboard power and adjust the power

supplies to +15 V and −15 V.

Next, turn off the power and insert the op amp into the breadboard,

straddling the central groove of a socket block, with pins 1–4 toward the

left and 5–8 toward the right. Note that the pins are delicate and are easily

bent or broken. If they are too bent to plug into the sockets, straighten them

carefully, preferably using needle-nose pliers. Run wire jumpers from the

V+pin to the +15 Vbus and fromthe V−pin to the −15 Vbus. Any point

of your circuit that is to be ‘grounded’ should be attached to the common

(or ground) bus.

Note: In the circuits shown below, the pin numbers have been omitted. It is good practice

for you to write in the pin numbers yourself before hooking up the circuits, to reduce

the possibility of confusion. Trying to work out pin numbers ‘on the ﬂy’ and keeping

themin your head instead of writing themdown is a common cause of errors in hooking

up circuits.

Also note: Even though the power connections are usually not shown on schematic

diagrams of op amp circuits, the positive and negative supplies must always be con-

nected, or the op amp won’t do anything!

7.1.2 An ideal op amp

For a (hypothetical) ideal op amp, zero volts between the inverting (marked

‘−’ in schematics) and noninverting (marked ‘+’) inputs would yield zero

volts at the output (relative to ground); in other words

r

the common-mode gain and DC offset of an ideal op amp are zero.

Moreover,

r

the differential gain and input impedance of an ideal op amp are inﬁnite

and the output impedance is zero;

r

the bandwidth (frequency range over which the op amp can correctly

operate) and slew rate (rate at which the output voltage can change) of

an ideal op amp are inﬁnite.

While, of course, it is impossible to build a circuit having these ideal

characteristics, one can come remarkably close. The above statements are

approximately true for practical op amps – for example, one can buy op

amps with gains of 10

5

to 10

6

.

88 Hands-on electronics

V

out

−

+

R

2

op amp

R

1

~

V

in

I

1

I

2

Fig. 7.2. Op amp inverting-ampliﬁer circuit. Note the negative feedback resulting from

the resistor that connects the output to the inverting input. Op amps are almost always

used with negative feedback.

We shall see next that these approximations lead to a very simple way

of analyzing op amp circuits.

7.1.3 Gain of inverting and noninverting amplifiers

Fig. 7.2 shows an op amp conﬁgured as an inverting ampliﬁer. The

key principle at work in this circuit is negative feedback. The idea is

that a fraction of the output signal is applied at the inverting input.

Since the gain of the op amp is large and the noninverting input is grounded,

any nonzero voltage at the inverting input will cause a large output

voltage of the opposite sign. If you think about it, you will see that the

only stable situation that can result is that the voltage difference between

the inverting and noninverting inputs is zero. In other words, the op

amp will do whatever is necessary to zero the voltage difference at its

inputs.

Once this principle is grasped, it is easy to compute the gain of the op

amp inverting ampliﬁer. Assuming the input currents of the op amp are

zero, all the current ﬂowing in through R

1

must ﬂow out through R

2

, i.e.

I

2

= I

1

. Assuming that the open-loop voltage gain (i.e. that without any

feedback) of the op amp is inﬁnite, the voltage difference at the op amp’s

inputs must be zero. Applying Ohm’s law to R

1

and R

2

, and designating

the voltage at the inverting input as V

−

,

V

−

= V

in

− I

1

R

1

= 0 (7.1)

⇒ I

1

= −

V

in

R

1

(7.2)

89 7 Introduction to op amps

V

out

= V

−

− I

2

R

2

= −I

1

R

2

(7.3)

⇒V

out

= −

R

2

R

1

V

in

. (7.4)

Thus, the closed-loop voltage gain (i.e. the gain with feedback) of this

circuit is

A

v

=

V

out

V

in

= −

R

2

R

1

. (7.5)

In a nutshell, since all of the current due to the input signal ﬂows around

the op amp, the output voltage is determined entirely by Ohm’s lawapplied

to R

1

and R

2

. Thus, if R

1

= 10 k, a gain of −10 can be achieved by choosing

R

2

= 100 k, and a gain of −1 results from choosing R

2

= 10 k.

Fig. 7.3 shows an op amp conﬁgured as a noninverting ampliﬁer. Again,

neglecting the tiny input currents of the op amp and assuming inﬁnite open-

loop gain, application of Ohm’s lawshows that the closed-loop voltage gain

of this circuit is

A

v

=

I

1

R

1

+ I

2

R

2

I

1

R

1

= 1 +

R

2

R

1

. (7.6)

As above, Eq. 7.6 follows fromthe assumptions that all of the current ﬂows

around the op amp and that feedback forces the inverting input to follow

the signal applied to the noninverting input. So if R

1

= 10 k, a gain of

11 can be achieved by choosing R

2

= 100 k, while a gain of 2 results from

choosing R

2

= 10 k.

V

out

+

−

R

2

R

1

V

in

op amp

I

2

I

1

~

Fig. 7.3. Op amp noninverting-ampliﬁer circuit.

90 Hands-on electronics

7.1.4 Op amp ‘golden rules’

We see from the above that to understand to a reasonable approximation

an op amp circuit with negative feedback, you need only two simple rules.

1. The inputs draw no current.

2. The output does whatever is necessary to maintain the two inputs at

equal voltages.

7.1.5 The nonideal op amp

Although the ideal op amp approximations are very close to reality, in most

situations, here are a few limitations that you should consider:

Input offset voltage

A small DC output voltage usually results even when the inputs are iden-

tical. The input offset voltage refers to the DC voltage that must be ap-

plied at the input to achieve exactly zero volts at the output. The nonzero

offset arises due to manufacturing limitations; however, most op amps

have pins that allow for external adjustment of the offset. The input offset

voltage must be considered when designing small-signal or high-gain cir-

cuits. For the 741C op amp, the input offset voltage is speciﬁed to be less

than 6 mV.

Input bias current

An ideal op amp would drawno current at its inputs; however, real op amps

require a small input bias current for proper operation. You can think of

this as the equivalent of the base current for a bipolar transistor, or the gate

current of a JFET. Although these currents are small, they are not zero. For

the 741C op amp, the input bias current is speciﬁed to be less than 500 nA.

Slew rate

The output voltage of a real op amp cannot change instantaneously. The

maximum rate at which the output can change is called the slew rate, and

is typically in the range of volts per microsecond. The slew rate can be

a serious limitation at large output amplitudes and high frequencies. The

741 slew rate is typically 0.5 V/s. High-speed op amps are available with

slew rates of 2000 V/s.

91 7 Introduction to op amps

7.2 Experiments

7.2.1 Testing open-loop gain

First, try a 741 in the open-loop circuit shown in Fig. 7.4. (As mentioned

above, open-loop means that there is no feedback connection between the

output and input; op amps are never actually used in this way.) The 1000-

to-1 attenuator in the input circuit means that, by adjusting the pot, you can

vary the input between −15 and +15 mV. Try to adjust the input so that

the output sits near zero volts.

**Vary the input in steps of 100 or 200 Vover a range that causes the out-
**

put to vary from its maximum negative voltage to its maximum positive

voltage, taking several readings of input and output voltage as you do

so. (You may ﬁnd that the output switches states within a single step!)

Approximately what input voltage would result in an output voltage near

zero? This is the input offset voltage of your 741.

**What are the op amp’s positive and negative output saturation voltages
**

(i.e., the maximum voltages it can output)?

If your chip has particularly high gain, you may ﬁnd that no setting of

the input voltage causes the output to sit near zero volts; however, you can

still set upper and lower limits on the input offset voltage by determin-

ing at what input voltages the output switches between its negative and

positive saturation voltages. Since your applied input voltages are so tiny,

the output voltage is very susceptible to input noise, and you may also

observe substantial drift of the output voltage with time. Still, you should

V

out

_

+

+15

−15

100 k

100

10 k

Fig. 7.4. Open-loop op amp test circuit.

92 Hands-on electronics

be able to estimate the gain roughly, or at least place a lower limit

on it.

**What gain do you observe? Is it consistent with the manufacturer’s spec-
**

iﬁcation ‘gain (typical) = 200 000’?

Youmayﬁndthat stabilityandnoise are improvedbykeepingthe connec-

tions between the noninverting input, the 100 resistor, and the inverting

input as short as possible – long wires in this current path act as an antenna

and pick up electromagnetic noise from the environment, as you will soon

observe.

7.2.2 Inverting amplifier

Now hook up the inverting ampliﬁer of Fig. 7.2, with R

1

= 10 k and

R

2

= 100 k. Using a 1 kHz sinusoidal input signal with an amplitude of

approximately 1 V, look at the input and output.

**Verify that the ampliﬁer inverts. Measure the gain and compare with
**

what you expect. Replace the 100 k feedback resistor with 10 k – what

gain do you predict and observe now? (Put the 10 k resistor back for the

following parts.)

Due to feedback, the output impedance of the inverting ampliﬁer is

extremely small: the ≈75 open-loop output impedance of the 741 is

multiplied by the ratio of the closed-loop gain to the open-loop gain, here

a factor of order 10

−5

.

**Conﬁrm the small size of the closed-loop output impedance by adding
**

a 100 load to ground at the output. To see that this gives sensitivity to

the output impedance, analyze the voltage-divider circuit consisting of

the output impedance in series with the load resistance.

**Do you expect the output amplitude to decrease measurably under load?
**

How big a decrease do you predict? Is this consistent with what you

observe? (You will need to use a small input signal – how small? – to

avoid running into the 741’s ≈25 mA output-current limit.)

Remove the 100 load and switch to a 10 kHz square wave as the

input. Compare the appearance of the output signal and the input signal.

For sufﬁciently large amplitude, you are likely to observe that the lead-

ing and trailing edges of the output square wave are not quite vertical.

**Measure the slope of the leading edge in volts per microsecond. This is
**

the slew rate of the 741C.

According to the manufacturer, the typical slew rate is 0.5 V/s. There

exist more expensive op amps with much higher slew rates.

93 7 Introduction to op amps

10 k

V

out

+

−

~

V

in

+15

−15

10 k

10 k

10 k

pot

R

2

R

1

I

1

I

2

I

1

+I

2

R

3

V

0

Fig. 7.5. Circuit for demonstrating a summing junction. Since the inverting input is held

near ground due to feedback, V

out

= (I

1

+ I

2

) · R

3

, where I

1

= V

in

/R

1

and I

2

= V

0

/R

2

.

If R

1

= R

2

= R

3

, V

out

= V

in

+ V

0

.

The inverting input is called a ‘virtual’ ground because it is kept at zero

volts by feedback. Thus, even though the open-loop input impedance of

the inverting input is very large, it acts here like a short circuit to the

noninverting input (ground). Since the input signal sees R

1

to ground, the

input impedance of the inverting ampliﬁer is equal to the value of R

1

.

Note that the virtual ground at the inverting input can also be used as

a ‘summing junction’: all currents arriving at that point are summed and

passed through the feedback resistor to the output. To demonstrate this,

feed an adjustable current into the summing junction, by adding a 10 k

pot with 10 k resistor in series with the wiper, as shown in Fig. 7.5, and

examine the ampliﬁer output for a small-amplitude 1 kHz sinusoidal input.

**What happens to the output DC offset as you adjust the pot, and why?
**

7.2.3 Noninverting amplifier

Set up the noninverting ampliﬁer circuit of Fig. 7.3 with R

1

= 10 k.

**With a 1 kHz sinusoidal input, measure the gain with R
**

2

= 100 k and

with R

2

= 10 k. Compare with what you expect. Verify that the ampliﬁer

is noninverting. (Leave the 100 k resistor in for the following parts.)

**Try to measure the input impedance by putting a 1 M resistor in series
**

with the input and looking at the signal before and after the resistor.

Explain your result.

The input impedance can be inferred by analyzing the voltage-divider cir-

cuit consisting of the 1 Mresistor in series with the input impedance. If your

answer is 10 M, consider that this is the input impedance of the attenuating

94 Hands-on electronics

scope probe! You can get around it by looking only at the output, with and

without the 1 M input resistor.

**What value do you get for Z
**

in

this way?

Again you are seeing the effect of negative feedback: even though the open-

loop input impedance of the noninverting input is only of order megohms,

it is, in principle, multiplied by the ratio of the open-loop gain to the closed-

loop gain, here a factor of order 10

5

. (In practice, Z

in

is limited by other

effects, such as the capacitance to ground at the noninverting input.)

**Be careful not to be confused by the DC shift in the output produced by
**

the 1 M input resistor! Explain how this shift results from the op amp’s

input bias current. What value for the input bias current is implied by

the observed DC shift?

The output impedance should of course be the same as for the inverting

ampliﬁer, since, as far as output impedance is concerned, it is the same

circuit!

**Explain this last statement. (Hint: what was the only thing you had to
**

change to make the noninverting amp from the inverting amp?)

7.2.4 Voltage follower

As shown in Fig. 7.6(a), connect the output of the op amp directly to the

‘−’ input, and connect the output of the function generator to the ‘+’ input

of the op amp.

**After turning on the power, conﬁrm that the input and output signals are
**

identical and that the voltage follower is noninverting. Record the input

and output amplitudes.

**Of what possible use is an ampliﬁer of unity gain that does not even
**

invert the signal?

As an illustration of the voltage follower’s usefulness, coil a long wire

(30 to 50 cm in length) around the outside of an AC power cord. (Any

power cord will do provided that it is plugged in!) The breadboard cord

is usually the most convenient. Leave one end of your coil ﬂoating, and

connect the other to any convenient spot on the breadboard.

**Using your scope probe, observe the waveform. Record and explain what
**

you see.

**Given that the scope probe has an input impedance of 10 M, estimate
**

the power of this signal (i.e., how much power is dissipated by the scope

probe?).

95 7 Introduction to op amps

V

out V

in

−

+

V

in

−

+

V

out

−

+

10 k

100 k

(a)

(b)

1 MΩ

Fig. 7.6. (a) Op amp voltage follower; (b) voltage follower as the input stage to an

inverting-op-amp circuit.

You can reproduce this effect more simply by touching the probe tip

with your ﬁnger. 60 Hz noise is pervasive throughout North America

(50 Hz in Europe) and is often the dominant background noise in elec-

tronic equipment.

**Try amplifying this low-power signal using the inverting-ampliﬁer
**

circuit previously constructed. Sketch the output and record your

observations.

**Now, instead of driving the ampliﬁer directly, insert a voltage follower as
**

shown in Fig. 7.6(b). Record the follower output as well as the ampliﬁer

output. If the ampliﬁer output saturates, choose a smaller feedback re-

sistor to reduce the gain of the inverting amp. Explain your observations.

7.2.5 Difference amplifier

Fig. 7.7 shows the 741 conﬁgured as a difference ampliﬁer, with the output

voltage equal to the difference of the two input voltages. A difference am-

pliﬁer is both an inverting amp and a noninverting amp. An inverting amp is

created if V

in+

is grounded, whereas a noninverting amp is created if V

in−

is

grounded. If both inputs are connected to signals, the difference between

the two signals is ampliﬁed. This follows since, if these inverting and non-

inverting ampliﬁers have equal gains, the output is proportional to the dif-

ference of the inputs. The gains are matched provided that R

1

/R

2

= R

3

/R

4

.

96 Hands-on electronics

V

out

V

in

_

V

in+

10 k pot

10 k

100 k

R

1

R

2

R

4

R

3

−

+

Fig. 7.7. Difference ampliﬁer. The parts of the pot on either side of the slider serve as R

3

and R

4

.

Difference ampliﬁers are often used in the life sciences where signals are

small and exist within a noisy environment, e.g. in the electrocardiograph

(ECG). Any background noise common to both inputs (common-mode

noise) is rejected, while the signal of interest (present at only one of the

inputs) is ampliﬁed and appears at the output. A potentiometer is often

used (as in Fig. 7.7) to tune the gain and common-mode rejection of the

ampliﬁer.

The quality of the ampliﬁer is measured (in part) by the common-mode

rejection ratio (CMRR), based on the ratio of the differential voltage gain

and common-mode voltage gain:

CMRR = 20 log (A

diff

/A

CM

). (7.7)

It is customary to give the CMRR in decibels, as shown in Eq. 7.7. The

741 general-purpose op amp is a differential ampliﬁer with a CMRR value

speciﬁed to be at least 70 dB. Precision op amps are commercially available

with CMRR values as high as 140 dB or more.

Build the circuit shown in Fig. 7.7. As always, be careful not to short

the potentiometer’s center tap to ground or power. You can estimate the

common-mode rejection ratio by measuring both the common-mode volt-

age gain and the differential voltage gain.

The common-mode voltage gain is determined by applying identical sig-

nals to both inputs and observing the output voltage: A

CM

= V

out

CM

/V

in

CM

.

Usinga 1kHz sine wave at maximumamplitude as your input, tune your dif-

ference amp to minimize the output amplitude (i.e. adjust the potentiometer

until the output amplitude is as small as possible). The best estimate for

the common-mode gain can be made using the averaging feature from the

97 7 Introduction to op amps

‘acquire’ menu of the TDS210 oscilloscope. It may be possible, depend-

ing on the quality of your op amp, to tune the potentiometer such that no

output signal is visible, even at the most sensitive scale.

Verify that R

1

/R

2

= R

3

/R

4

.

**Estimate (or set an upper limit on) the common-mode voltage gain.
**

The resistance ratios can be measured using your meter; however, be

sure to turn off the power and disconnect each component before making

your measurements – accurate resistance measurements cannot be guaran-

teed if the component remains connected to the circuit, since then you are

measuring a parallel combination with the other components rather than

just the resistance of interest. For example, when measuring the resistances

between the potentiometer center tap and ends, be sure to disconnect all

wires connecting the pot to the input signal, op amp, and ground. Replace

all resistors and wires when done.

To measure the differential voltage gain, ground one input while applying

a 1 kHz sine wave of amplitude 1 V or less at the other input. It shouldn’t

matter which input you ground. If you’re curious, try it both ways and see

if you get equal gains.

Measure the differential voltage gain.

**Explain why it doesn’t matter which input is grounded when measuring
**

the differential voltage gain.

Estimate the CMRR for this circuit.

**What is the input impedance at each input?
**

7.3 Additional experiments

7.3.1 Current source

An ideal current source would maintain a constant current through the load,

regardless of load resistance. Try the op amp current source of Fig. 7.8.

Vary the ‘load’ pot as you measure the load current with a digital multimeter

and the load voltage with a multimeter or scope; what do you observe?

Explainhowthis circuit works (use diagrams andequations as necessary).

What should the current be, and why?

**What is dI /dV? What is the ‘compliance’ (output-voltage range over
**

which dI /dV is small) of your current source?

**Compare the performance of this op amp current source to the transistor
**

current sources you built previously.

98 Hands-on electronics

0--10 k

V

out

−

+

+15

10 k

V

A

Fig. 7.8. Op amp current source.

7.3.2 Noninverting summing amp with difference amplifier

This circuit is fun to build and to observe in action. It also serves as an

excellent demonstration of the constructive and destructive interference

between two waveforms.

As stated above, difference ampliﬁcation is usually used to eliminate

common-mode noise. The circuit shown in Fig. 7.9 sums two waveforms

and then eliminates one using the difference amp. One waveform will be

from the function generator while the other will be a 60 Hz wave created

as in section 7.2.4. (If two function generators are available, feel free to

replace the voltage follower with the output from the second generator.)

First construct the voltage follower and summing ampliﬁers as shown in

Fig. 7.9. Set the function generator to a sine wave with a frequency between

59 and 61 Hz.

2

Adjust the generator amplitude to match the amplitude of

the voltage follower and observe the output of the summing amp. The

output should be the linear sum of the two input waveforms.

Have fun with the circuit by changing the amplitude and frequency.

Observe what happens with either a triangle- or square-wave input. Using

the ‘AC line’ as your trigger source may be useful here.

Nowadd the difference ampliﬁer. With the function-generator amplitude

set to zero, adjust the potentiometer until the difference amp output has

zero amplitude (as with the difference ampliﬁer previously built). Now in-

crease the amplitude of your function generator and observe the difference

2

If the local supply frequency is not 60 Hz, set the function generator frequency to be within 1 Hz of

your local value.

99 7 Introduction to op amps

V

in

−

+

−

+

−

+

R

3

R

2

R

1

R

2

R

1

R

3

10 MΩ

10 k pot CMRR Adjustment

2

1

3

~

Summing

Output

V

out

function

generator

R

1

= = = 10 k

(or similar)

Ω R

2

R

3

R

1

Fig. 7.9. Fancy summing circuit. Amp 1 is a voltage follower used to buffer the 60 Hz

pickup on a wire wrapped around an AC power cord. Amp 2 is a noninverting summing

ampliﬁer with unity gain. Amp 3 is a difference ampliﬁer with an adjustment to maximize

the CMRR.

ampliﬁer output. Be sure to switch your trigger source back to the appro-

priate input channel. Try changing the input frequency. Replace the 60 Hz

AC-line signal at the noninverting input of the difference amp with the

output from the function generator. Observe how the output changes.

**Explain how this circuit works using diagrams and equations as needed.
**

Explain why the summing amp isn’t inverting and why it has unity gain.

**Sketch the inputs and output of the summing amp for a function gener-
**

ator frequency near 60 Hz. Why doesn’t the output have a well deﬁned

amplitude?

**Sketch the inputs and output of the difference ampliﬁer. How does the
**

output change when the inputs are switched?

**Is the output inverted with respect to the original inputs? If so, why?
**

If so, what could you change to make the output of the difference amp

noninverted with respect to the original inputs?

8

More op amp applications

In Chapter 7 we studied some of the basic properties of operational ampli-

ﬁers. There are an enormous number of ways that op amps can be applied

to process analog signals. In this chapter we will explore several such ap-

plications: circuits that differentiate or integrate their input voltage as a

function of time, form the logarithm or exponential of their input voltage,

or rectify their input voltage. The op amp versions of these applications

come closer to the ideal than the passive versions of some of them that

you studied in earlier chapters. We will also see how to use feedback to

compensate for the limitations of discrete devices.

Apparatus required

Breadboard, dual-trace oscilloscope with two attenuating probes, one 741C

and one LF411 operational ampliﬁer, one 1 k, two 10 k, and one 100 k

1

4

W

resistor, 0.0047 F and 0.033 F capacitors, two 1N914 (or similar) silicon

signal diodes, 2N3904 and 2N3906 transistors.

8.1 Op amp signal processing

Recall that for an inverting ampliﬁer made from an op amp, with input

resistor R

i

and feedback resistor R

f

, the gain is −R

f

/R

i

(neglecting the

input offset voltage and offset and bias currents and taking the op amp

open-loop gain to be inﬁnite). We can generalize this result for devices

other than resistors, as illustrated in Fig. 8.1.

A

v

= −

Z

f

Z

i

. (8.1)

Eq. 8.1 is useful if we are analyzing circuit performance in the frequency-

domain for a sine-wave input, but often we are concerned with the response

101

102 Hands-on electronics

V

out

_

+

Z

f

Z

i

V

in

Fig. 8.1. Generalized op amp inverting-ampliﬁer circuit.

_

+

R

f

C

V

out

V

in

Fig. 8.2. Basic op amp differentiator.

in the time-domain to an arbitrary input waveform. Then we can analyze

the circuit using Ohm’s law. Since, to an excellent approximation, I

f

= I

i

,

the gain is determined by the current–voltage characteristics of the input

and feedback devices, as we will see in more detail below.

8.1.1 Differentiator

As shown in the circuit of Fig. 8.2, the basic op amp differentiator (not to

be confused with the difference ampliﬁer) is similar to the basic inverting

ampliﬁer studied in Chapter 7, except that the input element is a capacitor

rather than a resistor. Using the assumption that the output does whatever

necessary to maintain the two inputs at equal voltages, it is easy to show

that the output voltage is given by

V

out

= −I R

f

= −

dQ

dt

R

f

= −R

f

C

dV

in

dt

, (8.2)

since Q = CV

in

, where Q is the charge stored on the capacitor.

103 8 More op amp applications

_

+

C

R

s

R

f

Fig. 8.3. Improved op amp differentiator.

One problem with the basic circuit is that the capacitor’s reactance,

X

C

=

1

2π f C

, (8.3)

decreases with increasing frequency. Since here Z

i

= X

C

, Eq. 8.1 shows

that the output voltage of the basic differentiator increases with frequency,

making the circuit susceptible to high-frequency noise and prone to

oscillation.

Amore practical differentiator circuit is shown in Fig. 8.3, with a resistor

placed in series with the input capacitor to limit the high-frequency gain to

the ratio R

f

/R

s

. The output voltage as a function of time is still given by

Eq. 8.2, as long as the input frequency is small compared with

f =

1

2π R

s

C

. (8.4)

For input frequencies greater than this, the performance of the circuit

approaches that of an inverting ampliﬁer with voltage gain

A

v

= −

R

f

R

s

. (8.5)

8.1.2 Integrator

By interchanging the resistor and capacitor in the differentiator circuit of

Fig. 8.2, we obtain an op amp integrator. As shown in Fig. 8.4, the resistor

R

i

is the input element and the capacitor C is the feedback element. The

output voltage, as a function of time, is given by

V

out

= −

1

R

i

C

V

in

dt, (8.6)

104 Hands-on electronics

V

out

_

+

C

V

in

R

i

Fig. 8.4. Basic op amp integrator.

_

+

C

R

i

R

s

R

2

Fig. 8.5. Improved op amp integrator.

which is proportional to the time integral (area under the curve) of the input

waveform vs. time.

As in the case of the differentiator, a more practical integrator circuit

is shown in Fig. 8.5. The resistor R

s

across the feedback capacitor, called

a ‘shunt resistor’, is used to limit the low-frequency gain of the circuit.

If the low-frequency gain were not limited, the input DC offset, although

small, would be integrated over the integration period, possibly saturating

the op amp.

To minimize the DC offset voltage resulting from the input bias current,

the resistor R

2

should equal the parallel combination of the input and shunt

resistors:

R

2

=

R

i

R

s

R

i

+ R

s

. (8.7)

105 8 More op amp applications

Since the shunt resistor limits the circuit’s low-frequency gain, Eq. 8.6

is valid for input frequencies greater than

f =

1

2π R

s

C

. (8.8)

For input frequencies less than f , the performance of the circuit approaches

that of an inverting ampliﬁer with voltage gain

A

v

= −

R

s

R

i

. (8.9)

8.1.3 Logarithmic and exponential amplifiers

By using a diode as the input or feedback element, we obtain a circuit that

takes the logarithm (Fig. 8.6) or exponential (Fig. 8.7) of its input signal.

For the log ampliﬁer, we can analyze the circuit performance as follows;

the analysis of the exponential ampliﬁer is left as an exercise.

_

+

R

i

I

i

I

f

Fig. 8.6. Op amp logarithmic ampliﬁer.

V

out

_

+

V

in

R

f

Fig. 8.7. Op amp exponential ampliﬁer.

106 Hands-on electronics

The current I

f

through the feedback element equals the input current I

i

,

which is determined by the input voltage V

in

:

I

f

= I

i

=

V

in

R

i

. (8.10)

We can relate this to the output voltage V

out

using the exponential diode

current–voltage law:

I

f

= I

s

e

−eV

out

/nkT

−1

. (8.11)

The minus sign in the exponential reﬂects the fact that the anode of

the diode is connected to virtual ground; thus, for I

f

positive, V

out

is neg-

ative. The constant n has been introduced since (as we saw in Chapter 3)

the slope of the exponential for a silicon diode is not quite as steep as e/kT;

one ﬁnds experimentally that n ≈ 2 for silicon diodes and n ≈ 1 for

germanium. Thus the output voltage is

V

out

≈ −

nkT

e

ln

I

i

I

s

(8.12)

≈ −

nkT

e

(ln V

in

−ln I

s

R

i

). (8.13)

In practice, one should add additional components to compensate for

the temperature dependences in Eq. 8.13 (both the explicit kT factor and

the temperature dependence of I

s

). Often, a transistor is used in place of

a diode, since experimentally one ﬁnds that a transistor gives an accurate

exponential characteristic over a wider range of current.

8.2 Experiments

Complete at least the differentiator or integrator exercise, the logarithmic-

or exponential-ampliﬁer exercise, and the half-wave rectiﬁer and push–pull

driver exercises. (You should understand the theory of all of the circuits

discussed, including the ones you choose not to build.)

8.2.1 Differential and integral amplifiers

Differentiator

Set up the active differentiator of Fig. 8.3, with R

s

= 1 k, R

f

= 10 k, and

C = 0.033 F. Adjust the peak-to-peak voltage of the triangle-wave input

to 10 V, and use a frequency in the neighborhood of 500 Hz.

107 8 More op amp applications

**Measure carefullythe peak-to-peakvoltage V
**

p–p

andperiod T of the input

signal, and from them compute the slope of the input voltage vs. time,

dV

in

dt

=

2V

p–p

T

. (8.14)

Now look at the function-generator output on channel 1 of the scope and

the op amp output on channel 2, while triggering on channel 1. You should

see that the op amp output signal is a square wave that is 90

◦

out of phase

with the input, i.e., the output signal is a representation of the negative of

the time derivative of the input.

**Measure the step height in volts of the square-wave output of the
**

op amp.

The theoretical prediction is

step height = 2R

f

C

dV

in

dt

. (8.15)

Now change the input frequency from 500 Hz to 10 kHz. Be sure to

reduce the input amplitude to avoid saturating the output voltage. Record

the appearance of the output signal at this frequency. Measure the peak-to-

peak output voltage and determine the voltage gain.

**Sketch the input and output waveforms at 500 Hz and 10 kHz and com-
**

ment on your results.

**Compare your data with Eq. 8.15 and compare the measured gain at
**

10 kHz with the theoretical expectation.

**At what approximate frequency does this circuit cease to act as a differ-
**

entiator (i.e. approach the operation of an inverting ampliﬁer)?

Integrator

Set up the circuit shown in Fig. 8.5, with R

i

= R

2

= 10 k, R

s

= 100 k, and

C = 0.0047 F. Adjust the peak-to-peak voltage of the square-wave input

to 1 V and the frequency to 10 kHz. You should see an output signal that

is a triangle wave 90

◦

out of phase with the input square wave.

Derive Eq. 8.6.

**Measure the peak-to-peak voltage of the triangle wave and compare with
**

the value you would expect theoretically.

Now change the input frequency to 100 Hz.

**Record, describe, and explain the appearance of the output signal at this
**

frequency.

**Measure the peak-to-peak output voltage and determine the voltage gain,
**

comparing with what you would expect theoretically.

108 Hands-on electronics

**Sketch the input and output waveforms at 10 kHz and 100 Hz, and
**

comment on your results. At what approximate frequency will this circuit

cease to act as an integrator (i.e. approach the operation of an inverting

ampliﬁer)?

**Compare the output amplitudes with the theoretical expectation.
**

8.2.2 Logarithmic and exponential amplifiers

Set up the circuit of Fig. 8.6 with R

i

= 10 k or Fig. 8.7 with R

f

= 10 k.

Figure out how to apply an adjustable DC input voltage.

**Verify the logarithmic or exponential gain characteristic for several input
**

voltages that probe the full range of output voltage of which the circuit

is capable.

**Explain how the circuit works and derive an expression that relates the
**

input voltage to the output voltage.

**Make a semilog plot of your data and ﬁnd the experimental value of n
**

for your diode.

**The log of zero is undeﬁned and any number raised to the power zero
**

equals one. Why then do the logarithmic and exponential ampliﬁers give

an output of zero volts when the input is zero volts?

8.2.3 Op amp active rectifier

Fig. 8.8(a) shows a simple op amp half-wave rectiﬁer.

**Build it with R = 10 k and try it out with a low-frequency sine-wave
**

input, in the neighborhood of 100 Hz. Record the input and output

waveforms.

This circuit suffers froma drawback: there is effectively no feedback during

the half of the input-wave cycle when the diode is reverse-biased! (This is

obvious since the voltages at the inverting and noninverting inputs are not

equal during that time.) In other words, the reverse-biased diode has such a

large impedance that the gain becomes very large and the op amp saturates.

**To verify this, look at the waveform at the op amp’s output pin.
**

Because of the limited slew rate of the op amp, saturation limits the

circuit’s performance at high input frequencies. Try a high frequency (say

10 kHz) and record in detail what you see, both at the op amp output and

at the rectiﬁer output. Replace the 741 with a higher speed op amp such

as the LF411, and repeat your observations. Also, be sure to use a signal

109 8 More op amp applications

V

out

_

+

V

in

R

V

out

_

+

V

in

R

R

(a) (b)

Fig. 8.8. (a) Simple op amp half-wave rectiﬁer; (b) improved version.

or switching diode like the 1N914 – while switching diodes are designed

for high frequencies, rectiﬁer diodes (e.g., the 1N4001) tend to have large

junction capacitance and thus have poor performance at high frequency.

The circuit of Fig. 8.8(b) overcomes the slew rate limitation.

**Build it, try it out, and ﬁgure out why it has much better response at high
**

frequency. Explain how each circuit works (use diagrams if necessary)

and explain why one is better than the other.

Sketch V

in

, V

out

, and the op amp output for both half-wave rectiﬁer

circuits at 100 Hz and at 10 kHz.

**How is this active rectiﬁer better than a simple diode rectiﬁer? (For
**

example, what would be the response of a simple rectiﬁer to an input

signal of amplitude less than 0.7 V? How does this circuit respond to

such a signal?)

**Comment on the performance of the LF411 as compared with that of
**

the 741.

8.2.4 Op amp with push--pull power driver

A typical op amp (such as the 741) by itself does not have enough output-

current capability to drive an 8 load such as a speaker. The addition of

a push–pull driver stage to buffer the output is a common solution. When

large output power is needed (several watts) as in audio applications, power

Darlingtons are often employed as the push–pull transistors.

Recall that in a previous chapter you used a push–pull power driver

to drive the breadboard’s speaker from the function generator. This circuit

displayed crossover distortion since one transistor switched off 2V

BE

before

the other switched on. One solution to crossover distortion is to employ an

110 Hands-on electronics

V

in

+5

2N3904

2N3906

−5

Speaker

_

+

+5

−5

V

in

+5

2N3904

2N3906

−5

Speaker

_

+

+5

−5

(a)

(b)

Fig. 8.9. Op amp follower with push–pull output-buffer power driver with two feedback

arrangements: (a) feedback before and (b) feedback after power driver.

op amp to compare the output signal with the input signal and correct the

drive signal to the push–pull stage to compensate for the 2V

BE

gap.

**First hook up the circuit of Fig. 8.9(a) to see crossover distortion in
**

action. This circuit is susceptible to noise, so be neat and orderly and

keep the leads as short as possible. Use an audio-frequency sine wave,

in the vicinity of 1 kHz.

Record the input and output waveforms – how do they differ, and why?

**Rearrange the feedback loop to include the push–pull driver inside it
**

(Fig. 8.9(b)), and compare the input and output waveforms again. What

does the signal at the op amp output look like, and why?

**How much power is dissipated in the speaker assuming a sine wave of
**

amplitude 4 V? How does the peak current through the speaker compare

with the 741’s maximum output current?

**Sketch the op amp input, the op amp output, and the waveform at the
**

speaker for bothcircuits andexplainhowthe opampeliminates crossover

distortion.

Estimate the total power consumed by this circuit.

**What is the maximuminput amplitude that can be accurately reproduced
**

at the speaker (without clipping)?

111 8 More op amp applications

Logarithmic

Amplifier

Linear

Amplifier

(gain = A)

Exponential

Amplifier

V

in

V

out

V

out

= V

in

A

Fig. 8.10. Block diagram showing how to build an ‘exponentiator’: a circuit that creates

an output voltage equal to the input voltage raised to any desired power.

8.3 Additional exercises

The availability of log and antilog circuits allows elaborate mathematical

operations on voltages. For example, the square of a number can be found

by taking the logarithm of the number, multiplying by two, and then taking

the antilogarithm of the result (see Fig. 8.10); similarly, the product of two

numbers equals the antilog of the sum of their logarithms:

x

2

= log

−1

(2 log x) (8.16)

x × y = log

−1

(log x +log y). (8.17)

Choose an arithmetic function of your choice (other than addition, sub-

traction, and multiplication by a constant) and design a circuit using op

amps to perform that function. Make a plot of the output vs. input voltages

to verify that the circuit works correctly. Discuss the limitations of your

circuit.

9

Comparators and oscillators

In this chapter you will encounter some applications of positive feedback

in op amp and comparator circuits. You will see how uncontrolled feed-

back can cause unwanted oscillation, and howcontrolled positive feedback

(hysteresis) can be used to eliminate unwanted oscillation or produce in-

tentional oscillation. There is also an optional active-ﬁlter application at

the end.

Apparatus required

Breadboard, dual-trace oscilloscope with two attenuating probes, two 741

and one LF411 op amp, 311 comparator, 555 timer, one 100 , one 820 ,

two 1 k, two 3.3 k, three 10 k, one 100 k, one 1 M, and one 10 M

1

4

W

resistor, three 0.033 F, one 0.01 F, and one 1 F capacitor, one red LED,

and two 3.3 V Zener diodes.

9.1 Experiments

9.1.1 Op amp as comparator

Begin by wiring up a 741 in open-loop mode as you did in a previous lab

(Fig. 9.1(a)). With no negative feedback, the saturated output that results

allows the op amp to be used as a voltage comparator – a circuit that tells

you whether an input voltage is higher or lower than a ‘threshold’ voltage

(the threshold is ground in this case). Since op amps are not speciﬁcally

engineered for open-loop operation, it is not a very good voltage compara-

tor (in ways that we shall soon see), but in some situations (when high-

speed response and high sensitivity are not required) an open-loop-741

‘comparator’ is perfectly adequate.

**Start by applying a 1 kHz sine wave to the input and observe what the
**

circuit is doing.

113

114 Hands-on electronics

_

+

(b)

1

2

3

4

7

8

+15

–15

V

+

V

–

_

+

2

3

4

7

+15

–15

(a)

6

1 k

Fig. 9.1. (a) Poor comparator: 741 op amp used in open-loop mode; (b) 311 comparator.

Pinouts shown for eight-pin mini-DIP package.

**Now raise the input frquency to 100 kHz. Notice that the output ‘square
**

wave’ is not very square. Try an LF411 op amp in place of the 741.

**For each case studied, sketch the output waveform and measure the
**

output amplitude and DC offset. Explain your measurements.

**Contrast the performance of the LF411 and 741 when conﬁgured as com-
**

parators. What op amp limitation is responsible for the poor comparator

performance at high frequency?

Now substitute a 311 comparator for the op amp. (The pinouts are NOT

the same; see Fig. 9.1(b).) Whereas op amps are intended to be used with

negative feedback, the 311 is speciﬁcally designed for operation in open-

loop mode or with positive feedback.

The 311’s output stage differs from that of an op amp: to enhance ﬂexi-

bility, it has both a positive output (pin 7) and a negative output (pin 1). We

shall be using the positive output. In this conﬁguration, a ‘pull-up’ resistor

to a positive supply is required, in order to determine the output-voltage

level, and the negative output is generally connected to ground.

Note that the positive output is the collector of an NPNbipolar transistor.

As such, it cannot source current, but it can sink current. When the output

transistor is off (i.e., its base voltage is less than or equal to its emitter

voltage), the collector is pulled up to V

+

by the external pull-up resistor.

Conversely, when the base voltage is raised 0.7 Vabove the emitter voltage,

the transistor saturates and the output pin is pulled down close to the emitter

voltage. This output conﬁguration gives the 311 maximum ﬂexibility to

provide the various signal voltages used by digital logic chips, including

TTL, CMOS, and ECL logic levels.

1

**Start by choosing convenient values for V
**

+

and V

−

. Use an input fre-

quency of 100 kHz and observe the output.

1

Logic levels are discussed in section 10.1.1.

115 9 Comparators and oscillators

Change V

+

and V

−

. Record and explain any changes in the output

waveform.

**How does the output signal from the 311 differ from that of an op amp?
**

What is the 311’s slew rate?

9.1.2 Unintentional feedback: oscillation

An unfortunate side-effect of the 311’s fast response is its tendency to

oscillate when presented with a sufﬁciently small voltage difference be-

tween its inputs. The unavoidable capacitive coupling (typically a few

picofarads, depending on how the circuit is wired) between the output

and the input causes some feedback, so, when the output switches state, a

small transient signal is picked up at the input. There is also some feedback

due to the changing input base current when one input transistor switches

on and the other switches off. These effects can result in self-sustaining

oscillation.

The pickup occurs through what is essentially a voltage divider, con-

sisting of the large coupling impedance (i.e., small capacitance) between

the output and the input, together with whatever impedance to ground is

present at the input. The effect is thus mitigated if there is a low impedance

to ground at the input, and exacerbated if the impedance is high.

Try to make your 311 oscillate by feeding it a triangle wave with a gentle

slope dV/dt (i.e., small amplitude and/or lowfrequency). Use V

+

= +15 V

and V

−

=ground. To exacerbate the feedback, insert 10 k in series with the

function-generator source resistance as shown in Fig. 9.2. When looking

at the input signal, it is a good idea here to use the ground clips on your

_

+

1

2

3

4

7

8

+15

–15

1k

+15

10 k

C

High

Low

Oscillations

~

Fig. 9.2. 311 comparator with 10 k series input resistor. The capacitor shown connecting

the input and output represents the stray capacitance associated with the breadboard. (Do

not add a discrete capacitor!)

116 Hands-on electronics

scope probes, to suppress pickup of the large transient pulse due to the

15 V swing of the 311’s output. You’ll want to look at a low-to-high or

high-to-low transition and ‘zoom in’ by increasing the time sensitivity to

look for the rapid multiple transitions that indicate oscillation.

**Starting with an input frequency near 100 kHz and an amplitude of
**

several volts, gradually reduce the input amplitude and frequency until

oscillations are observed. To be able to see the rapid transitions of the

output, be sure the time scale on your oscilloscope is about 500 ns per

division or less.

**Sketch the observed oscillations. What is the time scale of the oscill-
**

ations – i.e., what is the period t for one cycle of the oscillation

waveform?

How small an input slope does it take to cause oscillation?

What causes the oscillations to stop?

**Why is a 741 conﬁgured as a comparator less likely to oscillate than
**

the 311?

9.1.3 Intentional positive feedback: Schmitt trigger

The oscillation problem can be eliminated by adding controlled positive

feedback (hysteresis) as shown in Fig. 9.3. The hysteresis also makes the

circuit less sensitive to noise.

The phenomenon of hysteresis should be familiar from your study of

magnetic materials in introductory physics. In the present context, the term

_

+

2

3

4

8

+15

—15

1k

+15

1k

R

2

R

1

10k

1

7

Fig. 9.3. Schmitt trigger using 311 comparator.

117 9 Comparators and oscillators

hysteresis refers both to the situation of having two different comparator

thresholds, depending on the history of the input signal, as well as to the

size of the voltage difference between the two thresholds. The threshold

voltage at the noninvertingterminal is foundbyapplyingthe voltage-divider

equation

V

in+

= V

out

R

2

R

1

+ R

2

. (9.1)

There are two cases to consider, since the output might be at +15 V or it

might be at ground – i.e., since V

out

has two possible states, V

in+

has two

possible states.

**Build the circuit and apply a signal with an amplitude of at least 1.5 volts.
**

Try to create output oscillations. Notice the rapid and clean transitions

at the output, independent of the input waveform or frequency.

**Sketch the output waveform. Is the output symmetric? If not, why not?
**

To see exactly what is happening, use the two-channel oscilloscope to

display the voltages at both comparator input terminals simultaneously. Be

sure to use DCcoupling for both channels, and set themto the same voltage

sensitivity and the same zero offset.

**Carefully sketch both input waveforms on the same graph and explain
**

how the circuit works. How does the hysteresis prevent oscillations?

**Based on the component values used, what do you predict for the com-
**

parator thresholds? What is the predicted amount of hysteresis in volts?

How do these compare with your observations?

**Record what you see as you vary the input amplitude. Below some input
**

amplitude the output stops switching. At what amplitude does this occur,

and why?

9.1.4 RC relaxation oscillator

By adding an RC network to your Schmitt trigger, construct a square-

wave oscillator, as shown in Fig. 9.4. Since a fraction of the output is fed

back into both the inverting and noninverting inputs, this circuit has both

negative and positive feedback. By connecting pin 1 to −15 V as shown,

you should get a symmetrical square-wave output with little DC offset. No

input from the function generator is required; the circuit should oscillate

spontaneously.

118 Hands-on electronics

_

+

2

3

4

8

+15

–15

1 k

+15

–15

1

7

0.01 F µ

R

2

R

1

3.3 k

10 k

10 k

C

V

out

R

Fig. 9.4. RC relaxation oscillator using comparator.

The expected frequency of oscillation is

f =

1

2RC ln

2R

1

R

2

+1

. (9.2)

**Sketch the output waveform and explain how this circuit works. (Hint: it
**

operates by repeatedly charging and discharging the capacitor between

the two threshold voltages of the Schmitt trigger.)

What are the threshold voltages?

Why does the circuit oscillate spontaneously?

**Derive Eq. 9.2 and compare with the observed frequency.
**

9.1.5 555 timer IC

Nowadays, one hardly ever builds a square-wave oscillator froman op amp

or comparator, since the 555 series of timer chips makes designing stable,

predictable oscillators easy. Note that a 555 is not an op amp or comparator,

but an oscillator ‘kit’, including two comparators, among other items (see

Fig. 9.5). The 555 can also be used as a timer, as we shall see below.

The 555 is an eight-pin IC powered from a positive voltage source. The

supply voltage can range from 4.5 to 16 V. As with a comparator, the

output is ‘digital’, with a high value near V

CC

and a low value near ground.

The output sinks current while low and sources current while high (up to

200 mAin either case). Also, as with a comparator, the 555 output slewrate

119 9 Comparators and oscillators

_

+

_

+

Output

Stage

Trigger

Threshold

Control

Voltage

Reset

Set

Output

Discharge

Reset

V

cc

Q

R

R

R

Flip-Flop

555 Timer

Fig. 9.5. Block diagram for the 555 timer IC.

is high, and the 555 can change states within 100 ns. Refer to a 555 data

sheet for additional details.

To see how the 555 works, ﬁrst examine the connections associated with

the two comparators. The top comparator output is high when the threshold

input is greater than

2

3

V

CC

, and low otherwise. The bottom comparator

output is high when the trigger input is less than

1

3

V

CC

, and low otherwise.

In general, trigger and threshold should be conﬁgured such that only

one of the comparators is high at any given moment. The outputs connect

to a ﬂip-ﬂop (described in detail in Chapter 11). A positive signal at set

causes Q to be near ground, while a positive signal at reset causes Q to

be high. A high value for Q turns on the transistor switch, which drives

the discharge pin toward ground. The output stage is an inverting buffer,

so set causes the output to go high, while reset causes the output to

go low.

Begin by connecting a 555 as shown in Fig. 9.6(a) and observe the

output. When the output is high, the capacitor is charging through R

A

and

R

B

. When the capacitor voltage V

C

exceeds

2

3

V

CC

, the discharge pin is

driven toward ground and the capacitor discharges across R

B

. The cycle

repeats once V

C

falls below

1

3

V

CC

. The frequency is predicted as

f =

1

0.7(R

A

+2R

B

)C

. (9.3)

120 Hands-on electronics

V

cc

Out

Discharge

Threshold

Trigger

Reset

10 k

10 k

0.01 F µ

R

B

C

+15

R

A

3

4 8

7

6

2

1

V

cc

Out

Discharge

Threshold

Trigger

Reset

10 MΩ

1 F µ C

+15

R

A

3

4 8

7

6

2

1

820

+15

(a) (b)

100 k

Red LED

Fig. 9.6. (a) 555 timer IC used as an oscillator; (b) 555 timer IC used as a one-shot or

timer.

**Sketch the output waveform and brieﬂy explain the operation of this
**

circuit. Is the output symmetric? If not, why not?

**Derive Eq. 9.3 and compare the measured output frequency with the
**

predicted oscillation frequency.

**Examine the voltage V
**

C

across the capacitor. Record its minimum and

maximum values. Do they make sense?

Try replacing R

B

with a short circuit – what happens? Explain why. Put

R

B

back for the next part.

Try changing V

+

to 5 V and observe how the output changes. To what

extent does the output frequency depend on supply voltage?

Now connect a 555 as shown in Fig. 9.6(b). The output should be a

‘one-shot’ pulse of duration

t = 1.1R

A

C. (9.4)

The output pulse is triggered by the push-button switch, which causes the

trigger input to go to ground. (Note: the output will remain high indef-

initely if the trigger input is held at ground, so one should ensure that

the trigger pulse is shorter than the desired output pulse!) Time the output

pulse by observing the LED.

**Brieﬂy explain the operation of this circuit. What prevents this circuit
**

from oscillating?

**Measure the output-pulse duration for several values of R
**

A

and C. Tab-

ulate your results.

**Derive Eq. 9.4. Are your data consistent with this expression? If not,
**

why not?

121 9 Comparators and oscillators

9.2 Additional experiments

9.2.1 Alarm!

You can conﬁgure the 555 to sound an alarmwhen prompted by an external

signal. The alarm is simply a 555 oscillator with the output connected to a

speaker. To prevent the alarm from sounding continuously, ground is ap-

plied to the reset input pin, which overrides the trigger and threshold

pins and forces the output near ground. When used in this way, the reset

line is said to ‘enable’ the 555, enabling oscillation when high while dis-

abling oscillation when low.

Build the circuit as shown in Fig. 9.7. Use a long wire, which simulates

a security loop. Cut or unplug the wire and hear the alarm!

The alarm can also be conﬁgured to sound for a speciﬁed duration using

two 555s by combining the timer from Fig. 9.6(b) with the alarm circuit.

Connect the alarm’s reset input to the timer’s output line. Trigger the timer

and the alarm will sound for the duration of the timer’s output signal.

V

cc

Out

Discharge

Threshold

Trigger

Reset

10 k

10 k

0.01 F µ

R

B

C

+15

R

A

3

4 8

7

6

2

1

100

1 k pot

(volume control)

Speaker

Fig. 9.7. 555 timer conﬁgured as an alarm.

122 Hands-on electronics

You can also make a pulsing alarm by using two oscillators. The ﬁrst

should oscillate with a period of a fewseconds. The output of this oscillator

is then connected to the reset line of the second oscillator. The second

oscillator’s output is connected to the speaker and can oscillate with any

audio frequency of your choice.

9.2.2 Sine/cosine oscillator

Perhaps surprisingly, the sine wave is one of the most difﬁcult waveforms

to produce. As you saw when you differentiated the sine output of the

PB-503’s built-in function generator, the PB-503 uses a piecewise-linear

approximation to a sine function.

The circuit of Fig. 9.8 should produce a better approximation. It should

oscillate at

f =

1

2π RC

, (9.5)

as long as R

1

< R. So you don’t expect it to work with R

1

=10 k, unless

Eq. 9.5 happens to be satisﬁed due to resistor manufacturing tolerances.

You can vary R

1

so as to satisfy Eq. 9.5 by using the 10 k pot for R

1

or by

adding other resistors in parallel with the 10 k.

Set up the circuit and apply power. Once you have it oscillating, look at

both outputs on the dual-trace oscilloscope.

Are the frequencies the same?

**What is the phase relationship?
**

_

+

0.033 F µ

R

R

1

10 k

10 k

C

_

+

10 k

0.033 F µ

C

R

0.033 F µ

C

3.3 V 3.3 V

sine

cosine

Fig. 9.8. Sine/cosine oscillator.

123 9 Comparators and oscillators

**Measure the peak-to-peak voltage of the cosine wave. Is it what you
**

would expect, considering the diodes being employed in the circuit?

Try to explain how the circuit works. What role is played by the two

back-to-back Zener diodes?

9.2.3 Active bandpass filter

Active ﬁlters have signiﬁcant advantages over passive ones, including

r

lower cost due to replacement of expensive inductors by capacitors phase-

shifted via feedback;

r

high input impedance and low output impedance;

r

ease of tuning over a wide frequency range.

They also have signiﬁcant disadvantages such as frequency response lim-

ited by the bandwidth of the op amp and the need to provide power to the

op amp. There are a variety of conﬁgurations that can be used for low-pass,

high-pass, and bandpass applications; for a more extensive discussion, see

e.g. Horowitz and Hill, The Art of Electronics, or Simpson, Introductory

Electronics for Scientists and Engineers.

Wire up the circuit of Fig. 9.9, apply power, and drive it with a sine wave

of about 1 V peak-to-peak. Vary the driving frequency until you ﬁnd the

maximum output amplitude. What are the center frequency f

0

and upper

and lower −3 dBpoints f

H

and f

L

of the frequency response? If you adjust

the output voltage to be 14.1 V peak-to-peak at f = f

0

, you can easily

ﬁnd the −3 dB points by varying the frequency until the output voltage

is 10.0 V peak-to-peak. Measure and graph the gain vs. frequency for a

reasonable frequency range.

_

+

0.033 F µ

R

2

R

1

3.3 k

100 k

C

op amp

R

1

0.033 F µ

3.3 k

C

V

in

V

out

Fig. 9.9. Active bandpass ﬁlter.

124 Hands-on electronics

Compare f

0

with the theoretical value

f

0

=

1

2πC

2

R

1

R

2

, (9.6)

and compare the voltage gain at the center frequency with the theoretical

value

A

0

= −

1

2

R

2

R

1

. (9.7)

The ‘quality factor’ for a bandpass ﬁlter is

Q =

f

0

f

H

− f

L

. (9.8)

Thus, a narrow passband corresponds to high Q and a broad passband to

low Q. Compare your observed Q to the theoretical value for this ﬁlter,

Q =

R

2

2R

1

. (9.9)

Try to explain how the circuit works.

10

Combinational logic

In this chapter you will be introduced to digital logic. You will build some

logic circuits out of discrete components and some out of integrated cir-

cuits, and familiarize yourself with the 7400 series of CMOS (comple-

mentary metal-oxide-semiconductor) andTTL(transistor–transistor logic)

integrated circuits and their basic operation.

Note

The kinds of things one thinks about in digital logic are almost completely

different from those in analog electronics.

Apparatus required

Breadboard, oscilloscope, multimeter, 100 , 330 , 1 k, 2.2 k, and 3.3 k

1

4

W resistors, two VP0610L and two VN0610L MOSFET transistors,

three 2N3904 transistors, three diodes, one LED, one red LED (optional),

74HC00, 7432, 7485, 7486 TTL or TTL-compatible logic chips, logic

switches, and logic displays.

10.1 Digital logic basics

In this section we introduce the 7400 series of CMOS and TTL digital-

logic chips. Unlike the analog ICs you’ve used up to now, which can output

any voltage within some range determined by the power-supply voltages,

digital-logic ICs employ only two ranges of output voltages, referred to as

logic levels, about which more below. These levels can be used to represent

true or false logical conditions or the zero and one of binary arithmetic.

The 7400 series is not the only logic series, nor are CMOS and TTL

the only types of logic circuitry; however, they are the most commonly

used. Other logic families include the CMOS 4000 series and the ECL

125

126 Hands-on electronics

(emitter-coupled logic) 10 000 and 100 000 series. Each logic family has

its own logic levels, speed, and recommended supply voltages.

The integrated circuits you will be using noware much more specialized

than the general-purpose 741 op amp and 555 timer. They feature much

higher bandwidth, with typical transition speeds of order volts/nanosecond

(in contrast to the volts/microsecond slew rate of the 741). While greater

complexity often means higher cost, the basic chips in the 7400 families

(such as the 74HC00, 74LS00, and 74ACT00) cost less than $0.50 each

in small quantities, with the more complex chips ranging toward several

dollars.

10.1.1 Logic levels

Digital chips employ two voltages to represent two possible states. These

voltages are called logic levels and can be used to represent the two states

of Boolean algebra

1

as well as the two digits of binary arithmetic. There

are three ways of referring to logic levels:

r

true and false,

r

zero and one, and

r

high and low.

In TTL logic, a voltage exceeding +2 V is called high, while a voltage

less than +0.8 V is called low. To ensure noise margin, TTL outputs are

guaranteed to put out at least +2.5 V in the high state and at most +0.4 V

in the low state (see Fig. 10.1). This means that, even in the presence of up

to 400 mV of noise, an output low will be recognized as low by the input of

the next logic circuit, and similarly for high. The comparable CMOS levels

are +3.5 and +1.5 V for the inputs and +4.5 and +0.5 V for the outputs.

While TTL chips are always powered froma +5 Vsupply, many CMOS

chips are tolerant of supply voltages ranging from+2 to +6 V. It is therefore

convenient to reference CMOS logic levels to the power-supply voltage

V

CC

. The minimum input voltage interpreted as a CMOS high (V

IH

) equals

0.7 × V

CC

, while the maximum input voltage interpreted as a CMOS low

(V

IL

) equals 0.3 × V

CC

. The output voltages V

OH

and V

OL

will vary with

supply voltage as well.

The ‘high/low’ nomenclature is unambiguous, since it directly charac-

terizes voltages, but the other two nomenclatures rely on a convention,

1

Also known as symbolic logic.

127 10 Combinational logic

5

4

3

2

1

0

Volts

CMOS:

C, HC, AC, and

AHC Series

CMOS & TTL:

HCT, ACT, AHCT,

S, F, LS, AS, and

ALS Series

Low-Voltage CMOS:

LVC and ALVC

Series

V

IL

V

IH

V

CC

V

CC

V

CC

V

IH

V

IH

V

IL

V

IL

Fig. 10.1. Logic levels for various 7400-family lines. V

CC

is the most positive voltage; V

IL

and V

IH

are the maximum input low and minimum input high voltages.

which can be assigned either of two ways. The more common convention

is positive logic: high=1=true, low=0=false. But there are occasionally

situations in which it is more convenient to employ negative logic, in which

high=0=false and low=1=true. (Another way to think about logic circuits

is interms of assertion-level logic, whichis a hybridof positive andnegative

logic that we will introduce in the next chapter.)

10.1.2 Logic families and history

As indicated in Table 10.1, CMOS and TTL chips come in a plethora of

types, each with its own speed, power dissipation, input load, and output-

current characteristics. This reﬂects the historical development of the var-

ious series. Since our initial purpose is to become familiar with the logic,

the details of speed and power are (for now) unimportant, but you will

need a general familiarity with them so as not to be in the dark when you

encounter them in future. Also there are restrictions on fanout (the number

of inputs that an output can drive) which matter when actually designing

circuits. For example, one LS-TTLoutput can drive twenty LS-TTLinputs,

but only four S-TTL inputs. In what follows we use ‘7400’ generically to

128 Hands-on electronics

Table 10.1. Common families within the 7400 series.

Family Year Brief description

TTL 1968 bipolar transistor–transistor logic

S 1974 TTL with Schottky transistors

LS 1976 low-power Schottky TTL

ALS 1979 advanced low-power Schottky

F 1983 fast TTL

HC 1975 high-speed CMOS

HCT 1975 high-speed CMOS (TTL compatible)

AC 1985 advanced CMOS

ACT 1985 advanced CMOS (TTL compatible)

LVC 1993 low-voltage CMOS

AHC 1996 advanced high-speed CMOS

refer to chips from any of these families. Unless otherwise speciﬁed, the

chip families you actually use will depend on what happens to be on hand

in the laboratory or (if you are working through this book on your own) on

what you happen to ﬁnd available.

Part numbers

The original TTL chips were the 7400 series and the corresponding ‘Mil

Spec’ (military speciﬁcation) 5400 series; these became popular in the

1970s. TTL chips are labeled with part numbers that begin with a letter

code (such as ‘SN’) that is typically different for each manufacturer (see

Fig. 10.2); then comes the ‘74’ that identiﬁes the device as belonging to the

7400 series; then there may be a letter code that identiﬁes the family; then

the number that identiﬁes the particular device (e.g. 00 for a quad NAND

gate, 01 for quad NAND with open-collector outputs, 02 for quad NOR

gate, 74 for dual D-type ﬂip-ﬂop, etc.); and ﬁnally there may be letters that

indicate package style, reliability, degree of testing by the manufacturer,

etc. (For example, the MC74LS00ND is a Motorola LS-TTL quad NAND

gate in the plastic dual-in-line package with 160 hour ‘burn-in’ testing.)

Pinouts and data sheets

Data sheets are available on the web or in data books produced by the chip

manufacturer. The sheets specify the function of each pin of the ICpackage

and provide detailed data on chip performance. In general, 7400-series

chips have compatible pinouts independent of the family. For example, the

129 10 Combinational logic

Pin #1

Pin #14 Pin #8

Pin #7

Manufacturer

Series

Family Chip Type

SN74LS00N

1A

1B

1Y

2A

2B

2Y 3A

3B

3Y

4A

4B

4Y

GND

V

CC

Fig. 10.2. Labeling of 7400-series chips.

pinout for a 74HC00 quad NAND IC is the same as the pinout for the

74LS00 quad NAND IC. It is always a good idea to review the data sheet

before using any logic chip.

Another consistency is in the pin numbering scheme. If you orient the

chip so that the pins are bending away from you and the end that has a

notch or a dot is pointing to the left, pin 1 is the one at the lower left of

the package. The numbering then proceeds sequentially around the chip in

a counterclockwise direction, such that the highest-numbered pin is at the

upper left. Almost always, when you orient the chip this way, the writing

on the top will be right-side-up (see Fig. 10.2).

10.1.3 Logic gates

There are six basic logic gates, as shown in Fig. 10.3. These gates are

sufﬁcient to implement all logic functions, although the more complex

functions are also available as specialized chips – multiplexers, decoders,

etc. (to be discussed later) – which can simplify design as well as reduce

cost. Although Fig. 10.3 shows only two-input gates, versions also exist

with three, four, or even eight inputs.

Note that NANDand NORare opposites to ANDand OR: NANDequals

not ANDwhile NORequals not OR. Also note that the NANDis a negative-

logic OR while the NOR is a negative-logic AND. (This follows from

DeMorgan’s theorem, about which more below.) The small circle shown

at the outputs of the NAND and NOR is shorthand for an inverter – a

NAND is equivalent to an AND followed by an inverter, and similarly

for NOR.

130 Hands-on electronics

AND OR

INVERTER NAND NOR

XOR

A B Out

H H

L

L

L L

H

H

H

L

L

L

AND

A B Out

H H

L

L

L L

H

H

H

H

H

L

OR

A B Out

H L

L

L

L L

H

H

H

H

H

H

NAND

A B Out

H L

L

L

L L

H

H

H

L

L

H

NOR

A B Out

H L

L

L

L L

H

H

H

H

H

L

XOR

A Out

L

H L

H

INVERTER

(NOT)

Fig. 10.3. Standard logic gates with truth tables.

After the inverter, the NAND is the simplest logic gate to construct from

discrete components, and historically it was the most commonly used gate.

NAND is a ‘universal’ logic function, in that the other logic functions can

all be created using NANDgates. For example, connecting together the two

inputs of a NAND creates a one-input gate – an inverter. Thus, if the output

of a NAND is connected to both inputs of a second NAND, the result is

equivalent to an AND gate.

10.1.4 Summary of Boolean algebra

Logic operations are best described using Boolean algebra. While a detailed

exposition of Boolean algebra is beyond the scope of this text, we give here

a brief introduction and mention a few useful points. Consider two logical

variables A and B, which can take on the values true and false. We can then

denote logic operations as follows:

r

the logical AND of A and B is denoted as A · B;

r

the logical OR of A and B is denoted as A + B;

r

the logical XOR of A and B is denoted as A ⊕ B;

r

the inverse of A is A, and the inverse of B is B;

r

the logical NAND of A and B is denoted as A · B;

r

the logical NOR of A and B is denoted as A + B.

As in any algebra, the rules of Boolean algebra allow theorems to be

derived starting from axioms. An alternative way to prove a theorem in

131 10 Combinational logic

A B A B

H L

L

L

L L

H

H

H

H

H

H

A B A+B

H L

L

L

L L

H

H

H

L

L

H

A B A+B

L L

H

H

H H

L

L

L

H

H

H

. .

A B A B

L L

H

H

H H

L

L

L

L

L

H

= =

Fig. 10.4. DeMorgan’s theorems expressed symbolically.

Boolean algebra is by ‘exhaustive demonstration’, i.e., to write down the

truth table – for every possible input state, work out the value of the output

and write it down.

Two relationships (known as DeMorgan’s theorems) are particularly

useful:

A · B = A + B. (10.1)

and

A + B = A · B (10.2)

These are illustrated schematically in terms of logic gates as well as with

truth tables in Fig. 10.4.

10.2 CMOS and TTL compared

10.2.1 Diode logic

We begin our consideration of the electronic implementation of logic gates

with the simplest example. Since diodes pass current in only one direction,

they can be used to perform logic, as shown in Fig. 10.5. Assuming logic

levels equal to either 0 or +5 V, it is easy to show that the output voltage is

near +5 V (actually one diode drop below +5 V) if both inputs are +5 V,

and zero otherwise. If we assume the positive-logic convention (high =

true, low=false), this is equivalent to a logic AND of the two inputs. (You

may want to verify this by building this circuit on a breadboard and trying

it out.)

132 Hands-on electronics

T

+5

3.3 k

A

B

Fig. 10.5. Two-input diode gate.

+5

3.3 k

A

B

1 k

100 2.2 k

T

Q

3

Q

2

Q

1

LED

330

+5

Fig. 10.6. Diode–transistor NAND gate using 2N3904s. This resembles the circuitry

actually used inside the 7400, except that in the 7400 a two-emitter transistor substitutes

for the input diodes. The resistor values are approximate and vary with TTL family. Also

shown is a simple LED logic-level indicator being driven by the NAND gate’s output.

10.2.2 Transistor–transistor logic (TTL)

Using the diode-logic concept as an input stage, TTL logic was developed

during the 1960s. A TTL-inspired NAND gate constructed using diodes

and bipolar transistors is shown schematically in Fig. 10.6. The operation

of the circuit can be analyzed as follows.

r

If either input is near ground, transistors Q

1

and Q

3

will be off. Q

2

will

consequently be on (saturated), which causes the output (T) to be about

two diode drops below V

CC

.

r

On the other hand, if both inputs become high, Q

1

’s base voltage in-

creases, causing Q

1

to turn on, which turns on Q

3

. Q

2

’s base voltage

133 10 Combinational logic

drops, turning Q

2

off. Since Q

3

is then in saturation, the output T is close

to ground.

Assuming the positive-logic convention, the output voltage thus represents

the logic NAND of the inputs.

The output structure is commonlyreferredto(fancifully) as a ‘totem-pole’

output, since it consists of a group of stacked components. Several im-

provements have been made since the original introduction of TTL logic,

resulting in numerous family lines. Many are listed in Table 10.1.

Quirks of TTL inputs and outputs

The gate circuit of Fig. 10.6 illustrates some peculiarities of TTL inputs

and outputs. In general, TTL outputs cannot be relied upon to source even

small amounts of current; however, they can sink tens of milliamperes of

current. Note how the LED logic indicator has been constructed to take

advantage of this – the LED is lit when the output is low! Note also that

TTL inputs source current while held low, but sink negligible amounts of

current while held high.

The input and output current speciﬁcations vary among the different

family lines. Refer to the data sheets for details.

Note that the LED logic indicators built into the PB-503 operate in

positive logic, whereas for TTL, home-built logic indicators (as inFig. 10.6)

operate in negative logic.

10.2.3 Complementary MOSFET logic (CMOS)

In the 1970s a new family of logic chips was developed using MOSFETs

instead of bipolar transistors. The basic MOSFET logic gate uses comple-

mentary N-channel and P-channel MOSFETs, and is thus called CMOS.

Fairchild ﬁrst introduced the 4000 series of CMOS logic, but the popularity

of TTL led manufacturers to develop TTL-like CMOS families, such as

the 74C, 74HC, 74HCT, 74AC, and 74ACT. The families with a ‘T’ in their

part numbers are fully compatible with TTL logic levels, while the ones

without a ‘T’ are logic- and pinout-compatible with TTL but operate with

CMOS logic levels. Since the 1990s, CMOS logic ICs have surpassed TTL

logic in popularity; however, TTL logic is still manufactured and is quite

common.

MOSFET operation resembles that of the JFETs studied previously. As

shown schematically in Fig. 10.7, in an N-channel MOSFET, a positive

134 Hands-on electronics

Drain Source

Gate

N-type N-type

N-type

P-type

Substrate

V

GS

Metal

Oxide

Semiconductor

+

Fig. 10.7. Schematic representation of an ‘enhancement-mode’ N-channel MOSFET.

gate–source voltage attracts current carriers into the channel between the

drain and source, allowing current to ﬂow. A gate–source voltage near

zero (or negative) closes the channel and prevents current ﬂow between

drain and source. When the channel is open, the drain–source resistance

is quite small (≈10–100 ), while the drain–source resistance is large (of

order megohms) when the channel is closed. The MOSFET thus acts like

a switch.

Notice that to turn on an N-channel MOSFET, the gate voltage is brought

positive with respect to the source. For a JFET, this would result in a

large current ﬂowing through the gate to the source since the gate–source

junction would act as a forward-biased diode. This does not occur in

MOSFETs since the gate and channel are separated by a layer of insu-

lating oxide. The oxide layer allows the electric ﬁeld to penetrate without

allowing current to pass. The input impedance for MOSFETs is conse-

quently even greater than for JFETs! Also notice that, unlike JFET opera-

tion, the drain–source channel is normally closed. A positive gate voltage

of a few to several volts is required to open the channel and allow current

to ﬂow.

For P-channel MOSFETs, the voltage relationships are reversed: the

channel is open when the gate–source voltage is negative and closed when

the gate–source voltage is near zero (or positive).

Combining an N-channel and P-channel MOSFET gives a comple-

mentary pair of switches that open and close opposite each other. Two

135 10 Combinational logic

V

cc

V

cc

MOSFET

Switch 1

MOSFET

Switch 2

V

out

V

out

V

in

D

S

G

G

S

D

P-channel

MOSFET

N-channel

MOSFET

V

cc

V

out

Fig. 10.8. Schematic representations of a CMOS inverter constructed using one

N-channel and one P-channel MOSFET.

MOSFETs can thus be combined in parallel to create an inverter, as shown

in Fig. 10.8. A NAND gate can be constructed by adding two more

MOSFETs (see Fig. 10.9). The inverter operates as follows.

r

When the input is high, the N-channel MOSFET is on and has a

low resistance between drain and source, while the P-channel MOSFET

is off and has a high resistance between drain and source. The

output is thus connected through a low-resistance path to ground and

goes low.

r

If the input is near ground, the P-channel MOSFET is on while the N-

channel MOSFET is off. The output now has a low-resistance path

to V

CC

and is pulled high.

A drawback is that if the input is at an intermediate voltage, then both

channels are partially open, which results in current ﬂowing through both

FETs from power to ground. While CMOS logic draws very little power

when in a steady state (i.e., all signals either high or low), it thus draws

much more power when signals are switching between high and low.

Another drawback of MOSFETs is their sensitivity to static-electricity

discharge. The oxide insulating layer between the gate and the channel

is usually quite thin and easily damaged. Static charge accumulating on

a human in cold, dry conditions can easily develop a potential of sev-

eral thousand volts. Although the total energy released is small, if this is

discharged in a spark to a MOSFET, it is sometimes sufﬁcient to blast a hole

136 Hands-on electronics

D

S

G

G

S

D

P-channel

MOSFET

N-channel

MOSFET

V

cc

C

B

A

A B C

H L

L

L

L L

H

H

H

H

H

H

LED

330

D

G

S

N-channel

MOSFET

S

G

D

P-channel

MOSFET

V

cc

Fig. 10.9. Schematic representation of a CMOS NAND gate with LED logic-level

indicator.

through the oxide layer, destroying the MOSFET. You’ve probably heard

of static-sensitive electronics such as computer memory cards. These cards

are constructed using MOSFETs.

To minimize the accidental destruction of components, we’ve chosen

medium-power MOSFETS for this lab – these are much more robust (and

less sensitive to static) than the tiny, high-speed MOSFETS used inside

high-density chips.

10.2.4 Powering TTL and TTL-compatible integrated circuits

Unlike the 741 and 311 chips that you’ve used previously, the logic chips

to be used here will require only a single +5 V supply rather than separate

positive and negative supplies. As in the case of the 741, the actual power

connections are often omitted in schematic diagrams, but you must not

forget to make those connections. Note that not all chips have the same

number of pins; furthermore, even among chips with the same number of

pins, not all follow the same convention in power-supply connections. You

must look up the pinout of each chip you employ!

137 10 Combinational logic

Be careful! Note that TTL and CMOS ICs are guaranteed to be destroyed if by mistake

you power them backwards! They may also be destroyed if you apply a voltage higher

than 7 V to any pin (5.5 V for the original 7400 TTL family).

10.3 Experiments

10.3.1 LED logic indicators and level switches

The PB-503 breadboard has eight built-in LED indicators. These are in-

tended for use as logic-level displays. A high level applied to the input

will light the LED – try connecting one to ground or +5 V and see that

it works. (If you ever ﬁnd a need for more than eight indicators, you

can augment the built-in indicators with individual LEDs in series with

several-hundred-ohm current-limiting resistors; this is also a good solution

if your breadboard lacks built-in logic indicators.)

**Drive an LED logic indicator with a variable voltage between 0 and 5 V.
**

Explore the threshold voltage and compare with the CMOS and TTL

logic levels.

Two single-pole double-throw(SPDT) switches are available at the lower

right-hand side of the PB-503. The PB-503 also provides a bank of eight

‘level’ switches in a unit located near the lower-left corner. These can be

used as logic-level switches, as shown in Fig. 10.10. The switches make a

connection either to power or to ground. The power voltage depends on the

setting of a ‘master’ switch located to the right of the eight-switch unit, and

allows for ﬂexibility when working with low-voltage CMOS ICs or other

nonstandard ICs. Older versions of the PB-503 have a bank of SPST DIP

1--100 k

SPST

Switch

+5

TTL or CMOS

LOGIC LEVEL

+5

TTL or CMOS

LOGIC LEVEL

SPDT

Switch

Fig. 10.10. Logic-level switch using either an SPST or SPDT switch and a pull-up resistor

as shown.

138 Hands-on electronics

switches instead of logic switches. (If your breadboard doesn’t have logic

switches, refer to Fig. 10.10 for details on howto use the SPST switch bank

as logic-level switches.)

**Figure out how to use the logic-level switches to turn an LED indicator
**

on and off. Explain how the indicator works.

CMOS inputs require well deﬁned input voltages; therefore, a CMOS

input must always be connected either to power or ground or to a CMOS

(or in some cases TTL) output. Unpredictable behavior (as well as exces-

sive current ﬂow from the power supply to ground) will result otherwise.

TTL is more forgiving, and SPST switches can easily be used as TTL

level switches, even without a pull-up resistor. As we will explore in more

detail later, an unconnected TTL input usually behaves as if connected

to a logic high: you assert a TTL low logic level at an input by con-

necting it to ground, whereas an open input acts like a high logic level.

With one end of the SPST switch connected to ground, you can thus

set the switch ‘on’ to assert a TTL low input level, or ‘off’ to assert a

TTL high.

10.3.2 MOSFETs

To demonstrate MOSFET behavior, construct the circuits shown in

Fig. 10.11.

**For both N- and P-channel MOSFETs, vary the gate voltage between 0
**

and 5 Vand measure the channel resistance as a function of gate voltage.

Note: if you try to measure the resistance directly with the meter, you

D

S

G

G

S

D

P-channel

MOSFET

N-channel

MOSFET

V

cc

V

cc

(5 V)

1 kΩ

V

out

V

in

V

out

1 kΩ

V

in

(5 V)

P

V

D

0

6

1

0

L

S

G

N

V

D

0

6

1

0

L

S

G

Fig. 10.11. Circuits for measuring the channel resistance as a function of gate voltage.

139 10 Combinational logic

will fail, since the quiescent current ﬂowing in the channel will confuse

the ohmmeter! Measure the output voltage and infer the resistance using

the voltage-divider equation.

**On a single graph, plot the channel resistance for both the N- and
**

P-channel MOSFETs vs. gate voltage. Comment on your results.

As discussed above, the complementary nature of P- and N-channel

MOSFETs which you’ve just demonstrated can be used to create logic

gates.

**Build the gate shown in Fig. 10.8 and determine its function.
**

You can easily display the output using an LED indicator. If you use

discrete components, don’t forget the current-limiting resistor (about

330 ) – connect the LED and resistor in series between the CMOS output

and ground. In either case, if the LED is on the output is high, and if the

LED is off the output is low.

**Apply CMOS logic levels to the input and note the output. Make a truth
**

table and verify that the gate inverts the logic level of the input.

**The TTL output of the function generator puts out a square waveform
**

whose low voltage level is near zero volts and whose high is near +5 V.

Use it to apply a signal to the input, and measure the output slew rate

(i.e., the transition speed from high to low and from low to high). How

does the output transition speed compare with the transition speed of the

input square wave?

**Measure the input and output impedances and compare your results with
**

your expectations. The output impedance should be measured for both

high and low output logic states, using a pull-up resistor to +5 V or a

pull-down resistor to ground as appropriate (try a 330 resistor). (If

you used a discrete-component LED indicator, you should disconnect it

to avoid confusion, or else ﬁgure out how to take it into account.)

The inverter can be converted to a NAND gate with the addition of two

more MOSFETs. Do so as shown in Fig. 10.9. Connect the output to a logic

indicator.

Using logic switches, verify the truth table.

**Tie one input to +5 V and the other to a TTL square wave. Measure the
**

transition speed of the output.

**Measure the output impedance with both inputs high as well as with
**

both inputs low. Compare these results with the values measured for the

inverter. Does the output impedance depend on the output logic level?

How does the output voltage level depend on the output load?

140 Hands-on electronics

10.3.3 CMOS NAND gate

Fromyour collection of chips, select a 74HC00 quad NANDgate. Referring

to the pinout diagram (Fig. 10.2.), you will note that the ground connection

is at pin 7 and +5 Vat pin 14, a typical although not universal conﬁguration.

Again, disaster will strike if you hook up the chip backwards and apply

power!

Carefully insert the 74HC00 so that it straddles the central groove of a

breadboard section. (When inserting a new chip into the breadboard, you

need to pay attention to all of the pins and make sure none of them is bent,

since a bent pin will fail to make contact.)

Check that the power is off. Then run jumpers to pins 7 and 14 of the

chip from the ground and +5 busses, respectively. For now we are going

to use just one of the four gates on this chip, the one whose output is at

pin 3 and whose inputs are at pins 1 and 2. Connect pin 3 to an LED logic

indicator. Connect pins 1 and 2 to level switches. Ground all other inputs.

After double-checking that the connections are correct, turn on the power.

Note: the 74HC00 can source only 5 mA of current; therefore, if you use

discrete components for your LED logic indicator, you will need to buffer

the output through a transistor. This is most simply done using an N-channel

MOSFET: connect the source to ground, the gate to pin 3 of the 74HC00,

and the LED and current limiting resistor in series from the drain to the +5

V supply.

**Connect the inputs to logic switches and try all four input combinations.
**

Record the truth table (in terms of logic levels).

**Connect pin1to+5Vanddrive pin2fromthe TTLoutput of the function
**

generator. Measure the output rise andfall times andpropagationdelayof

the 74HC00. Compare with the values listed on the 74HC00 data sheet.

For measurements of propagation delay, it is useful to deﬁne the transition

times of the input and output signals as the times at which they cross 2.5

V, which is about half-way between high and low. A good technique for

precise timing of logic signals is to set both scope channels to 1 or 2 V/

division and use the scope’s vertical-position knobs to put ground 2.5 V

below the center of the graticule; then you can easily measure the time at

which each signal crosses 2.5 V.

10.3.4 Using NANDs to implement other logic functions

As mentioned above, NANDs are a ‘universal’ logic function in the sense

that any Boolean-logic function can be constructed from them (the same

141 10 Combinational logic

is true of NOR and XOR). In the early days of logic chips, only NAND

was available, since AND and OR functions require more transistors to

implement.

**Construct the following two-input logic circuits using only NANDgates:
**

AND, OR, NOR.

To ﬁgure out the necessary logic, you can make use of DeMorgan’s the-

orems (Eqs. 10.1 and 10.2). Recall that you can use a NAND gate as an

inverter if needed – just connect the two inputs together, or tie one input

high.

**Write down your schematic diagramfor each circuit, indicating pin num-
**

bers next to each input and output that you use. These should be logic

diagrams, i.e., gate symbols arranged in a logical order according to their

function (like the schematics in the following chapters), not chip-level

diagrams (such as in Fig. 10.2).

**Drive each circuit using two switches on your breadboard and verify
**

that it gives the desired output for each of the four possible input states.

Record the truth table for each logic circuit.

Sometimes it is useful to have a gate that outputs a true if one or the

other input is true, but not both. Such a gate is called an exclusive-OR

(XOR) gate. In Boolean algebra the exclusive-OR of A and B is denoted

by A ⊕ B. As we shall see below, an XOR can be used to test two numbers

for equality; it also can be used to produce the sum bit for a 1-bit binary

addition.

**Construct an XOR circuit out of NANDs. It should turn on an LED
**

indicator when either input is high but not when both are.

It is easy to see how to do this with ﬁve gates; it can also be done with

only four. Be sure to show, both using Boolean algebra and using truth

tables for the intermediate signals in your circuit, that indeed you have

implemented an exclusive-OR.

**Show how to use an XOR to test two signals for equality. Try it out
**

and show that it works, displaying the XOR output with an LED logic

indicator. Does the LED light for equality or for inequality? Why?

10.3.5 TTL quad XOR gate

Use TTLchips for this exercise, as ﬂoating inputs will not work with CMOS

gates.

With a 7486 quad two-input XOR gate, you can test two 4-bit binary

numbers for equality.

142 Hands-on electronics

**Using all four XOR gates on the 7486, plus whatever additional gates
**

you need, design a circuit whose output indicates whether the two halves

of the 8-bit level switch are set equal (LED on if they are equal, off if

they are different).

You will ﬁnd that this is quite cumbersome using NANDgates, since equal-

ity corresponds to all four 7486 outputs being low. But it becomes much

simpler if you use OR gates such as the 7432.

**Record your schematic including pin numbers, build your circuit, and try
**

it out. Allow one or more inputs to ﬂoat by leaving them unconnected.

Is your observation consistent with the general property of 7400-series

TTL chips that ﬂoating input lines default to a high state?

This is a handy feature when testing circuits on a breadboard, but it is good

design practice not to rely on it – to be certain of reliable operation and

be maximally insensitive to noise, if you are using a gate in a circuit, you

should connect all of its inputs either to high or low or to outputs of other

gates.

**Display the output with a logic indicator and verify that your circuit
**

works for a few representative cases. Record the input and output in

each case.

10.4 Additional exercises

10.4.1 7485 4-bit magnitude comparator

This single chip does the work of the circuit you have just constructed

and more. If you have time, look up this chip in a logic data book or

on the web, familiarize yourself with its operation, and test it on your

breadboard. Describe how the function of this chip differs from the circuit

you constructed in section 10.3.5.

11

Flip-flops: saving a logic state

Next we turn to ﬂip-ﬂops. Also known grandiosely as bistable multivibra-

tors, these devices can remember their past. Their behavior thus depends

not only on their present input but also on their internal state. Circuits

containing ﬂip-ﬂops are termed sequential logic circuits, since their state

depends on the sequence of inputs that is presented to them. A truth table

is not sufﬁcient to describe the operation of a sequential circuit – you need

a timing diagram.

One category of sequential-logic circuits is the ﬁnite-state machine,

which goes through a predetermined sequence of states, advancing to the

next state each time it receives a clock pulse. We will encounter some

examples of state machines in this chapter, including divide-by-two and

divide-by-four counters. A useful tool in understanding a state machine is

a state diagram, showing the sequence of states through which the circuit

passes.

Some of these circuits will probably be the most complicated you have

wired up so far. Prepare carefully in advance and you will ﬁnd you can

complete these exercises easily; if you are unprepared, it is likely to take

you three to four times as long!

Apparatus required

The ideas explored in the following exercises apply equally whether CMOS

or TTL gates are used – use whichever is most convenient. You will need a

quad NAND (7400), a dual D-type ﬂip-ﬂop (7474), and a dual JK ﬂip-ﬂop

(74112). In addition, you will need a 74373, a breadboard, two 1 k

1

4

W

resistors, and a two-channel oscilloscope with two attenuating probes. (If

possible, avoid unnecessary complications by not mixing CMOS and TTL

chips within a single circuit.)

143

144 Hands-on electronics

11.1 General comments

11.1.1 Schematics

For each of the following exercises be sure to write down a complete

schematic diagram of each circuit you build, including pin numbers (but

power and ground connections need not be shown). Often when one wires

up a circuit off the top of one’s head, it fails to work. Writing down the

schematic showing all pin numbers is a powerful debugging tool, since it

makes incorrect connections much more obvious.

11.1.2 Breadboard layout

To keep your breadboard organized, it is a good idea to use one vertical bus

for power and another for ground, so that power and ground connections

to each chip can be made with short jumpers. A neat layout is easier to

debug than a messy one! Remember that the top and bottom busses on the

PB-503 are not connected internally.

11.1.3 Synchronous logic

The circuits in this chapter introduce the concept of synchronous logic

design. In an asynchronous circuit, signals can change at any time, requiring

the designer to work out in detail the signal propagation delays through all

possible paths tomake sure that the circuit will workas intended. Incontrast,

in a synchronous circuit, output signals change only in response to a clock

signal. Usually there is one common clock for an entire circuit, and its

transitions (low-to-high or high-to-low) – commonly referred to as rising

or falling ‘clock edges’ – are used to synchronize all other state changes

in the circuit. This approach leads to circuits that are easier to design and

analyze than asynchronous circuits. The synchronous approach is therefore

standard in microprocessors (for example) – we have all learned to rate the

processing power of computer chips in terms of their clock speed.

11.1.4 Timing diagrams

Timing diagrams are essential when analyzing synchronous logic circuits.

As shown in Fig. 11.1, timing diagrams for a synchronous circuit usually

show the clock signal, one or more inputs, and one or more outputs. Time

is shown on the x axis and signal logic levels on the y axis. Signal levels

145 11 Flip-flops

CLK

IN

OUT

t

PD

t

H

t

THL

t

SU

t

H

: hold time

t

SU

: setup time

t

TLH

: transition time (low--high)

t

PD

: propagation delay

t

TLH

t

THL

: transition time (high--low)

t

PW

t

PW

: pulse width

Fig. 11.1. Timing diagram with timing deﬁnitions for a rising-edge-triggered ﬂip-ﬂop.

Note that the rising edge of the clock signal causes the output to change while the falling

edge of the clock signal has no effect on the ﬂip-ﬂop.

are shown as either high or low – we tend not to worry in timing dia-

grams about analog details such as the exact voltage to which high and low

correspond.

Fig. 11.1 deﬁnes several important terms. The time between a clock edge

and the resulting changing edge of the output is deﬁned as the propagation

delay (t

PD

). Note howthe time is measured fromthe midpoint between logic

lowand high. The setup time (t

SU

) is the minimumtime that an input signal

must be stable preceding a clock edge. The hold time (t

H

) speciﬁes the

minimum time that the input signal must be stable following a clock edge.

The transition time measures the time required for a signal to transition

from logic low to high (t

TLH

) or from logic high to low (t

THL

).

These timing parameters are speciﬁed in the data sheets. Manufacturers

guarantee that the ICwill performcorrectly if the user satisﬁes the speciﬁed

minimum values. The specs are usually worst-case values. For example,

the measured propagation delay is almost always less than the maximum

value speciﬁed by the manufacturer. Although manufacturers often specify

the typical propagation delay as well, it is not guaranteed – safe designers

heed the minima and maxima, not the ‘typicals’.

11.2 Flip-flop basics

11.2.1 Simple RS latch

The circuit shown in Fig. 11.2 is the simplest ﬂip-ﬂop: an RS latch made

of cross-coupled NANDs (an equally simple RS latch can be made from

146 Hands-on electronics

Q

R

S

Q

Q

R

S

Q

7400

7400

7400

7400

1

4

1

4

1

4

1

4 * but note that the state after LL input

condition is removed depends on which

input signal goes high first; if both go

high “simultaneously,” state is undefined

Fig. 11.2. Simple RS latch made of two-input NANDs with state table.

NORs). R and S refer to reset and set (reset is also known as clear).

set turns Q on while reset (clear) turns Q off. Q is the opposite of Q,

except when both S and R are asserted.

Note on assertion-level logic notation

The inputs in Fig. 11.2 are active low – a low input forces a particular

output condition, while a high input does nothing. We have therefore taken

advantage of DeMorgan’s theorem to write the circuit in terms of negative-

logic OR gates, rather than blindly using the NAND symbol just because

the 7400 is called a NAND gate by its manufacturer. Since the OR gate

with inverted inputs has the same truth table as the NAND gate, it is just

as good a symbol for

1

4

of a 7400 as the NAND symbol. And it is actually

better to use the OR symbol here, since it makes the circuit’s operation

easier to see at a glance.

The use of DeMorgan’s theorem in this way is called assertion-level

logic notation, and it is a kind of hybrid between positive logic and negative

logic – the idea is always to choose the gate symbol that best clariﬁes the

operation of the circuit.

Build and test your ﬂip-ﬂop and record its state table (what its outputs

do for each of the four possible input states).

Keep in mind that what the ﬂip-ﬂop does for a given input may depend on

past history, so try a few different sequences of input states to make sure

you understand what it is doing. In other words, to see the entire state table

you need to try each input state for each of the ﬂip-ﬂop’s two internal states.

**How would the ﬂip-ﬂop’s operation be different if it were made of
**

positive-logic NORs? (You don’t need to build this circuit to answer

the question.)

Leave the RS latch in place for use later in this chapter.

147 11 Flip-flops

Q

R

S

Q

7474

1

2

D

clk

1

3

2

4

6

5

V = pin 14

Gnd = pin 7

cc

Q

R

S

Q

7474

1

2

D

clk

13

11

12

10

8

9

X = Don’t Care

* but note that the state after LL input

condition is removed depends on which

input signal goes high first; if both go

high “simultaneously,” state is undefined

Fig. 11.3. 7474 D-type ﬂip-ﬂop with state table.

CLK

IN

OUT

minimum setup time

not satisfied

output state

is not specified

preceding the first

rising clock edge

output is stable

between rising

clock edges

unpredictable output

due to changing input

coincident with clock edge

Fig. 11.4. Sample timing diagram for a (positive-edge-triggered) 7474 D-type ﬂip-ﬂop.

11.2.2 D-type flip-flop

In practice, RS latches are seldom used. The most commonly used ﬂip-

ﬂop is the clocked D-type, which remembers the state of its D input at

the time of a clock transition, but is insensitive to D at all other times.

We will use the 7474 D-type ﬂip-ﬂop (Fig. 11.3), which is sensitive to

rising clock edges (low-to-high transitions); in other words, it is positive-

edge-triggered. (Later in this chapter you will encounter a negative-edge-

triggered JK ﬂip-ﬂop.)

Note that, in addition to its ‘fancy’ D and clock inputs, the ’74 retains

the simple reset and set inputs of the RS latch (which are active lowsince

internally the ’74 is constructed from NANDs).

Test the reset and set inputs and explain what they do; then tie them

to +5 V to make sure they are inactive.

Next check out the clocked operation of the ’74 (illustrated in Fig. 11.4).

Provide the clocksignal usinga momentary-contact breadboard‘debounced

push-button’ and the D input using a breadboard logic switch, and display

148 Hands-on electronics

the Q and Q outputs using logic indicators. When using the debounced

push-buttons on the PB-503, be sure to add a pull-up resistor (as you did

for the logic-level switches; see Fig. 10.10).

**Showthat information presented on the Dinput is ignored except during
**

clock transitions by changing the state of D while the clock is low or

high. What happens?

**Now check whether the set and reset inputs take precedence over the
**

clock and D inputs: for example, try asserting reset (i.e. apply a low

level to it) and see whether you can clock in a high level applied at D.

**Disconnect the D input from the logic switch and connect the Q output
**

to the D input to make a toggling ﬂip-ﬂop. (Be sure to deassert set and

reset.) What happens now when you apply clocks?

**Try clocking the toggling ﬂip-ﬂop using a digital square wave from the
**

function generator, and use the scope to look at the input and output si-

multaneously. This is sometimes called a divide-by-two circuit – explain

what this means.

**Measure the ﬂip-ﬂop’s propagation delay, i.e., the time from the clock
**

transition to the change of output voltage – what is it? Is it about what

you would expect for chips of this family? How does it compare with

the manufacturer’s speciﬁcations? Be sure to trigger the scope on Q, not

the clock signal.

As mentioned above, a good technique for precise timing of digital signals

is to set both scope channels to 1 or 2 V/division and use the scope’s vertical-

position knobs to overlay both grounds below the center of the graticule;

then, you can easily measure the time at which each signal crosses the

midpoint between logic low and logic high (1.5 V for TTL and 2.5 V for

standard CMOS).

11.3 JK flip-flop

The JK ﬂip-ﬂop (Fig. 11.5) is slightly more complicated than the D ﬂip-

ﬂop; it can do everything a D can do plus more. The following exercises

use a negative-edge-triggered JK ﬂip-ﬂop with set and reset. Various

chips are available (e.g., the 74HC112 or the 74LS76) – which you use will

depend on what you have to hand; however, for the following exercises we

recommend using the 74112 JK ﬂip-ﬂop. Be sure to review the data sheet

to ensure that you have the correct pinout.

149 11 Flip-flops

Q

R

S

Q

74112

1

2

J

clk

15

1

3

4

6

5

V = pin 16

Gnd = pin 8

cc

K

2

Q

R

S

Q

74112

1

2

J

clk

14

13

11

10

7

9

K

12

Fig. 11.5. Pinout of the 74112 JK ﬂip-ﬂop.

Tie set and reset high and explore the JK’s clocked operation. (Note

that the 74112 senses its J and K inputs only at downward transitions of

the clock; hence, it is referred to as negative-edge-triggered.)

**Driving the clock from a debounced push-button and J and K from logic
**

switches, check the four possible input states, and write down the JK

state table. As in the case of the RS latch above, you need to try each

input state for both possible internal states.

**Nowadd an inverter (made froma NANDif you like) fromthe J to the K
**

input to make a D ﬂip-ﬂop. Drive J from a level switch and write down

the state table to verify that the circuit acts like a D ﬂip-ﬂop.

**Next remove the inverter and connect J and K together. Try it out and
**

write down the state table. What does this circuit do? How is it different

from the toggling D ﬂip-ﬂop?

11.4 Tri-state outputs

The ICs that we have used so far have two, and only two, valid output

states: high and low. Nowadays, it is common for some logic ICs to have

three output states.

There are situations in which the designer wishes the output to be neither

high nor low! Rather, in such situations, one wants to turn off the output

completely, i.e., the output should become a high impedance. The third

state is thus referred to as the ‘high-Z’ state. This feature allows multiple

outputs to be connected in parallel, as long as all but one of them are in

the high-Z state. A common example of the use of tri-state outputs is the

150 Hands-on electronics

connection of multiple memory chips (say 4 Mbytes each) in parallel to

build up a memory system (say 64 Mbytes).

We can explore the third output state using the 74373 octal D-type

transparent-latch IC. The ’373 is often used to drive a data bus, and is

equipped with three-state outputs to allow the bus to be driven in parallel

by multiple data sources (‘drivers’). The eight output pins are driven by

the IC when the output enable OE is held low, and they go into the high-Z

state when OE is high.

The ’373 is equipped with a latch enable LE instead of a clock. Data are

transferred from the input to the output while LE is high. The outputs are

latched (held constant) while LE is low. This type of ﬂip-ﬂop is called a

transparent latch.

**Begin by wiring the ’373 as shown in Fig. 11.6. (If logic-level switches
**

are available, use them for the LE, OE, and D inputs in place of the

resistor-SPST switch combinations shown.) Be sure LE is high and OE

is low. Choose any one of the D, Q pairs and apply a valid logic level to

the D input. Verify that the Q output follows the D input for both input

states. Measure the propagation delay between the input and output.

**Set D high and observe the output voltage (voltage at the output pin) as
**

you vary the potentiometer setting (there is no need to be excessively

quantitative here). Repeat your observation with Dlow. Explain why the

potentiometer has little effect on the output voltage.

**Set LElow. Does the output logic level varyas the input changes? Explain
**

your observations.

OE 1Q 1D 2D 2Q 3Q

6Q 7Q 6D 8D 8Q 7D

3D

V

CC

4D 4Q GND

5Q 5D LE

74373

+5 V

D Q

+5 V

+5 V

10 k pot

1 kΩ

1 kΩ

(a) (b)

Fig. 11.6. (a) Pinout and power connections for the 74373. (b) Input and output

connections for testing the tri-state output. Unlabeled resistors merely need to be large

enough to prevent excessive current ﬂow from +5 V to ground (e.g. 1 k or larger).

151 11 Flip-flops

**Set LE high and OE high. Observe the output voltage as you vary
**

the potentiometer. Explain your observations. Does the output behave

differently for input high and input low?

**Now set the input either high or low, record the input state, then set
**

LE low.

**Set OE low. The input state at the moment that LE was set low should
**

have been latched internally, independent of the state of OE. Does the

’373 remember the last input state correctly?

**In your own words and based on your observations, explain the operation
**

and features of the 74373.

11.5 Flip-flop applications

11.5.1 Divide-by-four from JK flip-flops

Ripple counter

Cascading two toggling ﬂip-ﬂops makes a divide-by-four circuit, otherwise

known as a two-bit counter.

**First make the two-bit asynchronous or ‘ripple’ counter shown in
**

Fig. 11.7. (This circuit is asynchronous in that it does not have a common

clock signal for all ﬂip-ﬂops.) Clock it froma push-button while looking

at the two Q outputs with logic indicators, and verify that it divides by

four, i.e., that the output square wave changes state at

1

4

the frequency

of the input clock. Write down its state diagram.

**Clock it with a square wave and look at the clock and the outputs on
**

the scope. To see the binary counting pattern, watch clock and Q

0

, then

Q

0

and Q

1

, always triggering on the slower waveform. Write down the

Q

0

R

S

Q

J

clk

K

Q

1

R

S

Q

J

K

+5

+5

Fig. 11.7. Divide-by-four ripple counter.

152 Hands-on electronics

Q

0

R

S

Q

J

clk

K

Q

1

R

S

Q

J

K

+5

+5

Fig. 11.8. Synchronous divide-by-four counter.

timing diagram and label the states with their numeric values (0 through

3), interpreting Q

0

as the low-order bit and Q

1

as the high-order bit of a

2-bit binary number.

**Turn the scope’s sweep rate up until you can see the ‘ripple’: Q
**

0

and Q

1

don’t change at the same time. Explain why not. (Be sure to look at both

edges of Q

1

.)

Synchronous counter

Now conﬁgure the circuit in its synchronous form (Fig. 11.8).

**Use the scope to conﬁrm that the ripple is gone, and make a timing
**

diagram showing this. You can see that, unlike D ﬂip-ﬂops, JK ﬂip-ﬂops

are natural for building synchronous counters – explain.

Keep your synchronous counter (and RS latch) for the next exercise.

11.5.2 Contact bounce

When a mechanical switch closes or opens, there is usually an effect called

‘contact bounce’. This means that the switch closes and opens repeatedly

for a period of order milliseconds until the contacts settle down. This makes

no difference if you are turning on a light, but if you are sending a clock

pulse to a counter made of high-speed devices, the counter can react to

each bounce and therefore count a random number of times.

First use the scope to observe contact bounce directly. Hook up an SPDT

(single-pole-double-throw) switch as in Fig. 11.9(a).

**Set the scope’s sweep rate in the vicinity of 0.1 to 1 ms/division, use
**

‘normal’ triggering, and look at the output of the switch as you open

and close the contact. Play with the trigger threshold and sweep rate

153 11 Flip-flops

+5

+5

÷ 4

+5

7400

7400

1

4

1

4

÷ 4

(a) (b)

Fig. 11.9. (a) Looking at contact bounce by driving a divide-by-four counter from a

switch. (b) A NAND latch is used as a debouncer.

to see if you can discern the bounces. Make a sketch of the observed

waveform.

**Next, clock your two-bit counter from the switch and see what happens.
**

Write down some typical sequences of states. Howis the contact bounce

affecting the sequence?

RS latch as debouncer

Now use the RS latch from the ﬁrst exercise as a switch debouncer

(Fig. 11.9(b)). (Although D and JK ﬂip-ﬂops are used for most ﬂip-ﬂop

applications, switch debouncing is one area in which RS latches continue

to hold their own.)

**Connect the switchtothe latchas shown, anduse the output fromthe latch
**

to clock the counter. Verify that the counting sequence is now correct.

Explain why this works. (The PB-503’s momentary-contact switches are

already debounced by RS latches built into the unit.)

11.5.3 Electronic coin toss

The RS-latch debouncer from the previous section can be combined with a

D-type ﬂip-ﬂop to create an electronic coin-toss game. To build the circuit,

connect the output of the RS-latch debouncer to the clock input of a D-type

ﬂip-ﬂop. Connect the digital output of the function generator to the Dinput

and connect the Q output to a logic indicator. Set the function-generator

154 Hands-on electronics

frequency to about 10 Hz. Operate the switch and observe the output. You

can assign the LED ‘on’ to be heads and ‘off’ to be tails.

**Write down the circuit diagram with pin numbers and explain how this
**

circuit works.

**Record a sequence of ten coin tosses. Is the sequence random? Repeat
**

another sequence of ten. How many tosses are required to determine if

the system is truly random? If you were an unscrupulous game designer,

how could you skew the ratio of heads to tails?

12

Monostables, counters, multiplexers,

and RAM

This chapter will introduce a variety of techniques that are important in

sequential-logic design. Such designs often make use of pulses of various

durations. Sometimes a logic pulse of a given width needs to be formed in

response to a particular input condition, e.g. to standardize a pulse from a

push-button. Monostable multivibrators are the usual solution. In addition

to monostables of a given logic family (such as the 74121, ’122, ’123, etc.),

there are also available the family of timer chips (such as the 555); the latter

are particularly useful when a long pulse of stable and reproducible width

is needed.

In this chapter you will also explore counters and their uses in timing

and addressing. As an example of the use of an address counter, you will

store and retrieve information in a small memory chip.

Be sure to write down the circuit’s schematic, with pin numbers, for ev-

ery circuit you build. You will ﬁnd the schematic especially useful should

your circuit not work. A simple review of the schematic will often re-

veal the source of the problem. Futhermore, a schematic is essential when

debugging subtle errors.

Apparatus required

Breadboard, oscilloscope, 7400 NAND, two 7490 and one 7493 counter,

74121 (or similar) one-shot, 74150 multiplexer, 7489, 74189, or 74219

RAM chip, two TIL311 displays, assorted resistors and capacitors.

Note

The circuits in this lab are rather involved, and many of the details of their

design are left for you to work out. You will need to work them out in

advance if you are to have any hope of completing the exercises in a timely

fashion!

155

156 Hands-on electronics

12.1 Multivibrators

Multivibrator circuits fall into three general categories.

r

Astable. These circuits have no stable state but keep changing from one

state to the other, hence the name multivibrator. They are very useful

as clocks or oscillators. (You used a 555 as an astable multivibrator in

chapter 9.)

r

Bistable. These circuits can be induced to go from one state to another,

and can remain in either state permanently after the input signals have

been removed. They are thus stable in both of their allowed states. The

more common term for bistable multivibrator is ﬂip-ﬂop.

r

Monostable. These circuits have only a single stable state. They can be

forced out of their stable state by a trigger pulse, but they return to it after a

very limited period of time. The primary use for monostables is to create

pulses of known duration from triggering pulses of shorter, longer, or

variable duration. Monostables are widely used to generate ‘gate’ signals

for counter circuits. They are considered to be hybrid analog/digital chips

in that the digital output is typically determined by the RC time constant

of an external analog circuit connected to the chip. Another term for

monostable multivibrator is one-shot, since it is a device that ‘shoots’

once (i.e., issues an output pulse) each time it receives an input signal.

Among monostables, the 555 timer is the best choice for pulse widths

ranging frommilliseconds to hours, and for applications in which the pulse

width must be stable to better than 0.05%. In digital designs, monostables

such as the 74121 are preferred for pulse widths ranging from about 40 ns

to 10 ms, and will operate up to tens of seconds, but their pulse widths are

not as predictable or stable.

12.2 Counters

In the last lab you wired up a divide-by-four circuit. That was of course

a 2-bit binary counter. Counters are so useful that IC manufacturers pro-

vide 4-bit (and more) counters as a single chip, with carry-in and carry-

out connections that allow them to be ‘cascaded’ in multiple stages for

8-bit, 12-bit, or greater range. Cascading means connecting multiple chips

together (as in the multiple digits of a car’s odometer) so that each chip

157 12 Monostables, counters, multiplexers, and RAM

counts when the preceding one ‘rolls over’ from its maximum count back

to zero. Counters are available in both binary and decimal versions and

in synchronous and asynchronous (‘ripple-through’) conﬁgurations, with

various arrangements of set, reset, and clock inputs.

Four-bit binary counters count from 0–15 and then ‘roll over’ to 0 again,

possibly issuing a carry-out signal to the next stage. (More speciﬁcally,

synchronous counters issue a carry-out, while for negative-edge-triggered

asynchronous counters, the high-order output bit from the preceding stage

serves to clock the next stage.)

Decimal counters work basically in the same way as binary counters,

except that they roll over (and possibly issue carry-out) at 9 rather than at

15; this makes them useful for driving decimal displays, which are easier

for humans to interpret than binary. Note that decimal counters are also

referred to as decade or BCD (‘binary-coded decimal’) counters. Some

are actually bi-quinary counters, i.e., a divide-by-ﬁve stage coupled to a

divide-by-two.

12.3 Experiments

12.3.1 Bi-quinary ripple counter

The 7490 (Fig. 12.1) is a negative-edge-triggered bi-quinary ripple counter.

It advances fromone state to the next on the falling edge of its clock input. It

consists of a 1-bit divide-by-two stage and a 3-bit divide-by-ﬁve stage that

can be cascaded two different ways. One way produces a decimal counting

sequence (0 through 9); the second produces a divide-by-ten sequence in

Q

A

R

9

Q

B

Q

C

Q

D

R

9

R

0

R

0

1

2

3

7

8 9 11 12

14

6

5

.

. 2

.

.

7490

GND: pin 10

+5: pin 5

Fig. 12.1. Pinout of 7490 decade counter.

158 Hands-on electronics

TIL 311

1

2

3

4

5

7 8

10

12

13

14 +5

D1

D0

DPL

LE

GND

DPR

BI

D2

D3

+5

Fig. 12.2. Pinout of TIL311 hex display: D0–D3 are data inputs, DPL and DPR connect to

LEDs for left and right decimal points, LE = high latches the input data, and BI = high

blanks the display.

which the high-order bit is a square wave at one-tenth the frequency of the

input clock.

**First ﬁgure out how to conﬁgure a 7490 as a divide-by-ten, clock it with
**

a digital square wave, and verify that the output is indeed a symmetrical

(i.e. high half the time and low half the time) square wave at one-tenth

the input frequency. Write down the state table and sketch the timing

diagram for the four output bits with respect to the clock input. (Also,

don’t forget to write down your complete schematic with pin numbers.)

**Next conﬁgure your 7490 as a decimal counter, so that as successive
**

clock pulses are applied it sequences through the states 0–9 in order

(0000, 0001, 0010, . . . , 1001).

**Display the state of the counter with a TIL311 hexadecimal LEDdisplay
**

(Fig. 12.2), as explained in the following paragraphs.

TIL311 numeric display

The TIL311 is a handy (but expensive) hexadecimal display that combines,

in a single package, 22 LEDs, each with its own driver circuit, a latch that

can store the four input bits, and a decoder that decides which LEDs to

turn on for a given input state. Note that, unlike most chips, the TIL311

has three notches on its package (rather than one), as indicated in Fig 12.2.

Also, pins 6, 9 and 11 are missing. The data sheet can be obtained from

the Texas Instruments website.

1

Although the TIL311 is a TTL device, it

1

http://www.ti.com/

159 12 Monostables, counters, multiplexers, and RAM

will display correctly the output of a CMOS chip, since CMOS logic levels

satisfy the TTL input criteria.

2

To display a 4-bit hexadecimal number, connect the digital signals for

the four bits to the pins labeled D0, D1, D2, D3, with D3 being the high-

order (2

3

) bit. Ground BI (blanking input – when high the display is blank)

and LE (latch enable – latches input when high). Since you don’t want

to display a decimal point, leave the DPL (decimal place left) and DPR

(decimal place right) pins open. If you wish to experiment with the decimal

place LEDs, be sure to use a current-limiting resistor in series with the input

pins. See the TIL311 data sheet for additional information concerning these

features.

**Clock your counter from a debounced switch and conﬁrm that it and the
**

display both work. What are the state table and timing diagram for the

four outputs?

**Try out the R
**

0

and R

9

inputs – what do they do?

**Add a second 7490 and TIL311 so that you can count from 0–99. Clock
**

your circuit from a digital square wave at several hertz and verify that it

works. Save it for use in the following sections.

12.3.2 Monostable multivibrator

The object of this exercise is to design a circuit that generates a pulse of

about 500 s duration. To determine experimentally whether your one-shot

is functioning properly, use its output to gate the clock to your two-digit

decimal counter, i.e., present the counter with a streamof clock pulses only

while the one-shot is ﬁring (see Fig. 12.3). The counter will count up the

number of clock pulses, which is proportional to the duration of the pulse

from the one-shot.

Fig. 12.4 shows the pinout of the 74121 and 74123 monostable multivi-

brators. For the RC timing network, use a conveniently sized resistor and

capacitor – the timing rules vary by family and type, so be sure to refer to

the correct data sheet for your one-shot. In brief, the predicted output pulse

width is given by

t

w

=

¸

¸

ln 2R

ext

C

ext

(74121)

K R

ext

C

ext

(74LS123)

R

ext

C

ext

(74HC123)

¸

¸

¸

. (12.1)

2

But the converse is not true: TTL logic levels do not satisfy the CMOS input criteria.

160 Hands-on electronics

Clock

Gate

Gated Clock

t

W

Fig. 12.3. Timing diagram for a gated clock signal. Notice how the gated clock signal is

simply the logical NAND of the gate and clock signals.

Q

Q

Trigger

A

1

B

A

2

6

11 10

1

5

4

3

Q

Q

Trigger

A

B

113

15 14

4

2

1

74121 74123

Clear

3

5 V 5 V

t

W

Trigger

Q

C

ext

C

ext

R

ext

R

ext

Fig. 12.4. Pinout of ’121 and ’123 one-shots with external RC timing network (see the

data sheets for details).

(In the above equation, K is a parameter specifed on the 74LS123 data

sheet.)

The output pulse begins following a rising edge at the trigger input. The

A and B inputs can be conﬁgured either to inhibit triggers, or to produce

a trigger from a rising B input or falling A inputs edge. Depending on the

speciﬁc chip used, there are either one or two A inputs (see Fig. 12.4).

There is always only one B input.

**Design and build logic that produces a gated clock signal as shown
**

in Fig. 12.3. Be sure to write down the schematic complete with pin

numbers.

Use a debounced switch to trigger the monostable. Clock the two-digit

counter with the gated clock signal. Choose an appropriate input clock

frequency for timing the duration of the one-shot pulse. (Calibrate your

time scale by using the scope to measure carefully the period of your digital

square-wave clock input.) The clock frequency should be low enough that

161 12 Monostables, counters, multiplexers, and RAM

the counter does not go past 99, but high enough that the width of the

one-shot pulse can be measured accurately.

**Build the gated counter as described above. Include a push-button reset
**

that zeros the counter. Record the complete circuit diagram including all

pin numbers.

**Reset the counter and trigger the monostable. What pulse width is im-
**

plied by the value of the counter, and why? What clock frequency did

you choose, and why?

**Repeat the measurement about ten times over a period of ﬁve minutes in
**

order to determine the reproducibility and stability of the output pulse

width. Plot your results as a histogram and compute the mean and r.m.s.

(root-mean-square) duration.

(Save your two-digit counter for use in the next section.)

Note on gating clocks

When gating a clock with a signal that is independent of the clock (e.g.

the push-button), a standard problem arises. If the signal from the push-

button arrives while a clock pulse is in progress, a pulse of substandard

width might be produced (see Fig. 12.5). Similarly, since the one-shot

pulse might end during a clock pulse, a substandard pulse might also be

produced then. Since substandard clock pulses might fail to meet setup

or hold requirements of ﬂip-ﬂops and counters, it is wise to avoid gating

clocks whenever possible. When gating a clock is necessary, one normally

uses a pulse synchronization circuit such as the 74120 or a pair of cascaded

ﬂip-ﬂops to ensure that signals used to gate clocks do not change state

during the clock pulse.

**Why is your circuit insensitive to this problem?
**

Clock

Gate

Gated Clock

substandard

clock pulses

Fig. 12.5. Substandard outputs can result when gating clock signals.

162 Hands-on electronics

The 74123 ICs have several additional features that we haven’t explored

here. For example, the ’123 is a retriggerable monostable equipped with

a clear input. The retrigger feature allows the output to persist longer

than the time speciﬁed by Eq. 12.1 through the application of additional

trigger edges while the output pulse is in progress. clear allows the

output to be prematurely terminated. See the data sheets for details and

operating rules.

12.3.3 Multiplexer and finite-state machine

A multiplexer (or ‘mux’) is a device that connects one of n inputs to a

single output, under control of an input number in the range 0 to n −1. It

can thus be used to select among n different input signals. It can also be

used to implement logic functions. For example, by connecting each of the

n inputs to low or high in a desired pattern, any desired 1-bit logic function

of the input number can be produced.

You can also use a mux plus a counter to generate an arbitrary timing-

pulse sequence: on each clock cycle, a different input will be selected, and

the output will be either high or low depending on the state of the corre-

sponding input. This is an example of a ﬁnite-state machine – it repeatedly

goes through a cycle of n internal states. Finite-state machines are often

useful in control applications (e.g., in deciding when to open the hot-water

valve in a washing machine).

Hook up the select inputs (A–D) of a ’150 16-to-1 multiplexer (Fig. 12.6)

to the outputs of the low-order counter chip from the previous exercise

(leave them connected to the hex display also). Note that the ’150 has an

enable input that needs to be held low. Since the ’150 is an inverting mux,

if you want its output to be high during counter state i , ground data input i .

**Which select input is high-order and which is low-order? What experi-
**

ment can you do to ﬁnd out? Do it and ﬁnd out.

**As an example of an arbitrary logic function, conﬁgure the ’150 to
**

identify which of the numbers from 0–9 are prime. Hard-wire the inputs

appropriately and connect the ’150 output to a logic indicator. Clock the

counter from a debounced push-button, and see if you are right – if not,

ﬁx it. Record the truth table for this function.

12.3.4 RAM

A random-access memory (RAM) is a chip containing a large number of

ﬂip-ﬂops, each designated by a unique numeric address. Each ﬂip-ﬂop can

163 12 Monostables, counters, multiplexers, and RAM

D15

D14

D13

D12

D11

D10

D9

D8

D7

D6

D5

D4

D3

D2

D1

D0

A B C D

W

Enable

74150

16

17

18

19

20

21

22

23

1

2

3

4

5

6

7

8

15 14 13 11

9

10

GND: pin12

+5: pin24

Fig. 12.6. Pinout of 74150 16-to-1 multiplexer.

A B C D

7489

12

10

6

4

3

2

1 15 14 13

GND: pin 8

+5: pin 16

11

9

7

5

WE

ME

DI

1

DO

3

DO

4

DO

1

DO

2

DI

2

DI

3

DI

4

Fig. 12.7. Pinout of 7489 16×4 RAM.

be accessed by address for reading or writing. Frequently, the ﬂip-ﬂops are

organized into multi-bit words, with each word separately addressable.

For example, the 7489 (Fig. 12.7), 74189, and 74219 are pin-compatible

64-bit RAM chips organized as sixteen words of 4 bits each. Each has

four address bits (labeled A through D) for selecting words 0–15, four data

inputs (DI

1

through DI

4

) for writing a value into the word being addressed,

and four data outputs (DO

1

through DO

4

) for reading the word being ad-

dressed. Of course, 64 bits in a chip is nothing nowadays, but it serves

conveniently to illustrate the random-access-memory principle using a

relatively simple chip.

164 Hands-on electronics

To write a 4-bit word into the memory, we (write enable) is brought

low. This causes the state of the inputs to be recorded in the word being

addressed. When we is high, the word being addressed is read nondestruc-

tively. Regardless of the state of we, the word being addressed appears at

the output. When the address bits change, the outputs settle to their new

value after a propagation delay called the read access time.

Since the RAM chip contains sixteen words, not ten, before hooking it

up replace your 7490 decimal counter with a 4-bit binary counter (e.g. the

7493). The 7490 and 7493 are pinout-compatible, but the ’93 counts from

0–15, allowing all sixteen words of memory to be addressed.

**Clock the 4-bit binary counter from a debounced push-button and verify
**

that it counts through all the hexadecimal (base 16) digits from0 (binary

0000) to F (binary 1111).

Open-collector outputs

RAM chips are designed for easy multiplexing with a minimum of addi-

tional components, since to increase the total amount of memory available

in a circuit one often wants to connect the outputs of multiple RAM chips

together. In the case of the TTL version of the 7489, this is accomplished

by making the data outputs open-collector (rather than the standard TTL

‘totem-pole’ output circuit). This means that the output transistors will not

operate properly unless a pull-up resistor to +5 is provided for each one.

Since we are not worrying about speed here, any convenient resistor in the

range of a few hundred ohms to 10 k is suitable. More modern memory

chips use three-state outputs, thus eliminating the need for pull-up resistors

and also improving the rise time when driving high capacitance.

The ‘master enable’ (me) signal is provided for use when the outputs of

multiple chips are connected together, to allow turning off the outputs of

all but one chip. The chip accepts input data and puts out output data only

when me = low. Therefore, be sure to ground me.

**Hookupthe counter’s outputs tothe address lines of the RAManddisplay
**

the output data with your second TIL311. Connect the data inputs and

we to level switches.

**Use the address counter and the write-enable switch to program your
**

memory to any desired sequence of hex digits (be sure to record what

sequence you choose). If you apply some ingenuity, you can spell out

messages using the letters A–F plus I (1) and O (0) (e.g., FEED B0B

A D10DE). Then clock the address counter with a digital clock at a

165 12 Monostables, counters, multiplexers, and RAM

frequency of a couple of hertz and watch your message appear! Record

the complete circuit diagram with pin numbers and explain how this

circuit works.

**How could you use the ’150 mux to shorten the sequence to any desired
**

fraction of the sixteen addresses? How could you use it to insert blank

spaces between words?

13

Digital↔analog conversion

In this chapter we will study simple techniques for generating and read-

ing voltage or current levels, i.e., converting between analog (voltage or

current) and digital (binary-number) information. The availability of high-

speed, easy-to-use, inexpensive digital⇒analog and analog⇒digital con-

verter chips has dramatically changed the way audio and video information

are recorded and processed, as well as howcomputers are used in laboratory

research and process control. The process of converting digital information

into voltages or currents whose magnitudes are proportional to the digitally

encoded numbers is called digital-to-analog (D/A) conversion. The reverse

process is called analog-to-digital (A/D) conversion. The devices that carry

out these conversions are called DACs and ADCs, respectively.

In this chapter, after building a simple DACfroma digital counter and an

op amp, you will continue your exploration of analog/digital conversion by

building a 4-bit tracking ADC. Having learned the basic operating princi-

ples, you’ll use an ADC080x 8-bit successive-approximation A/D chip to

digitize (i.e., convert to digital) an arbitrary AC signal. The original signal

will then be re-created from the digitized data using a DAC080x D/A chip.

This exercise will also allow you to explore the limitations of ADC and

DAC operations.

Please be sure to work through these circuits in advance, otherwise it

is highly unlikely that you will successfully complete the exercises in a

timely fashion! Carefully study the manufacturer’s data sheets which pro-

vide extensive details on operation and performance. As always, complete

schematic diagrams signiﬁcantly improve debugging efﬁciency.

Apparatus required

Breadboard, oscilloscope, 74191, TIL311, 311 comparator, 741 op amp,

resistors, capacitors, DAC0806 (or similar), ADC0804 (or similar), 7400,

7432, four 7474, 74112, 74138.

167

168 Hands-on electronics

13.1 A simple D/A converter fabricated from familiar chips

Recall that when an op amp is set up as an inverting ampliﬁer, the non-

inverting input is grounded, and the inverting input, which is tied to the

output through a feedback resistor, acts as a ‘virtual ground’. If a resistor

R is connected from a voltage V to the inverting input of the op amp, a

current V/R will ﬂow. If you double the resistance, half as much current

will ﬂow. Suppose you have four resistors with the resistances R, 2R, 4R,

and 8R. The corresponding current ﬂows will be in the proportion 8:4:2:1

(see Fig. 13.1(a)).

A 74191 counter has four outputs Q

3

through Q

0

, with Q

3

the MSB

(most signiﬁcant bit) and Q

0

the LSB (least signiﬁcant). In addition it

has four parallel-load inputs, count-enable and count-direction (up/down)

inputs, and ripple-clock and terminal-count outputs for use when cascading

multiple stages. If we feed the counter outputs to the inverting input of an

op amp through resistors R, 2R, 4R, and 8R (in order from MSB to LSB),

we get a 4-bit digital-to-analog converter. The current into the feedback

resistor will be proportional to the number that corresponds to the state

of the counter. Given a suitable feedback resistor such that the op amp

does not saturate, the output voltage will be proportional to this current. To

produce a desired output voltage, we can load into the counter any desired

value; we can also increment or decrement the counter to get a voltage that

changes in time in stepwise fashion (see Fig. 13.1(b)). This output can, of

course, be observed on an oscilloscope or other measurement device.

D1

D0

D2

D3

op amp

+

_

R

1

2R

4R

8R

R

Analog Out

0 Volts (0000)

−9.4 Volts (1111)

R = R

1

V = 5 V

H

V = 0 V

L

(b) (a)

Fig. 13.1. (a) Simple D/A converter; (b) output waveform resulting from input counting

sequence.

169 13 Digital↔analog conversion

To demonstrate D/A conversion, you will build such a 4-bit DAC. To

reduce the chances of hooking up the circuit incorrectly,

**Begin by setting up a 74191 counter and make sure it is working properly:
**

hookupits outputs toa TIL311display, clockit froma debouncedswitch,

and verify that it goes through all sixteen states in order. Test it counting

both up and down – you can control which way it counts using the ‘u/d’

input.

**Next, hook up the counter outputs to the summing junction of the op amp
**

through resistors, as described above. Use a 2.2 k resistor to connect to

Q

3

, a 4.7 k resistor for Q

2

, a 10 k resistor for Q

1

, and a 22 k resistor

for Q

0

, and connect a 3.3 k feedback resistor. Connect a 1 kHz digital

signal to the clock input of the counter and view the analog output on an

oscilloscope.

Of course, if we wanted to produce accurate analog output voltages,

we would need precision resistors, for example 2.50 k, 5.0 k, 10 k, and

20 k. Moreover, we would need to take into account the inevitable small

differences among the high and low levels of the counter outputs. We shall

not worry about these reﬁnements, since it is our intention here merely to

illustrate the basic idea of D/A conversion.

The output should be a ﬁfteen-step staircase waveform (Fig. 13.1b),

with each step having approximately the same height. To see a stable dis-

play of the waveform, you can trigger the scope using the falling edge of

the MSB.

**What full-scale output voltage do you expect (i.e. when the counter is at
**

15 in decimal or 1111 in binary)? What do you observe?

**Is the staircase rising or falling? Why is this? What simple change can
**

you make to reverse the direction of the staircase?

**What are the output voltages corresponding to states 4, 5, 6, 7, and 8 of
**

the counter? Measure the four resistances and the high and low voltage

levels of the four counter outputs (Q

0

–Q

3

), and explain each DACoutput

voltage.

**Write down a complete circuit diagram with pin numbers. Explain in
**

your own words how this circuit works.

Despite the common misconception that modern electronics is strictly

digital, analog electronics is still going strong. For all practical purposes,

our everyday world is analog. The digital representation of any waveform

(music, for example) is only an approximation. To smooth out the discon-

tinuities of digitized waveforms requires analog electronics.

170 Hands-on electronics

13.2 Tracking ADC

To measure an analog signal you need to invert the process of D/A conver-

sion. There are various ways of doing this, but, just as division is harder and

slower than multiplication, and taking the square-root harder and slower

than squaring, analog-to-digital conversion is harder and often slower than

digital-to-analog.

Given a DAC, a counter, and a comparator, a simple approach is to

increment the counter (starting fromzero) until the DACoutput crosses the

analog input. Using the comparator to compare the analog input to the DAC

output, you stop counting when the comparator output switches states. At

that point, the counter holds a digital approximation to the magnitude of

the input.

A simple variant of this circuit will follow (or ‘track’) changes in the

input voltage. You can turn your 4-bit counter/DAC into such a tracking

ADCby driving u/d froma comparator that compares the DACoutput with

the analog input voltage.

Use a potentiometer to make the analog input voltage: connect one end

to ground and the other to −15 V. The slider controls the input volt-

age, which you can vary between 0 V and −15 V. To stabilize the op-

eration of the circuit, use some hysteresis by connecting a series 10 k

resistor between the input voltage and the comparator noninverting in-

put and 1 M between the comparator output and the noninverting input

(see Fig. 13.2).

**Clock the counter at a few hertz and observe its state with the TIL311
**

as you vary the input voltage. What do you observe?

**How should the comparator inputs be conﬁgured: which signal should
**

go to the inverting and which to the noninverting input? Is the output

number ‘homing in’ on the expected value? If not, did you perhaps

connect the comparator backwards? Explain, and if you did it wrong the

ﬁrst time, ﬁx it.

Record the output numbers for a few different input voltages.

**Why is the output number never stable? How (if at all) does this affect
**

the precision of the voltage measurement?

Note that the trackingADCis slowat followinglarge input-voltage changes,

since it has to count through all the intermediate values, but it has good

performance if the input voltage changes gradually.

171 13 Digital↔analog conversion

Q

B

Q

A

Q

C

Q

D

op amp

+

_

Analog In

U/D

Load

Enable

74191

Clk

+5

comp

+5

1 MΩ

10 kΩ

22 kΩ

2.2 kΩ

4.7 kΩ

10 kΩ

3.3 kΩ

1 kΩ

Fig. 13.2. Simple A/D converter. (The polarity of the comparator inputs is left as an

exercise for the reader.)

13.3 080x ADC and DAC chips

13.3.1 Successive-approximation ADC

The technique just described is comparable to looking for a word in the

dictionary by looking at each word one after the other until you ﬁnd the

right one (or, if the word you’re looking for isn’t in the dictionary, until

you ﬁnd a word beyond the one you’re looking for). As we’ve seen, this is

a ﬁne algorithm for tracking a slowly changing signal, but if the DAC is

far from the signal voltage it takes a long time to home in.

A faster approach (and one used in many ADC chips) is based on the

binary (or logarithmic) search algorithm, in which at each step you reduce

the search range by half. For example, if you’re looking up a word in an

n-word dictionary, ﬁrst look at word n/2. If the word you want comes later

in the alphabet, next try word 3n/4; if your word comes earlier, next try

n/4. At the next step there are four possibilities: word n/8, 3n/8, 5n/8, or

7n/8 – and so on. This method will ﬁnd any word in at most log

2

n steps.

In the case of analog-to-digital conversion, the logarithmic search has

another name: successive approximation. If there are n output bits, there

are 2

n

possible output values, but instead of trying each value in succes-

sion, you try each bit in turn, starting from the MSB: ﬁrst generate a 1-bit

approximation to the value, then correct it to 2-bit accuracy, then 3-bit, and

so on.

172 Hands-on electronics

ADC080x

The ADC080x series of chips are inexpensive 8-bit successive-

approximation A/D converters. The logic inputs and outputs are compat-

ible with both TTL and CMOS, and the outputs have tri-state capability.

The chips are general-purpose ADCs that can be used as stand-alone

converters or interfaced with a computer or other logic system. They accept

differential inputs for increased common-mode-noise rejection capability.

The digitized output thus measures the voltage difference V

in+

− V

in−

.

The successive-approximation algorithm used in these chips requires

sixty-four clock cycles to complete a conversion. A few additional clock

cycles are used during startup and after the conversion to latch the data on

the output lines. The clock can originate from either an external or internal

(‘self-clocking’) source. The self-clocking option uses an on-chip oscillator

(withSchmitt-trigger timinginput), incombinationwithanexternal resistor

and capacitor that determine the period, as shown in Fig. 13.3.

All of the input and output control signals are active-low. There are three

input control lines, labeled cs, rd, and wr. An A/D conversion is started

V

CC

A GND

ADC080X

+5 V

WR

CLK R

D0

D7

D6

D5

D4

D3

D2

D1

D GND

DONE

CLK IN

V

ref

/2

V

in

(−)

V

in

(+)

CS

RD

C

R

CLK IN

CLK R

4

19

Outputs

Inputs

V

in

Internal Clock Osc.

T

CLK

= 1.1 RC

~

V

in

(a)

(b)

(MSB)

(LSB)

Fig. 13.3. (a) Pinout for the ADC080x series of A/D converters. (b) The on-chip

self-clocking conﬁguration. Note the locations of the most signiﬁcant bit (MSB) and the

least signiﬁcant bit (LSB). The ‘x’ in ADC080x means that multiple versions of this IC

exist (e.g. ADC0804).

173 13 Digital↔analog conversion

by bringing rd and wr low simultaneously. cs is used in microprocessor-

based applications – for this exercise it should be connected to ground. rd

is equivalent to output-enable – rd high puts the outputs into their high-

impedance state; when low, the output lines are active – therefore, connect

rd to ground. With cs tied to ground, wr simpliﬁes to A/D start (i.e.,

start the digitization now!). Once the digitization completes, the result is

latched onto the output bus and a done pulse is issued (pin 5).

Connecting done to wr (as shown in Fig. 13.3(a)) puts the ADC into

a ‘free-run’ mode, in which the completion of one conversion initiates the

next. Once triggered, the ADC will repeatedly digitize the input voltage

difference. To ensure free-run status following power-up, wr may need to

be brought low manually, so connect wr to a debounced push-button as

well as to done.

The chip has two grounds, called analog ground (a gnd) and digital

ground (d gnd). In many applications, the analog and digital grounds are

kept separate to reduce the analog noise introduced by rapidly changing

digital signals. We will ignore these effects and use a common ground for

both pin 8 and pin 10.

The range for the input voltage difference is determined by the input

voltage reference – we are using V

CC

. A digital output of 00000000 should

correspond to an input very close to zero volts, while the output 11111111

should correspond to a voltage near V

CC

. We shall use single-ended inputs,

so you should ground V

in−

.

(A more complete explanation of the ADC080x’s features can be found

in the manufacturer’s data sheets.)

**Connect an ADC080x chip as shown in Fig. 13.3(a). Connect pins 1, 2,
**

7, 8, and 10 to ground. Connect pin 4 to ground using a 50 pF capacitor.

Place a 10 k resistor between pins 4 and 19. Connect pins 3 and 5, and add

a connection to a push-button. Using the 1 k pot, apply a variable voltage

between +5 V and ground to pin 6. Connect pins 11–18 to LED logic

indicators. Connect pin 20 to +5 V power, and leave pin 9 unconnected.

**Adjust the input voltage and observe the digital output. What measured
**

input voltage corresponds to the binary output 00000001? What mea-

sured input voltage corresponds to the binary output 11111110? Measure

several other input-voltage values and plot the input voltage versus dig-

ital output. Is the plot linear? What is the input range, and how does the

ADCrespond to small excursions outside this range (not less than ground

and not more than V

CC

)? With what precision does the ADC measure

the input voltage?

174 Hands-on electronics

**Estimate the conversion time by looking at pin 5 with the oscilloscope.
**

(Conversion time is the amount of time an ADC requires to digitize an

input voltage.) What is the sampling rate? Does the conversion time

depend on the input voltage? Taking into consideration the value of the

external RC network, does the measured sampling rate agree with your

expectations? If not, why not? (Hints: parallel capacitors add linearly,

and some small stray capacitance is common on breadboards such as

the PB-503.)

**Observe, sketch, and explain the waveforms at pins 4 and 19. Replace
**

the 50 pF capacitor with a 100 pF capacitor. Explain what happens

to the waveforms. What happens if you remove the 100 pF capacitor

completely (i.e., do not use any external capacitor)? Estimate the value

of the stray capacitance. Replace the 50 pF capacitor.

**Write a complete circuit diagram with pin numbers, and explain in your
**

own words how this circuit works. Comment on your observations and

measurements.

Save your ADC circuit for use in the next part.

The DAC080x D/A converters

The DAC080x is a family of popular D/A converter chips, of which the

DAC0806 is the least expensive – it is an 8-bit DAC with 6-bit precision,

while the related DAC0807 and DAC0808 have 7- and 8-bit precision, re-

spectively. Our home-brew D/A converter above used 4 bits to distinguish

sixteen different voltage levels. With the DAC0806 we can distinguish not

sixteen but sixty-four different voltages (at least). If you need more accu-

racy, you can use the DAC0807, the DAC0808, or one of the more sophis-

ticated chips that are available. (You may use whichever of these chips is

available for this exercise.)

Chips of the DAC080x family output a current that is proportional to

the value of the digital input. Some D/A converter chips (e.g., the expen-

sive, and less readily available, NE5018) include an op amp on the same

chip. The DAC080x family does not! To convert the DAC’s output cur-

rent to an analog voltage, you can use an external op amp, as shown in

Fig. 13.4.

What exactly is the difference between the ‘top-of-the-line’ DAC0808

and the ‘budget’ DAC0806? Within the DAC0808 chip are eight resistors

of nominal resistance R, 2R, 4R, 8R, 16R, 32R, 64R, and 128R. For 8-bit

accuracy, these need to have better than 0.5% tolerance. The DAC0806

175 13 Digital↔analog conversion

GND

V

EE

A0

A7

A6

A5

A4 A3

A2

A1

V

CC

V

ref

(+)

V

ref

(−)

COMP

I

Q

N.C.

DAC0806

4.7 kΩ

+

_

Op Amp V

out

+5 V

−15V

C = 0.1 µF

4.7 kΩ

4.7 kΩ

Output

Inputs Inputs

(MSB) (LSB)

Fig. 13.4. Pinout for the DAC080x series of D/A chips (with an output op amp added).

Note that the bit order is the reverse of that used for the ADC080x, such that A0 is the

MSB and A7 the LSB.

also has eight resistors, but their tolerances are not as good, such that the

DAC0806 is not guaranteed to put out 256 distinguishable voltage levels,

but only sixty-four.

These chips also contain circuitry to buffer the input signals and stan-

dardize their voltage levels, to avoid inaccuracies due to voltage variations

in logic levels on different input lines. Consult the manufacturer’s data

sheet to see what else your D/A converter chip contains.

**Wire up a DAC080x chip as shown in Fig. 13.4. This conﬁguration will
**

give an output voltage between 0 and +5 V. Connect the outputs of the

ADC you built above to the inputs of the DAC.

**What is the output voltage that corresponds to the DAC digital input
**

00000001? What is the output voltage that corresponds to the digital

input 11111110? Take a few more data points and plot output voltage

versus digital input. Comment on your results.

**Measure the precision of your DAC. Does it match the speciﬁed
**

precision?

**Be sure to write down a complete circuit diagram with pin numbers.
**

Explain in your own words how this circuit works. Comment on your

observations and measurements.

176 Hands-on electronics

V

in

To A/D input

+5 V

R

R

C

R = 100 kΩ

C = 100 F µ

t

V

t

V

+

_

Op Amp

+

Fig. 13.5. Method for producing a DC-shifted waveform.

Measuring an AC input signal

Replace the DC input with an AC input. The most convenient AC input is a

waveform from your function generator. However, the function-generator

output is symmetric about ground. The waveform will thus be outside the

ADC range 50% of the time! You can DC-shift the waveform as shown in

Fig. 13.5. Warning: be careful not to exceed the ADCinput voltage range!

Input voltages larger than V

CC

or lower than ground may damage the ADC

chip.

**Explain how the DC shift occurs and the signiﬁcance of the component
**

values chosen. Suggest at least one other method for DC-shifting the

input waveform.

**Apply a sine wave of 2 Vamplitude centered at +2.5 Vto the A/Dinput.
**

Set the frequency to ∼0.5 Hz and try to measure the amplitude using the

LED indicators. You can ‘freeze’ the A/D output using the push-button

connected to pin 3. Explain how this works!

**Compare the input and output waveforms. How well do they agree?
**

Comment on your observations.

**Increase the input frequency (you may need to adjust the amplitude to
**

stay within the range of the A/D input – why?). What happens to the

output waveform as the input frequency approaches the sampling fre-

quency? What happens when the input frequency exceeds the sampling

frequency? Make several sketches of the input and output waveforms at

various frequencies, and comment on your observations. Explain the re-

lationship between sampling frequency and output frequency response.

177 13 Digital↔analog conversion

**Experiment with triangle and square waves of various frequencies.
**

Record and sketch a few input and output waveforms, and comment

on your results. Be careful not to exceed the allowed input range.

As discussed above, A/D and D/A conversions are merely approxima-

tions. Higher precision and higher sampling rates improve the approxi-

mation at the expense of increased cost and data size. For example, let’s

say we’re recording the sound track for a TV commercial. To digitize and

record a 30 s waveform with 8-bit precision and 10 kHz sampling rate

requires 2.4 Mbit of memory storage. Increasing the precision to 16-bit and

sampling rate to 100 kHz increases the required space by a factor of 20, to

48 Mbit.

13.4 Additional exercises

13.4.1 Digital recording

You can convert your A/D–D/A circuit into a digital recording and audio-

processing system by adding an audio input, a memory to store the digital

data, and an audio output. You can use speakers for both input and output,

buffered with suitable op amp circuits. The 32 k × 8 CY62256 memory

chips require a 15-bit address counter, which you can make from 7493s or

74191s. What other control circuitry do you need? By varying the clock

speed you can trade off ﬁdelity (sampling rate) for message length – e.g.

at 1 kHz sampling rate, you would be able to store 32 s worth of sound.

Try recording a sound sample, then playing it back at various speeds. You

can also try adding nonlinear gain elements (say a logarithmic ampliﬁer)

or ﬁltering to see how the sound is affected. Can you ﬁgure out a way to

program reverb?

For faithful recording, it is important that the input voltage be constant

during the entire time of the conversion. This can be accomplished using a

sample-and-hold ampliﬁer (SHA) to sample the input and hold it until the

conversion is complete. The National Semiconductor LF398 is an SHAIC.

Pin 1 is connected to +15 V and pin 4 is connected to −15 V. There

are separate analog and logic inputs, on pins 3 and 8, respectively, and

the output is at pin 5. Ground pin 7. An external capacitor (1000 pF) is

connected between pin 6 and ground. This capacitor is used to hold the

analog data until the ADC has had an opportunity to process it.

178 Hands-on electronics

If you choose to add a sample-and-hold, what additional control logic is

needed? Can you hear the difference it makes in ﬁdelity? How would you

describe the difference, and how would you explain it?

13.4.2 Successive-approximation ADC built from components

To see ﬁrst-hand how the successive-approximation algorithm works, you

can build an 8-bit ADC using TTL or CMOS parts plus a DAC and

comparator.

In our successive-approximation circuit (Fig. 13.7) the eight bits are

storedinfour ’74dual D-type ﬂip-ﬂopchips. The successive-approximation

algorithmconsists of trying each bit in both the 0 and 1 states, starting from

the MSB and ending with the LSB. Each bit starts out at 0 and is then set

to 1. If the DAC output exceeds the analog input, the bit is set back to 0,

otherwise it is left as a 1. To accomplish this, the circuit goes through a

sixteen-state cycle, in order both to set and (possibly) reset each of the eight

bits.

The sixteen-state cycle is provided using a ’191 four-bit binary counter.

The ’138 1-of-8 decoder routes clock pulses (how many?) to each ﬂip-ﬂop

in turn. The 311 comparator compares the DAC output with the analog

input, and its output is connected to the D input of each ﬂip-ﬂop. Note that

the DAC080x is a current-sinking DAC, and its output thus becomes more

negative as its digital input increases from 0 to 255. The connection shown

thus provides negative feedback: if the DAC output is too negative, the

currently addressed bit is set to 0 in order to raise the DAC output voltage,

and if the DAC output is too positive, the bit is set to 1 in order to lower it.

Timing and control logic

If you choose to undertake this exercise, you should ﬁrst spend some time

understanding the timing cycle and drawing a complete timing diagram

for the control logic (shown in Fig. 13.6 and the lower left-hand corner of

Fig. 13.7). Pay particular attention to the sequence and timing of signals at

the beginning and end of the conversion cycle, and see if you can ﬁgure out

the answers to the following questions. In what state does the counter start

out? In what state does it end? Exactly what function does each decoder

enable perform? (Hint: each performs a slightly different function.) What

happens if youattempt tostart a newconversioncycle while one is alreadyin

progress? Why is it important for the control ﬂip-ﬂops to be negative-edge

179 13 Digital↔analog conversion

Q

R

J

clk

K

Q

Q

J

K

+5

R

7 6 5 4 3 2 1 0

Start Convert

Done

74138

74191

E

1

Q

0

P

0

A

2

A

1

A

0

P

1

P

2

P

3

Q

1

Q

2

Q

3

E

2

E

3

LD

U/D

E

Fig. 13.6. Control logic for 8-bit successive-approximation ADC.

triggered: can the clock signal to the ’191 glitch (i.e. have a pulse of sub-

standard width – see discussion in section 12.3.2)? Why, or why not?

Begin by building and debugging the control logic by itself – but be

sure to leave room for the additional chips! (If you prefer, you may choose

to build and analyze the simpler control logic described below.) Try to

arrange the chips and connecting wires neatly, so that it is easy to see

where each wire goes – some color-coding could be helpful. In case a chip

needs to be replaced, try to avoid overly tight wiring across the top of

any chip.

Clock your control circuit with a digital square wave from the function

generator, and provide the ‘start convert’ signal with a debounced

push-button. You should be able to see the full timing sequence on the

oscilloscope. Trigger the scope on the output of the second ﬂip-ﬂop and

observe each of the other signals as you repeatedly issue ‘start convert.’

Verify that your timing diagram is correct.

Complete ADC circuit

Now add the rest of the circuit. Use the breadboard logic indicators to

display the data bits. Attach the reset input to the other debounced push-

button.

Measure the DC analog input voltage V

in

with a digital voltmeter. Try

several voltages over the full range and make a graph of your results.

180 Hands-on electronics

A0 A1 A2 A3 A4 A5 A6 A7

74112 74112

reset

Fig. 13.7. 8-bit successive-approximation ADC.

How good is your ADC? Is it linear? What is its zero offset (the number

it puts out for V

in

= 0), and what is its slope constant (volts per output

count)? What is its least count (the voltage change corresponding to one

ADC count)? What is its full-scale voltage?

Try raising the clock frequency. At what frequency does it stop working?

Does this make sense? What do you think limits the conversion speed of

this circuit? (Illustrate your answer with the relevant timing diagram.)

Will this ADC always work correctly the ﬁrst time after power is turned

on? Why, or why not?

181 13 Digital↔analog conversion

Simpler version of control logic

You might want to build and analyze a simpler version of the control logic

using parallel-output shift registers instead of the counter/decoder version

discussed above. The idea is to produce eight bits, of which seven are 1s

and one is a 0, and circulate them around an 8-bit shift register. Connect

each shift-register output both to clock and to set of a ﬂip-ﬂop, so that each

ﬂip-ﬂop is ﬁrst set, and then reset, by the clock edge if the analog output

is too big. This simple approach might or might not work, depending on

the internal timing of the ﬂip-ﬂop (clock and set are changing simulta-

neously) – the question is whether the set signal goes away within the

ﬂip-ﬂop soon enough so as not to override the clock edge. If it does not,

you can use the ‘diode trick’ to delay the clock relative to set. (The diode

trick consists of adding a series diode to shift a digital signal in voltage,

thereby changing the time at which it crosses threshold.)

Further reading

The following table indicates the sections or chapters in four popular

textbooks where you can ﬁnd additional background information for each

chapter of our text.

Expt D. & H. Barnaal H. & H. Simpson

1 1, 6 A1, A2 1 thru 1.02, 1.09, 1, App. C

1.11, 1.32, 12.01

2 2, 3 A1 1.02–03, 1.07–09, 2, 3, App. A

1.12–16, 1.18–20

3 4, 5 A1, A4 1.17, 1.25–28 4

4 8 A8 2 thru 2.13, 2.15–16 5

5 8 A8 3 thru 3.09 6

6 8 A4, A5, A8 2.18, 2.14

7 9 A6 4 thru 4.09, 4.11–12 9, 10

8 9 A6 4.09–10, 4.14, 4.19–20 10

9 10 A7, A1 Suppl. 4.23–24, (5) 10, 11

10 11 D1, D2 1.10, 8 thru 8.06, 8.08–10 12

11 12 D3, D4 8.07, 8.16–17 13

12 12, 13 D5 8.14, 8.18–26, 8.34 13

13 14 D6 4.16, 9.15–16, 9.20 15

Key:

D. & H.: A. James Diefenderfer and Brian E. Holton, Principles of Electronic Instrumen-

tation (Saunders, 1994);

Barnaal: Dennis Barnaal, Analog Electronics for Scientiﬁc Application and Digital

Electronics for Scientiﬁc Application (reissued by Waveland Press, 1989);

H. & H.: Paul Horowitz and Winﬁeld Hill, The Art of Electronics (2nd edition, Cambridge

University Press, 1989);

Simpson: Robert E. Simpson, Introductory Electronics for Scientists and Engineers

(2nd edition, Prentice-Hall, 1987).

183

Appendix A

Equipment and supplies

General equipment you will need:

one Global Specialties PB-503 powered breadboard, or equivalent;

one Tektronix TDS 210 Dual Trace Oscilloscope, or similar;

two oscilloscope probes with 10X attenuation;

one digital multimeter with probes;

one power tranformer (12.6 V each side of center tap);

four 50–100 cm banana leads (two red and two black).

Analog components

Resistors,

1

4

W Number required Capacitors Number required

33 1 50 pF 1

68 1 100 pF 1

100 3 300 pF 1

330 2 0.0047 F 1

560 1 0.01 F 1

820 1 0.033 F 3

1 k 2 0.1 F 1

2.2 k 1 0.47 F 1

3.3 k 3 1 F 1

4.7 k 3 100 F

b

1

10 k 7 1000 F

b

1

22 k 2

100 k 2

330 k 1

1 M 1

10 M 1

1 k

a

1

a

2 W resistor.

b

50 V capacitor.

185

186 Hands-on electronics

Diodes, transistors, analog IC’s

four Si signal diodes;

one 1 A Si rectiﬁer diode;

two 3.3 V 1 W Zener diodes;

one 5.1 V 1 W Zener diode;

one diode bridge element;

one red light-emitting diode;

three 2N3904 NPN transistors;

three 2N3906 PNP transistors;

two VP0610L MOSFETs;

two VN0610N MOSFETs;

two 2N5485 N-channel JFETs;

one 411 op amp;

one 555 timer;

three 741 op amps;

one 311 comparator.

Miscellaneous

four alligator clips;

two fat-pin adapter sockets.

Digital components

Component Number required

7400 1

7404 1

7432 1

7474 4

7485 1

7486 1

7489 1

7490 2

7493 1

74112 1

74121 1

74138 1

74150 1

74191 1

74373 1

DAC0806 1

ADC0804 1

TIL311 2

187 Appendix A. Equipment and supplies

Suppliers of parts

There are numerous companies selling electronic components and supplies. Most allow

customers to purchase small quantities directly over the web. Prices are reasonable and

service is excellent. Several (e.g., Digi-Key Corp) even have links to product data sheets as

part of their online catalog. Product information, availability, and pricing are easily found

through a few quick web searches. We’ve included a few URLs to help get you started.

1

At

the time of going to press, the parts and supplies needed to complete the exercises within

this book could be purchased from the companies below. Pricing and availability may vary,

so shop around!

RadioShack

http://www.radioshack.com/

Digi-Key Corp

http://www.digikey.com/

Newark Electronics

http://www.newark.com/

Tequipment

http://www.tequipment.net/

Electronix Express

http://www.elexp.com/

Arrow Electronics

http://www.arrow.com/

Jensen

http://www.jensentools.com/

1

The publisher has used its best endeavors to ensure that all URLs referred to in this book are correct

and active at the time of going to press. However, the publisher has no responsibility for the websites

and can make no guarantee that a site will remain live or that the content is or will remain appropriate.

Appendix B

Common abbreviations and circuit symbols

Order-of-magnitude prefixes

m = milli = 10

−3

= micro = 10

−6

n = nano = 10

−9

p = pico = 10

−12

f = femto = 10

−15

k = kilo = 10

3

(or kilohm = 10

3

)

M= mega = 10

6

(or megohm = 10

6

)

G = giga = 10

9

T = tera = 10

12

Mathematical symbols

∼ of order

≈ approximately equal to

≡ equals by deﬁnition

change in

⇒implies

Electrical terms

β = h

FE

= transistor current gain

ω = angular frequency

= ohm

A = ampere

AC = alternating current

C = coulomb

C = capacitance

dB = decibel

DC = direct current

F = farad

188

189 Appendix B. Common abbreviations and circuit symbols

f = frequency

g

m

= transconductance

H = henry

Hz = hertz

I = current

L = inductance

P = power

Q = quality factor (of a bandpass ﬁlter)

R = resistance

V = volt

V

CC

= most positive voltage in a circuit (positive supply voltage)

V

EE

= most negative voltage in a circuit (negative supply voltage)

X = reactance

Z = impedance

Electrical devices

ADC analog to digital converter

C symbol used in schematics for a capacitor

CMOS complementary-MOSFET integrated-circuit family

CRT cathode-ray tube

DAC digital to analog converter

ECL emitter-coupled-logic integrated-circuit family

FET ﬁeld-effect transistor

JFET junction FET

L symbol used in schematics for an inductor

MOSFET metal-oxide-semiconductor FET

op amp operational ampliﬁer

Q symbol used in schematics for a transistor; can also refer to the latched output

of a ﬂip-ﬂop or register

R symbol used in schematics for a resistor

SPDT single-pole-double-throw switch

SPST single-pole-single-throw switch

TTL transistor–transistor-logic integrated-circuit family

190 Hands-on electronics

Appendix C

RC circuits: frequency-domain analysis

In many freshman-physics textbooks, the frequency-domain analysis of RC circuits is not

explicitly treated; however, it is not particularly difﬁcult. Here is a detailed derivation.

At any moment of time, the charge Q stored on the capacitor is proportional to the voltage

V

C

across it:

Q = CV

C

. (C.1)

If the voltage across the capacitor is varying sinusoidally in time, it follows that the charge

must also vary sinusoidally. Then, since the current I ﬂowing onto one plate of the capacitor

is the time derivative of the stored charge, the current must also be a sinusoidal function,

but out of phase with the voltage by 90

◦

(since the derivative of the sine is the cosine, which

is out of phase with the sine by 90

◦

, and the derivative of the cosine is minus the sine).

Nowconsider a series RCcircuit beingdrivenbya sinusoidal ACvoltage source (Fig. C.1).

Since the resistor andcapacitor are inseries, theymust have the same current ﬂowingthrough

them; however, it is not necessarily in phase with the source voltage, V. Suppose (for the

sake of deﬁniteness) that

V = V

0

sin ωt, (C.2)

i.e. we have chosen the zero of time to be a moment when the voltage across the source

is zero. Then, allowing for an unknown phase difference between the current in the circuit

and the voltage applied by the source, we can write

I = I

0

sin (ωt +φ). (C.3)

Kirchhoff’s voltage law tells us that, at any moment of time, the applied voltage must

equal the sum of the voltage across the capacitor and that across the resistor:

V = V

R

+ V

C

. (C.4)

Now, by Ohm’s law, V

R

= I R, and we also have V

C

= Q/C = (1/C)

I dt. Substituting

these relations into Eq. C.4,

V = I R +

1

C

t

t

0

I dt (C.5)

= I

0

R sin (ωt +φ) +

1

C

t

t

0

I

0

sin (ωt +φ) dt, (C.6)

where we have made use of Eq. C.3.

191

192 Hands-on electronics

We can easily carry out the integration in Eq. C.6 using the substitution u = ωt +φ,

giving

V = I

0

R sin (ωt +φ) +

I

0

ωC

ωt +φ

ωt

0

+φ

sin u du (C.7)

= I

0

R sin (ωt +φ) −

I

0

ωC

[cos (ωt +φ) −cos (ωt

0

+φ)]. (C.8)

The constant of integration, cos (ωt

0

+φ), can be determined by the condition V(0) = 0,

which we assumed in writing Eq. C.2:

V(0) = 0 = I

0

R sin φ −

I

0

ωC

[cos φ −cos (ωt

0

+φ)], (C.9)

giving

cos (ωt

0

+φ) = cos φ −ωRC sin φ, (C.10)

thus

V = I

0

R sin (ωt +φ) −

I

0

ωC

[cos (ωt +φ) −(cos φ −ωRC sin φ)] (C.11)

= I

0

R[sin (ωt +φ) −sin φ] −

I

0

ωC

[cos (ωt +φ) −cos φ], (C.12)

which clearly satisﬁes V(0) = 0.

Eq. C.12 can be simpliﬁed using the trigonometric identities for sines and cosines of

sums:

V = I

0

R[sin ωt cos φ +cos ωt sin φ −sin φ]

−

I

0

ωC

[cos ωt cos φ −sin ωt sin φ −cos φ]. (C.13)

Gathering and separating terms in sin ωt and cos ωt , and using V = V

0

sin ωt , since sin ωt

and cos ωt are independent functions of time, we obtain two equations:

V

0

sin ωt =

I

0

R cos φ +

I

0

ωC

sin φ

sin ωt (C.14)

I

0

R sin φ −

I

0

ωC

cos φ =

I

0

R sin φ −

I

0

ωC

cos φ

cos ωt. (C.15)

Eq. C.15 states that a constant is equal to the same constant times a function of time. This

can be satisﬁed for all times only if the constant is zero,

1

thus

φ = tan

−1

1

ωRC

. (C.16)

Eq. C.14 can be simpliﬁed as

V

0

= I

0

R cos φ +

I

0

ωC

sin φ. (C.17)

1

Otherwise we could divide through by the constant to obtain cos ωt = 1, which clearly does not

describe the behavior of the circuit.

193 Appendix C. RC circuits: frequency-domain analysis

Fig. C.1. Series RC circuit.

V

0

I R

0

I R

0

I R cos

I

0

ωC

I

0

ωC

sin

φ

φ

φ φ

Fig. C.2. Right triangle represented by Eq. C.17, illustrating that V

0

= I

0

R cos φ +

I

0

ωC

sin φ.

This describes a right triangle with hypoteneuse of length V

0

and sides of length I

0

R

and I

0

/ωC (Fig. C.2), which is a useful way of visualizing the relationship among the

amplitudes of the source voltage, resistor voltage, and capacitor voltage. The relationship

is Pythagorean:

V

2

0

= (I

0

R)

2

+

I

0

ωC

2

. (C.18)

We thus have

I

0

=

V

0

R

2

+

1

ωC

2

. (C.19)

If we take the output as the resistor voltage, we get a high-pass ﬁlter:

V

out

= I

0

R =

V

0

1 +

1

ωRC

2

. (C.20)

If we take the output as the capacitor voltage, we get a low-pass ﬁlter:

V

out

=

I

0

ωC

=

V

0

(ωRC)

2

+1

. (C.21)

Appendix D

Pinouts

2N

E

3

9

0

4

B

C

2N

E

3

9

0

6

B

C

2

N

5

4

8

5

D

J

F

E

T

S

G D

0

6

1

0

L

S

G

M

O

S

F

E

T

V-

V+

1

2

3

4 5

6

7

8

741 or 411

op amp

311

comparator

+

_

2

3

4

6

7

V-

V+

+

_

2

3

4

1

7

8

GND

Trigger

Threshold Out

Discharge

Reset

V+

555 TIMER

1

2

3

4 11

12

13

14

GND

5

6

7 8

9

10

1

2

3

4 11

13

14

GND

5

6

7 8

9

10

+5

7400 Quad NAND

12

1

2

3

4 11

13

14

GND

5

6

7 8

9

10

+5

7404 Hex INVERTER

12

+5: pin 14

GND: pin 7

1

2

3

4

5

6 8

9

10

11

12

13

D Q

CLK

Q

R

S

D Q

CLK

Q

R

S

7474 D-Type

Flip-Flop

+5: pin 16

GND pin 8 :

15

3

1

4

5

6

J

Q

CLK

Q

R

S

74112 JK Flip-Flop

TIL 311

1

2

3

4

5

7 8

10

12

13

14 +5

D1

D0

DPL

LE

GND

DPR

BI

D2

D3

+5

A B C D

7489 16 x 4 RAM

12

10

6

4

3

2

1 15 14 13

GND: pin 8

+5: pin 16

11

9

7

5

WE

ME

DI

1

DO

3

DO

4

DO

1

DO

2

DI

2

DI

3

DI

4

D15

D14

D13

D12

D11

D10

D9

D8

D7

D6

D5

D4

D3

D2

D1

D0

A B C D

W

Enable

74150

Multiplexer

16

17

18

19

20

21

22

23

1

2

3

4

5

6

7

8

15 14 13 11

9

10

GND: pin12

+5: pin24

5

.

. 2

.

.

7490 Bi-Quinary Counter

+5: pin 5

GND: pin 10

2

3

6

7

14

12

8 9 11

1

R

0

R

9

Q

D

Q

C

Q

B

Q

A

R

9

R

0

Q

Trigger

A

1

B

A

2

6

11 10

1

5

4

3

74121

Monostable

C

ext

R

ext

Q

1

2

3

4 11

12

13

14

GND

5

6

7 8

9

10

1

2

3

4 11

12

13

14

GND

5

6

7 8

9

10

+5

7486 Quad XOR

+5 +5

2

K

14

11

13

10

9

7

J

Q

CLK

Q

R

S

12

K

5

CLK

4

Enable

U/D

14

Q

A

Q

B

Q

C

Q

D

D

A

D

B

D

C

D

D

RC

Max/Min

Load

2 3 6 7

12

13

11 9 10 1 15

74191 Four-Bit Counter

8

.

. 2

.

.

7493 Binary Counter

+5: pin 5

GND: pin 10

2

3

14

12

8 9 11

1

R

0

Q

D

Q

C

Q

B

Q

A

R

0

7432 Quad OR 7408 Quad AND

Control

+5: pin 16

GND pin 8 :

+5: pin 14

GND pin 7 :

194

195 Appendix D. Pinouts

Glossary of basic electrical and

electronic terms

ampere Basic unit of current: 1 ampere = 1 A = 1 coulomb/second.

angular frequency Rate of change of phase. Measured in radians per second: ω = 2π f .

anode The negative terminal

attenuation Decrease in voltage or current (also implies power reduction).

capacitor Device used to store charge and energy. The capacity of a capacitor is called the

capacitance. Capacitance C, charge Q, and voltage Vare related by the equation Q = CV.

cathode The positive terminal

cathode-ray tube A large vacuum tube in which the electron beam can be steered to create

a visible pattern on a phosphorescent screen.

charge A fundamental property of some elementary particles. Electrons have charge −1e,

and protons and holes have charge +1e, where e = 1.602 ×10

−19

coulombs.

common Voltage reference point (0 V). Also called ground.

compliance (usually of a current source): Range over which circuit performance is stable.

coulomb Unit of charge: 1 coulomb = 6.241 ×10

18

e, where e is the charge of the electron.

current Rate at which charge ﬂows. Deﬁned as the amount of charge that passes through a

given surface (such as the cross-section of a wire) per unit time. A convenient analogy

is the rate at which water ﬂows under a bridge or through a pipe. Positive current ﬂows

from points of higher voltage to points of lower voltage. (Due to Benjamin Franklin’s

choice for the deﬁnition of positive charge, this is opposite to the ﬂow of electrons.)

daraf Unit of inverse capacitance.

decibel Unit for specifying a voltage or power ratio on a logarithmic scale.

dynamic resistance Effective resistance of a nonlinear element (typically a PN junction),

such as a diode or transistor junction.

farad Unit of capacitance: 1 farad = 1 F = 1 coulomb/volt. The farad is a large unit; com-

monly available capacitors range in size from a few picofarads to many thousands of

microfarads.

feedback A design approach or situation in which an electronic signal communicates infor-

mation from the output of an electronic device or circuit to its input. Positive feedback

enhances a change at the output (i.e., a growing output with positive feedback grows even

larger), while negative feedback counteracts a change at the output.

frequency domain AC circuit analysis approach that focuses on circuit response to sine

waves vs. their frequency.

197

198 Hands-on electronics

gain Increase in voltage or current (also implies power ampliﬁcation).

gate A circuit that performs digital logic, such as an AND gate or a NOR gate.

ground Voltage reference point (0 V). Also called common.

henry Unit of inductance.

hertz Unit of frequency: 1 hertz = 1Hz = 1 cycle/second.

impedance Degree to which a circuit element impedes the ﬂow of current; includes both

resistive and reactive components. In the standard electrical-engineering notation, resis-

tance is a real quantity and reactance is imaginary, corresponding to their ±90

◦

phase

difference, thus impedance is given by

Z =

R +i

X.

jack Connector used to accept a plug; socket.

Kirchhoff’s current law The net current ﬂowing into or out of any point in a circuit is zero.

Kirchhoff’s voltage law The total voltage around any closed loop is zero.

mho Unit of transconductance; inverse of an ohm.

ohm Unit of resistance. 1 ohm = 1 = 1 volt/ampere.

plug Connector that plugs into a socket or jack.

quiescent Default voltage and/or current values when an input signal is absent.

reactance Capacitive (X

C

) and/or inductive (X

L

) component of a circuit element’s

impedance.

resistance Degree towhicha device impedes the ﬂowof DCcurrent; nonreactive component

of impedance. Measured in ohms. (For a nonreactive device, also the degree to which the

device impedes the ﬂow of AC current, i.e., for a resistor, Z = R.)

slew rate Rate at which an output voltage changes.

socket Connector used to accept a plug; jack.

The´venin equivalent A way to model complex circuits based on Th´ evenin’s theorem, which

reduces most circuits to a single ideal voltage source in series with a single impedance.

time domain AC circuit analysis approach that focuses on circuit response to an arbitrary

waveform vs. time.

volt Unit of electrostatic potential: 1 volt = 1 V = 1 joule/coulomb.

voltage Electrostatic potential. Voltage is deﬁned and measured with respect to a common

reference or ground point. When multiplied by the value of the charge, voltage gives the

potential energy of the charge with respect to that reference. Positive current ﬂows from

points of higher voltage to points of lower voltage (from larger potential energy to lower

potential energy).

V

CC

Most positive voltage in a circuit.

V

EE

Most negative voltage in a circuit.

Index

β, 48–50

h

FE

, 48–50

Q, 124

r

BE

, 53, 57

r

e

, 52, 57

V

BE

, 50, 52

V

CB

, 50

V

CE

, 51

−3 dB point, 26

2N3904 pinout, 54

2N3906 pinout, 54

2N5485 pinout, 69

311 comparator, 114

311 pinout, 114

555 timer, 118, 156

7400 IC series, 125

741 op amp, 85

741 pinout, 86

74121 monostable, 156, 159

74121 pinout, 159

74138 decoder, 178

74150 mux, 162

74150 pinout, 162

74191 counter, 168

7489 RAM, 163

7489 pinout, 163

AC coupling, 43

acceptor, 32

active bandpass ﬁlter, 123

active differentiator, 102, 106

active ﬁlter, 123

active integrator, 103, 107

active rectiﬁer, 108

ADC, 167

successive-approximation,

171

tracking, 170

addition, binary, 141

algebra, Boolean, 126, 140, 141

alternating current, 15

ammeter, ideal, 39

ampliﬁer, 50

common-emitter, 57

difference, 95

differential, 86

exponential, 105

grounded-emitter, 59

inverting, 168

op amp, 88

logarithmic, 105

noninverting, 89

op amp, 89

operational, 79, 85

amplitude, 13, 17, 18

analog, 167

analog information, 167

analog-to-digital conversion, 167

analog-to-digital converter, 167

anode, 35, 54

arithmetic, binary, 125, 126, 141

assertion-level logic, 127, 146

assertion-level logic notation, 146

astable multivibrator, 120, 156

asynchronous counter, 151, 157

attenuating probe, 10

attenuation, 10, 26, 77

attenuator, 76, 77, 91

bandpass ﬁlter, 123

bandwidth, 87

base, 48

BCD counter, 157

bi-quinary counter, 157

bias current, 94

binary addition, 141

binary arithmetic, 125, 126, 141

binary counter, 156, 157

binary search algorithm, 171

binary-coded decimal, 157

bipolar junction transistor, 47

bistable multivibrator, 143, 156

blocking capacitor, 56

Boolean algebra, 126, 140, 141

bounce, contact, 152

breadboard, 2

breadboard LED indicators, 137

breadboard level switches, 137, 138

breakpoint, 26

buffer, 55

push–pull, 62

199

200 Index

buffering, 55

capacitance code, 20

capacitance

parasitic, 56

stray, 115

capacitive reactance, 19

capacitor, 15, 16, 19, 20

blocking, 56

ceramic, 16

electrolytic, 16

mica, 16

paper, 16

polarized, 16

types of, 16

carry-in, 156

carry-out, 156, 157

cascaded counters, 156

cathode, 35, 54

clipping, 77

closed-loop, 89

CMOS, 125

CMOS ICs, powering, 136

CMOS logic, 133

CMOS TTL, 133

CMRR, 78, 96, 97

collector, 48

common-emitter ampliﬁer, 57

common-mode, 78

common-mode gain, 78

common-mode rejection, 78

common-mode rejection ratio, 78, 96

comparator, 113

311, 114

voltage, 113

magnitude, 142

compliance, 59, 71

contact bounce, 152

conversion

analog-to-digital, 167

digital-to-analog, 167

converter

analog-to-digital, 167

digital-to-analog, 167

counter, 151, 152, 156

74191, 168

asynchronous, 151, 157

BCD, 157

bi-quinary, 157

binary, 156, 157

decade, 157

cascaded, 156

decimal, 157, 158

four-bit, 157

negative-edge-triggered, 157

ripple, 151, 157

synchronous, 152, 157

two-bit, 151, 156

crossover distortion, 63, 109

CRT, 9

current mirror, 79

current source

FET, 70

op amp, 97

transistor, 59

current-source load, FET, 72

CY62256 memory, 177

D-type ﬂip-ﬂop, 147

DAC, 167

DAC0806, 174

DAC080x, 178

family, 174

Darlington, 61

data selector, 162

DC coupling, 79, 85

DC offset, 87

debouncer, switch, 153

debugging, 144

debugging digital logic, 144

decade counter, 157

decibel, 26, 83

decimal, binary-coded, 157

decimal counter, 157, 158

decoder, 178

74138, 178

delay, propagation, 148

DeMorgan’s theorem, 141, 146

diagram

state, 143

timing, 143

dielectric, 16

dielectric constant, 16

difference ampliﬁer, 95

differential ampliﬁer, 75, 76, 86

differential gain, 76, 77

differential signal, 75

differentiator, 15, 27, 75, 102

active, 102, 106

op amp, 102, 106

digital, 167

digital information, 167

digital logic, 125

debugging, 144

digital meter, 1

digital recording, 177

digital-to-analog conversion,

167

digital-to-analog converter, 167

diode, 31

gate-channel, 66

light-emitting, 60

zener, 123

diode characteristic, 31, 33, 34

diode constant, n, 106

diode drop, 37

diode logic, 131

diode test, multimeter, 54

diode-bridge rectiﬁer, 43

DIP IC package, pin numbers, 129

201 Index

display

logic-level, 137

TIL311, 158

distortion, 58

crossover, 63, 109

divide-by-four circuit, 151

divide-by-ten circuit, 158

divide-by-two circuit, 148

DMM, 1

donor, 32

driver, push–pull, 109

DVM, 1

dynamic resistance, 37, 52, 57

of diode, 37

of emitter, 52

FET source, 72

Ebers–Moll transistor model, 52

ECL, 125

electrolytic, 16

electrolytic capacitor, 16

emitter, 48

emitter follower, 55

emitter resistance, 52

equality tester, 141

exclusive-OR gate, 141

exponential ampliﬁer, 105

factor, quality (Q), 124

false, 126

feedback, 79

negative, 71, 75, 79, 88, 117

positive, 62, 113, 117

FET, 65

FET current source, 70

FET saturation, 67

ﬁlter, 123

active, 123

bandpass, 123

high-pass, 15, 28

low-pass, 15, 25, 28

ﬁnite-state machine, 143, 162

ﬂip-ﬂop, 143, 156

D-type, 147

JK, 148

toggling, 148

follower, voltage, 94

forward-bias, 34

four-bit counter, 157

frequency domain, 15, 101

function generator, 2, 13

gain, common-mode, 78

gain, differential, 77

gate current, 66

gate

exclusive-OR, 141

NAND, 140

OR, 142

XOR, 141

gate-channel diode, 66

golden rules, op amp, 90

ground, virtual, 93, 168

ground clip, 10

grounded-emitter ampliﬁer, 59

half-power frequency, 26

hexadecimal, 158, 164

hexadecimal display, TIL311, 158

high (logic level), 126

hysteresis, 116, 170

IC

digital, 85

linear, 85

ideal ammeter, 39

ideal op amp, 87

ideal rectiﬁer, 36

ideal voltmeter, 39

impedance

input, 45

output, 45

measuring, 46

indicators, LED, breadboard, 137

inductance, 19

inductive reactance, 19

inductor, 19

information

analog, 167

digital, 167

input bias current, 94

input impedance, 45

input offset voltage, 78, 91

integrated circuit

digital, 85

linear, 85

integrator, 15, 24, 103

active, 103, 107

op amp, 103, 107

internal state, 143, 146, 149, 162

inverter, 60, 141, 149

inverting ampliﬁer, 58, 88, 168

op amp, 88, 168

JFET, 65

JK ﬂip-ﬂop, 148

junction diode, 32

junction, summing, 93, 169

latch, RS, 145, 153

LED, 60

LED indicators, breadboard, 137

level switches, breadboard, 137,

138

LF398 SHA, 177

light-emitting diode, see LED

linear region, FET, 68

logarithmic ampliﬁer, 105

logarithmic search algorithm, 171

202 Index

logic

assertion-level, 127, 146

digital, 125

diode, 131

multiplexer, 162

negative, 127

positive, 127

sequential, 143

synchronous, 144

logic function, universal, 140

logic levels, 125, 126

TTL, 126

logic-level displays, 137

low (logic level), 126

machine, state, 143, 162

magnitude comparator, 142

margin, noise, 126

memory

random-access, 162

word-addressable, 163

meter, digital, 1

mho, 48

MKS, 16

momentary-contact switch, 153

monostable, 156, 159

monostable, 74121, 156, 159

monostable multivibrator, 156, 159

MOSFET, 65

MOSFET logic, 133

multimeter, 1

multimeter diode test, 54

multiplexer, 162

multiplexer logic, 162

multivibrator, 120, 143, 156, 159

astable, 120, 156

bistable, 143, 156

monostable, 156, 159

mux, 162

NAND, 140

NAND gate, 140

negative feedback, 75, 79, 88

negative logic, 127

negative-edge-triggered counter, 157

negative-edge triggering, 147, 149, 157

noise margin, 126

noninverting ampliﬁer, op amp, 89

NPN transistor, 48

offset voltage, 78

Ohm’s law, 6

one (logic level), 126

one-shot, 156, 159

op amp, 79, 85

differentiator, 102

golden rules, 90

ideal, 87

integrator, 103

inverting ampliﬁer, 88

noninverting ampliﬁer, 89

rectiﬁer, 108

signal-processing, 101

op-amp inverting ampliﬁer, 168

open-collector output, 164

open-loop, 88, 91, 113

operational ampliﬁer, see op amp

OR, 142

OR gate, 142

oscillation, 115

parasitic, 56

oscillator, square-wave, 117

relaxation, 117

sine–cosine, 122

oscilloscope, 8–10

cursors, 14

measurement, 13

triggering, 12

output impedance, 45

measurement of, 46

output, 3-state, see output, three-state

output, open-collector, 164

output, three-state, 164

output, totem-pole, 133, 164

output, tri-state, see output, three-state

parasitic capacitance, 56

parasitic oscillation, 56

passband, 124

peak-to-peak voltage, 17

permittivity, 16

phase shift, 25

pickup, 115

pin numbers, DIP IC package,

129

pinch-off, 66

pinch-off voltage, FET, 68

pinout

74121, 159

74150, 162

7489, 163

PN junction, 32, 47

PNP transistor, 48

positive feedback, 113

positive logic, 127

positive-edge triggering, 147

potentiometer, 6, 22, 23

power supplies, 40

powering CMOS ICs, 136

powering TTL ICs, 136

probe, 10

attenuating, 10

oscilloscope, 10

probe compensation adjustment,

10

propagation delay, 148

measurement of, 148

pull-up resistor, 114, 164

push–pull buffer, 62

push–pull driver, 109

203 Index

quality factor (Q), 124

quiescent voltage, 57

RAM, 162

7489, 163

random-access memory, see RAM

RC circuit, 15

RC timing network, 159

reactance, 19

capacitive, 19

inductive, 19

recording, digital, 177

rectiﬁcation, 31, 36

rectiﬁer, 36, 40

active, 108

full-wave, 43

half-wave, 40

ideal, 36

op amp, 108

reference lead, 10

register, shift, 181

regulator, 40

rejection ratio, common-mode, 96

relaxation oscillator, 117

resistance

dynamic, 37, 52, 57

static, 37

resistor

pull-up, 114, 164

shunt, 104

reverse saturation current, 33, 52

reverse-bias, 34

ripple counter, 151, 157

ripple voltage, 43

RS latch, 145, 153

sample-and-hold, 177

saturated switch, 60

saturation

bipolar-transistor, 59, 60

transistor, 68

saturation current, 33, 52

saturation drain current, FET, 68

saturation region, FET, 67

saturation voltage, 91

Schmitt trigger, 116

search algorithm

binary, 171

logarithmic, 171

sequential logic, 143

series, 5, 6, 8

SHA, 177

shift register, 181

short circuit, 7, 11

shunt resistor, 104

signal processing, op amp, 101

simple transistor model, 51

sine–cosine oscillator, 122

slew rate, 87

source follower, 71

SPDT switch, 152

speed, transition, 126

state, internal, 143, 146, 149, 162

state diagram, 143

state machine, 143, 162

state table, 146

static resistance, 37

stored charge, 61

stray capacitance, 115

successive-approximation ADC, 171

summing junction, 93, 169

switch, 152

momentary-contact, 153

SPDT, 152

level, breadboard, 137, 138

switch debouncer, 153

synchronous counter, 152, 157

synchronous logic, 144

table, state, 146

table, truth, 140, 141, 143, 146

tester, equality, 141

theorem, DeMorgan’s, 141, 146

Th´ evenin equivalent circuit, 45

three-state output, 164

three-terminal voltage regulators, 45

threshold voltage, 113

TIL311 display, 158

time constant, 24

time domain, 15, 102

timer, 118, 156, 555

timing diagram, 143

timing network, RC, 159

TO-92 case, 54

toggling ﬂip-ﬂop, 148

totem-pole output, 133, 164

tracking ADC, 170

transconductance ampliﬁer, 48

transconductance

FET, 68

transistor, 68

g

m

, 48

transformer, 40

transistor, 48

transistor, ﬁeld-effect, 65

transistor, junction, 47

transistor, simple model, 51

transistor action, 48

transistor current source, 59

transistor model, Ebers–Moll, 52

transistor saturation, 68

transition speed, 126

tri-state, 149, 150

tri-state output, see output, three-state

trigger, Schmitt, 116

triggering

negative-edge, 147, 149, 157

positive-edge, 147

true (logic level), 126

truth table, 140, 141, 143, 146

204 Index

TTL, 125

TTL families, 128, 133

TTL history, 128, 133

TTL ICs, powering, 136

TTL logic levels, 126

two-bit counter, 151, 156

universal logic function, 140

virtual ground, 93, 168

voltage

peak-to-peak, 17

quiescent, 57

threshold, 113

voltage comparator, 113

voltage-divider, 15, 22, 23, 26

voltage droop, 43

voltage follower, 94

voltage regulation, 44

voltmeter, ideal, 39

word addressing, 163

XOR, 141

XOR gate, 141

Zener diode, 35, 123

zero, 126

This page intentionally left blank

Hands-On Electronics

Packed full of real circuits to build and test, Hands-On Electronics is a unique introduction to analog and digital electronics theory and practice. Ideal both as a college textbook and for self-study, the friendly style, clear illustrations and construction details included in the book encourage rapid and effective learning of analog and digital circuit design theory. All the major topics for a typical one-semester course are covered, including RC circuits, diodes, transistors, op amps, oscillators, digital logic, counters, D/A converters and more. There are also chapters explaining how to use the equipment needed for the examples (oscilloscope, multimeter and breadboard), together with pinout diagrams for all the key components referred to in the book.

Kaplan and Christopher G.Hands-On Electronics A One-Semester Course for Class Instruction or Self-Study Daniel M. White Illinois Institute of Technology .

no reproduction of any part may take place without the written permission of Cambridge University Press. or will remain.cambridge. Singapore.org Information on this title: www. New York.org/9780521815369 © Cambridge University Press 2003 This book is in copyright. Madrid. Melbourne. . United Kingdom Published in the United States of America by Cambridge University Press. First published in print format 2003 - - - - - - ---- eBook (EBL) --- eBook (EBL) ---- hardback --- hardback ---- paperback --- paperback Cambridge University Press has no responsibility for the persistence or accuracy of s for external or third-party internet websites referred to in this book. Subject to statutory exception and to the provision of relevant collective licensing agreements. São Paulo Cambridge University Press The Edinburgh Building. and does not guarantee that any content on such websites is. accurate or appropriate. New York www. Cambridge.cambridge. Cape Town. Cambridge .

5 Triggering 1.1 Multimeter 1.1.3 Measuring resistance 1. resistance and Ohm’s law 1.2 Types and values of capacitors v .6 Additional features 1 1 2 4 5 8 8 10 11 11 12 12 13 15 15 17 19 2 RC circuits 2.Contents List of ﬁgures List of tables About the authors To the Reader Acknowledgments Introduction page xi xv xvi xvii xviii xix 1 Equipment familiarization: multimeter.2. review of AC circuits 2.3 Oscilloscope 1.2 Display 1.2.1 Use of capacitors.4 Horizontal sweep 1. and oscilloscope 1.3.1 Measuring voltage 1.3.3.3.1.2 Measuring current. breadboard.1 Probes and probe test 1.3 Vertical controls 1.2 Breadboard 1.3.3.2.1 Review of capacitors 2.

1 3.1.5 RC circuit as integrator 2.1 Checking transistors with a meter 4.2.1 DC voltage divider 2.6 Low-pass ﬁlter 2.3.2.4 RC circuit 2.7 Semiconductor basics Types of diodes Rectiﬁcation Diode action – a more sophisticated view Measuring the diode characteristic Exploring rectiﬁcation Input and output impedance 31 31 35 36 37 38 40 45 4 Bipolar transistors 4.2 Emitter follower 4.1.3 Common-emitter ampliﬁer 4.2.7 RC circuit as differentiator 2.3 Ebers–Moll transistor model 4.2 Review of current.2.5 3.2.3 Potentiometer as voltage divider 2.vi Contents 2.1 Destructive demonstration of resistor power rating 2.3.4 3.5 Transistor switch 4. voltage.4 Collector as current source 4.1 Basic deﬁnitions 4.9 Summary of high.2 AC voltage divider 2.3 3.3 Additional exercises 4.2 Experiments 4.3.1 Darlington connection 47 47 50 51 52 54 54 55 57 59 60 61 61 . and power 2.2.8 High-pass ﬁlter 2.1 Bipolar-junction-transistor basics 4.1.2 3.2 Simplest way to analyze transistor circuits 4.6 3.and low-pass ﬁlters 20 21 22 23 23 24 24 25 27 28 28 3 Diodes 3.

1 The 741 operational ampliﬁer 7.1 Differential ampliﬁer 6.4 Input offset voltage 6.5 Common-mode gain 6.3 Gain of inverting and noninverting ampliﬁers 7.3 Improved current mirror 6.1 741 pinout and power connections 7.3 Source follower 5.1.2 An ideal op amp 7.2.3.2.1.2.2.3 Measuring the differential gain 6.4 Wilson current mirror 75 75 76 76 77 78 78 79 79 80 82 82 85 85 86 87 88 90 90 7 Introduction to operational amplifiers 7.1.1.1 FET characteristics 5.5 The nonideal op amp .2 Exercises 5.1 Field-effect transistors 5.1 Operating principle 6.2.4 JFET ampliﬁer 65 65 66 68 69 69 70 71 73 6 Transistors III: differential amplifier 6.1.2 Modeling FET action 5.2 FET current source 5.2 Op amps and their building blocks 6.1 FET characteristics 5.2 Push–pull driver 4.vii Contents 4.1.1.3 Common-base ampliﬁer 62 63 5 Transistors II: FETs 5.2.1.1.1.2 Differential ampliﬁer with current-source loads 6.1.4 Op amp ‘golden rules’ 7.2 Expected differential gain 6.2.1.1 Current mirror 6.2.3.

2.3 Intentional positive feedback: Schmitt trigger 9.1.5 Difference ampliﬁer 7.2.1 Alarm! 9.1 Testing open-loop gain 7.3.1.1 Op amp signal processing 8.3 Additional experiments 7.2 Inverting ampliﬁer 7.2.2 Noninverting summing amp with difference ampliﬁer 91 91 92 93 94 95 97 97 98 8 More op amp applications 8.1.2.2 Experiments 7.2.2 Unintentional feedback: oscillation 9.3 Noninverting ampliﬁer 7.2 Sine/cosine oscillator 9.4 RC relaxation oscillator 9.2 Integrator 8.2.4 Voltage follower 7.1 Differentiator 8.1 Current source 7.1.5 555 timer IC 9.2.3.2 Logarithmic and exponential ampliﬁers 8.1.2 Additional experiments 9.2.1.2.3 Logarithmic and exponential ampliﬁers 8.2.1 Op amp as comparator 9.1.1 Differential and integral ampliﬁers 8.1 Experiments 9.4 Op amp with push–pull power driver 8.2.2 Experiments 8.viii Contents 7.3 Active bandpass ﬁlter 113 113 113 115 116 117 118 121 121 122 123 .1.3 Additional exercises 101 101 102 103 105 106 106 108 108 109 111 9 Comparators and oscillators 9.3 Op amp active rectiﬁer 8.2.

5 TTL quad XOR gate 10.2 CMOS and TTL compared 10.1 Simple RS latch 11.3.4 Tri-state outputs 143 144 144 144 144 144 145 145 147 148 149 .3 Complementary MOSFET logic (CMOS) 10.1 7485 4-bit magnitude comparator 125 125 126 127 129 130 131 131 132 133 136 137 137 138 140 140 141 142 142 11 Flip-flops: saving a logic state 11.4 Powering TTL and TTL-compatible integrated circuits 10.3 JK ﬂip-ﬂop 11.4 Summary of Boolean algebra 10.1 Logic levels 10.4 Timing diagrams 11.3.1 Diode logic 10.2.1.2.2 Logic families and history 10.3 Experiments 10.1 Digital logic basics 10.1.2.2 Breadboard layout 11.4 Additional exercises 10.ix Contents 10 Combinational logic 10.2 Transistor–transistor logic (TTL) 10.2.4.1.2.4 Using NANDs to implement other logic functions 10.1.1.1.1.1 General comments 11.3.1 Schematics 11.3 Synchronous logic 11.1.1 LED logic indicators and level switches 10.3 Logic gates 10.2.2 Flip-ﬂop basics 11.2 MOSFETs 10.3.2 D-type ﬂip-ﬂop 11.3 CMOS NAND gate 10.3.

1 A simple D/A converter fabricated from familiar chips 13.4 Additional exercises 13.4. counters.4 RAM 155 156 156 157 157 159 162 162 13 Digital↔analog conversion 13.3 Electronic coin toss 151 151 152 153 12 Monostables. multiplexers.3 Multiplexer and ﬁnite-state machine 12. and RAM 12.5 Flip-ﬂop applications 11.3.2 Counters 12.4.1 Digital recording 13.3.3.5.3 Experiments 12.3.1 Divide-by-four from JK ﬂip-ﬂops 11.2 Tracking ADC 13.x Contents 11.5.1 Successive-approximation ADC 13.2 Contact bounce 11.2 Successive-approximation ADC built from components Further reading Appendix A Equipment and supplies Appendix B Common abbreviations and circuit symbols Appendix C RC circuits: frequency-domain analysis Appendix D Pinouts Glossary of basic electrical and electronic terms Index 167 168 170 171 171 177 177 178 183 185 188 191 194 197 199 .2 Monostable multivibrator 12.1 Multivibrators 12.1 Bi-quinary ripple counter 12.3 080x ADC and DAC chips 13.3.5.

High-pass ﬁlter or voltage differentiator.s. Three schematics representing a resistive voltage divider. An example of how to insert a diode bridge into a breadboard.2 3.4 2.1 2.9 3. Full-wave rectiﬁcation using diode bridge. Representation of physical diodes and symbols used in circuit diagrams.8 3. Construction and circuit symbols and biasing examples for NPN and PNP junction transistors.1 xi Illustration showing many of the basic features of the PB-503 powered Protoboard.3 2. Power transformer with half-wave rectiﬁcation.and low-pass RC ﬁlters. Representation of a junction between P-type and N-type semiconductor material. Measuring the forward characteristic of a diode. periodic waveform. Circuit demonstrating destructive power loading.5 2.7 3.6 3.2 1. Power transformer supplies Vout ≈ 25 V r. Representation of an arbitrary. Half-wave rectiﬁer with ﬁlter capacitor. The voltage-divider concept for RC circuits.10 3.1 1. Relationships among input voltages and capacitor and resistor voltages for high.4 2.6 3. Illustration of the Tektronix TDS 210 digital oscilloscope.Figures 1. Typical current–voltage characteristics for germanium and silicon diodes.3 1.1 3. Measuring current.5 3.3 3.12 4.2 2. Diode circuit symbol and biasing. Measuring voltage.4 3.m. Full-wave rectiﬁcation with ﬁlter capacitor. Complete rectiﬁer circuit. page 3 5 5 9 18 21 22 24 27 29 33 33 34 35 39 41 42 42 43 44 45 46 48 .11 3.

Driving loudspeaker with push–pull buffer.10 4.12 5. JFET ampliﬁer.6 4.11 4.4 6.3 7.2 6.4 4. Common-base ampliﬁer. Difference ampliﬁer.1 7.5 7.7 4.2 4. Circuit for demonstrating summing junction. Diagram of 8-pin DIP 741 package showing ‘pinout’. Differential ampliﬁer and function generator with 100-to-1 attenuator. Schematic representation of JFET operation. Op amp voltage follower and voltage follower as the input stage to an inverting-op-amp circuit.xii List of figures 4. Darlington pair.7 7. Differential ampliﬁer with current-mirror load. Emitter follower. Source follower with current-source load. Transistor current source.4 5.9 4. Common-emitter ampliﬁer.1 Schematic representation of how an NPN transistor operates.5 5.3 5.2 5. Characteristic curves for an NPN bipolar transistor.1 6.2 7. Transistor as back-to-back diodes.7 5. Fancy summing circuit. Self-biasing JFET current source. TO-92 pinout. Circuit for measuring the common-source characteristic curves.5 4.5 7.3 4.6 5. Source follower. Current mirror. Op amp current source. Construction and circuit symbols of JFETs. Open-loop op amp test circuit. Transistor switch. 49 51 55 55 56 57 59 60 62 63 64 66 67 67 70 71 72 73 73 76 79 80 81 82 86 88 89 91 93 95 96 98 99 102 .6 7. Emitter follower with optional load circuit for measurement of Z out . Current sink for differential ampliﬁer.1 5.4 7. Differential ampliﬁer with Wilson-current-mirror load. Idealized common-source characteristic curves for a JFET.9 8. Op amp inverting-ampliﬁer circuit.8 7. Generalized op amp inverting-ampliﬁer circuit.8 4. Op amp noninverting-ampliﬁer circuit.8 6.3 6.

5 9. Standard logic gates with truth tables.7 9.2 9. Schematic representations of a CMOS inverter constructed using one N-channel and one P-channel MOSFET.5 8. Sine/cosine oscillator.9 10. Simple and improved versions of an op amp half-wave rectiﬁer. 555 timer conﬁgured as an alarm.10 10.1 10. Block diagram for the 555 timer IC. Circuits for measuring the channel resistance as a function of gate voltage. Schmitt trigger using 311 comparator. Basic op amp integrator. Timing diagram with timing deﬁnitions for a rising-edge-triggered ﬂip-ﬂop. RC relaxation oscillator using comparator. 311 comparator with 10 k series input resistor. Block diagram showing how to build an ‘exponentiator’. Poor comparator and 311 comparator. Two-input diode gate. Improved op amp differentiator. Op amp exponential ampliﬁer.9 10.3 8. Op amp logarithmic ampliﬁer.xiii List of figures 8. 555 timer IC used as an oscillator and as a one-shot or timer.11 11. Active bandpass ﬁlter.1 Basic op amp differentiator.3 9.4 8.8 9. Op amp follower with push–pull output-buffer power driver with two feedback arrangements.6 9.4 9.7 8. Logic-level switch using either an SPST or SPDT switch and a pull-up resistor. De Morgan’s theorems expressed symbolically.7 10.1 9. Improved op amp integrator. Logic levels for various 7400-family lines.5 10.6 10. Schematic representation of an ‘enhancement-mode’ N-channel MOSFET.3 10.8 8. Diode–transistor NAND gate using 2N3904s.4 10.9 8.8 10.2 8. Labeling of 7400-series chips. 102 103 104 104 105 105 109 110 111 114 115 116 118 119 120 121 122 123 127 129 130 131 132 132 134 135 136 137 138 145 .2 10.10 9.6 8. Schematic representation of a CMOS NAND gate with LED logic-level indicator.

2 12. Simple D/A converter and output waveform resulting from input counting sequence.9 12.4 11. 8-bit successive-approximation ADC.5 12. Pinout of the 74112 JK ﬂip-ﬂop.7 13. Simple A/D converter. Synchronous divide-by-four counter.17. 146 147 147 149 150 151 152 153 157 158 160 160 161 163 163 168 171 172 175 176 179 180 193 193 . Control logic for 8-bit successive-approximation ADC.3 11.7 C. Right triangle to illustrate Eq.2 11. Sample timing diagram for a (positive-edge-triggered) 7474 D-type ﬂip-ﬂop.3 13. Pinout and power connections for the 74373 and input and output connections for testing the tri-state output. Looking at contact bounce by driving a divide-by-four counter from a switch. Pinout of 7489 16×4 RAM.5 11. 7474 D-type ﬂip-ﬂop with state table.1 12.1 13.xiv List of figures 11. Pinout for ADC080x series of A/D converters and the on-chip self-clocking conﬁguration.3 12. Pinout of 7490 decade counter. Method for producing a DC-shifted waveform.2 13. Pinout for DAC080x series of D/A chips.8 11.1 C. Series RC circuit. Pinout of 74150 16-to-1 multiplexer. C. Pinout of ’121 and ’123 one-shots with external RC timing network. Substandard outputs resulting from gating clock signals.5 13. Timing diagram for a gated clock signal.4 13.2 Simple RS latch made of two-input NANDs with state table.6 12.7 11.6 13. Divide-by-four ripple counter.6 11. Pinout of TIL311 hex display.4 12.

1 3. A sample of commercially available diodes. Color code for nonprecision resistors.2 2.1 1. page 2 7 16 36 50 128 xv . Some typical dielectric materials used in capacitors.1 Digital multimeter inputs. Common families within the 7400 series. A sample of commercially available bipolar transistors.1 4.Tables 1.1 10.

He also serves as Principal Investigator of the Illinois Consortium for Accelerator Research. and co-editor of three books on heavy-quark physics and related ﬁelds. He has authored or co-authored over 100 scientiﬁc articles in the ﬁeld of high-energy particle physics. in Physics from the University of Minnesota in 1990. Kaplan received his Ph. Dr Christopher G. He received his Ph. He has taught electronics laboratory courses for non-electrical-engineering majors over a ﬁfteen-year period at Northern Illinois University and at Illinois Institute of Technology. He is the author or co-author of over 150 scientiﬁc papers and one encyclopedia article. His thesis experiment discovered the b quark. Over more than twenty-ﬁve years in experimental particle physics he has often been responsible for much of his experiments’ custom-built electronic equipment. He has been interested in electronics since high school.About the authors Dr Daniel M. White is Assistant Professor of Physics at Illinois Institute of Technology. during the junior year of which he designed a computer based on DTL integrated circuits. where he is currently Professor of Physics and Director of the Center for Accelerator and Particle Physics. and he has devoted much of his career to experimentation at the Fermi National Accelerator Laboratory on properties of particles containing heavy quarks. xvi . Dr White is an enthusiastic and dedicated teacher who enjoys helping students to overcome their fear of electronics and to gain both conﬁdence and competence.D. and his current research interests involve neutrinos and hyperons.D. in Physics in 1979 from the State University of New York at Stony Brook.

To that end. our goal is that by the end of the book. or to interface two pieces of equipment that may not have been designed for that purpose. We expect few of you to have much familiarity with such physical theories as electromagnetism or quantum mechanics. you were ever bitten by the ‘ham radio’ bug. Others may have ‘played around’ with such stuff if. testing. so much the better. but unfamiliarity with physical theory should not prevent you from building or using electronic circuits and instruments. you will be able to design and build any little analog or digital circuit you may ﬁnd useful. for example. We use electronic equipment all the time in our work and recreation. this sequence of laboratory experiments has been designed to introduce you to the fundamentals of modern analog and digital electronics. xvii . so the thrust of this course will be from phenomena and instruments toward theory. or process-control settings. A basic knowledge of electronics will also help you to understand and appreciate the quirks and limitations of instruments you will be using in research. If your curiosity is aroused concerning theoretical explanations. or at least understand it well enough to have an intelligent conversation about the problem with an electrical engineer. Scientists and engineers need to know a bit of electronics. not the other way round. for example to modify or repair some piece of equipment. development.To the Reader Some of you may be encountering electronic circuits and instruments for the ﬁrst time. In either case.

we thank our wives and children for their support and patience.Acknowledgments We are grateful to Profs Carlo Segre and Tim Morrison for their contributions and assistance. It is to them that we dedicate this book. Finally. and especially to the IIT students without whom this book would never have been possible. xviii .

a hands-on education in the techniques of electronics is much more valuable than a blackboardand-lecture approach. that to someone who will be working with electronic instrumentation. mechanical engineering. Each experiment can be completed in about four hours (with one or two additional hours of preparation). In curricula that normally include one year of laboratory instruction in electronics. The book is also suitable for self-study by a person who has access to the necessary equipment and wants a hands-on introduction to the subject. Analog Electronics for Scientiﬁc Application and Digital Electronics for Scientiﬁc Application (reissued by Waveland Press. it may be suitable for the ﬁrst part of a two-semester sequence. 1989). The appendices suggest sources for equipment and supplies. 1994). and experience at IIT has borne out. This book differs from existing books of its type in that it is faster paced and goes into a bit less depth. xix . 1989). Horowitz and Hill’s comprehensive The Art of Electronics (Cambridge University Press. We feel strongly. provide tables of abbreviations and symbols. and list recommendations for further reading. with the second part devoted to computers and computer interfacing – this scheme has the virtue of separating the text for the more rapidly changing computer material from the more stable analog and digital parts.Introduction This book started life as the laboratory manual for the course Physics 300. offered every semester at Illinois Institute of Technology to a mix consisting mostly of physics. ‘Instrumentation Laboratory’. and aeronautical engineering majors. in order to accommodate the needs of a onesemester course covering the elements of both analog and digital electronics. which includes chapter-by-chapter correspondences to some popular electronics texts written at similar or somewhat deeper levels to ours: the two slim volumes by Dennis Barnaal. Diefenderfer and Holton’s Principles of Electronic Instrumentation (Saunders. Certainly it is a better learning process than simply reading a book and working through problems.

Combinational logic. There are other texts that put integrated circuits. and direct.and alternating-current circuits. The book includes step-by-step instructions and explanations for the following experiments: 1. After the hard uphill slog. simple AC circuits – then proceeds towards greater complexity by introducing nonlinear devices (diodes). RC circuits. with the three transistor chapters.xx Introduction and Simpson’s Introductory Electronics for Scientists and Engineers (2nd edition. 7. before discrete devices. 5. Transistors III: differential ampliﬁer. 9. as well as with elementary college physics (including electricity. although these topics are reviewed in the text). 10. More op-amp applications. Introduction to operational ampliﬁers. Transistors I. Multimeter. . Ohm’s law. 4. The order we have chosen for our subject matter begins with the basics – resistors. or digital circuits. magnetism. There is also a glossary of terms and pinout diagrams for transistors and ICs used within. it’s smooth sailing from there (hold onto your seatbelts!). We have tried these approaches on occasion in our teaching and found them wanting. before the complexities of analog devices. with their simpler rules. Flip-ﬂops: saving a logic state. 8. 6. Prentice-Hall. and oscilloscope. We have chosen to discuss transistors before devices made from them (operational ampliﬁers. Diodes and power supplies. then active devices (bipolar and ﬁeld-effect transistors). digital circuitry) so that the student can understand not only how things work but also why. comparators. Comparators and oscillators. 1987). breadboard. A course based on this book thus builds to a pinnacle of intellectual challenge towards the middle. 3. Transistors II: FETs. 11. with their greater ease of use. 2. The reader is presumed to be familiar with the rudiments of differential and integral calculus. Only by considering ﬁrst the discrete devices from which integrated circuits are made can the student understand and appreciate the remarkable properties that make ICs so versatile and powerful.

obviously this is also the case for Chapter 13. multiplexers. These thirteen experiments ﬁt comfortably within a sixteen-week semester. counters. 13. Chapter 6. and RAM. ‘Transistors III’. we adhere strictly to the MKS standard. which has no subsequent experiment. Monostables. one or two experiments may easily be omitted to leave a couple of weeks at the semester’s end for independent student projects. . has been designed so that no subsequent experiment depends on it. the standard system of units for electronics is the MKS system. If you or your instructor prefers. As you work through the exercises. Finally. Digital↔analog conversion. ‘Digital↔analog conversion’. Although you may occasionally run across other unit systems. Key concepts for each exercise will be denoted by the symbol ‘ r’. To this end.xxi Introduction 12. you will ﬁnd focus questions and detailed instructions indicated by the symbol ‘ ’.

.

two alligator clips. one digital multimeter. The multimeters we use have various input jacks that accept ‘banana’ plugs. breadboards.1. They allow measurement of voltage. The input jacks are described in Table 1. 1 . red and black banana leads. Note: to obtain the highest measurement precision. Multimeters usually have a selector knob that allows you to select what is to be measured and to set the full-scale range of the display to handle inputs of various size. and resistance.1 Multimeter You are probably already familiar with multimeters. and oscilloscope In this chapter you will become acquainted with the ‘workhorses’ of electronics testing and prototyping: multimeters. 1. Depending on how you conﬁgure the meter and its leads. You will ﬁnd these to be indispensable aids both in learning about and in doing electronics. and oscilloscopes. one powered breadboard. Just as with wristwatches and clocks. it displays r the voltage difference between the two leads. in recent years digital meters (commonly abbreviated to DMM for digital multimeter or DVM for digital voltmeter) have superseded the analog meters that were used for the ﬁrst century and a half or so of electrical work. Apparatus required One dual-trace oscilloscope. two 10X attenuating scope probes.1 Equipment familiarization: multimeter. or r the resistance connected between the leads. set the knob to the lowest setting for which the input does not cause overﬂow. breadboard. r the current ﬂowing through the meter from one lead to the other. current. and you can connect the meter to the circuit under test using two banana-plug leads.

The breadboards we use represent a great step forward in convenience. etc. logic displays. two debounced push-button switches.1. The description will thus be of some use for users of other breadboard models as well. 1. making an electrical connection.. but also power supplies. a function generator. 1. Input jack COM V mA 10 A a Purpose reference point used for all measurements input for voltage or resistance measurements input for current measurements (low scale) input for current measurements (high scale) Limitsa 1000 V DC/750 V AC 200 mA 10 A For the BK Model 2703B multimeters used in the authors’ labs. Fig. many other breadboards share some. To avoid damaging the meter. any suitable breadboard will do. The exercises that follow were designed using the Global Specialties PB-503 Protoboard. if not all. be sure to read the safety warnings in its data sheet or instruction booklet. a small 8 speaker. and several on–off switches. several LED logic indicators. (For simplicity. the contacts press against it. provided you have a function generator and two variable power supplies. Additional components that you will need along the way (that are built into the PB-503) include a 1 k and a 10 k potentiometer.1 displays many of the basic features of the PB-503. Digital multimeter inputs. switches. some PB-503 features that will be used in experiments in later chapters have been omitted. The PB-503’s sockets are designed for a maximum wire thickness of 22 AWG (‘American Wire Gauge’) – anything thicker (i.) While the following description is speciﬁc to the PB-503. If you do not have access to a PB-503. with smaller .2 Breadboard ‘Breadboard’ may seem a peculiar term! Its origins go back to the days when electronics hobbyists built their circuits on wooden boards. since they include not only sockets for plugging in components and connecting them together. of these features.e.2 Hands-on electronics Table 1. The breadboard’s sockets contain spring contacts: if a bare wire is pushed into a socket.

1. Note that each vertical column is broken into halves with no built-in connection between the top and bottom. The three supplies.3 1 Equipment familiarization +15 −15 +5 V +15 V −15 V Voltage Adjustment Knobs Horizontal Row (Group of 5) Amplitude Slider Frequency Slider Function Generator Analog Digital 8Ω Speaker Push Button De-Bounced Switches Vertical Column (Group of 25) 10 k pot 1 k pot Logic Switch Bank SPDT Switches Fig. +15 V (yellow jack).1. see Fig. The PB-503 sockets are internally connected in groups of ﬁve (horizontal rows) or twenty ﬁve (vertical columns. Each power supply connects to a ‘banana’ jack and also to a row of sockets running along the top edge of the unit. AWG number) may damage the socket so that it no longer works reliably for thin wires. have a common . and −15 V (blue jack). 1. with internal connections shown for clarity.1). +5 V (red jack). Illustration showing many of the basic features of the PB-503 powered Protoboard.

To practice measuring voltages. 1 If you wonder what we mean by ‘within reason’. usually a local ground.1 Measuring voltage Voltage is always referenced to something.2). 1. Physics. Next you will connect the meter’s ‘voltage’ jack to the point of interest. In each case set the meter’s range for the highest precision (i.e. measure and record the voltage between each power supply jack and ground. it is said to ‘ﬂoat’ with respect to ground (i. For the following exercises you will measure voltage with respect to the breadboard ground. it’s always a good idea to use color coding to help keep track of which lead is connected to what. twenty million volts – if you’re interested. you will ﬁrst connect the ‘common’ jack of the meter to the breadboard common (i. see e. within reason.2. Use a black bananaplug lead to connect the ‘common’ input of the meter to the ‘ground’ jack of the breadboard (black banana jack labeled with a ‘ ’ or ‘ ’ symbol). . New York. say. breadboard ground).e. C. which is also the common ground for the three power supplies. The +15 V and −15 V supplies are actually adjustable.g. 1. The meter will then tell you the voltage with respect to ground at this one point. Warning: This is not true for most AC-powered meters and oscilloscopes.4 Hands-on electronics ‘ground’ connection (black jack). 1988).1 one may connect the DMM’s common jack to any arbitrary voltage with respect to the breadboard ground). VI–8 for more information on this. pp. When connecting things. ask yourself what bad thing would happen if you connected the DMM common to. Ohanian. 2. It is therefore possible to measure the voltage drop across any circuit element by simply connecting the DMM directly across that element (see Fig. vol. one setting above overﬂow). from less than 5 volts to greater than 15 volts.. Adjust the +15 V and −15 V supplies over their full range and record the minimum and maximum voltage for each. Carefully set the +15 V supply to a voltage half-way between its minimum and maximum for use in the next part. ‘Interlude VI’ (Norton.e.. H. To measure a voltage. using the knobs provided. Use a red banana-plug lead with the ‘V’ input of the meter. 2nd edition. esp.. Since the DMM is battery powered.

) (b) A drawing of the same circuit showing how the DMM leads should be conﬁgured to measure current. Note carefully the differences between Fig.2 Measuring current. (a) Schematic diagram of series circuit consisting of power supply. Note that the meter is connected in series with the resistor.3. . Measuring voltage.2 and Fig.2. (Note that the center tap of the potentiometer is left unconnected in this exercise – accidentally connecting it to power or ground could lead to excessive current ﬂow and burn out the pot. resistance and Ohm’s law Current is measured by connecting a current meter (an ammeter.011 A A + _ A mA COM VΩ Ground (Common) Fig. 1.3.3). or a DMM in its ‘current’ mode) in series with the circuit element through which the current ﬂows (see Fig. Measuring current.2. (a) An arbitrary circuit diagram is shown as an illustration of how to use a voltmeter. 10 k potentiometer. 1. and multimeter. Note that the meter measures the voltage drop across both the resistor and capacitor (which have identical voltage drops since they are connected in parallel). 1.5 1 Equipment familiarization (a) Multimeter V (b) Power Supply DMM + Power Supply + _ A mA COM VΩ Ground (Common) 1 µF Fig. Notice how the meter is connected in parallel with the resistor. 1. (a) Potentiometer Slider (Center Tap) (b) Power Supply DMM + Power Supply Multimeter 0. 1. (b) A drawing of the same circuit showing how the leads for a DMM should be connected when measuring voltage. 1.

The ‘10 k pot’ (as it is called for short) is located near the bottom edge of the breadboard.3) consisting of an adjustable power supply. A potentiometer is a type of resistor that has an adjustable ‘center tap’ or ‘slider’. ﬁnd a 10 k pot on your breadboard if it has one. Resistors come in various sizes according to their power rating. otherwise you will have to purchase a separate 10 k pot. If you don’t have a PB-503 breadboard. encased in an insulating coating. carbon ﬁlm.2). turn off the breadboard power to avoid burning anything out if you happen to make a mistake in hooking up the circuit. r Be careful to keep any exposed bits of metal from touching each other and making a ‘short circuit’! Note that most of the exposed metal on 2 3 Of course. but also at an adjustable point along the resistive material. 1. or wound-up wire. By pushing wires into the sockets you can make a series circuit (Fig. with wire leads sticking out the ends.3 Inside the breadboard’s case.6 Hands-on electronics Recall that Ohm’s law relates current I . the ends of the pot (as well as the center tap) connect to sockets as labeled on the breadboard’s front panel. The common sizes are 1 W. 1 W. the existence of other materials (namely semiconductors) for which the I –V relationship is nonlinear makes electronics much more interesting and underlies the transformation of daily life brought about by electronics during the twentieth century. and the multimeter (conﬁgured to measure current). be sure to observe the following warnings: r First.1) This is not a universal law of electrical conduction so much as a statement that there exist certain materials for which current is linearly proportional to voltage. 1 W. You can easily verify this linear relationship between voltage and current using the ﬁxed 10 k (10 000 ohm) resistance provided between the two ends of one of the breadboard’s ‘potentiometers’. and can be adjusted by means of a large black knob. allowing electrical connections to be made not only at the two ends.2 Materials with such a linear relationship are used to fabricate ‘resistors’: objects with a known and stable resistance. 1 W. 8 4 2 and 2 W. . metal ﬁlm. But. Often the resistance is indicated by means of colored stripes according to the resistor color code (Table 1. the 10 k pot. Usually they are little cylinders of carbon. (1. voltage V . before doing so. You can attach alligator clips to the meter leads to connect them to the wires. and resistance R according to V = IR.

7 1 Equipment familiarization Table 1. For example. or 20% tolerance). Use Ohm’s law to predict the current that will ﬂow around the circuit if you use the power supply that you set to its midpoint in the previous exercise. have someone check your circuit before turning on the power. Stripe: Black Brown Red Orange Yellow Green Blue Violet Gray White Gold Silver None Stripe 2 Stripe 1 Stripe 3 1 0 10 20 30 40 50 60 70 80 90 2 0 1 2 3 4 5 6 7 8 9 3 100 101 102 103 104 105 106 107 108 109 4 (tolerance) 5% 10% 20% Tolerance Stripe the breadboard (screw heads for example) has a low-resistance path to ground.2. rendering it useless! If in doubt. Color code for nonprecision resistors (5. plus or minus the tolerance in column 4. for 51 ‘green--brown--black’. for 330 ‘orange--orange--brown’. etc. multiplied by the value in column 3. The resistance in ohms is the sum of the values in columns 1 and 2. 10. What current should ﬂow if the supply is set to its minimum voltage? What is the current if the supply is set to its maximum voltage? . you can easily burn out the pot. the color code for a 1 k resistor would be ‘brown--black--red’. r If you accidentally connect power or ground to the potentiometer’s center tap.

If you don’t have a TDS210. Do the two measurements add up to the total you measured above? They should – explain why. In this and the following part. Due to manufacturing tolerances. any dual-trace oscilloscope. the pot should connect only to the meter. 1.3 Oscilloscope With its many switches and knobs. measure the currents for these three voltages. and compare with your predictions. 1. Set the meter for resistance and measure and record the resistance between the two ends of your 10 k pot.4). By what percentage does it differ from the nominal 10 k value? Does the measured value agree more closely with the slope you previously measured than with the nominal value? Explain.3 Measuring resistance Now turn off the breadboard power and disconnect your series circuit. analog or digital. The TDS210 is not entirely as it appears. 30 MHz or higher. current to 10 k ? 1.8 Hands-on electronics Now turn on the breadboard power. the rest of this laboratory session will be devoted to becoming acquainted with such an instrument and seeing some of the things it can do. In the past you may have used an oscilloscope that displayed voltage as a function of time on a . The oscilloscope we use is the Tektronix TDS210 (illustrated in Fig. While the description below may not correspond exactly to your scope. can be used for these labs as long as the bandwidth is high enough – ideally. you will probably ﬁnd that it is not exactly 10 k . measure and record the resistance between the center tap and each end. yet the scope is an essential tool for electronics troubleshooting and you must become familiar with it. Make a graph of voltage vs. Now connect the meter between the center tap and one end of the pot. a modern oscilloscope can easily intimidate the faint of heart.2. with careful study of its manual you should be able to ﬁgure out how to use your scope to carry out these exercises. What resistance do you observe? What happens to the resistance as you turn the potentiometer’s knob? Leaving the knob in one place. Is the relationship linear? How close is the slope of voltage vs. current from these measurements. Accordingly.

the input signal is sampled.). Note and remember the location of the ‘autoset’ button – when all else fails. you can learn about multiple traces and triggering. One of your ﬁrst objectives will be to set up the scope to do some of the things for which you may already have used simpler scopes. and stored in memory. 1. In order to have something to look at on the scope. The digitized signal can then be displayed on a computer screen. etc. it does not contain a CRT (part of the reason it is so light and compact). The basic features to be used in this tutorial are marked. But start by using the built-in ‘calibrator’ signal provided by the scope on a metal contact labeled ‘probe comp’ (or something similar). After that. try autoset! cathode-ray tube (CRT). analog scopes have been largely superseded by digital devices such as the TDS210 (although low-end analog scopes are still in common use for TV repair. digitized. you can use your breadboard’s built-in function generator.9 1 Equipment familiarization Seconds per Division MEASURE CURSOR 60 MHz 1 GS/s MENU Menu Options Menu Options AUTOSET HORIZONTAL CONTROLS TDS 210 Trigger Info TRIGGER LEVEL TRIGGER MENU Volts per Division Menu Options Menu Options Menu Options Horizontal and Vertical Info Trigger Level CH 1 CALIBRATION CONTACT POINT MENU OPTION BUTTONS CH 2 SEC/DIV VERTICAL CONTROLS VOLTS/DIV Fig. sinusoidal waves. In a digital scope. most oscilloscopes were purely ‘analog’ devices: an input voltage passed through an ampliﬁer and was applied to the deﬂection plates of a CRT to control the position of the electron beam. In the past few years. often located near the lower right-hand corner of the display screen. .4. and triangular waves of adjustable amplitude and frequency. Illustration of the Tektronix TDS210 digital oscilloscope. Until the 1990s. A digital scope operates on the same principle as a digital music recorder. The position of the beam was thus a direct analog of the input voltage. While the TDS210 can perform a similar function. a device capable of producing square waves.

on the tip assembly. It is useful for reducing noise when looking at high-frequency .) Note that each probe also has an alligator clip (sometimes referred to as the ‘reference lead’ or ‘ground clip’)..4).e.10 Hands-on electronics Note that a leg folds down from the bottom of the scope near the front face. which has circuitry inside that divides the signal voltage by ten. not square). try ‘autoset’. The screw is usually located on the assembly that connects the probe to the scope. The probe’s coaxial connector slips over the ‘ch 1 ’ or ‘ch 2 ’ input jack and turns clockwise to lock into place. so we recommend that you use it. which needs to be tuned for minimum distortion. and have equal calibrations. Sketch the observed waveform. carefully adjust the probe compensation using a small screwdriver. are properly compensated. Display the calibrator square-wave signal on the scope. or. The manufacturer therefore provides a ‘compensation adjustment’ screw. your scope should also have a built-in ‘calibrator’ circuit that puts out a standard square wave you can use to test the probe (see Fig. 1. (Consult your oscilloscope user manual for more information about carrying out a probe test. 1.3.1 Probes and probe test Oscilloscopes come with probes: cables that have a coaxial connector (similar to that used for cable TV) on one end. This adjusts the viewing angle for greater comfort when you are seated at a workbench. allowing you to grab the calibrator-signal contact with a metal hook or ‘grabber’. Some scopes sense the nature of the probe and automatically correct for this factor of ten. we generally use a ‘10X’ attenuating probe. To increase the scope’s input impedance and affect the circuit under test as little as possible. for connecting to any desired point in the circuit to be tested. If the signal looks distorted (i. (If you have trouble achieving a stable display. An attenuating scope probe can distort a signal. This connects to the shield of the coaxial cable. As mentioned above.) Check your other probe. occasionally. and a special tip on the other. others (such as the TDS210) need to be told by the user what attenuation setting is in use. for connecting to the scope. The probe tip has a springloaded sheath that slides back. Make sure that both probes work.

These adjust the sensitivity (volts per vertical division on the screen) and offset (the vertical position on the screen that corresponds to zero volts). vertical scale factors. 1. your manual will explain the corresponding settings for your scope. The difference between the two measurements is the voltage across the element.11 1 Equipment familiarization (time intervals of order nanoseconds) or low-voltage signals. 1.4). which is grounded via the third prong of the AC power plug. if you have a different model. For each.) Warning: A short circuit will occur if the probe’s reference lead is connected anywhere other than ground. you should study the description in your oscilloscope user’s manual. The ‘ch 1 ’ and ‘ch 2 ’ menu buttons can be used to turn the display of each . it must never be allowed to touch a point in a circuit other than ground! Otherwise you will create a short circuit by connecting multiple points to ground. the reference lead should remain ﬁrmly attached to ground and should not be moved! (Alternatively. The following exercises will give you practice in understanding the various settings.3. Explain brieﬂy the various pieces of information displayed around the edges of the screen. Since it is connected directly to the scope’s case. During this process. which could damage circuit components. The description below is speciﬁc to the TDS210. etc. This is no trouble if you are measuring a voltage with respect to ground.3 Vertical controls There is a set of ‘vertical’ controls for each channel (see Fig. 1.3. But if you want to measure a voltage drop between two points in a circuit. neither of which is at ground.2 Display Your oscilloscope user’s manual will explain the information displayed on the scope’s screen. Record the various settings: timebase calibration. ﬁrst observe one point (with the probe) and then the other. you can use two probes and conﬁgure the scope to subtract one input from the other.

The level knob sets the threshold voltage for triggering. they also select which control settings are programmed by the push-buttons just to the right of the screen. 1. so that many images of the signal occurring one after another can be superimposed in the same place on the screen. Since setting up the trigger can be tricky. The position knob moves the image horizontally on the screen. known as the trigger ‘threshold’. What happens when you adjust the position knob? The volts/div knob? 1. The sec/div knob sets the sensitivity of the horizontal axis. i. Display a waveform from the calibrator on channel 1.e. using trigger control knobs and switches).3. How many periods of the square wave are you displaying on the screen? How many divisions are there per period? What time interval corresponds to a horizontal division? Explain how these observations are consistent with the known period of the calibrator signal.5 Triggering Triggering is probably the most complicated function performed by the scope. Normally. To create a stable image of a repetitive waveform.4). the scope must ‘trigger’ its display at a particular voltage. Adjust the sec/div knob to display a larger number of periods. the TDS210 provides an automatic setup feature (via the autoset button) which can lock in on .3. or the 120 V AC power line. the scope displays voltage on the vertical axis and time on the horizontal axis. Now what is the time per division? How many divisions are there per period? 1.4 Horizontal sweep To the right of the vertical controls are the horizontal controls (see Fig. an external trigger signal. for some scope models. and control various other triggering features as well. You can select whether triggering occurs when the threshold voltage is crossed from below (‘rising-edge’ triggering) or from above (‘falling-edge’ triggering) using the trigger menu (or.12 Hands-on electronics channel on or off. You can also select the signal source for the triggering circuitry to be channel 1. the interval of time per horizontal division on the screen. channel 2. The display is synchronized whenever the input signal crosses that voltage.

be sure to ask for help. or study the oscilloscope manual more carefully. frequency. and the triggering to produce a stable display. adjust the level knob in each direction until the scope just barely stops triggering. How then can you derive the amplitude from something the scope does measure? Using the measurement features.3.13 1 Equipment familiarization almost any repetitive signal presented at the input and adjust the voltage sensitivity and offset. determine the amplitude. Particularly useful are the digital measurement features. After getting a stable display of the calibrator signal. with one channel looking at the output of the function generator and the other looking at the scope’s calibrator signal. . if the trigger threshold is set too high or too low)? How does this depend on whether you select ‘normal’ or ‘auto’ trigger mode? Why? (If you ﬁnd this confusing. What is the range of trigger level that gives stable triggering on the calibrator signal? How does it compare with the amplitude of the calibrator waveform? Does this make sense? Explain. What do you see on the screen if you trigger on channel 1? On channel 2? What do you see if neither channel causes triggering (for example. Next connect the scope probe to the breadboard’s function generator – you can do this by inserting a wire into the appropriate breadboard socket and grabbing the other end of the wire with the scope probe’s grabber.) 1. and period of a waveform of your choice from the function generator.6 Additional features The TDS210 has many more features than the ones we’ve described so far. The function generator’s amplitude and frequency are adjusted by means of sliders and slide switches. Look at each of the waveforms available from the function generator: square. The scope does not measure amplitude directly. You can use them to measure the amplitude. period. and triangle. Try out the frequency and voltage controls and explain how they work. Make sure the vertical sensitivity and offset are adjusted for each channel so that the signal trace is visible. and frequency of a signal. Display both scope channels. Adjust the function generator’s frequency to about 1 kHz. the time sensitivity. sine. Push the measure button to program these.

) . Use the cursors to measure the half-period of the signal you just measured. (A feature that comes in particularly useful on occasion is signal averaging. to be suppressed relative to signal.14 Hands-on electronics You can also use the on-screen cursors to make measurements. this is programmed via the acquire button and allows noise. Explain how you made these measurements and what your results were. which tends to be random in time. which is usually periodic.

However. The ﬁrst is referred to as the time-domain approach. a capacitor consists of two parallel conductors separated by an insulating gap.1 Review of capacitors As you may recall from an introductory physics course. since it considers the voltage across the capacitor as a function of time. which are open circuits for DC. whereas the second is more useful when analyzing low-pass and high-pass ﬁlters. Then we apply it to the analysis of RC circuits as ﬁlters. breadboard. since it gives us a simple way to understand circuits containing more than one component in series. for voltages that change with time. a simple series circuit with a capacitor and a resistor can output the time derivative or integral of an input signal. and 10 k resistors.2 RC circuits Capacitors are not useful in DC circuits since they contain insulating gaps. Both approaches are valid – in fact. 15 . or can ﬁlter out low-frequency or high-frequency components of a signal. they are mathematically equivalent – but the ﬁrst is more useful when using capacitors as integrators or differentiators. 2. The capacitance. But before plunging into the world of time-varying voltage and current (i. frequency.. 68 0. Note that the series RC circuit can be analyzed in two different ways: r via the exponential charging/discharging equation. alternating-current circuits).e. Apparatus required Oscilloscope. and r as an AC voltage divider. digital multimeter. we explore the voltage-divider idea using direct current. since it focuses on the ﬁlter attenuation vs.01 F ceramic capacitor. and the second as the frequency-domain approach.

A. and the applied voltage should be more positive at the positive end than at the negative end – if you reverse-voltage a polarized capacitor it can burn out. manufacturers wind ribbon-shaped capacitors up into small cylinders and use insulators of high dielectric strength. of the insulating material: C =κ 0 A/s. categorized according to the type of dielectric used. abbreviated F (1 farad ≡ 1 coulomb per volt). with dielectric constant of order 1. s in meters. where. Electrolytic and tantalum capacitors are polarized.1. To achieve the substantially larger capacitances (of order microfarads) often found in electronic circuits. in the MKS system of units. A is in meters squared. .0 1. Material Vacuum Air (at STP) Paper Mica Ceramic Dielectric constant (κ) 1.1 gives dielectric constants for some typical dielectrics used in capacitors.4 ≈100 C. and C in farads. 0 = 8. (The constant of proportionality is the so-called permittivity of free space and has the value −12 F/m). Capacitors thus come in a variety of types. that chemically form an extremely thin insulating layer when a voltage is applied. is proportional to the area of the conductors. and inversely proportional to their separation. mica. the capacitance is ∼ picofarads.5 5. Some typical dielectric materials used in capacitors. and ceramic capacitors are unpolarized and can be hooked up in either direction. which means that they have a positive end and a negative end. or even explode! Paper.854 × 10 The farad is an impractically large unit: for a conductor area of 1 cm2 and separation of 1 mm. κ. The large dielectric constants of the polarized dielectrics permit high capacitance values – up to millifarads in a several-cubic-centimeter can. multiplied by the dielectric constant.00054 3. s. such as ceramics or (in the so-called electrolytic capacitors) special dielectric pastes.16 Hands-on electronics Table 2. Table 2. which determines how much capacitance can be squeezed into a small volume.

it is sufﬁcient to focus on two: square waves and sine waves. symmetric with respect to ground). In other words. the current into or out of a capacitor is equal to the time derivative of the charge stored on it.1 Use of capacitors. A square wave of amplitude V0 is a signal that oscillates back and forth between two voltage levels. You can see the resemblance between Eq. f . Current can ﬂow into or out of a capacitor. which. as well as their angular frequency ω = 2π f . 1. There are an inﬁnite variety of AC waveforms. and period.1. by Fourier decomposition. which of course are related by T = 1/ f. and C is the capacitance. namely AC circuits. spending 50% of the time at each level.17 2 RC circuits 2. Capacitors are thus useful only in circuits in which voltages or currents are changing in time.) A periodic waveform is also characterized by its amplitude. but to understand how capacitors are used. T .1) where Q is the charge stored on the capacitor at a given time. Most of AC circuit analysis therefore concerns itself with the response of circuits to sine waves. any periodic waveform can be represented as a sum of sine waves of various amplitudes and frequencies. is the maximum voltage that it reaches. (2. 2. these can be characterized by their frequency. We will consider the response of circuits to periodic waveforms. review of AC circuits The fundamental rule governing the behavior of capacitors is Q = C V. Note that the peak-to-peak voltage is twice the amplitude: Vp–p = 2A. assuming the wave is pure AC (i.e. whereas for a capacitor it is the charge itself.1). The key difference is that. one at +V0 and one at −V0 . then T is in seconds and ω is in radians per second. it is the time derivative of the charge that is proportional to voltage. A sine wave is a particularly important case because. You have already encountered a square wave in the previous chapter – the scope’s calibrator signal. V is the voltage across the capacitor at that time. for a resistor.1 and Ohm’s law (Eq. (If f is expressed in hertz ≡ cycles per second. . but only to the extent that the charge on the capacitor is changing.

periodic waveform with period T .18 Hands-on electronics T V0 V p--p V0 Fig. i. The phase φ determines the voltage at t = 0: V (0) = V0 sin φ. this is only an idealized approximation. i.3) Now suppose such a voltage signal is applied to a capacitor. at t = 0 the voltage is zero.9) Of course. there is no possibility of a voltage drop anywhere in the circuit. A sine wave can be represented mathematically by V (t) = V0 sin(ωt + φ). We can write an ‘Ohm’s-law equivalent’ for a capacitor.1 Thus. we choose φ = 0.7) (2. for simplicity. Since we are assuming there is no resistance in the circuit.8) We see that the current is also a sine wave.e.4) (2.1: I = dQ dt dV =C dt d = C (V0 sin ωt) dt = ωC V0 cos ωt = ωC V0 sin (ωt + 90◦ ).5) (2. 1 (2. 2. (2. 2.1. this describes a voltage with amplitude V0 changing sinusoidally in time at the rate of ω radians per second. (2. . at any moment in time. (2.e. as long as it is understood that we are talking about sinusoidal waveforms only: V0 = I0 X C .2) If t is in seconds. since in any real circuit there is at least the resistance of the wires. but shifted in time with a phase shift of 90◦ . the voltage across the capacitor must equal the voltage out of the signal source. amplitude V0 and peak-to-peak voltage Vp–p . Representation of an arbitrary.6) (2. the current leads the voltage by 90◦ . and in practice any signal source has some internal resistance. To ﬁnd the resulting current we can differentiate Eq.

. and have negligible resistance. A small capacitor that says just ‘10’ on it is 10 picofarads. Thus. A 10 millifarad cap would be labeled ‘10000M’. but if you understand capacitors you will have very little difﬁculty in applying inductors. bandpass ﬁlters. in keeping with our intuition that for DC a capacitor should look like an open circuit (inﬁnite resistance). most manufacturers don’t want to print Greek letters.2 Types and values of capacitors For some reason the various manufacturers’ conventions for marking capacitors are particularly confusing – probably it has to do with the fact that many small-value capacitors are physically too small to permit much printing on them. For completeness. In the interests of time we omit inductor exercises from our course. an inductor can thus be thought of as the opposite of a capacitor. 2. The other important number is the maximum operating voltage.1) relates the voltage across a capacitor to the integral of the current. the current through an inductor lags the voltage across it by 90◦ .10) Just as capacitors often employ a dielectric. expensive. With respect to its function in a circuit. Inductors are coils of wire and satisfy the equation V =L dI . and these are the two commonly used units. inductors are often wound on a ferrite core to increase their inductance. radio-frequency circuits. which is usually printed on the capacitor if there is room. light. heavy. e.g. Some common sense is required. inductors tend to be large. they ﬁnd important use in ﬁltering applications. where L is the inductance of an inductor. Note that it is frequency-dependent.1.19 2 RC circuits where I0 is the amplitude of the current sine wave and X C ≡ 1/ωC is the capacitive reactance of the capacitor. The reactance is thus the effective resistance of the capacitor. A physically large capacitor that says ‘10M’ on it is usually 10 microfarads. we mention here the inductive reactance X L ≡ ωL. where the current through a capacitor leads the voltage across it. cheap. not 10 millifarads (for some reason. so they use ‘M’ instead of ‘ ’). crossover circuits for hi-ﬁ speakers. 2. while at high frequency it should approach a short circuit (zero resistance). Nevertheless. and have appreciable resistance. dt (2. Whereas capacitors are relatively small. Keep in mind that 1 farad is a huge unit! Most capacitors are in the picofarad and microfarad ranges. while the capacitor equation (Eq. and so forth. Note that the inductor equation relates the voltage across an inductor to the derivative of the current through it.

etc. the cross-section of a wire) is deﬁned as the net charge passing through that surface per unit time. voltage. either with a color code or with numbers that mean the same thing. or by plugging it into a capacitance meter. The potential energy (with respect to some reference point) is equal to the voltage multiplied by the charge. Current refers to the motion of charges. the third is the power of 10. let us brieﬂy review power dissipation and component ratings – you need to understand these to avoid damaging components. When in doubt. whereas just 470 means 470 pF! Note that there is no ambiguity if it says 471 – since normal capacitors are not manufactured with enough precision to warrant a third signiﬁcant digit. The voltage drop across any circuit element is directly proportional to the change in energy of a charge as it traverses the circuit element. you can always check it out by putting it in an RC circuit with a known R value and measuring the time constant (see below). This is sometimes ambiguous – for example. 2. if you have one. a capacitor that says ‘470’ could be 470 pF or 47 × 100 = 47 pF! Usually the clue is the presence of a letter.g. the second is the ones.20 Hands-on electronics Some small capacitors are labeled like resistors. following the capacitance code. 1 volt = 1 joule/coulomb. Voltage is related to potential-energy difference. The ﬁrst digit of this capacitance code is the tens.2 Review of current. K for ±10%. Speciﬁcally. The current through a given surface (e. otherwise known as watts. The product of voltage and current has units of joules/second. The unit for current is the ampere: 1 ampere = 1 coulomb/second. and power Before we get started on RC circuits. – so that ‘470 K’ means 47 pF ± 10%. that indicates the tolerance – J for ±5%. If the voltage drop across a circuit element equals the change in potential energy per unit charge. M for ±20%. the ‘1’ must be the power of ten. and the current equals the amount of charge . and the units are picofarads.

21 2 RC circuits moving through the element per unit time. Exceeding these ratings can have disastrous effects on your circuit.12) Resistors. Do not touch the resistor with your ﬁngers. relays.11) For resistive elements (or when an effective resistance can be deﬁned). + 15 V 68 Ω 1/4 watt Resistor Fig. 2. (2. Note that the resistor will heat up rapidly. Be sure that the power is turned off.11 can be combined with Ohm’s law to give: P = IV = I 2 R = V 2 /R. Keep the body of the resistor well above the breadboard. This circuit can be used to demonstrate destructive power loading. . are rated (in part) by their maximum allowed power.2. and may even cause a ﬁre! To illustrate this point. 2. then their product equals the power released within the device! The power dissipated within any device is given by P = IV. (2.2. transistors. Be sure to turn off the power as soon as the resistor begins to smoke.2 using a 1 watt carbon-ﬁlm resistor. 2. integrated-circuit chips. etc. Record your observations and comments. Eq. 2. care must be taken to prevent burns. 4 Turn on the power and observe the effect on the resistor. diodes.. our ﬁrst exercise will deliberately lead to the destruction of a carbon-ﬁlm resistor. Remove the destroyed resistor using pliers or a similar tool. The resistor in the following exercise will become very hot and may even catch ﬁre (brieﬂy). and construct the circuit shown in Fig.1 Destructive demonstration of resistor power rating Caution: In the following exercise.

What is the minimum resistor value that can be safely used in this circuit? (Assume that only 1 watt resistors are available. In all cases you can show using Ohm’s law that Vout = Vin R2 /(R1 + R2 ). A resistive voltage divider is simply two resistors in series (see Fig. so you will need to become thoroughly familiar with it. and a smaller voltage. Note that the far right representation is implemented using a potentiometer. results at the junction between them. and you will now try this out using the breadboard’s 10 k pot. 2. but in the wrong circumstances they can still cause trouble.3. and to make sure you are not exceeding the ratings. damage was nevertheless done! Because your body’s resistance is large. 2. Vin . Turn off the power before hooking up the circuit! 2. Warning: You can easily burn out the pot in this exercise if you are not careful! 1. 2. the output voltage is variable and ranges between ground and Vin (depending on the position of the slider).3 Potentiometer as voltage divider The voltage-divider idea is very useful in analyzing almost all circuits.22 Hands-on electronics Calculate the power that was dissipated by the resistor before it burned out. low voltages can’t give you a shock. Three schematics representing a resistive voltage divider. The key to safe work in electronics is always to estimate power dissipations in components before turning on the power. Note that even though the voltage was low and the current was well under 1 A.3). If you accidentally connect the pot’s slider to ground while one end is connected to the supply voltage (or vice versa) you can easily burn out R1 R2 V V R1 V V ~ R2 ~ V potentiometer Fig. Vout . A potentiometer can be used as a variable voltage divider.) 4 Calculate the current that ﬂowed through the resistor (before it burned out). A voltage difference. . In this case. is applied across the two.

Without moving the slider or changing the supply voltage. otherwise you will have to use a separate 10 k pot.s. R1 + R2 (2. R2 /(R1 + R2 ). voltage.23 2 RC circuits the pot – brieﬂy explain why this is true. using the meter to tell you when you get there. compare for DC and AC? 2 The 10 k pot is located near the center of the bottom edge of the PB-503 breadboard and is adjusted by means of a large black knob. use the voltmeter setting to adjust the variable positive power supply to +10 V.3. 2. values there? How does the pot’s voltage-division ratio. and r. amplitude. turn on the power. (Hint: how much power can be dissipated in the pot in such a situation?) 3. Then. amplitude.s. connect one end of the pot to ground. Connect the meter (on the voltage setting) between the slider and ground. since on these settings a meter can act as a low impedance (short circuit). Look at the function generator’s output signal with the scope and measure its peak-to-peak voltage.m. and measure the voltage between the slider and ground.) Look at the signal at the pot’s center tap – what are the peak-to-peak. What are the values of R1 and R2 ? Using the voltage-divider equation Vout = Vin R2 . and r.2 AC voltage divider Now verify that a resistive voltage divider works the same way for AC as for DC. you can easily burn out the pot. Apply a sinusoidal signal from the function generator to the pot in place of the +10 V DC. ﬁnd a 10 k pot on your breadboard if it has one. turn off the power.m. Next. If you don’t have a PB-503. Double-check your connections. and connect the other end to +10 V.1 DC voltage divider First.2 Turn the knob to set the slider exactly half-way between the ends. How close to this prediction is your measured voltage? What is the percentage error? 2.13) explain why the predicted output voltage is +5 V.3. use the ohmmeter setting of your multimeter to measure the resistance between the slider and each end of the 10 k pot. . (The scope’s measure menu is useful here. If you connect the multimeter on a current or resistance setting between the slider and some other point in the circuit while the circuit is powered.

2. (Be sure to set the time and voltage scales sufﬁciently sensitive to yield an accurate measurement – what settings should you use. What are the amplitudes of the input and output waveforms? Sketch the output waveform.5 RC circuit as integrator Now switch the function-generator frequency to 50 kHz.24 Hands-on electronics R Vin C V R ~ C V Fig. RC. Ground one end of the capacitor.14) where t = 0 corresponds to a rising or falling edge of the square wave.e. or as a voltage integrator.4. You may recall from your general physics that it obeys the equation V (t) = V0 e−t/RC . This can be explained . and determine RC using the scope – you should ﬁnd your oscilloscope’s ‘cursor’ feature useful here. Observe what happens to the output waveform’s shape and amplitude. connect the other end to the resistor. 2. V (t) = 0. and connect the other end of the resistor to a 500 Hz square wave from the function generator (see Fig. what do you predict for RC? Is your measurement consistent with this prediction? What are the tolerances of the components you are using – does this explain any discrepancy? 2. Display the input signal (output of the function generator) on channel 1 of the scope. Show that the time to fall to 37% of the peak value (i.4 RC circuit Now hook up a 10 k resistor and a 0.4). and why?) Based on the nominal component values. This circuit is also known as a low-pass ﬁlter. The voltage-divider concept works perfectly well for RC circuits.37 V0 ) is the ‘time constant’. and the output signal (at the junction of the resistor and capacitor) on channel 2.01 F ceramic capacitor in series. (2. 2.

25

2 RC circuits

**quantitatively from Eq. 2.1 Vout (t) = Q(t) C 1 t = I (t)dt C 0 = ≈ 1 C 1 RC
**

t 0

(2.15) (2.16) (2.17) (2.18)

Vin − Vout dt R

t

Vin dt,

0

where Vout is the voltage across the capacitor, and the approximation is Vin . valid as long as Vout What does Eq. 2.18 predict when Vin is a constant, as it is during half of each period of the square wave? Carefully measure and sketch the output waveform. Compare your observations with your expectations based on Eq. 2.18 and explain your results. What output amplitude would you expect at 25 kHz? Change the input frequency and see if your prediction is correct. At approximately what frequency will Eq. 2.18 cease to predict the output waveform accurately? Change the input frequency and test your prediction.

**2.6 Low-pass filter
**

Now switch from a 50 kHz square wave to a 50 kHz sine wave. Since Eq. 2.18 should still apply, the output waveform should be the integral of a sine wave, i.e. a cosine wave. What does this imply about the phase shift between input and output? Measure the phase shift: 360◦ multiplied by the time t between the zero crossing of the input signal and the zero crossing of the output signal, divided by the period, or φ = 360◦ t T . (2.19)

(The cursors are useful here.) Is the measured phase shift consistent with your prediction? Does the voltage across the capacitor lag or lead the current through it? Explain.

26

Hands-on electronics

The other way to analyze this circuit is as an AC voltage divider, using Eq. 2.13 with R2 replaced by the reactance of the capacitor, X C , and R1 + R2 replaced by the total impedance, Z , of the resistor and capacitor in series. Since the voltage across the capacitor is 90◦ out of phase with the current through the resistor, these add in Pythagorean fashion: Vout = Vin = Vin = Vin XC Z XC R2 + 1 1 + (ω RC)2

2 XC

(2.20) (2.21) .

(2.22)

We see that the attenuation (Vout /Vin ) depends on the frequency. What attenuation do you observe at 50 kHz? Calculate what you expect, and compare. At the breakpoint or half-power frequency f 0 , ω RC = 1, and thus the √ attenuation is 1/ 2 = 0.707. This is also referred to as the −3 dB point, since it is the frequency at which the output voltage is attenuated by 3 dB. The breakpoint is a convenient way to parametrize simple ﬁlters. (The decibel is a logarithmic measure of the ratio of two signals: number of dB = 20 log A2 , A1

where A1 is the amplitude of the ﬁrst signal and A2 is the amplitude of the second; in this case A1 = Vin is the amplitude of the output signal and A2 = Vout is the amplitude of the input signal.) By varying f until the output amplitude is 70.7% of the input amplitude, measure f 0 . Calculate what you expect, and compare with your measurement. What are the attenuation and phase shift at low frequency, say 50 Hz? Compare with the predictions of Eq. 2.22. Compare the phase shifts with φ = arctan R . XC (2.23)

The phase shift at low frequency is easy to understand: in the limit of DC the capacitor must act like an open circuit, i.e. inﬁnite impedance, and thus does not affect the output signal. Conversely, in the high-frequency limit the capacitor must look like a short circuit to ground, so the output signal goes to zero and the phase shift becomes dominated by the capacitor.

27

2 RC circuits

2.7 RC circuit as differentiator

Now interchange the capacitor and resistor so that the input signal is applied at the capacitor (see Fig. 2.5). Drive the circuit with a 50 Hz square wave. What waveform do you see at the output? What are the input and output amplitudes? You can think of the shape of the output in terms of the exponential RC charging/discharging curve, with f 1/RC, or you can think of it as an approximation to the derivative of the input signal. Mathematically, the derivative of an ideal square wave would be inﬁnite at the voltage steps and zero in between, but of course an electrical signal can never be inﬁnite! In this circuit the voltage spikes are limited in size to twice the input amplitude. Using Eq. 2.16, Vout = IR dQ =R dt d(Vin − Vout ) = RC dt dVin , ≈ RC dt (2.24) (2.25) (2.26) (2.27)

Vin . So, indeed, the where the approximation is again valid when Vout circuit puts out an approximation to the time derivative of the input signal. You can see why the approximation of Eq. 2.27 breaks down in the case of a square wave, since at the rising and falling edges of the square wave Vout > Vin . What does Eq. 2.27 imply if the input is a triangle wave? Try it out and compare quantitatively with what you expect. What does Eq. 2.27 imply if the input is a sine wave? Try it out and compare quantitatively with what you expect. Sketch the output waveform.

V C R V C V

~

R

Fig. 2.5. High-pass ﬁlter or voltage differentiator.

28

Hands-on electronics

If you are surprised at all the wiggles on the output signal, you can verify that they are real (as opposed to noise) using the signal-averaging feature of the scope’s acquire menu. You’ve discovered a poorly kept secret of function-generator design! The sine waveform is rather difﬁcult to generate, and most function generators actually use an approximation to it that is piecewise-linear around the peaks and valleys. The derivative of a piecewise-linear function is a series of steps and plateaus.

**2.8 High-pass filter
**

What attenuation and phase shift do you observe with a 50 Hz sine wave as input? What about with a 50 kHz sine wave? Why do these phase shifts make sense? Should the breakpoint frequency be any different in this conﬁguration than in the low-pass ﬁlter? Check it and make sure. Compare your measurements with R (2.28) Vout = Vin Z R = Vin (2.29) 2 R2 + X C = Vin ω RC 1 + (ω RC)2 . (2.30)

Show that well below the breakpoint frequency, Eq. 2.30 predicts that the output amplitude should increase linearly with frequency. Take a few measurements to demonstrate that this prediction is correct.

**2.9 Summary of high- and low-pass filters
**

For reference, here once again are the key equations describing high-pass and low-pass RC-ﬁlter operation in the frequency domain. High-pass: Vout = Vin ω RC 1 + (ω RC)2 1 . ω RC , (2.31) (2.32)

φ = arctan

.34) φ = arctan ω RC. the breakpoint frequency is the point at which the two terms in the Fig. while the phase shift goes to zero in the ﬁrst case and to 90◦ in the second. and those on the right for a frequency well above f 0 . Right triangles depicting the relationships among input voltages (always represented by the hypoteneuse of the triangle) and capacitor and resistor voltages for (a) high-pass and (b) low-pass RC ﬁlters. that at high frequency. the voltage-division ratio goes to unity in the ﬁrst case and to zero in the second case.29 2 RC circuits Low-pass: Vout = Vin 1 1 + (ω RC)2 . in both cases. Note. In each case. 2. Also. as expected.6. (2.33) (2. the center diagram shows the isosceles triangle representing the case f = f 0 . the triangles on the left are for a frequency well below f 0 .

37) 2π RC These relationships are all illustrated in Fig.6.35) ω0 f0 = (2. . for lowand high-pass ﬁlters. for a ﬁxed input voltage. (2. which shows.30 Hands-on electronics square-root are equal: ω0 RC = 1 (2. how the phase and magnitudes of the voltages across the resistor and capacitor are related. See Appendix C for further discussion of frequency-domain analysis of RC circuits. 2.36) 2π 1 = . in three frequency regimes: f = f 0 and frequencies that are well below and well above f 0 .

one 100 and one 10 k 1 W resistor. one 1 k 2 W resistor. no current 31 . We’ve seen that resistors have a simple linear relationship between the voltage across them and the current through them (Ohm’s law). but we think you’ll agree that the idea as just stated is simple enough – it just takes some getting used to! As we’ll see. power transformer 4 with 12. output on each side of the center tap. Insulators (such as diamond) possess very few free charge carriers – all the electrons are tightly bound to the crystal lattice.m.1 Semiconductor basics Current will ﬂow through a material provided that there are charge carriers free to move and an electric ﬁeld to move them. one diode bridge element. Apparatus required Breadboard. and one 1000 F electrolytic capacitor. Mathematically this may seem much more complicated than Ohm’s law. Conductors (such as copper) have lots of charge carriers (electrons) ready to move in response to the slightest electric ﬁeld. one or two multimeters. even in the presence of a strong electric ﬁeld.6 V r. an important consequence of the exponential characteristic is that diodes conduct much more readily in one direction than in the other. 3. This makes them ideally suited for rectiﬁcation: the conversion of AC into DC. one 1N914 (or similar) silicon signal diode. On the other hand.3 Diodes In this chapter we will explore semiconductor diodes and some circuits using them. so that.s. one 1N4001 (or similar) 1 A silicon rectiﬁer diode. oscilloscope. one 100 F electrolytic capacitor. r diodes have an exponential relationship between current and voltage.

If an ‘acceptor’ material such as boron is substituted for silicon. On the other hand. i. the ‘depletion region’ (region around the junction devoid of charge carriers) increases in size. If an external ﬁeld is applied in the same direction as the internal ﬁeld. these doped semiconductors are electrically neutral: the ‘extra’ electrons in N-type material are compensated for by the additional protons in the atomic nuclei of the donors. In a silicon crystal. Phosphorus has ﬁve valence electrons. and boron has three. This creates a net charge build-up around the junction (see Fig. the deliberate inclusion of impurities within the semiconductor lattice. Thus. the result is a junction diode.) The ‘extra’ electrons within N-type material can move under the inﬂuence of an electric ﬁeld. the dominant charge carriers are electrons. electrons from neighboring atoms can jump into the holes. despite their names. Electrons from the N-region migrate across the junction into the P-region.e. The holes can migrate in the direction of an electric ﬁeld.32 Hands-on electronics ﬂows. which are used to make covalent bonds with neighboring silicon atoms. The conductivity of a semiconductor can be enhanced through doping... moving the holes from one place to another. while the ‘missing’ electrons in P-type are compensated for by the missing protons in the acceptor nuclei. if an external ﬁeld is applied opposite to the internal ﬁeld. For P-type material. a ‘hole’ appears in the electron structure of the lattice. has four valence electrons. for example. such as phosphorus. Silicon. P-type material has positive charge carriers.1) – positive in the N-region and negative in the P-region – leading to an internal electric ﬁeld as shown. Electrons ﬂow into the N-type material from the metal contact . i.e. in such a way that the crystalline structure is preserved across the junction. so current does not ﬂow. If a junction between P-type and N-type semiconductor material is created within a single crystal. The charge motion is thus due to the motion of the holes. the junction region becomes devoid of charge carriers and thus acts as an insulator. whereas doping with an acceptor creates a ‘P-type’ semiconductor. Once the holes are ﬁlled. N-type material has negative charge carriers. Doping silicon with a donor material creates an ‘N-type’ semiconductor. free charge carriers ﬂow toward the junction. an extra valence electron becomes available that is loosely bound to the lattice. 3. Semiconductors (such as silicon and germanium) are somewhere in between. ﬁlling holes as they go. if a silicon atom is replaced with a ‘donor’ material. (Note that. preventing further current ﬂow.

For such a diode the current I ﬂowing through the device is given approximately by I = Is (eeV /nkT − 1) (3.1) where Is (sometimes called IR or I0 ) is the ‘reverse saturation current’. e is the electron charge. we’ll assume for now that n = 1.33 3 Diodes Internal Electric Field _ + Hole Free Electron Silicon. 3. the holes from the P-type material meet electrons from the N-type material and combine. Diode circuit symbol and biasing. Aluminum Arsenic. For simplicity.1. .2). New holes are created within the P-material as electrons jump from the semiconductor to the metal contacts. This dependence of current on voltage is illustrated in Fig. 3. V is the voltage across the junction.2. k is Boltzmann’s constant. At the junction. A PN junction thus allows current to ﬂow easily in one direction but blocks current ﬂow in the reverse direction. Representation of a junction between P-type and N-type semiconductor material.3. 3. 3. and T is the junction temperature in kelvin. Phosphorus P-type N-type Fig. (see Fig. n is an empirical constant between 1 and 2. Free electrons from the N-region will migrate into the P-region. Electric Field Free Electron Hole Electric Field _ _ _ + _ + _ + _ + + + + N-type P-type N-type I Is _ P-type Is I Depletion Region + V R V + R Forward-Biased Diode Circuit Reverse-Biased Diode Circuit Fig. Germanium Boron. combining with holes.

note that the current scales in the forward and reverse directions differ by a factor of 10 000.3) . and I ≈ Is eeV /kT .4 −0. this corresponds to V > 0 in Fig. (3. and the reverse current is almost constant. 3.4 0. If a large enough reverse voltage is applied.3. the diode is said to be ‘reverse-biased’. 3. the exponential term is negligible. 3. with I ≈ −Is .6 0.2) When the diode is reverse-biased and |V | is greater than ≈100 mV or so. this corresponds to V < 0 in Fig. when V is greater than ≈100 mV or so.3. When the P-type material is at a more positive voltage than the N-type material.2 0. Typical current–voltage characteristics for germanium and silicon diodes.8 Vdiode (Volts) Reverse Breakdown Fig.34 Hands-on electronics Forward Current 8 (mA) Ge 6 Si 4 2 −20 −18 −0. the ‘1’ in Eq. the diode is said to be ‘forward-biased’. When the P-type material is more negative than the N-type material. Some useful approximations In the forward-biased case. the junction breaks down and allows a large reverse current to ﬂow (the ‘Zener effect’).2 Reverse Current (µA) 0.1 becomes negligible compared with the exponential term. (3.3. 3.2 −0. and that the voltage scale changes at large reverse voltage.

Schottky diodes are used in applications requiring high speed and low capacitance. to distinguish it from the ‘anode’ (the P-type end).2 Types of diodes In addition to standard junction diodes. Their forward characteristics are similar to those of junction diodes. 3.3 to 75 volts. green. LEDs are junction diodes typically made from gallium arsenide phosphide (GaAsP).4). To distinguish the ends from each other.4. 3.35 3 Diodes 3. most diodes look like a little cylinder with wires sticking out the two ends (Fig. and blue LEDs are commercially available. This migration creates a potential barrier across the boundary. which then behaves in a similar fashion to a PN junction. Electrons from the N-type material migrate into the metal. and switching speed (time required for the diode to Fig. Schottky diodes are manufactured by bonding a metal conductor to an N-type semiconductor. Schottky diodes. Zener diodes are designed to operate at and around their reverse-breakdown (‘Zener’) voltage. however. yellow. Zener diodes are used in reverse-biased mode. Representation of physical diodes along with the symbols used in circuit diagrams. orange. reversebreakdown voltage. the manufacturer often prints the diode circuit symbol on the diode body. Physically. Zener diodes are manufactured with controlled reverse-breakdown properties. light-emitting diodes (LEDs). forward voltage drop. leakage current (reverse saturation current). . Alternatively. They act very much like silicon junction diodes except that they emit light when conducting forward current and have forward voltage drops about twice as large as silicon diodes. and Zener diodes are also common. Diodes are manufactured with speciﬁed values for maximum current. The Zener voltage is determined during the manufacturing process by adjusting the semiconductor doping. sometimes a ring is marked around the body close to the ‘cathode’ (the N-type end). While reverse breakdown typically destroys a standard junction diode. red. Infrared. In general. Typical Zener voltages range from 3.

It would offer inﬁnite resistance when reverse-biased.2 @ 0. even for volts of reverse voltage (as long as the reverse voltage is kept small enough to avoid breakdown).0 40 200 1.1 C (pF) 4.0 Diode 1N914 1N4001 1N4004 1N5402 FR601 MBD301 1N4733A Type small signal rectiﬁer rectiﬁer rectiﬁer fast rectiﬁer Schottky Zener 0. of which a tiny sampling is given here.. The junction capacitance is sometimes listed as well.0 0.28 1.0 8. An ideal rectiﬁer would offer zero resistance when forward-biased. Pmax (W) 0.1 VFmax @ IF (V @ A) 1.0 0. a . and reversebreakdown voltage.0 @ 0. A few examples are given in Table 3. A diverse selection of diodes is commercially available. semiconductor diodes are universally used for the purpose.0 switch from forward to reverse bias or vice versa).0 3.1 @ 3. 3. independent of the amount of forward current ﬂow.01 1.1. maximum forward voltage at a speciﬁed forward current. with only a few hundred millivolts of forward voltage over a wide range of forward current.01 1. regardless of the size of the applied reverse voltage. You can see from Fig. maximum forward current. and with reverse current measured in nanoamps. i. the voltage across it would be zero.3 Rectification A rectiﬁer is a device that converts AC to DC by blocking the ﬂow of current in one direction.3 @ 6. Nowadays.6 @ 0.0 1. maximum power dissipation.e.2 VBR(R) (V) 75 50 400 200 50 30 5.0 6. Diodes are commonly rated by their switching speed.0 1.. For the typical range of currents encountered in most electronic circuits (a few to hundreds of milliamps). no current would ﬂow. Rectiﬁcation used to be almost the exclusive province of vacuum tubes. i.0 1.36 Hands-on electronics Table 3.e.3 that real diodes typically approximate the ideal reasonably well. with the exception of the ‘detector crystals’ (naturally occurring semiconductor diodes) used in crystal sets in the early days of radio.0 8.1 @ 1.0 1.3 1.1. 3.1 @ 1.5 IFmax (A) 0.

. dV kT thus.7) so the dynamic resistance of a forward-biased diode can be simply approximated by dV 25 mV ≈ .6 × 10−19 C e = = 39 V−1 ≈ .a more sophisticated view If we want. We can ﬁnd the dynamic resistance of a diode by differentiating Eq. the resistance is dynamic rather than having a constant (or static) value. While the static resistance of a device is the voltage across the device divided by the current through it (R = V /I . You can see that.38 × 10 25 mV (3. 3. since V /I is not constant.4) kT /e .e. dI I (3. i. . 1 1.6) I where the approximation is valid for forward-biasing such that the ‘1’ in Eq. Ohm’s law). (3.37 3 Diodes handy approximation is that a forward-biased germanium diode has about a 300 mV voltage drop across it. Note that at T = 300 K (room temperature). they become different.5) (3. the static and dynamic resistances are the same. but for a device whose V –I curve is nonlinear. 3. 3.4 Diode action -.1: e dI = Is eeV /kT . Note in particular that a nonlinear device does not have a static resistance. for a resistor.8) These results – the exponential dependence of current on voltage and the consequent dynamic-resistance formula – are important to remember. the dynamic resistance is the slope dV /dI of the V –I curve at any point. while a forward-biased silicon diode has about a 600 mV drop. the dynamic resistance is kT /e dV = dI Is eeV /kT ≈ (3. we can think of a diode as a resistor whose resistance depends on the current ﬂowing through it.1 is negligible. −23 J/K) × 300 K kT (1.

do not let the forward current exceed 50 mA. 3. 3.5. For example.1) is implied by your measured slope? Now reverse the diode – for a reverse voltage of 5 V what reverse current do you observe? Is it consistent with the range of Is expected for a silicon diode? .8.5 Measuring the diode characteristic As before. kT (3. For the circuit shown in Fig. resistance that is in series with the junction resistance just described. Plot your results for I vs. V .9) Plot ln I vs.1? When unity is negligible in comparison with the exponential term of Eq.2 may be re-expressed as ln I ≈ ln Is + e V.1. 3. For the portion of the characteristic curve close to the origin. estimate the maximum current in your circuit and the maximum power dissipation in the 100 resistor – is the resistor’s 1 W rating safe for its worst-case power dissipation? Caution: 4 These small diodes are easily damaged by overcurrent. forward-biased as shown in Fig.38 Hands-on electronics as they also characterize the behavior of transistors. increase the voltage across the diode starting from 0 V in steps of 100 mV and record the diode current in each case. If you don’t have a second meter available. To be on the safe side. so in practice the dynamic resistance is somewhat larger than that given by Eq. 3. 3. (This model of diode behavior neglects some features that matter in practice. 3. be careful not to burn out your breadboard’s 1 k pot: hook up the following circuits with the power off.5. Is ln I approximately linear in V ? How does its slope compare with e/kT ? What value for n (from Eq. the microamp range of the multimeter will be required to measure the forward current. V on a linear scale – do they seem qualitatively consistent with the functional shape of Eq. or static. the semiconductor has some ‘ohmic’. With a 1N914 (or similar) silicon diode. and double-check each circuit before powering it up. Eq.) 3. 3. you can use your oscilloscope to measure the diode voltage.

39 3 Diodes 1.5. the DMM is equivalent to an ideal ammeter in series with a small input impedance. the DMM or oscilloscope looks like an ideal voltmeter in parallel with a large input impedance. Most ammeters will also have a series fuse to protect the meter. 3. (a) Measuring the forward characteristic of a diode. Disconnect the scope or meter – now how much current do you observe? Reconnect it and disconnect the diode – how much current ﬂows with the scope or meter alone? What do you infer to be the input resistance of the scope or meter? Explain by applying Ohm’s law to relate the voltage being measured to the current you observe.2 mA A mA COM VΩ + 1 k pot 100 Ω 0.64 V +5 V A mA COM VΩ (a) Ammeter + 1 k pot 100 Ω mA DMM or Oscilloscope Rin fuse Zin +5 V V Ground (Common) (b) Fig. If the reverse current seems to be much bigger than you expect. (b) When used to measure current. When measuring voltage. while an ideal ammeter would have zero resistance? . Keep this experience in mind – it is often necessary to consider the effect of your measuring device on the circuit being studied. 3.5(b)). consider that you have a voltage-measuring device (a scope or voltmeter) in parallel with the diode (Fig. Can you see why an ideal voltmeter would have inﬁnite resistance.

RL . value of the sinusoidal voltage.s.. the transformer’s output voltage depends on 1 In North America. the power is supplied by batteries. 3. Set up the circuit shown in Fig. twice this. However.m. Observe the sinusoidal voltage waveform across RL . a DC power supply. Measure the amplitude V0 and the r.6 Exploring rectification Next we take up the basic principles of rectiﬁcation. Almost all electronic equipment requires power from a steady voltage source. Caution: In using a power transformer. the most convenient power source is the 120 V 60 Hz AC line. as you can easily verify from the deﬁnition of the root-mean-square by integrating over the sine wave. Since the windings of the transformer have some ohmic resistance. (3. which may be substituted according to your local supply voltage and frequency. (3. the supply voltage from a standard wall socket is 120 V.11) You will probably ﬁnd Vrms > 25 V. i. the discussion is equally valid for other values. bear in mind that an especially large transient current sometimes flows when the line cord is first plugged in. and the supply frequency is 60 Hz.e. there is a power transformer that steps down the 120 V AC to a more convenient voltage. For portable equipment. it is safest to disconnect the leads at the transformer. and a regulator that maintains the output voltage at the desired level.s. . Do not permit powered lines to dangle loosely.10) and the peak-to-peak voltage is. when reconﬁguring your circuit.) Within most electronic equipment using the AC line. not at the breadboard. a rectiﬁer that converts the alternating voltage from the transformer to a DC voltage. of course. Check the relation √ 2Vrms = V0 . Attach banana-plug leads to the transformer’s secondary only after you are sure your circuit will not damage any of the equipment.m. or Vp−p = 340 V.6(a) using a 10 k resistor as the load. voltage. You will probably blow fewer fuses if you leave the power transformer plugged in at all times.40 Hands-on electronics 3. the amplitude being √ V0 = 2 × 120 V = 170 V.1 (120 V is in fact the r.

m. 3. Caution: The 100 F electrolytic capacitor is polarized -.12) Add a ﬁlter capacitor in parallel with the load as shown in Fig.s. Add a 1N4001 diode to give the half-wave rectiﬁer of Fig.7(a) with RL = 10 k.41 3 Diodes Transformer fuse (a) 120 VAC Center Tap RL Vout (b) Vrms V0 V0 Vp--p Fig.6.2 V r.be careful not to hook it up backwards! The negative terminal should be labeled with a ‘−’ sign. Check that for a half-wave rectiﬁer Vav = V0 . the current drawn. By how much has the amplitude decreased? Is this the amount you expect? Explain.s.. 3. Measure the amplitude V0 using the oscilloscope (due to the rectiﬁcation it is now equal to the peak-to-peak voltage).8(a). and its 25. (a) Power transformer supplies Vout ≈ 25 V r. Measure the average voltage Vav across the load with a DC voltmeter. nominal output voltage is for substantially higher current than is drawn by the 10 k load. Compare the amplitude of the half-rectiﬁed waveform with the amplitude of the unrectiﬁed waveform measured above. . 3. Observe and record the voltage waveform. (b) waveform produced by the circuit in (a). π (3.m.

7. (b) waveform produced by circuit shown in (a). . Transformer Rectifier Diode fuse (a) 120 VAC + 100 µF RL Vout Ripple Voltage (b) 1 s 60 Ripple Voltage (c) 1 s 60 Fig. (a) Power transformer with half-wave rectiﬁcation. (b) waveform produced by circuit shown in (a). (a) Half-wave rectiﬁer with ﬁlter capacitor. 3.42 Hands-on electronics Transformer Rectifier Diode fuse (a) 120 VAC RL Vout (b) V0 Vav Vp--p Fig.8. (c) simple approximation to waveform produced by circuit shown in (a). 3.

3. illustrated in Fig.e.9). Diode Bridge ~ _ ~ + Fig. The terminals are labeled: ‘∼’ marks the two terminals that should be connected to the transformer secondary. 3. An example of how to insert a diode bridge into a breadboard. These bridges have four diodes within. The average current through the load can be determined using the known resistance RL and the average DC voltage across it as measured with a voltmeter. Using these assumptions.. 3. and compare with the peak-to-peak ripple voltage as measured.8(c). making the bridge-rectiﬁer approach particularly convenient. 3. Replace the 100 F electrolytic capacitor with a 1000 F capacitor. and the fundamental capacitor equation Q = C V . equal to the average load current.43 3 Diodes What is the voltage rating of your capacitor? Make sure it is sufﬁcient for the voltage that will be applied! Observe and record the output-voltage waveform across the load resistance RL and measure the peak-to-peak ‘ripple voltage’ (i. Here is that rare situation – measuring accurately a small AC signal on top of a large DC offset – in which you should use the AC-coupling feature of the scope’s vertical menu. If you are troubled with noise. the amount by which the output voltage is varying. Diode-bridge rectiﬁers are available as a single encapsulated unit. This can be accomplished using two diodes and the transformer’s center tap or by using a diode bridge. A simpliﬁed analysis approach for predicting the expected output waveform. Measure the ripple voltage and compare with your expectations. and ‘+’ and ‘−’ denote the positive and negative outputs (see Fig.8(b)). Is the observed percentage discrepancy within the tolerances of your components? Explain. . Calculate the output-voltage droop in each cycle.9. see Fig. is to assume that the capacitor charges up to the peak voltage instantaneously and discharges at a uniform rate dQ/dt. you may want to trigger the scope on ‘line’ and employ signal averaging. Do you expect the ripple voltage to increase or decrease? Explain. Full-wave rectiﬁcation should decrease the ripple by about a factor of two.

keeping in mind that the ﬁlter-capacitor discharge time is now one-half of the AC cycle. as shown in Fig. 3. 3. To make this circuit into a ‘complete’ power supply. Record the peak-to-peak ripple voltage. Repeat the ripple voltage calculations for these two values of RL . 3. It should show essentially infinite resistance between the terminals marked ‘∼’.check it before placing it in service.44 Hands-on electronics Transformer Diode Bridge + fuse 120 VAC ~ _ ~ RL Vout Fig. employ feedback to make the output voltage and ripple less dependent on the load resistance. Repeat these measurements for RL = 1 k.11(a) – again.9. Measure the average output voltage across the 10 k load. with the long dimension of the package running along the groove.10 with RL = 10 k. Add a ﬁlter capacitor to the full-wave rectiﬁer as shown in Fig. This could be done using Zener diodes. but a more effective technique is a transistorized regulating .10. 3. that is. Set up the bridge rectiﬁer circuit of Fig. (Insert the rectiﬁer package straddling the central groove of a breadboard socket unit. and why? Use socket adapters to handle the fat leads of the 2 W resistor. As before. observe and record the voltage waveform across RL .) Four rectiﬁer diodes can be used if a bridge rectiﬁer is not available. be careful not to connect the capacitor backwards. remember to measure it for both orientations of the terminals -. Caution: A defective bridge element can blow the power transformer fuse -. Caution: What power rating must the 1 k resistor have. When using an ohmmeter to check the resistance. This form of power supply is very common. the resistance could be different in each direction. one would want to regulate the output. Full-wave rectiﬁcation using a diode bridge.since you are dealing with diodes.

(b) waveform produced by circuit shown in (a). 3. an input also has some small capacitance and inductance and hence is more completely characterized by its impedance vs. (3. Integrated three-terminal voltage regulators (such as the 7800 and 7900 series) have made this particularly simple. Outputs can also be characterized by their impedance (see Fig. 3.13) . From Z out = − Vout / Iout .11. frequency. (a) Full-wave rectiﬁcation with ﬁlter capacitor.5.45 3 Diodes Transformer Bridge Rectifier + fuse (a) 120 VAC + ~ _ ~ 1000 µF RL Vout Ripple Voltage (b) 1 s 120 Fig.7 Input and output impedance Input and output impedance are key ideas that are used all the time in analyzing circuits. which takes into account both the resistance and the capacitive and inductive reactances. circuit. 3.12). In practice. A good way to think about the effect of an instrument on the circuit to which it is connected is via the instrument’s Th´ venin equivalent. The e Th´ venin equivalent of the multimeter (when set to measure voltage) is a e large resistor in parallel with an ideal voltmeter (Fig. You’ve already encountered the input impedance of the scope or voltmeter in section 3.5). 3. You’ve already taken data that determine the output impedance of your ﬁltered full-wave rectiﬁer circuit.

If the output impedance has negligible frequency dependence. and your data on Vout vs. in ohms. in which case the function generator’s Th´ venin-equivalent circuit consists of an ideal AC voltage source (one e having zero internal resistance) in series with a single resistor. The output impedance is measured by observing the output voltage as a function of the output current: Z out = − Vout / Iout . by measuring its sine-wave output amplitude. determine the output impedance of your breadboard’s function generator. Z out .46 Hands-on electronics Output Impedance Zout V0 Ideal Voltage Source + I RL Vout Complete Rectifier Circuit Fig. it can be approximated as a pure resistance. 3.12. . compute the circuit’s output impedance. As another example. ﬁrst with no load. and then with a load resistance of 1 k to ground. and function generator. RL . Check the function generator’s Z out both at low and high frequencies (say 50 Hz and 50 kHz) – do you observe any appreciable frequency dependence? Sketch schematic diagrams (with component values labeled) of the Th´ venin-equivalent circuits of your voltmeter. full-wave-rectiﬁed power e supply. A rectiﬁer circuit can be modeled (as a Th´ venin equivalent) using an ideal e voltage source in series with an output impedance.

oscilloscope. two 10 k. 22 k. but it is something you must master.4 Bipolar transistors Invented in 1947.e.) A bipolar junction transistor consists of two PN junctions sandwiched very close together within a single crystal of semiconductor (Fig.g. 1 k. Apparatus required Breadboard.3 k. 1N914 (or similar) silicon signal diode. transistors (and integrated circuits made from them) have been the basis for the explosive proliferation of electronic devices that revolutionized so much of life in the latter half of the twentieth century. and 100 k 1 W resistors. and compare it with the descriptions in other books.1(a)). an understanding of how they work remains valuable. (If you want more of the background detail on semiconductor physics. 4.1 Bipolar-junction-transistor basics Why and how transistors work is a bit subtle and can easily confuse the beginning student. two 330 . 47 . or any textbook on modern physics. 1 F capacitor. since transistors form the basis of a large class of integrated circuits. You may also want to re-read both our description and others after you’ve had some experience building and analyzing transistor circuits. and one each of 100 . red light-emitting diode (LED). individually packaged) transistors are now used mainly in special situations (e. This chapter will introduce you to some basic bipolar-junction-transistor circuits. Study the following description carefully. Although discrete (i. This will be the subject of the next few chapters. two multimeters. 4 4. good places to look are Simpson’s Introductory Electronics for Scientists and Engineers. 2N3904 and 2N3906 transistors. where high power or speed is required). 3.

in the circuit symbol. Note that. .) The ratio of collector current to base current is called β (or hfe ) and is typically in the range 20 to 300. You can tell whether a transistor in a schematic diagram is PNP or NPN by the direction of the arrow. in regions known as the emitter and collector.1(b). The circuit symbols for NPN and PNP junction transistors are shown in Fig. For PNP. a transistor is a voltage-controlled current source: small changes in the base voltage cause large changes in collector current. The region common to the two junctions. To understand the operation of an NPN transistor in more detail. may be of either N-type or P-type material. As above. The simplest way to think of transistor action is as current ampliﬁcation: a small current ﬂowing into the base controls a large current ﬂowing into the collector. with both currents ﬂowing in through the emitter. is called a transconductance ampliﬁer. called the base. (a) Construction and (b) circuit symbols and biasing examples for NPN and PNP junction transistors.48 Hands-on electronics (a) E N P N C E P N P C B E RE B C RC RE E B C B RC (b) Fig. Such a device. Both the base and collector currents ﬂow out from the emitter. This thin region is surrounded by material of the opposite type. The transconductance (gm ) for a given device is deﬁned as the change in output current per change in control voltage and has units of (ohm)−1 (otherwise known as a mho). Wire leads are attached to the three regions. 4.1. it is convenient to consider the ﬂow of electrons. the current directions are opposite: a small current ﬂowing out from the base controls a large current ﬂowing out from the collector. 4. however. since electrons are the ‘majority carriers’ in the N-type regions. the arrow on the emitter lead points in the direction of positive current ﬂow. More precisely. (This description assumes an NPN transistor. in which an input voltage controls an output current. Note that the ﬂow of electrons is of course opposite in direction to the ﬂow of conventional positive current.

This results (seemingly paradoxically) in a large ﬂow of electrons through the (reverse-biased) base–collector junction. the emitter lead injects electrons into the emitter.2). The large hollow arrow represents the ﬂow of electrons from the emitter to the collector. The base current results from the small fraction of electrons entering the base that combine . and only ≈1% emerge as base current. the construction of the transistor results in large numbers of the ‘wrong’ current carrier entering the base and then continuing ‘downhill’ into the collector. At the same time. They do this even though the base–collector junction is reverse-biased.2. In normal transistor operation. 4. While (as just stated) the base–collector bias inhibits the ﬂow of holes from the base into the collector.49 4 Bipolar transistors Electric Field N-type Depletion Region P-type Free Electron Electric Field N-type Hole Collector Emitter IC (lightly doped) Base (heavily doped) IE IB + + VCB VBE Fig. the base–emitter diode is forward-biased and the base–collector diode is reverse-biased (Fig. Essentially. The depletion region between the base and the collector extends essentially throughout the thin base region (creating an electric ﬁeld as shown in Fig. 4. which ﬂow across the (forward-biased) base–emitter junction. Typically. ≈99% of electrons entering the base from the emitter continue into the collector. 4. This is the essence of transistor action. the electrons with which the base is now ﬁlled are drawn by the electric ﬁeld through the junction and into the collector. a current that is easily controlled by small changes in base voltage. which pulls electrons (emitted into the base by the emitter) across the base and into the collector. but with the current carriers and directions reversed. Schematic representation of how an NPN transistor operates.2) and blocks the ﬂow of majority current carriers – holes ﬂowing from base to collector and electrons ﬂowing through the collector to the base. the following description applies also to PNP transistors. External bias voltages create an electric ﬁeld.

1. It is worth keeping in mind that β is not well controlled in the transistor manufacturing process.05 0. ICmax (A) 0. and the relationships among them: r VBE ≡ VB − VE = potential of base relative to emitter. r VCB ≡ VC − VB = potential of collector relative to base. β depends signiﬁcantly on temperature.06 0. A diverse selection of bipolar transistors is commercially available. we need to deﬁne three voltage differences. .0 8.06 0.50 Hands-on electronics Table 4.20 0.90 Part # 2N3904 2N3906 2N5089 2N2369 2N5415 TIP102 MJ10005 Type NPN PNP NPN NPN PNP NPN NPN Application gen’l purpose gen’l purpose gen’l purpose switching power pwr Darlington pwr Darlington hfe 100–300 100–300 450–1800 40–120 30–150 1000–2000 50–600 with holes. voltage capability. (For examples of the range of variation.14 1.) In addition.1.20 0. maximum current. Because β is so variable. it varies from transistor to transistor even if they are of the same type. small changes in the base–emitter voltage difference have a large effect on the collector current. It is desirable to have as large a ratio of collector current to base current as possible.625 0. Transistors are commonly rated by their speed (e. the base is made of lightly doped material.g. and power capability .1 Basic definitions To discuss transistor action quantitatively. Although β is approximately constant for a given transistor. a transistor can be used as an ampliﬁer: a device in which a small signal controls a large signal. the current-ampliﬁer picture of transistor action is less useful than the transconductance-ampliﬁer picture. and collector-to-emitter voltage.625 0. typical hfe .69 6. toggle frequency f T ).4 10 80 175 Cost ($US) 0.2 1. collector current.625 0. it is good practice to design transistor circuits in such a way that their proper functioning does not depend strongly on its exact value. Since β can vary over a substantial range. of which a small sampling is given here. Since the base–emitter junction obeys the exponential diode law for current as a function of voltage.99 0. 4. in addition to being as thin as possible.1.40 0. Thus. To achieve this. three currents.0 20 fT (MHz) 300 250 50 500 Pmax (W) 0. see Table 4.

However. 4. r associated identity (from Kirchhoff’s current law): IE = IC + IB . Fig.1) For an NPN transistor in its normal operating mode. The relationships between these voltages and currents are usually expressed in terms of characteristic curves.1. all the above potential differences are positive. Characteristic curves for an NPN bipolar transistor. r IB = current ﬂowing into base.51 4 Bipolar transistors r VCE ≡ VC − VE = potential of collector relative to emitter.2 Simplest way to analyze transistor circuits In the following section we consider a mathematical model (the Ebers– Moll model) that gives a good approximation to transistor performance. We will see from the Ebers–Moll model that. r IE = current ﬂowing out of emitter. 4.3. r associated identity (from Kirchhoff’s voltage law): VCE = VBE + VCB . Study their shapes carefully and refer to them as you perform the following exercises. 4.3 displays sets of representative curves for an arbitrary bipolar transistor. . r IC = current ﬂowing into collector. if a transistor Breakdown IC (mA) 40 Saturation Region Linear Region IB = 350 µA IC (mA) 40 Linear Region IE = 35 mA IE = 25 mA IE = 15 mA IE = 9 mA IE = 3 mA Breakdown 30 IB = 250 µA 30 20 IB = 150 µA IB = 90 µA IB = 30 µA 20 10 10 0 5 10 15 20 VCE (Volts) 0 5 10 15 20 VCB (Volts) Fig. you don’t need the model to understand transistor operation in most circuits. all the above currents are positive.2) For an NPN transistor in its normal operating mode. (4. (4.

re . a changing voltage at the base causes matching voltage changes at the emitter. r the collector assumes any voltage required by Ohm’s law as applied to the rest of the circuit. r the collector current nearly equals the emitter current. and 4. In other words.3. 4. the base is a point of high impedance.2. described in sections 4. we next consider the Ebers–Moll model. Recalling the deﬁnition of dynamic resistance. the base current is related to the collector current by IB = IC . and 4.52 Hands-on electronics is on and conducting milliamperes of collector current. of the emitter.5.2. Is is the reverse saturation current.8.3 Ebers–Moll transistor model To see why the simple picture just described is valid. We can thus use it to derive a useful expression for the dynamic resistance. (4. the exponential usually dominates and the ‘1’ can be neglected. Thus.4) Dynamic resistance of emitter Since the emitter current equals the collector current to a good approximation.3) As with Eq. Then come back and study the next section.2. These three rules are all you need to know in most situations. as mentioned above. Furthermore. 4.2. re is the partial derivative of emitter . the base voltage controls the emitter voltage: r the emitter voltage follows the base.) Also. Eq.3 also detemines the emitter current. (Suggestion: You may want to skip the next section for now and use these rules to analyze the circuits of Figs. Since β is large. Since the collector is of even higher impedance than the base.1. In this model. which is based on the physical description in section 4. (In practice. 3. 4. β (4.4. its base–emitter voltage difference VBE is approximately constant at about 700 mV.7.1.) 4. the amount of collector current that ﬂows is determined by the amount of forward-bias that is applied to the base–emitter diode. whereas the emitter is a point of low impedance. We thus have IC = Is eeVBE /kT − 1 . 4.1 and provides a reasonably good description of transistor action.

7) This tells us how the base loads the circuit that is driving it.3 at ﬁxed base voltage. IC 25 mV =β . the emitter voltage would change in response to a change in emitter current – for example. but appears to have a higher resistance (by a factor β) when viewed from the base end. e/kT = 39 V−1 . we differentiate Eq. so it is easy to apply a signal voltage to the base. which is of order 100. We see that the base–emitter junction appears to have a low resistance when viewed from the emitter end. Although rBE is only a few hundred ohms. Some useful approximations Since at room temperature.53 4 Bipolar transistors voltage with respect to emitter current: re ≡ ∂ VE . in practice a reasonable approximation is re = rBE 25 mV . To determine re .) Dynamic resistance of base On the other hand. its voltage hardly depends on the current ﬂowing through it. This comes about because the base current is smaller than the emitter current by the factor β.8) (4. 4.6) (In addition there is the ohmic resistance of the emitter. IC (4. e IC (4. the factor β applies also to any resistor .9) Since the emitter acts as a low impedance (only a few ohms for typical collector-current values).3 and 4. ∂ IB e IC (4. 4. we can ﬁnd from Eqs. ∂ IE (4.4 the dynamic resistance rBE to the emitter as seen at the base: rBE = ∂ VB kT 1 =β . which is typically a few ohms. But the base acts as a high impedance. how the emitter voltage would differ if the emitter were driving a small load resistance as opposed to a large one. if we ﬁx the emitter voltage.5) The dynamic resistance tells us how. giving re = kT 1 . for ﬁxed base voltage.

e. As mentioned above.54 Hands-on electronics RE that is in series with the emitter.1 Checking transistors with a meter Since a transistor is constructed as a pair of back-to-back PN junctions. To tell which pin of the transistor is which. Test the base–collector and base–emitter junctions of a 2N3904 (NPN) and 2N3906 (PNP).and reverse-biased directions. Often this can be done using an ohmmeter. However. each junction should show about 700 mV in the forward direction. they provide a diode test function that measures the forward voltage corresponding to a forward current of about 600 A. If your transistors are good.2. To test a diode or transistor junction. and record your readings in both the forward. 4. the apparent resistance seen at the base is β(re + RE ) = rBE + β RE .4. 4. . Often re is much smaller than RE and can be neglected. which shows the pinout for the TO-92 plastic case in which these transistors are packaged. in many practical transistor applications (including the circuits you will build in this chapter) a simple approximation is sufﬁcient: treat the base–emitter voltage difference VBE as constant at about 700 mV and re as constant at a few ohms. and an out-of-range indication in the reverse direction.2 Experiments 4. We saw in the case of the silicon diode that a crude approximation in which the forward diode drop is taken to be approximately constant at 600 mV is adequate for most applications. refer to Fig. a quick way to test a transistor is to verify its junction resistances in the forward.and reverse-biased directions. which sends current through the device under test and measures the resulting voltage. instead. i. our digital multimeters are not designed for this kind of measurement. connect it between the ‘V ’ and ‘common’ jacks with clip leads and set the meter’s selector knob to the position marked with the diode symbol. which is typically tens to hundreds of kilohms. The diode is forward-biased if its anode is connected to ‘V ’ and its cathode to ‘common’. This reﬂects the fact that the order of magnitude for IC in a typical small-signal-transistor circuit is several milliamperes.

. power ampliﬁcation can occur through an increase in either quantity.) It is called an emitter follower because the voltage at the emitter follows voltage changes at the base.3). and low output impedance. 4. Although it has voltage gain (Vout /Vin ) of about unity (actually slightly less because of the logarithmic dependence of VBE on IC implied by Eq. which is what is meant by buffering. TO-92 pinout. Transistor as back-to-back diodes. (b) Emitter-follower model used for input-impedance measurements. (It is also a close relative of the common-emitter ampliﬁer that we will study next.55 4 Bipolar transistors (a) E B NPN C E B PNP C (b) 2N 4 390 TO-92 case 2N 6 390 E B C E B C Fig. 4. You can determine IB and IC (within the uncertainties due to the resistor tolerances) by measuring the voltage drops 1 Since power is the product of voltage and current..7 V. 4. The value for Z in is found using the voltage-divider equation. 4. (a) Emitter follower. and can thus provide current gain (i.e. (a) +15 V (b) Emitter Follower 330 10 k 2N3904 Vin RB Vin RB Vout RE 3.4.2 Emitter follower This simple transistor circuit (shown in Fig. as well as to change its DC voltage level by VBE . 4.3 k Z in −15 V Fig.1 Start by grounding Vin .5.5(a)) is often used to ‘buffer’ an AC signal.2. output current > input current). with VBE almost constant at about 0. it is still an ampliﬁer: it has high input impedance.

Derive an approximate β value for your transistor by computing the ratio of collector current to base current. whether yours does could also depend on details of the wiring arrangement. The 330 base resistor is in series with the input resistance rBE + β RE .0 µF 330 RL Vout 330 RL ~ −15 V Fig.6(b)).56 Hands-on electronics across RB and RE . The blocking capacitor allows the 330 load to affect the AC signal voltage without changing the DC biasing of the transistor. Measure the input impedance. . Explain using the voltage-divider idea how this measures the input impedance. (When you’re done with this exercise. restore the 330 base resistor for use in the following parts.3 k 1. (a) Emitter follower with optional load circuit for measurement of Z out . replace the base resistor with 10 k and measure the small decrease in amplitude from one side of RB to the other (Fig. (‘Parasitic’ refers to capacitance that is there inevitably. It is there to prevent parasitic oscillation. rather than by design. To do so. 4. due to the proximity of the leads to each other. 4. which might otherwise occur due to the inductance of the wire jumpers and the ‘parasitic’ capacitive coupling between the emitter and the base.) Measure the output impedance. so it has little effect on the signal. To do so. try omitting the base resistor and see what happens. Not all 2N3904s are guaranteed to oscillate in this circuit.) Measure the DC voltage shift VBE between the base and emitter. 4. (a) 330 +15 V (b) Emitter Follower Vin 2N3904 Zout Vout 3. make sure both scope-probe compensations are properly adjusted. 4.6. add a blocking capacitor and load resistor as shown in Fig. What are the input and output amplitudes? (If they seem to differ signiﬁcantly.6(a) and infer the output impedance from the small decrease in output amplitude (see Fig.5(b)). (b) Emitter follower modeled as an ideal voltage source in series with an output impedance.) If you’re curious about this. Now drive Vin with a sine wave from the function generator and compare the input and output signals.

7. and the output impedance should equal the dynamic resistance of the emitter (as described in section 4. . be sure to use a small signal amplitude (about 1 V) so as not to apply too large a reverse voltage to the capacitor.3 Common-emitter amplifier The common-emitter ampliﬁer is a very common transistor ampliﬁer conﬁguration.m. as you’ll see.e. The predictions are easy: 1. and use a high enough frequency so that the capacitor causes negligible attenuation – about 10 kHz. the base current is small enough that the approximation is a good one. When using the blocking capacitor.2. emitter..57 4 Bipolar transistors Explain what would happen to the DC bias voltage of the emitter if the capacitor were omitted.3). 4. Common-emitter ampliﬁer. but. but that is not how it gets its name – the name reﬂects the idea that the emitter is in common between the input circuit and the output circuit.7. 4. Note that polarized capacitors can be safely reverse-voltaged by a volt or two r. but typically not by more than 15% of their voltage rating. and collector. given the large value of β. Predict and measure the quiescent DC voltages (i. VCC = +15 V R1 1 µF RC R1 = 100 k R2 = 10 k Vout 2N3904 RC = 10 k RE = 1 k Vin R2 RE Fig. the voltages when no input signal is present) at the base. How do your measured impedances compare with what you expect? The input impedance should equal β times the emitter resistor. Apply Ohm’s law to the base-bias resistive voltage divider to determine the quiescent base voltage: VB = VCC R2 /(R1 + R2 ). Construct the common-emitter ampliﬁer shown in Fig.1. This is an approximation since it neglects the base current. 4.s.

and base currents. taking into account the transistor’s expected VBE drop: IE = (VB − VBE )/RE . . Compared to your measurements. Since the emitter follows the base. by what percentages are your voltage predictions wrong? Is this as expected given the resistor tolerances and uncertainties in β and VBE ? Using the measured voltages. 3. (4. (4. Assume that the base current ﬂows through the Th´ venin e equivalent of the base-bias voltage divider (i.11) Measure the voltage gain ( Vout / Vin ) and compare with what you expect. the input impedance of the base is in parallel with R2 ).e. You know the collector current well enough since it equals the emitter current to a good approximation. a voltage change at the base causes a larger voltage change at the collector: Vout = − IC · RC = − IE · RC IE = VE /RE = VB /RE = Vin /RE .. So as not to exceed the available output-voltage range of the circuit. Is the ampliﬁer inverting? Look at the signal at the emitter and explain what you see. Then apply Ohm’s law to the emitter resistor to determine the emitter current. Then apply Ohm’s law to the collector resistor to determine the quiescent collector voltage: Vout = VCC − IC RC . Vout = − Vin RC /RE . Calculate the change in quiescent base voltage if you take the base current into account. use a high enough frequency that the input high-pass ﬁlter does not attenuate the signal too much – it is a little tricky to estimate the breakpoint frequency of the ﬁlter because three resistances in parallel need to be taken into account: those of the base-bias voltage divider as well as the input impedance of the base. be careful to keep the input amplitude less than about 700 mV.10) Therefore. predict the collector.58 Hands-on electronics 2. You can understand how the circuit ampliﬁes by applying Ohm’s law to the emitter and collector resistors. emitter. but be sure not to exceed the 1 V reverse-voltage capability of the input capacitor. (You can check what happens to the output waveform as you exceed this amplitude.) Also.

4. the voltage gain is greater. At the pot setting where the current source starts to fail (output current starts to vary rapidly with load resistance). 4. using a second meter or the scope to make various voltage measurements. the current-source behavior stops since . Try a triangle-wave input – can you see any distortion in the output waveform? There should be some due to the variation of re with collector current. measure the collector voltage. independent of its voltage. This means that the collector is a good approximation to a current source: a device that outputs a constant current.8. for RE = 0. but so is the distortion. the collector should act as a very high impedance. i. How big should the effect be according to the Ebers– Moll model? In the grounded-emitter ampliﬁer. be careful to connect it properly so as not to burn it out!) set to 0 .e.59 4 Bipolar transistors +15 mA 330 +5 2N3904 1k Fig.2.4 Collector as current source Since the base–collector junction is reverse-biased.8. since re alone appears between the emitter and ground. You can verify this using the circuit of Fig.. How much collector current should ﬂow? Is this conﬁrmed by your measurement of the emitter voltage? Slowly increase the load resistance until the output current starts to decrease. Start with the load resistance (10 k pot – as always. Transistor current source. how does the collector voltage compare with the base voltage? You should be observing transistor saturation: when |VCE | < |VBE |. but the effect is small since re is in series with the 1 k emitter resistor. 4. What is the compliance of your current source (the range of output voltage over which the current is approximately constant)? Where the current source starts to fail.

) Note that the transistor in this common-emitter connection is an ‘inverter’. when the transistor is on. 4.9. the voltage across it is small. the collector-base junction becomes forward-biased and its impedance decreases.9. take a few measurements of β as you turn up the load resistance and the transistor becomes more and more saturated. Make a graph of β vs. Transistor switch. Another way of thinking about the same phenomenon is that when the transistor saturates. By measuring the voltage drop across the base resistor.5 Transistor switch Transistor saturation is put to good use in the saturated switch. Construct the circuit of Fig. It has the virtue of dissipating very little power.2. 4. You should see the light-emitting diode (LED) turn on when you connect the input to +5 V and off when you leave it open or connect it to ground. β decreases sharply. This circuit is called a saturated switch since the transistor goes into saturation (|VCE | < |VBE |) when turned on. since. Often the cathode is indicated by a ﬂat spot on the rim of the plastic that encapsulates the diode. . and some manufacturers make the anode lead slightly longer than the cathode lead. (If you are unsure which terminal of the LED is the anode and which is the cathode. in the sense that a high voltage level at its input (the open end of the series base resistor) causes a low voltage level at its output (the collector).60 Hands-on electronics +5 Red LED 330 10 k ground or +5 2N3904 Fig. 4. or feel free to try it both ways. thus the base current increases as the base ‘steals’ current from the collector. VCE . you can check it with a meter.

4.1 Darlington connection To provide high input impedance and reduce the input base current. High-speed switching transistors (such as the 2N2369) are manufactured to minimize this effect and can operate at frequencies as high as 1400 MHz. It takes time for the transistor to switch states since this saturation charge must ﬂow in or out of the base through the input resistor. one can cascade two transistors in series.) What is your transistor’s saturation voltage VCE(sat) ? What is the ‘on voltage’ across the LED? Approximately what minimum value of β must the transistor have to be sure of saturating when +5 V is applied at the input? Drive the switch from the ‘TTL’ output of the function generator (‘digital’ square-wave with a low voltage level near zero and a high level near +5 V) at 100 kHz and use the dual-trace oscilloscope to measure the turn-on and turn-off delays in nanoseconds. With the input grounded. when the transistor is off. Build the circuit of Fig.) The relatively slow turn-on and turn-off delays of the saturated switch are due to the charge stored in the base when the transistor saturates.10.3 Additional exercises The following optional exercises offer additional practice. what are the base and collector currents? (You don’t need to bother with an ammeter for this measurement – just look at the voltage drops across the base and collector resistors. 4.3.61 4 Bipolar transistors whereas. 4. When the transistor is on. the current through it is essentially zero. This Darlington transistor pair acts like a single transistor whose current gain is the product of the two β’s and whose VBE drop is the sum of the two VBE ’s.e. i. what quiescent currents do you observe through RB and RE ? What does this imply for the combined β value of the Darlington pair? Now apply an input signal – what do you see at the output? What is the DC voltage drop from input to output? . (Trigger the scope with the function generator while looking at both the function generator signal and the collector voltage. ‘buffer’ the input with an emitterfollower stage.

Darlington pair. What is the attenuation due to the 8 load? Is it consistent with the measurement of the function generator’s output impedance you made in section 3. and why? Darlington pairs are available encapsulated in three-lead packages.10. and one NPN. the TIP110 50 W power Darlington has a minimum combined β value of 500.62 Hands-on electronics +15 10 k Q1 Q2 RB Q1 = Q2 = 2N3904 Vin Vout 3. to compensate for their low β (β ≈ 20 is not uncommon). This consists of two emitter followers.3. Go back to 1 kHz and add a push–pull buffer as shown in Fig. What minimum value does this imply for the input impedance? What input impedance do you expect.11. The input impedance should be so big that you can’t measure any decrease in signal amplitude across the 10 k resistor – check this assumption. for example the 2N6426 with combined β value of about 100 000. 4.2 Push–pull driver To provide low output impedance.3 k RE −15 Fig.7? Since the speaker impedance may depend on frequency.) You should see a larger output amplitude and hear a louder tone from the speaker. 4. With the amplitude set to maximum and the frequency at 1 kHz. arranged so that the PNP conducts during one half of the output period (when Vout < 0) and the NPN during during the other half (Vout > 0). drive your breadboard’s speaker (nominal impedance = 8 ) from the function generator. The Darlington connection is particularly useful for power transistors. . For example. measure the function-generator amplitude both with and without the speaker connected. First. (Be sure you have adjusted the supply voltages to not more than ±5 V so as not to overheat the transistors – they are rated for 200 mA collector current and 625 mW power dissipation. one PNP. 4. a push–pull buffer stage is often used. repeat the measurement at 10 kHz and compare.

(3) adding a few-hundredpicofarad cap to ground at the output. 4.3. For high-power applications.12.63 4 Bipolar transistors +5 2N3904 Vin 2N3906 Speaker −5 Fig. Since the diode and base–emitter voltage drops are unlikely to be exactly the same.3 Common-base amplifier Build the common-base ampliﬁer of Fig. Driving loudspeaker with push–pull buffer. try (1) rearranging your circuit so that the wire jumpers are as short as possible. The output waveform will display ‘crossover distortion’ – what does it look like and why does it occur? (Hint: for a transistor to be on. if it does. (2) adding a 330 resistor in series with the input. Two methods can be used to alleviate the crossover distortion: r shifting the bias points of the two transistors apart to minimize the portion of the cycle when neither transistor is on.11. The diode at the base should bias the emitter approximately at ground. a power-Darlington push–pull stage is often used. and why? What is the minimum input amplitude required for an audible output? Explain. r using feedback to apply a signal to the bases that compensates for the distortion. the input will have a small DC offset – how big is it? . 4. 4. Predict and measure the transistor’s quiescent currents and bias voltages. The circuit may oscillate because of inadvertent positive feedback from output to input. We will explore these techniques in Chapter 8. and if that’s not enough to stabilize it. there must be about 700 mV between base and emitter – is there a time during the cycle when neither transistor is on?) By how much does it reduce the output amplitude.

Common-base ampliﬁer. and the ampliﬁer is noninverting.12. 4. What do you predict for the input and output impedances? You can measure the input impedance easily using the 400 output impedance of the function generator: how much smaller does the function-generator output become when you connect it to the ampliﬁer input? What input impedance does this imply for the ampliﬁer? Explain. the input and output currents are almost equal.64 Hands-on electronics +15 10 k 10 k Vout 2N3904 Vin 100 22 k −15 Fig. Note that. in contrast to the common-emitter ampliﬁer. . Connect a small sine-wave input and determine the voltage gain.

two 2N5485 JFETs. oscilloscope. 65 . Fundamentally. The channel can be made of either N-type or P-type material (Fig. one 100 k. Qualitatively. In both types. Also. two 1 k.5 Transistors II: FETs In this chapter we introduce the ﬁeld-effect transistor (FET). ﬁeld-effect transistors (FETs) are threeterminal semiconductor devices capable of power gain. and you will ﬁnd that knowledge useful later on. Apparatus required Breadboard. due to the variability of FET parameters. 1. 0. a conducting channel between the drain and source terminals is controlled by a voltage applied to the gate terminal.3 k. 5. one 3. there are two types of FETs: junction FETs and metaloxide-semiconductor FETs (MOSFETs). however.0 F and 100 F electrolytic capacitors. and one 1 M 1 W re4 sistors. FET operation is easier to explain than that of bipolar transistors. but we will concentrate for today on junction FETs (JFETs). N-channel is more common since the conductivity of N-type semiconductor (in which electrons carry the current) is higher than that of P-type (in which holes do).1 Field-effect transistors Like bipolar junction transistors. they operate much like junction transistors. A majority of today’s integrated circuits are built using FETs of one type or another. it is essential that you master the basics of FET operation. many people ﬁnd FETs more difﬁcult to use.1 F ceramic capacitor. but they have much higher input impedance and lower transconductance and voltage gain. They come in a confusing variety of types. multimeter. two 10 k.1). they have a larger variation in their ‘VBE ’ equivalent (called VGS ) than bipolar transistors. 5. one 1N4733 Zener diode. As with bipolar technologies.

to distinguish the source from the drain.1). gate and source of an FET play similar roles to the collector. thus. i. The gate region of a JFET consists of material of opposite type to that of the channel. As the reverse-bias is increased. depending only slightly on the voltage VDS between the drain and source. Since the gate–channel diode is normally reverse-biased. Unlike the bipolar case. However. 5. Construction and circuit symbols of JFETs (note that other variants of these symbols are also used). gate. the source and drain are roughly interchangeable and it is possible for an FET to be used ‘backwards’. 5. even though the channel is physically spread out between the source and drain. the gate terminal is drawn at the source end. the arrow on the gate symbol indicates the direction of forward-bias. The voltage at which this occurs is called VP or VGS(off) . In our preferred symbol for JFETs (Fig.2).1. and emitter (respectively) of a bipolar transistor.1 FET characteristics The simplest way to think of FET action is as a voltage-controlled current source.1. 5. the gate current is extremely tiny (typically ∼ nanoamperes). In a JFET.66 Hands-on electronics G P D N P G N S D P N-Channel P-Channel G N N G P S D S D S G N-Channel JFET G P-Channel JFET Fig. the drain current ID is approximately constant for a given gate– source voltage VGS . 5. more and more current carriers are repelled out of the channel until it is ‘pinched off’ and the drain–source current drops to zero (see Fig. the channel conducts unless it is turned off by an applied reverse-bias voltage between the gate and the channel. JFETs are normally used with the gate–channel diode reverse-biased. so that for all . the gate and channel form a diode.e. As for any diode. Note that the drain.

2. This is called the saturation region of the FET characteristic (see Fig. 5. ID (mA) Saturation Region IDSS VGS = 0 V 20 Linear Region 15 VGS = −1 V 10 VGS = −2 V 5 VGS = −3 V 0 5 10 15 20 VDS (Volts) Fig.3). practical purposes the drain and source currents are equal (ID = IS ). Schematic representation of JFET operation: (a) gate–channel diode slightly reverse-biased.67 5 Transistors II: FETs VGS + Drain Gate Source (a) Gate VGS VDS + + VGS + Drain Gate Source (b) Gate VGS VDS + + Fig. 5. Idealized common-source characteristic curves for a JFET. 5. .3. (b) gate–channel diode highly reverse-biased (VGS ≥ VP so that channel is pinched off). The voltage-controlled current-source behavior occurs as long as the drain– source voltage VDS is sufﬁciently high.

the CA3140 MOSFET-input op amp with its teraohm input impedance. 5. and thus can give higher gain in ampliﬁcation applications. but that of an FET only as the square root of ID .2) As in the case of bipolar transistors. This has led to the common practice of combining FETs with bipolar transistors in analog integrated circuits to exploit the advantages of both. . the slope of the I –V characteristic is controlled by the gate–source voltage. IDSS and VP are temperature-dependent and vary substantially even among devices of the same type.1) where VP is the pinch-off voltage and IDSS is the saturation drain current for VGS = 0 (i. For FETs operated in the saturation region. This is the linear region of the FET characteristic. useful for automatic gain control (AGC) and modulation applications. Don’t confuse FET saturation with bipolar-transistor saturation – they are entirely different phenomena! For example. whereas FET saturation occurs at large VDS .g. this is only a model and should not be expected to be exact. a JFET behaves like a voltagecontrolled resistor rather than a current source.1. the relationship is quadratic: VGS VP 2 ID = IDSS 1 − . the transconductance is √ proportional to ID : gm = ID / VGS . Like the parameter β for bipolar transistors. Thus.68 Hands-on electronics For VDS smaller than a volt or two. Since the transconductance of a bipolar transistor increases linearly with IC ((gm )bipolar = 1/re ). bipolar transistors typically have higher transconductance than FETs for a given current. (5.2 Modeling FET action Recall that for bipolar transistors the collector current varies exponentially with the base–emitter voltage. (5. recall that bipolar-transistor saturation occurs at small VCE . so good designs minimize the dependence of circuit performance on them. e.e. i. gate shorted to source).e.

Measure IDSS . set the other end to −5 V.2. 5.1. 5. adjust VGS while measuring the drain current.1. The common-source characteristic curves for the 2N5485 can be measured using the circuit shown in Fig.4. 5. 5. Use the diode-test feature of the multimeter and verify this picture using a 2N5485 JFET (the pinout for the 2N5485 is shown in Fig. The pinch-off voltage will be equal to this gate voltage. Use one pot to adjust the gate . These curves illustrate the basic dependences among ID . Note: the pinout of the 2N5485 does not correspond to that of the 2N3904. If the meter reading ﬂuctuates. the drain current now equals (by deﬁnition) IDSS . 5. Use an ohmmeter to show that the drain and source are connected. more so than bipolar transistors.1.69 5 Transistors II: FETs 5. According to Eq. and VGS .2 Exercises 5. connected via the channel. ground the source. Explain why this will stabilize your measurement.1. try connecting the gate and source together using the breadboard and a small piece of wire. Plot ID versus VGS . set the drain voltage to 10 V. VDS . Why is this true? Verify Eq. To measure IDSS and VP . Adjust the pot until the gate is at ground while leaving the drain voltage at 10 V. With the drain voltage set to 10 V. the gate–source and gate–drain connections are PN junctions. and will be useful while performing the remaining exercises. and apply a variable negative voltage to the gate. What resistance do you measure? Next verify Eq. 5. Since JFETs are notoriously variable. adjust the gate voltage until the drain current drops to zero. and connect the slider to the gate). Using an ammeter to measure the drain current. Measure VP . the drain and source of an N-channel JFET occupy opposite ends of a single N-type region. try to use the same JFET for all exercises. whose emitter and collector are distinct N-type regions separated by the P-type base.4).1 FET characteristics As shown in Fig. This can be accomplished using either the 1 k or 10 k potentiometer as a voltage divider (ground one end of the pot. Unlike an NPN transistor.

record and plot your measurements. and IDSS .2 FET current source Adding a resistor improves the JFET’s current-source performance compared with that of the bare JFET. Repeat your measurement procedure for several VGS values between VP and zero. What is VGS for each data point? . This is the ‘linear region’. Record and plot your measurements. Plot the data on a single graph and clearly label each curve. Hook up the circuit shown in Fig. the saturation region.5 and measure the dependence of the drain current on the drain–source voltage as you adjust the pot. resulting in approximately constant current.5 V more positive than VP (keep in mind that VP is negative!). the current should then ‘saturate’ at an approximately constant value.2. Use the scope probes to measure the drain and gate voltages while using a meter to measure the drain current.70 Hands-on electronics (a) (b) 0–15 V 85 54 2N T E JF A D −5 to 0 V G S D S G Fig. Circuit for measuring the common-source characteristic curves. As you further increase VDS . Since VGS > VP . voltage and the other to adjust the drain voltage. Using the pot. JFET saturation occurs because the increasing drain voltage creates an increasing depletion region between the gate and drain. the current should increase linearly with drain voltage. For VDS less than a few volts. 5.4. with an equilibrium (of sorts) created. The size of the depletion region (and thus the resistance of the channel) increases approximately linearly with drain–source voltage difference. indicating the linear region. 5. adjust VGS to be 0. 5. in which the JFET acts as a voltagecontrolled variable resistor. Slowly increase the drain voltage from zero to 15 V while measuring the drain current. the channel will never be pinched off completely.

6. 5. suppose ID increases. with ID determined by IDSS and VP according to Eq. If the input voltage increases.1. the FET is thus in the saturation region.71 5 Transistors II: FETs open +15 10 k pot A 2N5485 1k RS Fig. . This is because negative feedback is at work. the 1 M resistor is small enough to bias the gate very near ground. the channel opens. Given the nanoampere gate current. increasing the magnitude of VGS and moving the FET closer to pinch-off.2. how constant is the current? Calculate the approximate output impedance in the saturation region (see Eq. Self-biasing JFET current source. and ID increases. Even though VGS is not exactly constant as VDS is varied. You can compute VGS from ID and the known resistance of RS .4. Quiescently. You should see the drain current start to vary substantially as you make the transition from the saturation region to the linear region. What is the compliance? Within the saturation region. Build the circuit of Fig. For example. Compare the performance of this current source with that of the bare JFET and of the bipolar current source that you built in section 4.13). The operation of this circuit is reasonably straightforward.2.2. 5. this circuit actually works better (has larger output impedance) than one in which VGS is held constant. 5. thus decreasing ID .2. then. so does the drop across RS .3 Source follower As with the emitter follower from section 4. 3.5. even though the voltage gain is less than or equal to unity. a JFET source follower provides power ampliﬁcation via current buffering. 5. and try it out with a 1 kHz sine wave of small amplitude. VGS moves closer to zero.

VGS1 . since the dynamic resistance of the source (= 1/gm ) forms a voltage divider with RS . the channel closes and Vout drops.6. this should also be true for Q 1 and R1 . 5. What are ID .) Try it out. What is the DC offset at the output? Measure the voltage gain. You can improve the source follower by providing it with much higher load resistance. derive a value for gm and compare with that of a bipolar transistor at the same current. Note that if the two 2N5485s approximately match in their characteristics. When Vin decreases. Thus the output voltage must follow the source voltage of Q 1 . but was much smaller due to the bipolar transistor’s larger value of gm .) Draw a diagram of this voltage divider. it is actually a current sink. and VGS2 ? . Vout follows the source. Furthermore. (Since we are using N-channel JFETs. Measure the voltage gain and the DC offset from input to output.7. Thus. as in the clever circuit of Fig.1 µF 2N5485 Vin Vout 1 MΩ RS 1k Fig. a current-source load is often used. Since an ideal current source would have inﬁnite resistance. From your observed attenuation. You should see that the voltage gain is less than unity. (This effect was also present for the bipolar transistor. 5. but the DC offset is small: the constant current due to Q 2 creates a constant voltage drop across R1 . since the gate of Q 2 is at the same voltage as the bottom of R2 .72 Hands-on electronics +15 0. it can be constructed by adding another FET. Explain the operation of this circuit in your own words. Source follower. not only is the voltage gain unity. to the extent that the two FETs (and the two resistors) match.

5.73 5 Transistors II: FETs +15 0. the 2N3958 dual JFET. predict the quiescent gate–source bias voltage and output voltage for this . Using the common-source curves that you measured in section 5.2. 5. The principle at work is essentially the same as for the common-emitter ampliﬁer: a varying input voltage controls a varying current.3 k RD Vout 0.g. 5. which the drain resistor ‘converts’ to an output voltage.0 µF Fig. JFET ampliﬁer.7.4 JFET amplifier Construct the ampliﬁer shown in Fig. +15 3. e. Source follower with current-source load.1 µF Vin + 1 MΩ 1k RS 1.1 µF Vin 1 MΩ Q1 2N5485 R1 1k Vout Q2 2N5485 R2 1k −15 Fig. Such a circuit is often used in the input stage of an oscilloscope.8. 5.2. The offset can be much improved by using a matched FET pair.1.8.

when would you choose a bipolar transistor over a JFET or vice versa? Feel free to include any general comments you have on the experience you’ve gained from the last few chapters. For example. Switch the input to a triangle wave and adjust the frequency widely. then VG = VGS . The voltage gain (A) is deﬁned as the ratio of the output amplitude to the input amplitude. ID · RD . The source capacitor is used to ‘ﬁx’ the source voltage even as the drain current ﬂuctuates due to the AC input.74 Hands-on electronics ampliﬁer. How reproducible are the results? Comment on the design and operation of simple transistor and JFET circuits. (This trick can also be used to increase the voltage gain of the bipolar-transistor common-emitter ampliﬁer. and since A = gm · RD . As discussed previously. . measure the voltage gain and compare with the expected gain. If the source voltage remains ﬁxed. How much power is dissipated by the FET? What is the input impedance? Is this an inverting or noninverting amp? Explain why. How does this change things? Try several different 2N5485s and record the voltage gain and quiescent drain current and output voltage. What is the predicted voltage gain for this ampliﬁer? Using a 1 kHz small-amplitude sine-wave input. gm = VD = ID / VGS .) This implies that the gain will be frequency-dependent. Explain what you see. Power the ampliﬁer and compare the measured values with your predictions. Replace the source capacitor with a 100 F capacitor.

2 k. We will also look at some other circuits that serve as building blocks for op amps. three 2N3904 and three 2N3906 transistors. it ampliﬁes the voltage difference of its two inputs. Apparatus required Breadboard. one 5. Probably the most surprising thing about op amps is their very large voltage gain. as it is the basis of the operational ampliﬁer (or op amp).6 Transistors III: differential amplifier In this chapter we will study the transistor differential-ampliﬁer circuit. two 22 k one each of 560 . multimeter. ﬁve 10 k. i. differential ampliﬁers make it easy to build circuits that use negative feedback.1 V Zener diode. the two circuits perform entirely different operations. This is useful in two important ways: 1. Don’t confuse the differential ampliﬁer with the differentiator: although the names sound similar. What does each do? 75 . 2. 2. one of the most useful devices for analog signal processing. and 100 k 1 W resistors. This is a very important transistor circuit. As we will see in future chapters. A differential ampliﬁer can be used to amplify a differential signal (the voltage difference between the two inputs) while suppressing any noise that is common to the two inputs.1 Differential amplifier A differential ampliﬁer is an ampliﬁer for differential input signals. three 100 . usually exceeding 100 000.e. oscilloscope. 4 6. This chapter will give you a clearer idea how such performance is achieved.

This causes a current I to ﬂow across the two emitter resistors.2 Expected differential gain The differential voltage gain can be understood as follows.1 Operating principle Fig.1. (a) Differential ampliﬁer. as explained in more detail below. the current through R1 is shared equally between Q 1 and Q 2 . Since these are in series with the dynamic emitter resistances of the two transistors (each of which has an approximate value re = 1/gm = 25 mV/IC according to the Ebers–Moll model). a differential output signal. 6. I . (b) function generator with 100-to-1 attenuator.76 Hands-on electronics (a) 10 k RL +15 RL 10 k (b) Vout (−) Q1 2N3904 100 RE R1 −15 10 k 100 RE Q2 Vout (+) Vin(−) ~ Function generator 10 k Vin(+) Vin(+) 100 Fig. Suppose we apply a differential signal consisting of equal and opposite changes in the base voltages of Q 1 and Q 2 : Vin+ = − Vin− ≡ V. some collector current shifts from one transistor to the other.1.1(a) shows a differential ampliﬁer made of two NPN transistors. This change.e. If a differential input signal is applied to the bases of Q 1 and Q 2 . You can think of it as a ‘current divider’. 6. (6. (6. a similar voltage change occurs between the two emitters.1) Since the emitter voltages follow the base voltages.1. 6. 6. i. causes a positive voltage change at one collector and a negative voltage change at the other.2) . we have I = V /(RE + re ). Quiescently.

. you can display the functiongenerator output on the scope and calculate from it the size of the actual input signal to the ampliﬁer. Now that you know the attenuation. Compare your observed attenuation with the theoretical value. beyond which the output cannot go. and observe the ampliﬁer outputs at the collectors of Q 1 and Q 2 . In other words. 6. the current through R1 is constant. 6. The ampliﬁer circuit clips its output when all of the available current has been switched to one transistor or the other. To avoid these problems. you will need to use a small enough amplitude out of the function generator so that the ampliﬁed signal is not ‘clipped’ (cut off at the top or bottom). At what output voltages does clipping set in? Compare with what you expect – you can estimate the amount of current that is available to either transistor by measuring the voltage drop across R1 . check it against your schematic (Fig. run the function-generator output through a 100-to-1 attenuator (voltage divider) made from a 10 k resistor in series with a 100 resistor to ground (Fig. This means that the current due to the differential signal must add to the collector current of one transistor and subtract from that of the other transistor. Because it has a large gain.3 Measuring the differential gain Wire up the circuit. a small differential input voltage applied to the two bases causes a large differential output signal at the two collectors. and try it out. it does not change the voltage at the top of R1 . Onehalf of the ampliﬁed differential signal appears at each collector. However.1). thus. compare with what you expect. Measure the attenuation (voltage-division ratio) by setting the function generator for a large amplitude and measuring the signal amplitudes before and after the voltage divider.77 6 Transistors III: differential amplifier Since a purely differential signal raises the voltage at one emitter by the same amount that it lowers the voltage at the other emitter. Try it and see. Measure the differential voltage gain Adiff = (Vout+ − Vout− )/ (Vin+ − Vin− ) for a few different input amplitudes and frequencies.1. Connect the output of your attenuator to the base of Q 1 .1(b)). 6. this determines maximum and minimum voltages. it puts out a rather noisy signal with a relatively large DC offset. when the function generator is set to a small amplitude. so the differential voltage gain should be RL /(RE + re ).

since often the noise is in common on both inputs and can be subtracted away by a differential ampliﬁer. Vin+ = Vin− ).e. but you will probably ﬁnd that. .1. How big an input offset voltage do you obtain this way? 6. they are not. identical signals applied to both inputs. the two emitter resistors. it can still be a nuisance in practice. To obtain exactly equal outputs you would have to input a small voltage difference. What do you observe at the outputs? You should see almost identical signals on both outputs. Measure how much current your current sink sinks. Test the common-mode rejection of your differential ampliﬁer: Connect both inputs to the same sine wave from the function generator (i. called the input offset voltage. It could be reduced by increasing the size of R1 .e. and compare. 6. What common-mode gain (ACM + = Vout +/ Vin + and ACM − = Vout −/ Vin −) do you observe for each output. but that would reduce the amount of current ﬂowing through Q 1 and Q 2 . due to small mismatches between the two collector resistors. This feature (called common-mode rejection) is useful when sensing a small signal in the presence of noise. calculate what you expect.78 Hands-on electronics 6.3) What value of CMRR do you observe? Although the common-mode gain of this circuit is small. what are the output voltages? For an ideal differential ampliﬁer. A better solution is to replace R1 with a current source (shown in Fig. i. it is a current ‘sink’ since positive current ﬂows into the collector as indicated.1. 1 Strictly speaking. and the two transistors. ACM (6. increasing re and reducing the differential gain.5 Common-mode gain It is desirable for a differential ampliﬁer to be insensitive to commonmode input. they should be equal. which you can estimate as the output-voltage difference divided by the voltage gain.1 Try this. and why? Common-mode rejection is usually speciﬁed in terms of the commonmode rejection ratio expressed in decibels: CMRR = 20 log10 Adiff .4 Input offset voltage If you ground both inputs.2).

.1 Current mirror To achieve high gain. This determines the current out of Q 4 by Ohm’s law applied to R. What output signals do you observe now for a common-mode input? What is the common-mode gain now? It should be very much smaller than previously – if the output signal is so small that you have trouble measuring it. Op amps are manufactured as integrated circuits.1 V −15 2.2. it is held at a VBE drop below the positive supply. 2 To set an upper limit. you can at least set an upper limit on it and on the common-mode gain.2.2 Op amps and their building blocks An operational ampliﬁer is a differential ampliﬁer with a single-ended output and as high a differential gain as possible (typically >105 ). A current mirror is a convenient conﬁguration for this purpose. They are typically used with DC coupling and with negative feedback from output to input. assume that the output signal equals the precision of your measurement. in op amps the emitter resistors are typically omitted. Their internal design includes level-shifting circuitry so that the single output is at ≈0 V if the two input voltages are equal. 6. and the collector resistors are replaced by current sources. Build the PNP current mirror of Fig.3. Current sink for differential ampliﬁer. 6. 6. The current out of Q 4 ‘programs’ an approximately equal current out of Q 5 as follows: since Q 4 ’s collector is connected to its base. 6.2 Save your three-transistor differential ampliﬁer for use below.2 k Fig.79 6 Transistors III: differential amplifier ~ 2 mA 560 2N3904 1N4733 5.

But ﬁrst.4). Monitor the collector voltage as you adjust the load resistance. and remove the 100 emitter resistors (see Fig. as shown in Fig. How does the output current vary with collector voltage? What is the approximate dynamic resistance of the output? Is this a better or a worse current source than the ones you built in previous chapters? 6. You can explore this using a variable load. the currents through Q 1 and Q 2 will be equal. since β increases with VCE . the gain would be 10 M/2re . 6. and thus their collector currents will match if they have matching values of Is and β and are at the same temperature.2. as we will see below. both transistors have the same value of VBE . 6. the current mismatch between Q 2 and Q 5 must ﬂow through the 10 M scope-probe input impedance. . If the inputs balance. Also. hook up your simple current mirror to your three-transistor differential ampliﬁer. there will always be a slight current mismatch since the programming current includes the base currents of Q 4 and Q 5 and the output current does not. 6. Explain the last statement. In practice. Current mirror.3. the current mismatch will depend on the output voltage. Connect a scope probe to the collector of Q 2 . Since the bases are connected together. But if a differential signal is present at the input. replacing the 10 k collector resistors.2 Differential amplifier with current-source loads Various tricks can be used to improve the performance of the current mirror. You should then see an enormous differential gain! If Q 5 were an ideal current source. and so will the currents through Q 4 and Q 5 .80 Hands-on electronics +15 2N3906 Q4 Q5 Iout R 10 k Iprog 10 k pot 1k load Fig.3. and VBE at a given collector current depends slightly on VCE (called the ‘Early effect’).

2.1 V −15 2. What gain do you observe? Compare with the gain you would expect if Q 2 ’s effective load were 10 M.2 k Vin(−) offset adjust +15 1 k pot 22 k Fig. with one end of the pot connected through a resistor to +15 V and the other end connected through a second resistor to −15 V. the collector of Q 5 will have an output impedance less than 10 M. 6. Observe how the output changes and suggest an explanation for your observation. Also note that a gain value approaching 1000 is not unreasonable. but to the wiper of the 1 k pot.81 6 Transistors III: differential amplifier +15 2N3906 Q4 Q5 Vout Q1 2N3904 Q2 Vin(+) 560 Q3 −15 22 k 1N4733 5.4.1? The sensitivity of this circuit is easily demonstrated. you will want to arrange for the quiescent collector voltage of Q 2 to be around 7 V – you can adjust it by hooking up the base of Q 2 not to ground. . and so the gain will be lower. Try doing the same for the other transistor.4. Also. the output voltage will be sensitive to small variations at either input. In practice. Due to the high gain of this ampliﬁer. You may ﬁnd it desirable to increase the attenuation of your input voltage divider from 100 to 1000 here. as shown in Fig. Try warming either Q 4 or Q 5 by gently squeezing the transistor between your ﬁnger and thumb. Differential ampliﬁer with current-mirror load. it is likely that you need to ﬁne-tune the offset voltage using the 1 k pot. What effective load resistance do you infer? Is this consistent with the dynamic resistance of the current-mirror output you observed in section 6. to avoid clipping of the output signal. 6. If you observe a DC output signal near either ground or +15 V.

Differential ampliﬁer with Wilson-current-mirror load. What effective load resistance do you infer now? +15 VBE Q4 VBE Q5 Q6 Vout Q1 Q2 Vin(+) 560 Q3 2 mA Vin(−) offset adjust −15 22 k 1 k pot 22 k +15 1N4733 5. Do this by adding a third PNP transistor as shown in Fig. This clever circuit beats the Early effect by ﬁxing Q 5 ’s VCE – at the same time. it also symmetrizes the base-current mismatch of the simple current mirror.5.2 k Fig. Try this to see by how much the gain increases.1 V −15 2.4 Wilson current mirror You can do even better by converting the simple current mirror to a Wilson current mirror.82 Hands-on electronics 6.2. 6. Note that you will need to re-tune the collector voltage of Q 2 by adjusting the offset voltage. Explain.3 Improved current mirror You can improve the output impedance of your current sources by adding a small resistor in series between the positive supply and the emitter of each PNP transistor. What effective load resistance do you infer now? 6. Try this with two 100 resistors and see by how much the gain increases.2. 6. .5.

To achieve low output impedance. of course. Higher input impedance can be achieved by using Darlington transistor pairs in place of the input transistors. or by using FET inputs instead of bipolar transistors.83 6 Transistors III: differential amplifier IC op amps have even higher gain than this. as well as higher input impedance. the output is usually buffered with an additional transistor stage. The gain can be increased further by adding a second stage of ampliﬁcation after the differential pair. .

.

. Apparatus required Breadboard.g. whose outputs have typically only two states). The 741 is a very popular and successful design. one 1 M 1 W resistors.1 The 741 operational amplifier The IC we shall be using is a general purpose op amp designated by the number 741. oscilloscope. the 741 is available in various package styles and is rated for use in various temperature ranges. Thus. but the 741s made by different manufacturers are all electrically compatible. an IC op amp is often the device of choice in scientiﬁc instrumentation. multimeter. 7. three 10 k. and 4 four more 10 k resistors (optional). The one we use is the 85 .7 Introduction to operational amplifiers An operational ampliﬁer is a high-gain DC-coupled ampliﬁer with differential inputs and single-ended output. two 100 k. e. Each manufacturer has a different system of nomenclature for ICs. In this chapter we will introduce the op amp and its most common applications. Nowadays they are packaged as integrated circuits (ICs). one 100 . National Semiconductor calls the 741 an LM741. It is available from most manufacturers of linear integrated circuits (chips that produce an output proportional to their inputs. Fairchild a A741. two 741 op amps. etc. To add complication. useful for signals from DC to beyond audio frequency (though in recent years FET-input op amps such as the LF411 have been gaining on the 741 in popularity). one further 741 (optional). Op amps were originally developed as vacuum-tube circuits to be used for analog computation. as opposed to digital ICs. and their use avoids the necessity of coping with the messy internal details of ampliﬁer circuitry. Such devices can closely approximate the behavior of an ideal ampliﬁer.

there is a dot next to pin 1. The package is a rectangular piece of black plastic containing the silicon chip itself as well as the ﬁne gold wires that connect the chip to the contact pins. i. 741C (commercial temperature range. . and one capacitor. 0 to 70◦ C) in the eight-pin mini-DIP (dual in-line pin) plastic package. this is not to be confused with a differentiator! Recall that a differentiator outputs the time derivative of its input signal. 7. 7. as we shall see.) 7. the 741 actually has on its single silicon crystal a multiple-stage twoinput DC ampliﬁer consisting of twenty transistors. eleven resistors. Often. Look at the top face of the package and orient yourself as to which pin is pin 1 – looking at the top.1. while pins 5–8 are on the opposite edge.86 Hands-on electronics offset null 1 inverting 2 input noninverting 3 input V− 4 741 op amp (top view) − + 8 no connection 7 V+ 6 output 5 offset null Fig. Diagram of eight-pin DIP 741 package showing ‘pinout’.e.1 shows the layout of the eight-pin DIP package. You can ﬁnd the manufacturer’s data sheet for the 741 on the web or in just about any electronic textbook or linear-IC data book.1 741 pinout and power connections Fig. the Pentium chip had 3. 7. as well as in many other applications.1 Its output voltage is proportional to the voltage difference between the two inputs.1.1. Pins 1–4 are located along one long edge of the package. this is nothing – for example. The end at which pin 1 is located is indicated by a special mark – depending on the manufacturer. not necessarily the mark shown in Fig. in addition to (or instead of) the notch at the ‘pin 1’ end of the package. A differential ampliﬁer is convenient when one wants to study the difference between two almost identical signals. (Of course. of course. not at all what a differential ampliﬁer does. This package is convenient because the two rows of pins easily straddle the groove running down the center line of a breadboard socket block. As shown on the data sheet.3 million transistors. it is a ‘differential’ ampliﬁer. 1 Nowadays. you see the pins bending away from you and the pin numbers increasing in the counterclockwise direction.

1. Also note: Even though the power connections are usually not shown on schematic diagrams of op amp circuits. If they are too bent to plug into the sockets. r the bandwidth (frequency range over which the op amp can correctly operate) and slew rate (rate at which the output voltage can change) of an ideal op amp are inﬁnite. While. The above statements are approximately true for practical op amps – for example. . Note: In the circuits shown below. Next. turn off the power and insert the op amp into the breadboard. with pins 1–4 toward the left and 5–8 toward the right.87 7 Introduction to op amps The 741C is rated for maximum supply voltages of ±18 V. straighten them carefully. turn on the breadboard power and adjust the power supplies to +15 V and −15 V. and the recommended range is ±(≤15) V. preferably using needle-nose pliers. To be on the safe side. the pin numbers have been omitted. Note that the pins are delicate and are easily bent or broken. it is impossible to build a circuit having these ideal characteristics. to reduce the possibility of confusion. or the op amp won’t do anything! 7.2 An ideal op amp For a (hypothetical) ideal op amp. Run wire jumpers from the V+ pin to the +15 V bus and from the V− pin to the −15 V bus. in other words r the common-mode gain and DC offset of an ideal op amp are zero. one can come remarkably close. It is good practice for you to write in the pin numbers yourself before hooking up the circuits. of course. zero volts between the inverting (marked ‘−’ in schematics) and noninverting (marked ‘+’) inputs would yield zero volts at the output (relative to ground). Trying to work out pin numbers ‘on the ﬂy’ and keeping them in your head instead of writing them down is a common cause of errors in hooking up circuits. one can buy op amps with gains of 105 to 106 . r the differential gain and input impedance of an ideal op amp are inﬁnite and the output impedance is zero. straddling the central groove of a socket block. Any point of your circuit that is to be ‘grounded’ should be attached to the common (or ground) bus. the positive and negative supplies must always be connected. before you begin to build your circuit. Moreover.

If you think about it.e. Note the negative feedback resulting from the resistor that connects the output to the inverting input. 7.2 shows an op amp conﬁgured as an inverting ampliﬁer.e. any nonzero voltage at the inverting input will cause a large output voltage of the opposite sign. 7. Since the gain of the op amp is large and the noninverting input is grounded. We shall see next that these approximations lead to a very simple way of analyzing op amp circuits. Assuming that the open-loop voltage gain (i. and designating the voltage at the inverting input as V− . In other words. 7. Op amps are almost always used with negative feedback.1. i.2) . I2 = I1 . Assuming the input currents of the op amp are zero. the op amp will do whatever is necessary to zero the voltage difference at its inputs. Applying Ohm’s law to R1 and R2 .1) (7. Op amp inverting-ampliﬁer circuit. you will see that the only stable situation that can result is that the voltage difference between the inverting and noninverting inputs is zero. The idea is that a fraction of the output signal is applied at the inverting input. that without any feedback) of the op amp is inﬁnite. all the current ﬂowing in through R1 must ﬂow out through R2 . it is easy to compute the gain of the op amp inverting ampliﬁer.3 Gain of inverting and noninverting amplifiers Fig. V− = Vin − I1 R1 = 0 ⇒ I1 = − Vin R1 (7. Once this principle is grasped.88 Hands-on electronics R2 I2 R1 I1 − op amp Vin ~ Vout + Fig.2. The key principle at work in this circuit is negative feedback. the voltage difference at the op amp’s inputs must be zero.

.6 follows from the assumptions that all of the current ﬂows around the op amp and that feedback forces the inverting input to follow the signal applied to the noninverting input. since all of the current due to the input signal ﬂows around the op amp. I1 R1 R1 (7. R1 (7.89 7 Introduction to op amps Vout = V− − I2 R2 = −I1 R2 ⇒ Vout = − R2 Vin . Eq.3 shows an op amp conﬁgured as a noninverting ampliﬁer. application of Ohm’s law shows that the closed-loop voltage gain of this circuit is Av = I1 R1 + I2 R2 R2 =1+ .6) As above. R1 I1 R2 I2 − op amp + Vin Vout ~ Fig. Again. So if R1 = 10 k. a gain of 11 can be achieved by choosing R2 = 100 k. the gain with feedback) of this circuit is Av = Vout R2 =− . Thus. if R1 = 10 k. while a gain of 2 results from choosing R2 = 10 k. the closed-loop voltage gain (i. Vin R1 (7. a gain of −10 can be achieved by choosing R2 = 100 k. and a gain of −1 results from choosing R2 = 10 k. 7. Fig.4) Thus.5) In a nutshell. neglecting the tiny input currents of the op amp and assuming inﬁnite openloop gain.e. the output voltage is determined entirely by Ohm’s law applied to R1 and R2 . 7. Op amp noninverting-ampliﬁer circuit. 7.3) (7.3.

most op amps have pins that allow for external adjustment of the offset. The input offset voltage must be considered when designing small-signal or high-gain circuits. The output does whatever is necessary to maintain the two inputs at equal voltages. the input offset voltage is speciﬁed to be less than 6 mV. . You can think of this as the equivalent of the base current for a bipolar transistor.4 Op amp ‘golden rules’ We see from the above that to understand to a reasonable approximation an op amp circuit with negative feedback. For the 741C op amp. 1. The input offset voltage refers to the DC voltage that must be applied at the input to achieve exactly zero volts at the output. The slew rate can be a serious limitation at large output amplitudes and high frequencies. High-speed op amps are available with slew rates of 2000 V/ s. in most situations. Although these currents are small. real op amps require a small input bias current for proper operation. Slew rate The output voltage of a real op amp cannot change instantaneously. The inputs draw no current.1. however.5 V/ s.90 Hands-on electronics 7. 2. For the 741C op amp. the input bias current is speciﬁed to be less than 500 nA. The nonzero offset arises due to manufacturing limitations. you need only two simple rules. The 741 slew rate is typically 0. The maximum rate at which the output can change is called the slew rate. however. they are not zero. 7.5 The nonideal op amp Although the ideal op amp approximations are very close to reality.1. and is typically in the range of volts per microsecond. here are a few limitations that you should consider: Input offset voltage A small DC output voltage usually results even when the inputs are identical. or the gate current of a JFET. Input bias current An ideal op amp would draw no current at its inputs.

Still. 7. . Open-loop op amp test circuit. 7. Vary the input in steps of 100 or 200 V over a range that causes the output to vary from its maximum negative voltage to its maximum positive voltage.91 7 Introduction to op amps 7. op amps are never actually used in this way.e. you can vary the input between −15 and +15 mV.1 Testing open-loop gain First. you may ﬁnd that no setting of the input voltage causes the output to sit near zero volts.2 Experiments 7.. taking several readings of input and output voltage as you do so.4. however. try a 741 in the open-loop circuit shown in Fig. What are the op amp’s positive and negative output saturation voltages (i. you should +15 10 k 100 k −15 _ Vout + 100 Fig. the output voltage is very susceptible to input noise. Since your applied input voltages are so tiny.2.4. Try to adjust the input so that the output sits near zero volts. (You may ﬁnd that the output switches states within a single step!) Approximately what input voltage would result in an output voltage near zero? This is the input offset voltage of your 741.) The 1000to-1 attenuator in the input circuit means that. open-loop means that there is no feedback connection between the output and input. and you may also observe substantial drift of the output voltage with time. (As mentioned above. the maximum voltages it can output)? If your chip has particularly high gain. by adjusting the pot. you can still set upper and lower limits on the input offset voltage by determining at what input voltages the output switches between its negative and positive saturation voltages.

) Due to feedback.2.92 Hands-on electronics be able to estimate the gain roughly. Measure the slope of the leading edge in volts per microsecond.5 V/ s. Using a 1 kHz sinusoidal input signal with an amplitude of approximately 1 V. Measure the gain and compare with what you expect. 7. analyze the voltage-divider circuit consisting of the output impedance in series with the load resistance. Compare the appearance of the output signal and the input signal. the 100 resistor. as you will soon observe. According to the manufacturer.) Remove the 100 load and switch to a 10 kHz square wave as the input. For sufﬁciently large amplitude. the output impedance of the inverting ampliﬁer is extremely small: the ≈75 open-loop output impedance of the 741 is multiplied by the ratio of the closed-loop gain to the open-loop gain. What gain do you observe? Is it consistent with the manufacturer’s speciﬁcation ‘gain (typical) = 200 000’? You may ﬁnd that stability and noise are improved by keeping the connections between the noninverting input. with R1 = 10 k and R2 = 100 k. To see that this gives sensitivity to the output impedance. Conﬁrm the small size of the closed-loop output impedance by adding a 100 load to ground at the output. and the inverting input as short as possible – long wires in this current path act as an antenna and pick up electromagnetic noise from the environment. Replace the 100 k feedback resistor with 10 k – what gain do you predict and observe now? (Put the 10 k resistor back for the following parts.2. Verify that the ampliﬁer inverts.2 Inverting amplifier Now hook up the inverting ampliﬁer of Fig. This is the slew rate of the 741C. . Do you expect the output amplitude to decrease measurably under load? How big a decrease do you predict? Is this consistent with what you observe? (You will need to use a small input signal – how small? – to avoid running into the 741’s ≈25 mA output-current limit. here a factor of order 10−5 . There exist more expensive op amps with much higher slew rates. look at the input and output. or at least place a lower limit on it. you are likely to observe that the leading and trailing edges of the output square wave are not quite vertical. the typical slew rate is 0. 7.

3 with R1 = 10 k. it acts here like a short circuit to the noninverting input (ground). Since the input signal sees R1 to ground. The inverting input is called a ‘virtual’ ground because it is kept at zero volts by feedback. 7. Compare with what you expect. consider that this is the input impedance of the attenuating .2. Since the inverting input is held near ground due to feedback.5.93 7 Introduction to op amps +15 10 k pot −15 V0 R2 10 k I2 R1 10 k I1 R3 10 k I1+I2 − Vout + Vin ~ Fig. (Leave the 100 k resistor in for the following parts. as shown in Fig. Explain your result.3 Noninverting amplifier Set up the noninverting ampliﬁer circuit of Fig. If R1 = R2 = R3 . 7. Note that the virtual ground at the inverting input can also be used as a ‘summing junction’: all currents arriving at that point are summed and passed through the feedback resistor to the output. Verify that the ampliﬁer is noninverting. If your answer is 10 M. To demonstrate this. where I1 = Vin /R1 and I2 = V0 /R2 .) Try to measure the input impedance by putting a 1 M resistor in series with the input and looking at the signal before and after the resistor. Vout = (I1 + I2 ) · R3 . With a 1 kHz sinusoidal input. Vout = Vin + V0 . What happens to the output DC offset as you adjust the pot. feed an adjustable current into the summing junction. the input impedance of the inverting ampliﬁer is equal to the value of R1 . 7. measure the gain with R2 = 100 k and with R2 = 10 k. even though the open-loop input impedance of the inverting input is very large. Thus. by adding a 10 k pot with 10 k resistor in series with the wiper. The input impedance can be inferred by analyzing the voltage-divider circuit consisting of the 1 M resistor in series with the input impedance.5. and examine the ampliﬁer output for a small-amplitude 1 kHz sinusoidal input. Circuit for demonstrating a summing junction. and why? 7.

2. After turning on the power.94 Hands-on electronics scope probe! You can get around it by looking only at the output. .6(a). and connect the output of the function generator to the ‘+’ input of the op amp. and connect the other to any convenient spot on the breadboard. What value do you get for Z in this way? Again you are seeing the effect of negative feedback: even though the openloop input impedance of the noninverting input is only of order megohms. here a factor of order 105 . Given that the scope probe has an input impedance of 10 M . Using your scope probe. multiplied by the ratio of the open-loop gain to the closedloop gain. such as the capacitance to ground at the noninverting input. conﬁrm that the input and output signals are identical and that the voltage follower is noninverting. it is the same circuit! Explain this last statement.4 Voltage follower As shown in Fig. (Hint: what was the only thing you had to change to make the noninverting amp from the inverting amp?) 7. Z in is limited by other effects. it is.. estimate the power of this signal (i. Leave one end of your coil ﬂoating. Record the input and output amplitudes. Record and explain what you see. Of what possible use is an ampliﬁer of unity gain that does not even invert the signal? As an illustration of the voltage follower’s usefulness. (In practice. What value for the input bias current is implied by the observed DC shift? The output impedance should of course be the same as for the inverting ampliﬁer.e. with and without the 1 M input resistor. coil a long wire (30 to 50 cm in length) around the outside of an AC power cord.) Be careful not to be confused by the DC shift in the output produced by the 1 M input resistor! Explain how this shift results from the op amp’s input bias current. since. (Any power cord will do provided that it is plugged in!) The breadboard cord is usually the most convenient. connect the output of the op amp directly to the ‘−’ input. in principle. as far as output impedance is concerned. how much power is dissipated by the scope probe?). observe the waveform. 7.

7. . Now. 7. 7. if these inverting and noninverting ampliﬁers have equal gains. Try amplifying this low-power signal using the inverting-ampliﬁer circuit previously constructed. Explain your observations. whereas a noninverting amp is created if Vin− is grounded. If both inputs are connected to signals. instead of driving the ampliﬁer directly.6.95 7 Introduction to op amps − Vin + (a) 100 k Vout − Vin + 1 MΩ 10 k − + Vout (b) Fig. 60 Hz noise is pervasive throughout North America (50 Hz in Europe) and is often the dominant background noise in electronic equipment. with the output voltage equal to the difference of the two input voltages. The gains are matched provided that R1 /R2 = R3 /R4 . choose a smaller feedback resistor to reduce the gain of the inverting amp. the difference between the two signals is ampliﬁed.5 Difference amplifier Fig. the output is proportional to the difference of the inputs. A difference ampliﬁer is both an inverting amp and a noninverting amp. 7. You can reproduce this effect more simply by touching the probe tip with your ﬁnger.6(b). Sketch the output and record your observations. (b) voltage follower as the input stage to an inverting-op-amp circuit.7 shows the 741 conﬁgured as a difference ampliﬁer. (a) Op amp voltage follower. If the ampliﬁer output saturates. This follows since.2. Record the follower output as well as the ampliﬁer output. insert a voltage follower as shown in Fig. An inverting amp is created if Vin+ is grounded.

96

Hands-on electronics

R2 100 k

Vin_ Vin+

10 k pot

R1 10 k

−

Vout

+

R3 R4

Fig. 7.7. Difference ampliﬁer. The parts of the pot on either side of the slider serve as R3 and R4 .

Difference ampliﬁers are often used in the life sciences where signals are small and exist within a noisy environment, e.g. in the electrocardiograph (ECG). Any background noise common to both inputs (common-mode noise) is rejected, while the signal of interest (present at only one of the inputs) is ampliﬁed and appears at the output. A potentiometer is often used (as in Fig. 7.7) to tune the gain and common-mode rejection of the ampliﬁer. The quality of the ampliﬁer is measured (in part) by the common-mode rejection ratio (CMRR), based on the ratio of the differential voltage gain and common-mode voltage gain: CMRR = 20 log (Adiff /ACM ). (7.7)

It is customary to give the CMRR in decibels, as shown in Eq. 7.7. The 741 general-purpose op amp is a differential ampliﬁer with a CMRR value speciﬁed to be at least 70 dB. Precision op amps are commercially available with CMRR values as high as 140 dB or more. Build the circuit shown in Fig. 7.7. As always, be careful not to short the potentiometer’s center tap to ground or power. You can estimate the common-mode rejection ratio by measuring both the common-mode voltage gain and the differential voltage gain. The common-mode voltage gain is determined by applying identical signals to both inputs and observing the output voltage: ACM = VoutCM /VinCM . Using a 1 kHz sine wave at maximum amplitude as your input, tune your difference amp to minimize the output amplitude (i.e. adjust the potentiometer until the output amplitude is as small as possible). The best estimate for the common-mode gain can be made using the averaging feature from the

97

7 Introduction to op amps

‘acquire’ menu of the TDS210 oscilloscope. It may be possible, depending on the quality of your op amp, to tune the potentiometer such that no output signal is visible, even at the most sensitive scale. Verify that R1 /R2 = R3 /R4 . Estimate (or set an upper limit on) the common-mode voltage gain. The resistance ratios can be measured using your meter; however, be sure to turn off the power and disconnect each component before making your measurements – accurate resistance measurements cannot be guaranteed if the component remains connected to the circuit, since then you are measuring a parallel combination with the other components rather than just the resistance of interest. For example, when measuring the resistances between the potentiometer center tap and ends, be sure to disconnect all wires connecting the pot to the input signal, op amp, and ground. Replace all resistors and wires when done. To measure the differential voltage gain, ground one input while applying a 1 kHz sine wave of amplitude 1 V or less at the other input. It shouldn’t matter which input you ground. If you’re curious, try it both ways and see if you get equal gains. Measure the differential voltage gain. Explain why it doesn’t matter which input is grounded when measuring the differential voltage gain. Estimate the CMRR for this circuit. What is the input impedance at each input?

7.3 Additional experiments

7.3.1 Current source An ideal current source would maintain a constant current through the load, regardless of load resistance. Try the op amp current source of Fig. 7.8. Vary the ‘load’ pot as you measure the load current with a digital multimeter and the load voltage with a multimeter or scope; what do you observe? Explain how this circuit works (use diagrams and equations as necessary). What should the current be, and why? What is dI /dV ? What is the ‘compliance’ (output-voltage range over which dI /dV is small) of your current source? Compare the performance of this op amp current source to the transistor current sources you built previously.

98

Hands-on electronics

V

0--10 k

+15

10 k

− +

A Vout

Fig. 7.8. Op amp current source.

7.3.2 Noninverting summing amp with difference amplifier This circuit is fun to build and to observe in action. It also serves as an excellent demonstration of the constructive and destructive interference between two waveforms. As stated above, difference ampliﬁcation is usually used to eliminate common-mode noise. The circuit shown in Fig. 7.9 sums two waveforms and then eliminates one using the difference amp. One waveform will be from the function generator while the other will be a 60 Hz wave created as in section 7.2.4. (If two function generators are available, feel free to replace the voltage follower with the output from the second generator.) First construct the voltage follower and summing ampliﬁers as shown in Fig. 7.9. Set the function generator to a sine wave with a frequency between 59 and 61 Hz.2 Adjust the generator amplitude to match the amplitude of the voltage follower and observe the output of the summing amp. The output should be the linear sum of the two input waveforms. Have fun with the circuit by changing the amplitude and frequency. Observe what happens with either a triangle- or square-wave input. Using the ‘AC line’ as your trigger source may be useful here. Now add the difference ampliﬁer. With the function-generator amplitude set to zero, adjust the potentiometer until the difference amp output has zero amplitude (as with the difference ampliﬁer previously built). Now increase the amplitude of your function generator and observe the difference

2

If the local supply frequency is not 60 Hz, set the function generator frequency to be within 1 Hz of your local value.

99

7 Introduction to op amps

R2 function generator

R1

~

R1

− +

2

Summing Output R3 R3

−

R2 R1

−

Vin

+

1

+

3

Vout

10 MΩ

R1 = R2 = R3 = 10 kΩ (or similar)

10 k pot

CMRR Adjustment

Fig. 7.9. Fancy summing circuit. Amp 1 is a voltage follower used to buffer the 60 Hz pickup on a wire wrapped around an AC power cord. Amp 2 is a noninverting summing ampliﬁer with unity gain. Amp 3 is a difference ampliﬁer with an adjustment to maximize the CMRR.

ampliﬁer output. Be sure to switch your trigger source back to the appropriate input channel. Try changing the input frequency. Replace the 60 Hz AC-line signal at the noninverting input of the difference amp with the output from the function generator. Observe how the output changes. Explain how this circuit works using diagrams and equations as needed. Explain why the summing amp isn’t inverting and why it has unity gain. Sketch the inputs and output of the summing amp for a function generator frequency near 60 Hz. Why doesn’t the output have a well deﬁned amplitude? Sketch the inputs and output of the difference ampliﬁer. How does the output change when the inputs are switched? Is the output inverted with respect to the original inputs? If so, why? If so, what could you change to make the output of the difference amp noninverted with respect to the original inputs?

.

1 is useful if we are analyzing circuit performance in the frequencydomain for a sine-wave input. In this chapter we will explore several such applications: circuits that differentiate or integrate their input voltage as a function of time. two 1N914 (or similar) silicon signal diodes. The op amp versions of these applications come closer to the ideal than the passive versions of some of them that you studied in earlier chapters. the gain is −Rf /Ri (neglecting the input offset voltage and offset and bias currents and taking the op amp open-loop gain to be inﬁnite). 8. Zi (8. We can generalize this result for devices other than resistors. There are an enormous number of ways that op amps can be applied to process analog signals. 8. Av = − Zf . form the logarithm or exponential of their input voltage. with input resistor Ri and feedback resistor Rf . one 1 k. or rectify their input voltage. 0.0047 F and 0.1) Eq. and one 100 k 1 W 4 resistor.1 Op amp signal processing Recall that for an inverting ampliﬁer made from an op amp.8 More op amp applications In Chapter 7 we studied some of the basic properties of operational ampliﬁers.1. two 10 k. one 741C and one LF411 operational ampliﬁer. Apparatus required Breadboard. dual-trace oscilloscope with two attenuating probes. as illustrated in Fig. but often we are concerned with the response 101 . 2N3904 and 2N3906 transistors.033 F capacitors. 8. We will also see how to use feedback to compensate for the limitations of discrete devices.

1. to an excellent approximation. it is easy to show that the output voltage is given by Vout = −I Rf = − dQ dVin Rf = −Rf C . where Q is the charge stored on the capacitor. except that the input element is a capacitor rather than a resistor. Generalized op amp inverting-ampliﬁer circuit. . as we will see in more detail below. Basic op amp differentiator. in the time-domain to an arbitrary input waveform. If = Ii . 8. 8. the gain is determined by the current–voltage characteristics of the input and feedback devices.1. dt dt (8.102 Hands-on electronics Zf Vin Zi _ Vout + Fig. Since. the basic op amp differentiator (not to be confused with the difference ampliﬁer) is similar to the basic inverting ampliﬁer studied in Chapter 7.2.1 Differentiator As shown in the circuit of Fig. Using the assumption that the output does whatever necessary to maintain the two inputs at equal voltages. Rf C Vin _ Vout + Fig.2. 8. Then we can analyze the circuit using Ohm’s law. 8.2) since Q = C Vin .

As shown in Fig. 2π Rs C (8. Eq. The output voltage.103 8 More op amp applications Rf C Rs _ + Fig. making the circuit susceptible to high-frequency noise and prone to oscillation. The output voltage as a function of time is still given by Eq.3. with a resistor placed in series with the input capacitor to limit the high-frequency gain to the ratio Rf /Rs . the resistor Ri is the input element and the capacitor C is the feedback element.1. as long as the input frequency is small compared with f = 1 .2. 8. 8. Improved op amp differentiator. (8. 8.5) 8. 2π f C (8.3.6) .4) For input frequencies greater than this. as a function of time. the performance of the circuit approaches that of an inverting ampliﬁer with voltage gain Av = − Rf .2 Integrator By interchanging the resistor and capacitor in the differentiator circuit of Fig. Rs (8.4. Since here Z i = X C .3) decreases with increasing frequency. 8. A more practical differentiator circuit is shown in Fig. 8.1 shows that the output voltage of the basic differentiator increases with frequency. One problem with the basic circuit is that the capacitor’s reactance. XC = 1 .2. we obtain an op amp integrator. is given by Vout = − 1 Ri C Vin dt. 8.

the input DC offset. is used to limit the low-frequency gain of the circuit. To minimize the DC offset voltage resulting from the input bias current. time. The resistor Rs across the feedback capacitor.7) .104 Hands-on electronics C Ri Vin _ Vout + Fig. a more practical integrator circuit is shown in Fig. 8. although small. Ri + Rs (8. Basic op amp integrator. Improved op amp integrator.5. 8.4. called a ‘shunt resistor’. would be integrated over the integration period. C Rs Ri _ + R2 Fig. 8. As in the case of the differentiator. which is proportional to the time integral (area under the curve) of the input waveform vs.5. If the low-frequency gain were not limited. possibly saturating the op amp. the resistor R2 should equal the parallel combination of the input and shunt resistors: R2 = Ri Rs .

105 8 More op amp applications Since the shunt resistor limits the circuit’s low-frequency gain. 8. 8. Op amp logarithmic ampliﬁer. we can analyze the circuit performance as follows.3 Logarithmic and exponential amplifiers By using a diode as the input or feedback element. Op amp exponential ampliﬁer.6. the performance of the circuit approaches that of an inverting ampliﬁer with voltage gain Av = − Rs . we obtain a circuit that takes the logarithm (Fig.8) For input frequencies less than f . For the log ampliﬁer. Rf Vin _ Vout + Fig. 8. Ri (8.1.7. the analysis of the exponential ampliﬁer is left as an exercise. If Ri Ii _ + Fig.6) or exponential (Fig. 8.9) 8. 8. Eq. .6 is valid for input frequencies greater than f = 1 . 2π Rs C (8.7) of its input signal.

which is determined by the input voltage Vin : If = Ii = Vin . the logarithmicor exponential-ampliﬁer exercise. for If positive. .10) We can relate this to the output voltage Vout using the exponential diode current–voltage law: If = Is e−eVout /nkT − 1 . since experimentally one ﬁnds that a transistor gives an accurate exponential characteristic over a wider range of current. including the ones you choose not to build.2.2 Experiments Complete at least the differentiator or integrator exercise.11) The minus sign in the exponential reﬂects the fact that the anode of the diode is connected to virtual ground. (8. Adjust the peak-to-peak voltage of the triangle-wave input to 10 V.033 F. with Rs = 1 k. one should add additional components to compensate for the temperature dependences in Eq.3. a transistor is used in place of a diode. Vout is negative.13) In practice. e (8.13 (both the explicit kT factor and the temperature dependence of Is ). and the half-wave rectiﬁer and push–pull driver exercises. and use a frequency in the neighborhood of 500 Hz. 8.12) (8. Rf = 10 k. thus. Often.106 Hands-on electronics The current If through the feedback element equals the input current Ii . and C = 0.1 Differential and integral amplifiers Differentiator Set up the active differentiator of Fig. 8. Ri (8. (You should understand the theory of all of the circuits discussed. one ﬁnds experimentally that n ≈ 2 for silicon diodes and n ≈ 1 for germanium. The constant n has been introduced since (as we saw in Chapter 3) the slope of the exponential for a silicon diode is not quite as steep as e/kT . 8. Thus the output voltage is Vout ≈ − ≈− nkT Ii ln e Is nkT (ln Vin − ln Is Ri ).) 8.

with Ri = R2 = 10 k.e.15) Now change the input frequency from 500 Hz to 10 kHz. 8. comparing with what you would expect theoretically. and explain the appearance of the output signal at this frequency. You should see that the op amp output signal is a square wave that is 90◦ out of phase with the input. Derive Eq. describe.0047 F. At what approximate frequency does this circuit cease to act as a differentiator (i. Compare your data with Eq. dVin 2Vp–p = . 8. Measure the step height in volts of the square-wave output of the op amp. Record the appearance of the output signal at this frequency.e.107 8 More op amp applications Measure carefully the peak-to-peak voltage Vp–p and period T of the input signal.. . Adjust the peak-to-peak voltage of the square-wave input to 1 V and the frequency to 10 kHz.5.15 and compare the measured gain at 10 kHz with the theoretical expectation. and from them compute the slope of the input voltage vs. Be sure to reduce the input amplitude to avoid saturating the output voltage. Measure the peak-to-peak voltage of the triangle wave and compare with the value you would expect theoretically. Rs = 100 k. Sketch the input and output waveforms at 500 Hz and 10 kHz and comment on your results. Now change the input frequency to 100 Hz. The theoretical prediction is step height = 2Rf C dVin . while triggering on channel 1.6. dt (8. Measure the peak-topeak output voltage and determine the voltage gain. approach the operation of an inverting ampliﬁer)? Integrator Set up the circuit shown in Fig. i.14) dt T Now look at the function-generator output on channel 1 of the scope and the op amp output on channel 2. time. Record. (8. Measure the peak-to-peak output voltage and determine the voltage gain. the output signal is a representation of the negative of the time derivative of the input. You should see an output signal that is a triangle wave 90◦ out of phase with the input square wave. and C = 0. 8.

8.8(a) shows a simple op amp half-wave rectiﬁer. approach the operation of an inverting ampliﬁer)? Compare the output amplitudes with the theoretical expectation.2. be sure to use a signal .e. The log of zero is undeﬁned and any number raised to the power zero equals one.3 Op amp active rectifier Fig. the reverse-biased diode has such a large impedance that the gain becomes very large and the op amp saturates. in the neighborhood of 100 Hz. Because of the limited slew rate of the op amp.6 with Ri = 10 k or Fig. and comment on your results.7 with Rf = 10 k. look at the waveform at the op amp’s output pin. Why then do the logarithmic and exponential ampliﬁers give an output of zero volts when the input is zero volts? 8. Build it with R = 10 k and try it out with a low-frequency sine-wave input. both at the op amp output and at the rectiﬁer output. Figure out how to apply an adjustable DC input voltage. Verify the logarithmic or exponential gain characteristic for several input voltages that probe the full range of output voltage of which the circuit is capable. Replace the 741 with a higher speed op amp such as the LF411. saturation limits the circuit’s performance at high input frequencies. At what approximate frequency will this circuit cease to act as an integrator (i. and repeat your observations. Also. Make a semilog plot of your data and ﬁnd the experimental value of n for your diode. Record the input and output waveforms. To verify this. 8. 8.) In other words. Explain how the circuit works and derive an expression that relates the input voltage to the output voltage.2.108 Hands-on electronics Sketch the input and output waveforms at 10 kHz and 100 Hz. 8. Try a high frequency (say 10 kHz) and record in detail what you see.2 Logarithmic and exponential amplifiers Set up the circuit of Fig. This circuit suffers from a drawback: there is effectively no feedback during the half of the input-wave cycle when the diode is reverse-biased! (This is obvious since the voltages at the inverting and noninverting inputs are not equal during that time.

109 8 More op amp applications (a) Vout R (b) R Vout _ Vin R Vin _ + + Fig.. rectiﬁer diodes (e. or switching diode like the 1N914 – while switching diodes are designed for high frequencies. Recall that in a previous chapter you used a push–pull power driver to drive the breadboard’s speaker from the function generator.g. power Darlingtons are often employed as the push–pull transistors. This circuit displayed crossover distortion since one transistor switched off 2VBE before the other switched on. 8. Explain how each circuit works (use diagrams if necessary) and explain why one is better than the other. what would be the response of a simple rectiﬁer to an input signal of amplitude less than 0.2.7 V? How does this circuit respond to such a signal?) Comment on the performance of the LF411 as compared with that of the 741. (a) Simple op amp half-wave rectiﬁer. (b) improved version.4 Op amp with push--pull power driver A typical op amp (such as the 741) by itself does not have enough outputcurrent capability to drive an 8 load such as a speaker. Build it. and ﬁgure out why it has much better response at high frequency. 8. The addition of a push–pull driver stage to buffer the output is a common solution. try it out. When large output power is needed (several watts) as in audio applications. and the op amp output for both half-wave rectiﬁer circuits at 100 Hz and at 10 kHz. How is this active rectiﬁer better than a simple diode rectiﬁer? (For example. The circuit of Fig.8. the 1N4001) tend to have large junction capacitance and thus have poor performance at high frequency.8(b) overcomes the slew rate limitation. One solution to crossover distortion is to employ an . Vout . 8. Sketch Vin .

9. Record the input and output waveforms – how do they differ. and why? Rearrange the feedback loop to include the push–pull driver inside it (Fig. Use an audio-frequency sine wave. in the vicinity of 1 kHz. This circuit is susceptible to noise. op amp to compare the output signal with the input signal and correct the drive signal to the push–pull stage to compensate for the 2VBE gap. First hook up the circuit of Fig. 8. Estimate the total power consumed by this circuit. and the waveform at the speaker for both circuits and explain how the op amp eliminates crossover distortion. What is the maximum input amplitude that can be accurately reproduced at the speaker (without clipping)? . and compare the input and output waveforms again. and why? How much power is dissipated in the speaker assuming a sine wave of amplitude 4 V? How does the peak current through the speaker compare with the 741’s maximum output current? Sketch the op amp input. 8.9(a) to see crossover distortion in action.110 Hands-on electronics (a) +5 2N3904 +5 _ Vin + −5 2N3906 −5 Speaker (b) +5 2N3904 +5 _ Vin + −5 2N3906 −5 Speaker Fig. Op amp follower with push–pull output-buffer power driver with two feedback arrangements: (a) feedback before and (b) feedback after power driver. What does the signal at the op amp output look like. the op amp output. so be neat and orderly and keep the leads as short as possible. 8.9(b)).

10). similarly. and multiplication by a constant) and design a circuit using op amps to perform that function.111 8 More op amp applications A Vout = V in Vin Logarithmic Amplifier Linear Amplifier (gain = A) Exponential Amplifier Vout Fig. 8. .3 Additional exercises The availability of log and antilog circuits allows elaborate mathematical operations on voltages. subtraction. Discuss the limitations of your circuit. Make a plot of the output vs. the product of two numbers equals the antilog of the sum of their logarithms: x 2 = log−1 (2 log x) x × y = log−1 (log x + log y).17) Choose an arithmetic function of your choice (other than addition. the square of a number can be found by taking the logarithm of the number. (8. 8. 8. and then taking the antilogarithm of the result (see Fig. multiplying by two. Block diagram showing how to build an ‘exponentiator’: a circuit that creates an output voltage equal to the input voltage raised to any desired power. input voltages to verify that the circuit works correctly. For example.16) (8.10.

.

three 10 k. but in some situations (when highspeed response and high sensitivity are not required) an open-loop-741 ‘comparator’ is perfectly adequate. three 0. two 3. 555 timer. 113 . two 741 and one LF411 op amp. and one 10 M 1 W 4 resistor. 311 comparator. two 1 k. one 0. There is also an optional active-ﬁlter application at the end. Start by applying a 1 kHz sine wave to the input and observe what the circuit is doing.3 k.1(a)).033 F. one 820 . one 100 k. and two 3. 9. Apparatus required Breadboard.3 V Zener diodes. one red LED. one 1 M. 9. With no negative feedback.1. it is not a very good voltage comparator (in ways that we shall soon see). and how controlled positive feedback (hysteresis) can be used to eliminate unwanted oscillation or produce intentional oscillation.1 Experiments 9.9 Comparators and oscillators In this chapter you will encounter some applications of positive feedback in op amp and comparator circuits. dual-trace oscilloscope with two attenuating probes.1 Op amp as comparator Begin by wiring up a 741 in open-loop mode as you did in a previous lab (Fig. Since op amps are not speciﬁcally engineered for open-loop operation. the saturated output that results allows the op amp to be used as a voltage comparator – a circuit that tells you whether an input voltage is higher or lower than a ‘threshold’ voltage (the threshold is ground in this case).01 F. You will see how uncontrolled feedback can cause unwanted oscillation. and one 1 F capacitor. one 100 .

114

Hands-on electronics

V+ +15

3 6 3 2 4 8 7 1 4

(a)

2

+15

(b)

_ +

_ +

7

1k

–15

–15

V–

Fig. 9.1. (a) Poor comparator: 741 op amp used in open-loop mode; (b) 311 comparator. Pinouts shown for eight-pin mini-DIP package.

Now raise the input frquency to 100 kHz. Notice that the output ‘square wave’ is not very square. Try an LF411 op amp in place of the 741. For each case studied, sketch the output waveform and measure the output amplitude and DC offset. Explain your measurements. Contrast the performance of the LF411 and 741 when conﬁgured as comparators. What op amp limitation is responsible for the poor comparator performance at high frequency? Now substitute a 311 comparator for the op amp. (The pinouts are NOT the same; see Fig. 9.1(b).) Whereas op amps are intended to be used with negative feedback, the 311 is speciﬁcally designed for operation in openloop mode or with positive feedback. The 311’s output stage differs from that of an op amp: to enhance ﬂexibility, it has both a positive output (pin 7) and a negative output (pin 1). We shall be using the positive output. In this conﬁguration, a ‘pull-up’ resistor to a positive supply is required, in order to determine the output-voltage level, and the negative output is generally connected to ground. Note that the positive output is the collector of an NPN bipolar transistor. As such, it cannot source current, but it can sink current. When the output transistor is off (i.e., its base voltage is less than or equal to its emitter voltage), the collector is pulled up to V+ by the external pull-up resistor. Conversely, when the base voltage is raised 0.7 V above the emitter voltage, the transistor saturates and the output pin is pulled down close to the emitter voltage. This output conﬁguration gives the 311 maximum ﬂexibility to provide the various signal voltages used by digital logic chips, including TTL, CMOS, and ECL logic levels.1 Start by choosing convenient values for V+ and V− . Use an input frequency of 100 kHz and observe the output.

1

Logic levels are discussed in section 10.1.1.

115

9 Comparators and oscillators

Change V+ and V− . Record and explain any changes in the output waveform. How does the output signal from the 311 differ from that of an op amp? What is the 311’s slew rate?

9.1.2 Unintentional feedback: oscillation An unfortunate side-effect of the 311’s fast response is its tendency to oscillate when presented with a sufﬁciently small voltage difference between its inputs. The unavoidable capacitive coupling (typically a few picofarads, depending on how the circuit is wired) between the output and the input causes some feedback, so, when the output switches state, a small transient signal is picked up at the input. There is also some feedback due to the changing input base current when one input transistor switches on and the other switches off. These effects can result in self-sustaining oscillation. The pickup occurs through what is essentially a voltage divider, consisting of the large coupling impedance (i.e., small capacitance) between the output and the input, together with whatever impedance to ground is present at the input. The effect is thus mitigated if there is a low impedance to ground at the input, and exacerbated if the impedance is high. Try to make your 311 oscillate by feeding it a triangle wave with a gentle slope dV /dt (i.e., small amplitude and/or low frequency). Use V+ = +15 V and V− = ground. To exacerbate the feedback, insert 10 k in series with the function-generator source resistance as shown in Fig. 9.2. When looking at the input signal, it is a good idea here to use the ground clips on your

C

+15 +15 10 k

3 2 8 7 1 4

Oscillations

High

1k

_ +

~

Low

–15

Fig. 9.2. 311 comparator with 10 k series input resistor. The capacitor shown connecting the input and output represents the stray capacitance associated with the breadboard. (Do not add a discrete capacitor!)

116

Hands-on electronics

scope probes, to suppress pickup of the large transient pulse due to the 15 V swing of the 311’s output. You’ll want to look at a low-to-high or high-to-low transition and ‘zoom in’ by increasing the time sensitivity to look for the rapid multiple transitions that indicate oscillation. Starting with an input frequency near 100 kHz and an amplitude of several volts, gradually reduce the input amplitude and frequency until oscillations are observed. To be able to see the rapid transitions of the output, be sure the time scale on your oscilloscope is about 500 ns per division or less. Sketch the observed oscillations. What is the time scale of the oscillations – i.e., what is the period t for one cycle of the oscillation waveform? How small an input slope does it take to cause oscillation? What causes the oscillations to stop? Why is a 741 conﬁgured as a comparator less likely to oscillate than the 311?

9.1.3 Intentional positive feedback: Schmitt trigger The oscillation problem can be eliminated by adding controlled positive feedback (hysteresis) as shown in Fig. 9.3. The hysteresis also makes the circuit less sensitive to noise. The phenomenon of hysteresis should be familiar from your study of magnetic materials in introductory physics. In the present context, the term

+15 +15

3 8 7 2

_ +

1k

1 4

—15

1k R2

10k

R1

Fig. 9.3. Schmitt trigger using 311 comparator.

117

9 Comparators and oscillators

hysteresis refers both to the situation of having two different comparator thresholds, depending on the history of the input signal, as well as to the size of the voltage difference between the two thresholds. The threshold voltage at the noninverting terminal is found by applying the voltage-divider equation Vin+ = Vout R2 . R1 + R2 (9.1)

There are two cases to consider, since the output might be at +15 V or it might be at ground – i.e., since Vout has two possible states, Vin+ has two possible states. Build the circuit and apply a signal with an amplitude of at least 1.5 volts. Try to create output oscillations. Notice the rapid and clean transitions at the output, independent of the input waveform or frequency. Sketch the output waveform. Is the output symmetric? If not, why not? To see exactly what is happening, use the two-channel oscilloscope to display the voltages at both comparator input terminals simultaneously. Be sure to use DC coupling for both channels, and set them to the same voltage sensitivity and the same zero offset. Carefully sketch both input waveforms on the same graph and explain how the circuit works. How does the hysteresis prevent oscillations? Based on the component values used, what do you predict for the comparator thresholds? What is the predicted amount of hysteresis in volts? How do these compare with your observations? Record what you see as you vary the input amplitude. Below some input amplitude the output stops switching. At what amplitude does this occur, and why?

9.1.4 RC relaxation oscillator By adding an RC network to your Schmitt trigger, construct a squarewave oscillator, as shown in Fig. 9.4. Since a fraction of the output is fed back into both the inverting and noninverting inputs, this circuit has both negative and positive feedback. By connecting pin 1 to −15 V as shown, you should get a symmetrical square-wave output with little DC offset. No input from the function generator is required; the circuit should oscillate spontaneously.

118 Hands-on electronics R 10 k +15 +15 3 0. The supply voltage can range from 4. but an oscillator ‘kit’. The output sinks current while low and sources current while high (up to 200 mA in either case). as with a comparator.1.01 µF _ + 8 7 1k C 2 Vout 1 4 –15 –15 R1 10 k R2 3. RC relaxation oscillator using comparator.5 to 16 V.2 and compare with the observed frequency. the output is ‘digital’. The 555 can also be used as a timer. As with a comparator.4. among other items (see Fig.2) Sketch the output waveform and explain how this circuit works. the 555 output slew rate . 9. including two comparators. with a high value near VCC and a low value near ground. 9. The expected frequency of oscillation is f = 1 2RC ln 2R1 R2 +1 .5).) What are the threshold voltages? Why does the circuit oscillate spontaneously? Derive Eq. as we shall see below. since the 555 series of timer chips makes designing stable. 9. (9.5 555 timer IC Nowadays. predictable oscillators easy. Also. Note that a 555 is not an op amp or comparator. (Hint: it operates by repeatedly charging and discharging the capacitor between the two threshold voltages of the Schmitt trigger. one hardly ever builds a square-wave oscillator from an op amp or comparator. The 555 is an eight-pin IC powered from a positive voltage source. 9.3 k Fig.

3) .6(a) and observe the output. Refer to a 555 data sheet for additional details. Block diagram for the 555 timer IC. The top comparator output is high when the threshold input is greater than 2 VCC . 9.5. 9. trigger and threshold should be conﬁgured such that only one of the comparators is high at any given moment. and the 555 can change states within 100 ns. while reset causes the output to go low.7(RA + 2RB )C (9. so set causes the output to go high. A positive signal at set causes Q to be near ground. When the output is high. The cycle repeats once VC falls below 1 VCC . The bottom comparator 3 output is high when the trigger input is less than 1 VCC . When the capacitor voltage VC exceeds 2 VCC . the capacitor is charging through RA and RB . while a positive signal at reset causes Q to be high. The frequency is predicted as 3 f = 1 . Begin by connecting a 555 as shown in Fig. the discharge pin is 3 driven toward ground and the capacitor discharges across RB . which drives the discharge pin toward ground.119 9 Comparators and oscillators Vcc R Threshold + _ 555 Timer Output Stage Output Flip-Flop Reset Control Voltage R Set + _ Q Discharge Trigger R Reset Fig. 3 In general. is high. The output stage is an inverting buffer. The outputs connect to a ﬂip-ﬂop (described in detail in Chapter 11). and low otherwise. To see how the 555 works. A high value for Q turns on the transistor switch. 0. ﬁrst examine the connections associated with the two comparators. and low otherwise.

4. Sketch the output waveform and brieﬂy explain the operation of this circuit. The output should be a ‘one-shot’ pulse of duration t = 1. 9. Is the output symmetric? If not.120 Hands-on electronics +15 10 k RA 7 8 4 +15 10 MΩ RA 7 8 4 Vcc Discharge Reset Vcc Discharge Threshold Trigger 1 Reset +15 10 k RB 6 2 Out Threshold Trigger 1 3 6 Out 3 100 k 2 820 0. Do they make sense? Try replacing RB with a short circuit – what happens? Explain why. so one should ensure that the trigger pulse is shorter than the desired output pulse!) Time the output pulse by observing the LED. why not? Derive Eq.6. 9. Brieﬂy explain the operation of this circuit.6(b). which causes the trigger input to go to ground.4) The output pulse is triggered by the push-button switch. To what extent does the output frequency depend on supply voltage? Now connect a 555 as shown in Fig. What prevents this circuit from oscillating? Measure the output-pulse duration for several values of RA and C. Examine the voltage VC across the capacitor. 9. Try changing V+ to 5 V and observe how the output changes. Tabulate your results. Derive Eq. Record its minimum and maximum values. Are your data consistent with this expression? If not.01 µF C 1 µF C Red LED (a) (b) Fig.1RA C. (b) 555 timer IC used as a one-shot or timer. (9. 9. (a) 555 timer IC used as an oscillator. (Note: the output will remain high indefinitely if the trigger input is held at ground. why not? .3 and compare the measured output frequency with the predicted oscillation frequency. Put RB back for the next part.

Trigger the timer and the alarm will sound for the duration of the timer’s output signal. Cut or unplug the wire and hear the alarm! The alarm can also be conﬁgured to sound for a speciﬁed duration using two 555s by combining the timer from Fig. the reset line is said to ‘enable’ the 555. The alarm is simply a 555 oscillator with the output connected to a speaker. enabling oscillation when high while disabling oscillation when low. 555 timer conﬁgured as an alarm. .1 Alarm! You can conﬁgure the 555 to sound an alarm when prompted by an external signal. 9. Connect the alarm’s reset input to the timer’s output line.121 9 Comparators and oscillators 9.01 µF C Fig. Build the circuit as shown in Fig. which simulates a security loop. 9. +15 8 4 10 k RA 7 Vcc Discharge Reset 10 k RB 6 Threshold Trigger 1 Out 3 100 2 1 k pot (volume control) Speaker 0.7. ground is applied to the reset input pin.2.2 Additional experiments 9.7. To prevent the alarm from sounding continuously. which overrides the trigger and threshold pins and forces the output near ground.6(b) with the alarm circuit. 9. Use a long wire. When used in this way.

The second oscillator’s output is connected to the speaker and can oscillate with any audio frequency of your choice. look at both outputs on the dual-trace oscilloscope. unless Eq.8.5 happens to be satisﬁed due to resistor manufacturing tolerances. . The circuit of Fig.5 by using the 10 k pot for R1 or by adding other resistors in parallel with the 10 k. Are the frequencies the same? What is the phase relationship? sine 0. 9.8 should produce a better approximation.5) as long as R1 < R. Set up the circuit and apply power. You can vary R1 so as to satisfy Eq. The output of this oscillator is then connected to the reset line of the second oscillator. 9. Sine/cosine oscillator.122 Hands-on electronics You can also make a pulsing alarm by using two oscillators. Once you have it oscillating. 9.2. The ﬁrst should oscillate with a period of a few seconds. the PB-503 uses a piecewise-linear approximation to a sine function. 9. It should oscillate at f = 1 . So you don’t expect it to work with R1 = 10 k.3 V 3.033 µF 3. 9. 2π RC (9.033 µF R Fig.033 µF _ 10 k 10 k C _ cosine R1 + R + 10 k C 0. As you saw when you differentiated the sine output of the PB-503’s built-in function generator.3 V C 0.2 Sine/cosine oscillator Perhaps surprisingly. the sine wave is one of the most difﬁcult waveforms to produce.

.033 µF 100 k R2 3. Measure and graph the gain vs. Active bandpass ﬁlter. or Simpson.0 V peak-to-peak. Vary the driving frequency until you ﬁnd the maximum output amplitude.033 µF C 0. What role is played by the two back-to-back Zener diodes? 9. see e. Introductory Electronics for Scientists and Engineers. 0. frequency for a reasonable frequency range. There are a variety of conﬁgurations that can be used for low-pass.3 Active bandpass filter Active ﬁlters have signiﬁcant advantages over passive ones. and drive it with a sine wave of about 1 V peak-to-peak. Wire up the circuit of Fig. you can easily ﬁnd the −3 dB points by varying the frequency until the output voltage is 10.123 9 Comparators and oscillators Measure the peak-to-peak voltage of the cosine wave.2.9. r high input impedance and low output impedance. They also have signiﬁcant disadvantages such as frequency response limited by the bandwidth of the op amp and the need to provide power to the op amp. including r lower cost due to replacement of expensive inductors by capacitors phaseshifted via feedback. and bandpass applications. considering the diodes being employed in the circuit? Try to explain how the circuit works. r ease of tuning over a wide frequency range. apply power.1 V peak-to-peak at f = f 0 . The Art of Electronics. Horowitz and Hill.3 k C R1 + Fig. Is it what you would expect. 9.3 k Vin _ op amp Vout R1 3. What are the center frequency f 0 and upper and lower −3 dB points f H and f L of the frequency response? If you adjust the output voltage to be 14.9. 9.g. high-pass. for a more extensive discussion.

Compare your observed Q to the theoretical value for this ﬁlter. (9.6) and compare the voltage gain at the center frequency with the theoretical value 1 R2 A0 = − . .7) 2 R1 The ‘quality factor’ for a bandpass ﬁlter is Q= f0 .9) Try to explain how the circuit works.124 Hands-on electronics Compare f 0 with the theoretical value f0 = 1 2πC 2 . fH − fL (9. Q= R2 . a narrow passband corresponds to high Q and a broad passband to low Q. 2R1 (9.8) Thus. R1 R2 (9.

10

Combinational logic

In this chapter you will be introduced to digital logic. You will build some logic circuits out of discrete components and some out of integrated circuits, and familiarize yourself with the 7400 series of CMOS (complementary metal-oxide-semiconductor) and TTL (transistor–transistor logic) integrated circuits and their basic operation. Note The kinds of things one thinks about in digital logic are almost completely different from those in analog electronics. Apparatus required Breadboard, oscilloscope, multimeter, 100 , 330 , 1 k, 2.2 k, and 3.3 k 1 W resistors, two VP0610L and two VN0610L MOSFET transistors, 4 three 2N3904 transistors, three diodes, one LED, one red LED (optional), 74HC00, 7432, 7485, 7486 TTL or TTL-compatible logic chips, logic switches, and logic displays.

**10.1 Digital logic basics
**

In this section we introduce the 7400 series of CMOS and TTL digitallogic chips. Unlike the analog ICs you’ve used up to now, which can output any voltage within some range determined by the power-supply voltages, digital-logic ICs employ only two ranges of output voltages, referred to as logic levels, about which more below. These levels can be used to represent true or false logical conditions or the zero and one of binary arithmetic. The 7400 series is not the only logic series, nor are CMOS and TTL the only types of logic circuitry; however, they are the most commonly used. Other logic families include the CMOS 4000 series and the ECL

125

126

Hands-on electronics

(emitter-coupled logic) 10 000 and 100 000 series. Each logic family has its own logic levels, speed, and recommended supply voltages. The integrated circuits you will be using now are much more specialized than the general-purpose 741 op amp and 555 timer. They feature much higher bandwidth, with typical transition speeds of order volts/nanosecond (in contrast to the volts/microsecond slew rate of the 741). While greater complexity often means higher cost, the basic chips in the 7400 families (such as the 74HC00, 74LS00, and 74ACT00) cost less than $0.50 each in small quantities, with the more complex chips ranging toward several dollars.

10.1.1 Logic levels Digital chips employ two voltages to represent two possible states. These voltages are called logic levels and can be used to represent the two states of Boolean algebra1 as well as the two digits of binary arithmetic. There are three ways of referring to logic levels: r true and false, r zero and one, and r high and low. In TTL logic, a voltage exceeding +2 V is called high, while a voltage less than +0.8 V is called low. To ensure noise margin, TTL outputs are guaranteed to put out at least +2.5 V in the high state and at most +0.4 V in the low state (see Fig. 10.1). This means that, even in the presence of up to 400 mV of noise, an output low will be recognized as low by the input of the next logic circuit, and similarly for high. The comparable CMOS levels are +3.5 and +1.5 V for the inputs and +4.5 and +0.5 V for the outputs. While TTL chips are always powered from a +5 V supply, many CMOS chips are tolerant of supply voltages ranging from +2 to +6 V. It is therefore convenient to reference CMOS logic levels to the power-supply voltage VCC . The minimum input voltage interpreted as a CMOS high (VIH ) equals 0.7 × VCC , while the maximum input voltage interpreted as a CMOS low (VIL ) equals 0.3 × VCC . The output voltages VOH and VOL will vary with supply voltage as well. The ‘high/low’ nomenclature is unambiguous, since it directly characterizes voltages, but the other two nomenclatures rely on a convention,

1

Also known as symbolic logic.

127

10 Combinational logic

Volts 5

VCC

VCC

4

VIH

3

VCC

VIH

2

VIL

VIH

1

VIL

VIL

0

CMOS: C, HC, AC, and AHC Series CMOS & TTL: HCT, ACT, AHCT, S, F, LS, AS, and ALS Series Low-Voltage CMOS: LVC and ALVC Series

Fig. 10.1. Logic levels for various 7400-family lines. VCC is the most positive voltage; VIL and VIH are the maximum input low and minimum input high voltages.

which can be assigned either of two ways. The more common convention is positive logic: high=1=true, low=0=false. But there are occasionally situations in which it is more convenient to employ negative logic, in which high=0=false and low=1=true. (Another way to think about logic circuits is in terms of assertion-level logic, which is a hybrid of positive and negative logic that we will introduce in the next chapter.)

10.1.2 Logic families and history As indicated in Table 10.1, CMOS and TTL chips come in a plethora of types, each with its own speed, power dissipation, input load, and outputcurrent characteristics. This reﬂects the historical development of the various series. Since our initial purpose is to become familiar with the logic, the details of speed and power are (for now) unimportant, but you will need a general familiarity with them so as not to be in the dark when you encounter them in future. Also there are restrictions on fanout (the number of inputs that an output can drive) which matter when actually designing circuits. For example, one LS-TTL output can drive twenty LS-TTL inputs, but only four S-TTL inputs. In what follows we use ‘7400’ generically to

128

Hands-on electronics Table 10.1. Common families within the 7400 series. Family TTL S LS ALS F HC HCT AC ACT LVC AHC Year 1968 1974 1976 1979 1983 1975 1975 1985 1985 1993 1996 Brief description bipolar transistor–transistor logic TTL with Schottky transistors low-power Schottky TTL advanced low-power Schottky fast TTL high-speed CMOS high-speed CMOS (TTL compatible) advanced CMOS advanced CMOS (TTL compatible) low-voltage CMOS advanced high-speed CMOS

refer to chips from any of these families. Unless otherwise speciﬁed, the chip families you actually use will depend on what happens to be on hand in the laboratory or (if you are working through this book on your own) on what you happen to ﬁnd available. Part numbers The original TTL chips were the 7400 series and the corresponding ‘Mil Spec’ (military speciﬁcation) 5400 series; these became popular in the 1970s. TTL chips are labeled with part numbers that begin with a letter code (such as ‘SN’) that is typically different for each manufacturer (see Fig. 10.2); then comes the ‘74’ that identiﬁes the device as belonging to the 7400 series; then there may be a letter code that identiﬁes the family; then the number that identiﬁes the particular device (e.g. 00 for a quad NAND gate, 01 for quad NAND with open-collector outputs, 02 for quad NOR gate, 74 for dual D-type ﬂip-ﬂop, etc.); and ﬁnally there may be letters that indicate package style, reliability, degree of testing by the manufacturer, etc. (For example, the MC74LS00ND is a Motorola LS-TTL quad NAND gate in the plastic dual-in-line package with 160 hour ‘burn-in’ testing.) Pinouts and data sheets Data sheets are available on the web or in data books produced by the chip manufacturer. The sheets specify the function of each pin of the IC package and provide detailed data on chip performance. In general, 7400-series chips have compatible pinouts independent of the family. For example, the

four. 10.3. Also note that the NAND is a negativelogic OR while the NOR is a negative-logic AND. (to be discussed later) – which can simplify design as well as reduce cost.1. the writing on the top will be right-side-up (see Fig. Almost always. 10. If you orient the chip so that the pins are bending away from you and the end that has a notch or a dot is pointing to the left.2. when you orient the chip this way. Note that NAND and NOR are opposites to AND and OR: NAND equals not AND while NOR equals not OR. such that the highest-numbered pin is at the upper left. Another consistency is in the pin numbering scheme.3 shows only two-input gates. 10. Although Fig. versions also exist with three. decoders. 10. although the more complex functions are also available as specialized chips – multiplexers. pinout for a 74HC00 quad NAND IC is the same as the pinout for the 74LS00 quad NAND IC. (This follows from DeMorgan’s theorem. as shown in Fig. and similarly for NOR.3 Logic gates There are six basic logic gates. or even eight inputs.2). etc. The numbering then proceeds sequentially around the chip in a counterclockwise direction. about which more below. pin 1 is the one at the lower left of the package. It is always a good idea to review the data sheet before using any logic chip. These gates are sufﬁcient to implement all logic functions. 10.129 10 Combinational logic Pin #14 Pin #8 1A 1B 1Y VCC 4B 4A 4Y 3B 3A 3Y SN 74 LS 00 N 2A 2B 2Y Pin #1 Series Manufacturer Family Pin #7 GND Chip Type Fig.) The small circle shown at the outputs of the NAND and NOR is shorthand for an inverter – a NAND is equivalent to an AND followed by an inverter. Labeling of 7400-series chips. .

if the output of a NAND is connected to both inputs of a second NAND.3. and historically it was the most commonly used gate. As in any algebra.130 Hands-on electronics AND A L L H H NAND A L L H H OR A L L H H NOR A L L H H XOR A L L H H B Out L H L H L L L H B Out L H L H H H H L B Out L H L H L H H H B Out L H L H H L L L B Out L H L H L H H L AND OR XOR INVERTER (NOT) A Out L H H L NAND NOR INVERTER Fig. While a detailed exposition of Boolean algebra is beyond the scope of this text. Thus. Standard logic gates with truth tables. r the logical XOR of A and B is denoted as A ⊕ B. the rules of Boolean algebra allow theorems to be derived starting from axioms. connecting together the two inputs of a NAND creates a one-input gate – an inverter. Consider two logical variables A and B. For example. 10. the NAND is the simplest logic gate to construct from discrete components. in that the other logic functions can all be created using NAND gates. and the inverse of B is B. After the inverter. An alternative way to prove a theorem in . We can then denote logic operations as follows: r the logical AND of A and B is denoted as A · B. r the logical OR of A and B is denoted as A + B. we give here a brief introduction and mention a few useful points. r the inverse of A is A. 10. which can take on the values true and false. r the logical NOR of A and B is denoted as A + B. the result is equivalent to an AND gate. NAND is a ‘universal’ logic function.1.4 Summary of Boolean algebra Logic operations are best described using Boolean algebra. r the logical NAND of A and B is denoted as A · B.

and A+B = A·B (10. Assuming logic levels equal to either 0 or +5 V.e. 10. and zero otherwise. = Fig. Two relationships (known as DeMorgan’s theorems) are particularly useful: A · B = A + B. 10. as shown in Fig. it is easy to show that the output voltage is near +5 V (actually one diode drop below +5 V) if both inputs are +5 V. low = false).5.1 Diode logic We begin our consideration of the electronic implementation of logic gates with the simplest example.) . i. they can be used to perform logic.2. = Boolean algebra is by ‘exhaustive demonstration’. to write down the truth table – for every possible input state. work out the value of the output and write it down..2 CMOS and TTL compared 10.2) (10. 10.4. this is equivalent to a logic AND of the two inputs. DeMorgan’s theorems expressed symbolically. A H H L L B A+B H L H L H H H L A L L H H B A+B L H L H H L L L A H H L L B AB H L H L H L L L .131 10 Combinational logic A L L H H B AB L H L H H H H L .4. (You may want to verify this by building this circuit on a breadboard and trying it out. 10. Since diodes pass current in only one direction. If we assume the positive-logic convention (high = true.1) These are illustrated schematically in terms of logic gates as well as with truth tables in Fig.

6. 10. The resistor values are approximate and vary with TTL family. TTL logic was developed during the 1960s. which turns on Q3 . Q2 ’s base voltage . transistors Q1 and Q3 will be off. Also shown is a simple LED logic-level indicator being driven by the NAND gate’s output.2 Transistor–transistor logic (TTL) Using the diode-logic concept as an input stage. r If either input is near ground.3 k 2.5. Q1 ’s base voltage increases. causing Q1 to turn on. Two-input diode gate. Diode–transistor NAND gate using 2N3904s.3 k T A B Fig. A TTL-inspired NAND gate constructed using diodes and bipolar transistors is shown schematically in Fig.132 Hands-on electronics +5 3. if both inputs become high. except that in the 7400 a two-emitter transistor substitutes for the input diodes. which causes the output (T) to be about two diode drops below VCC . 10.2 k 100 +5 Q2 330 Q1 A B 1k Q3 LED T Fig.2. r On the other hand. This resembles the circuitry actually used inside the 7400. +5 3. 10.6. 10. The operation of the circuit can be analyzed as follows. Q2 will consequently be on (saturated).

74HC. 74HCT. such as the 74C. The input and output current speciﬁcations vary among the different family lines. and 74ACT. MOSFET operation resembles that of the JFETs studied previously. Fairchild ﬁrst introduced the 4000 series of CMOS logic. 10. the output T is close to ground. TTL outputs cannot be relied upon to source even small amounts of current. 74AC. 10.7. turning Q2 off. home-built logic indicators (as in Fig. Refer to the data sheets for details. since it consists of a group of stacked components. CMOS logic ICs have surpassed TTL logic in popularity. Assuming the positive-logic convention. however. Quirks of TTL inputs and outputs The gate circuit of Fig.133 10 Combinational logic drops. a positive . but sink negligible amounts of current while held high. resulting in numerous family lines. Many are listed in Table 10. but the popularity of TTL led manufacturers to develop TTL-like CMOS families. they can sink tens of milliamperes of current. 10. Since Q3 is then in saturation. Note how the LED logic indicator has been constructed to take advantage of this – the LED is lit when the output is low! Note also that TTL inputs source current while held low. and is thus called CMOS. Since the 1990s. while the ones without a ‘T’ are logic.3 Complementary MOSFET logic (CMOS) In the 1970s a new family of logic chips was developed using MOSFETs instead of bipolar transistors.1. As shown schematically in Fig. The families with a ‘T’ in their part numbers are fully compatible with TTL logic levels. whereas for TTL. Several improvements have been made since the original introduction of TTL logic.2.and pinout-compatible with TTL but operate with CMOS logic levels. in an N-channel MOSFET. The basic MOSFET logic gate uses complementary N-channel and P-channel MOSFETs. however.6 illustrates some peculiarities of TTL inputs and outputs. The output structure is commonly referred to (fancifully) as a ‘totem-pole’ output.6) operate in negative logic. In general. Note that the LED logic indicators built into the PB-503 operate in positive logic. TTL logic is still manufactured and is quite common. the output voltage thus represents the logic NAND of the inputs. 10.

Combining an N-channel and P-channel MOSFET gives a complementary pair of switches that open and close opposite each other.7. When the channel is open. A positive gate voltage of a few to several volts is required to open the channel and allow current to ﬂow. The oxide layer allows the electric ﬁeld to penetrate without allowing current to pass. This does not occur in MOSFETs since the gate and channel are separated by a layer of insulating oxide. Two . Schematic representation of an ‘enhancement-mode’ N-channel MOSFET. gate–source voltage attracts current carriers into the channel between the drain and source. The input impedance for MOSFETs is consequently even greater than for JFETs! Also notice that. 10. allowing current to ﬂow. For a JFET. The MOSFET thus acts like a switch. the drain–source channel is normally closed. the drain–source resistance is quite small (≈10–100 ). while the drain–source resistance is large (of order megohms) when the channel is closed. For P-channel MOSFETs. the gate voltage is brought positive with respect to the source.134 Hands-on electronics Metal VGS + Oxide Semiconductor Source Gate Drain N-type P-type N-type N-type Substrate Fig. the voltage relationships are reversed: the channel is open when the gate–source voltage is negative and closed when the gate–source voltage is near zero (or positive). this would result in a large current ﬂowing through the gate to the source since the gate–source junction would act as a forward-biased diode. Notice that to turn on an N-channel MOSFET. unlike JFET operation. A gate–source voltage near zero (or negative) closes the channel and prevents current ﬂow between drain and source.

10. The output now has a low-resistance path to VCC and is pulled high. all signals either high or low). while the P-channel MOSFET is off and has a high resistance between drain and source. if this is discharged in a spark to a MOSFET. dry conditions can easily develop a potential of several thousand volts. Schematic representations of a CMOS inverter constructed using one N-channel and one P-channel MOSFET. Static charge accumulating on a human in cold. r When the input is high.e. it is sometimes sufﬁcient to blast a hole . it thus draws much more power when signals are switching between high and low.135 10 Combinational logic Vcc Vcc Vcc S G P-channel MOSFET D Vin D G N-channel MOSFET S MOSFET Switch 2 Vout Vout Vout MOSFET Switch 1 Fig. the N-channel MOSFET is on and has a low resistance between drain and source. r If the input is near ground.8. A NAND gate can be constructed by adding two more MOSFETs (see Fig.9). the P-channel MOSFET is on while the Nchannel MOSFET is off. Another drawback of MOSFETs is their sensitivity to static-electricity discharge.8. 10. MOSFETs can thus be combined in parallel to create an inverter. then both channels are partially open. A drawback is that if the input is at an intermediate voltage. The inverter operates as follows. While CMOS logic draws very little power when in a steady state (i. The output is thus connected through a low-resistance path to ground and goes low. 10. as shown in Fig.. Although the total energy released is small. which results in current ﬂowing through both FETs from power to ground. The oxide insulating layer between the gate and the channel is usually quite thin and easily damaged.

but you must not forget to make those connections. the logic chips to be used here will require only a single +5 V supply rather than separate positive and negative supplies. Note that not all chips have the same number of pins. To minimize the accidental destruction of components.4 Powering TTL and TTL-compatible integrated circuits Unlike the 741 and 311 chips that you’ve used previously. 10. 10. through the oxide layer. You must look up the pinout of each chip you employ! .136 Hands-on electronics Vcc P-channel MOSFET G Vcc A L L H B L H L H C H H H L S S G P-channel MOSFET H D D C A G D N-channel MOSFET S D G N-channel MOSFET S 330 B LED Fig. we’ve chosen medium-power MOSFETS for this lab – these are much more robust (and less sensitive to static) than the tiny. destroying the MOSFET. You’ve probably heard of static-sensitive electronics such as computer memory cards. Schematic representation of a CMOS NAND gate with LED logic-level indicator. furthermore. As in the case of the 741.2.9. even among chips with the same number of pins. These cards are constructed using MOSFETs. not all follow the same convention in power-supply connections. high-speed MOSFETS used inside high-density chips. the actual power connections are often omitted in schematic diagrams.

3 Experiments 10. you can augment the built-in indicators with individual LEDs in series with several-hundred-ohm current-limiting resistors.) Drive an LED logic indicator with a variable voltage between 0 and 5 V. and allows for ﬂexibility when working with low-voltage CMOS ICs or other nonstandard ICs.5 V for the original 7400 TTL family).10. A high level applied to the input will light the LED – try connecting one to ground or +5 V and see that it works. The switches make a connection either to power or to ground. 10. Explore the threshold voltage and compare with the CMOS and TTL logic levels.1 LED logic indicators and level switches The PB-503 breadboard has eight built-in LED indicators.10. These can be used as logic-level switches. These are intended for use as logic-level displays. 10. as shown in Fig. (If you ever ﬁnd a need for more than eight indicators.137 10 Combinational logic Be careful! Note that TTL and CMOS ICs are guaranteed to be destroyed if by mistake you power them backwards! They may also be destroyed if you apply a voltage higher than 7 V to any pin (5. 10. Older versions of the PB-503 have a bank of SPST DIP +5 +5 1--100 k TTL or CMOS LOGIC LEVEL SPST Switch SPDT Switch TTL or CMOS LOGIC LEVEL Fig. . Logic-level switch using either an SPST or SPDT switch and a pull-up resistor as shown. this is also a good solution if your breadboard lacks built-in logic indicators.3. The power voltage depends on the setting of a ‘master’ switch located to the right of the eight-switch unit. The PB-503 also provides a bank of eight ‘level’ switches in a unit located near the lower-left corner. Two single-pole double-throw (SPDT) switches are available at the lower right-hand side of the PB-503.

construct the circuits shown in Fig. or ‘off’ to assert a TTL high. Circuits for measuring the channel resistance as a function of gate voltage. Unpredictable behavior (as well as excessive current ﬂow from the power supply to ground) will result otherwise. Note: if you try to measure the resistance directly with the meter. you can thus set the switch ‘on’ to assert a TTL low input level.) Figure out how to use the logic-level switches to turn an LED indicator on and off.3. 10. (If your breadboard doesn’t have logic switches. For both N.11.138 Hands-on electronics switches instead of logic switches.2 MOSFETs To demonstrate MOSFET behavior. therefore. 10.10 for details on how to use the SPST switch bank as logic-level switches. even without a pull-up resistor. vary the gate voltage between 0 and 5 V and measure the channel resistance as a function of gate voltage.and P-channel MOSFETs. and SPST switches can easily be used as TTL level switches. Explain how the indicator works. With one end of the SPST switch connected to ground. CMOS inputs require well deﬁned input voltages. whereas an open input acts like a high logic level. TTL is more forgiving. 10. an unconnected TTL input usually behaves as if connected to a logic high: you assert a TTL low logic level at an input by connecting it to ground. refer to Fig. a CMOS input must always be connected either to power or ground or to a CMOS (or in some cases TTL) output. 10.11. you Vcc (5 V) Vcc (5 V) 1 kΩ S Vin G D P-channel MOSFET Vout Vin G D 06 PV 10L D G S Vout N-channel MOSFET 061 NV 0L 1 kΩ S G D S Fig. . As we will explore in more detail later.

If you use discrete components. Compare these results with the values measured for the inverter. Using logic switches. the complementary nature of P. Use it to apply a signal to the input. Measure the transition speed of the output.. Do so as shown in Fig. Connect the output to a logic indicator.) The inverter can be converted to a NAND gate with the addition of two more MOSFETs. and measure the output slew rate (i. Build the gate shown in Fig. you should disconnect it to avoid confusion. As discussed above. Does the output impedance depend on the output logic level? How does the output voltage level depend on the output load? . The TTL output of the function generator puts out a square waveform whose low voltage level is near zero volts and whose high is near +5 V. the transition speed from high to low and from low to high).and P-channel MOSFETs vs. plot the channel resistance for both the N. Measure the output impedance with both inputs high as well as with both inputs low. The output impedance should be measured for both high and low output logic states. using a pull-up resistor to +5 V or a pull-down resistor to ground as appropriate (try a 330 resistor).139 10 Combinational logic will fail. How does the output transition speed compare with the transition speed of the input square wave? Measure the input and output impedances and compare your results with your expectations. don’t forget the current-limiting resistor (about 330 ) – connect the LED and resistor in series between the CMOS output and ground. In either case. verify the truth table. 10. or else ﬁgure out how to take it into account. 10. and if the LED is off the output is low. You can easily display the output using an LED indicator. if the LED is on the output is high. gate voltage. since the quiescent current ﬂowing in the channel will confuse the ohmmeter! Measure the output voltage and infer the resistance using the voltage-divider equation. Tie one input to +5 V and the other to a TTL square wave.e. Make a truth table and verify that the gate inverts the logic level of the input.9. Apply CMOS logic levels to the input and note the output.and N-channel MOSFETs which you’ve just demonstrated can be used to create logic gates. Comment on your results.8 and determine its function. On a single graph. (If you used a discrete-component LED indicator.

Compare with the values listed on the 74HC00 data sheet. (When inserting a new chip into the breadboard. Connect pins 1 and 2 to level switches. Connect pin 1 to +5 V and drive pin 2 from the TTL output of the function generator.3. A good technique for precise timing of logic signals is to set both scope channels to 1 or 2 V/ division and use the scope’s vertical-position knobs to put ground 2. This is most simply done using an N-channel MOSFET: connect the source to ground. the gate to pin 3 of the 74HC00.2. if you use discrete components for your LED logic indicator. Connect pin 3 to an LED logic indicator.3 CMOS NAND gate From your collection of chips. Referring to the pinout diagram (Fig. Then run jumpers to pins 7 and 14 of the chip from the ground and +5 busses. a typical although not universal conﬁguration. Note: the 74HC00 can source only 5 mA of current. For now we are going to use just one of the four gates on this chip. Record the truth table (in terms of logic levels).3. since a bent pin will fail to make contact.5 V. then you can easily measure the time at which each signal crosses 2. select a 74HC00 quad NAND gate.). it is useful to deﬁne the transition times of the input and output signals as the times at which they cross 2. therefore. which is about half-way between high and low. Ground all other inputs. For measurements of propagation delay. disaster will strike if you hook up the chip backwards and apply power! Carefully insert the 74HC00 so that it straddles the central groove of a breadboard section. 10. you will note that the ground connection is at pin 7 and +5 V at pin 14. turn on the power.140 Hands-on electronics 10. Connect the inputs to logic switches and try all four input combinations.) Check that the power is off. respectively. NANDs are a ‘universal’ logic function in the sense that any Boolean-logic function can be constructed from them (the same .4 Using NANDs to implement other logic functions As mentioned above.5 V.5 V below the center of the graticule. Measure the output rise and fall times and propagation delay of the 74HC00. the one whose output is at pin 3 and whose inputs are at pins 1 and 2. After double-checking that the connections are correct. 10. you will need to buffer the output through a transistor. and the LED and current limiting resistor in series from the drain to the +5 V supply. Again. you need to pay attention to all of the pins and make sure none of them is bent.

2). Drive each circuit using two switches on your breadboard and verify that it gives the desired output for each of the four possible input states.3. it can also be done with only four. It is easy to see how to do this with ﬁve gates.1 and 10. you can test two 4-bit binary numbers for equality. Record the truth table for each logic circuit. but not both. since AND and OR functions require more transistors to implement.5 TTL quad XOR gate Use TTL chips for this exercise. 10. NOR. an XOR can be used to test two numbers for equality. To ﬁgure out the necessary logic. 10. or tie one input high. Be sure to show.. that indeed you have implemented an exclusive-OR. displaying the XOR output with an LED logic indicator. With a 7486 quad two-input XOR gate. gate symbols arranged in a logical order according to their function (like the schematics in the following chapters). you can make use of DeMorgan’s theorems (Eqs.e. Construct an XOR circuit out of NANDs.2). Sometimes it is useful to have a gate that outputs a true if one or the other input is true. it also can be used to produce the sum bit for a 1-bit binary addition. i. . OR. only NAND was available. Write down your schematic diagram for each circuit. as ﬂoating inputs will not work with CMOS gates. Recall that you can use a NAND gate as an inverter if needed – just connect the two inputs together. Show how to use an XOR to test two signals for equality. These should be logic diagrams. Such a gate is called an exclusive-OR (XOR) gate. indicating pin numbers next to each input and output that you use. Construct the following two-input logic circuits using only NAND gates: AND. both using Boolean algebra and using truth tables for the intermediate signals in your circuit. Does the LED light for equality or for inequality? Why? 10.141 10 Combinational logic is true of NOR and XOR). In the early days of logic chips. Try it out and show that it works. As we shall see below. In Boolean algebra the exclusive-OR of A and B is denoted by A ⊕ B. not chip-level diagrams (such as in Fig. It should turn on an LED indicator when either input is high but not when both are.

build your circuit. If you have time. off if they are different). Display the output with a logic indicator and verify that your circuit works for a few representative cases. plus whatever additional gates you need. But it becomes much simpler if you use OR gates such as the 7432. Is your observation consistent with the general property of 7400-series TTL chips that ﬂoating input lines default to a high state? This is a handy feature when testing circuits on a breadboard.1 7485 4-bit magnitude comparator This single chip does the work of the circuit you have just constructed and more.5. 10. you should connect all of its inputs either to high or low or to outputs of other gates. .4. look up this chip in a logic data book or on the web. You will ﬁnd that this is quite cumbersome using NAND gates. but it is good design practice not to rely on it – to be certain of reliable operation and be maximally insensitive to noise.3. familiarize yourself with its operation. since equality corresponds to all four 7486 outputs being low.4 Additional exercises 10. Allow one or more inputs to ﬂoat by leaving them unconnected. and test it on your breadboard.142 Hands-on electronics Using all four XOR gates on the 7486. and try it out. Record the input and output in each case. Describe how the function of this chip differs from the circuit you constructed in section 10. design a circuit whose output indicates whether the two halves of the 8-bit level switch are set equal (LED on if they are equal. Record your schematic including pin numbers. if you are using a gate in a circuit.

Their behavior thus depends not only on their present input but also on their internal state. which goes through a predetermined sequence of states. you will need a 74373. showing the sequence of states through which the circuit passes. In addition. a dual D-type ﬂip-ﬂop (7474). including divide-by-two and divide-by-four counters. One category of sequential-logic circuits is the ﬁnite-state machine. advancing to the next state each time it receives a clock pulse. We will encounter some examples of state machines in this chapter. Circuits containing ﬂip-ﬂops are termed sequential logic circuits. if you are unprepared. (If possible. two 1 k 1 W 4 resistors.11 Flip-flops: saving a logic state Next we turn to ﬂip-ﬂops. these devices can remember their past. Prepare carefully in advance and you will ﬁnd you can complete these exercises easily. You will need a quad NAND (7400). it is likely to take you three to four times as long! Apparatus required The ideas explored in the following exercises apply equally whether CMOS or TTL gates are used – use whichever is most convenient. and a two-channel oscilloscope with two attenuating probes.) 143 . since their state depends on the sequence of inputs that is presented to them. A useful tool in understanding a state machine is a state diagram. and a dual JK ﬂip-ﬂop (74112). A truth table is not sufﬁcient to describe the operation of a sequential circuit – you need a timing diagram. Also known grandiosely as bistable multivibrators. avoid unnecessary complications by not mixing CMOS and TTL chips within a single circuit. Some of these circuits will probably be the most complicated you have wired up so far. a breadboard.

Often when one wires up a circuit off the top of one’s head. Signal levels . This approach leads to circuits that are easier to design and analyze than asynchronous circuits. The synchronous approach is therefore standard in microprocessors (for example) – we have all learned to rate the processing power of computer chips in terms of their clock speed. In an asynchronous circuit. it is a good idea to use one vertical bus for power and another for ground.1.1.1. requiring the designer to work out in detail the signal propagation delays through all possible paths to make sure that the circuit will work as intended.3 Synchronous logic The circuits in this chapter introduce the concept of synchronous logic design. As shown in Fig. In contrast. signals can change at any time. Writing down the schematic showing all pin numbers is a powerful debugging tool.4 Timing diagrams Timing diagrams are essential when analyzing synchronous logic circuits. and one or more outputs. 11.2 Breadboard layout To keep your breadboard organized.1.1.144 Hands-on electronics 11. one or more inputs. A neat layout is easier to debug than a messy one! Remember that the top and bottom busses on the PB-503 are not connected internally. 11. Usually there is one common clock for an entire circuit.1 General comments 11. and its transitions (low-to-high or high-to-low) – commonly referred to as rising or falling ‘clock edges’ – are used to synchronize all other state changes in the circuit. 11. in a synchronous circuit. timing diagrams for a synchronous circuit usually show the clock signal. it fails to work. output signals change only in response to a clock signal.1 Schematics For each of the following exercises be sure to write down a complete schematic diagram of each circuit you build. so that power and ground connections to each chip can be made with short jumpers. including pin numbers (but power and ground connections need not be shown). Time is shown on the x axis and signal logic levels on the y axis. 11. since it makes incorrect connections much more obvious.

145 11 Flip-flops tH : hold time OUT tTLH tPW tSU : setup time tH tTHL tPD tPW : pulse width tPD : propagation delay IN tSU tTLH : transition time (low--high) CLK tTHL : transition time (high--low) Fig. For example. Timing diagram with timing deﬁnitions for a rising-edge-triggered ﬂip-ﬂop.1 deﬁnes several important terms. Although manufacturers often specify the typical propagation delay as well. 11. Note how the time is measured from the midpoint between logic low and high. are shown as either high or low – we tend not to worry in timing diagrams about analog details such as the exact voltage to which high and low correspond. The time between a clock edge and the resulting changing edge of the output is deﬁned as the propagation delay (tPD ). it is not guaranteed – safe designers heed the minima and maxima.2 is the simplest ﬂip-ﬂop: an RS latch made of cross-coupled NANDs (an equally simple RS latch can be made from . the measured propagation delay is almost always less than the maximum value speciﬁed by the manufacturer. The setup time (tSU ) is the minimum time that an input signal must be stable preceding a clock edge. These timing parameters are speciﬁed in the data sheets. The hold time (tH ) speciﬁes the minimum time that the input signal must be stable following a clock edge. The transition time measures the time required for a signal to transition from logic low to high (tTLH ) or from logic high to low (tTHL ).1 Simple RS latch The circuit shown in Fig. Note that the rising edge of the clock signal causes the output to change while the falling edge of the clock signal has no effect on the ﬂip-ﬂop. 11. 11. not the ‘typicals’. Fig.2 Flip-flop basics 11. The specs are usually worst-case values.1.2. 11. Manufacturers guarantee that the IC will perform correctly if the user satisﬁes the speciﬁed minimum values.

Keep in mind that what the ﬂip-ﬂop does for a given input may depend on past history.2. We have therefore taken advantage of DeMorgan’s theorem to write the circuit in terms of negativelogic OR gates. R and S refer to reset and set (reset is also known as clear).146 Hands-on electronics S 1 4 7400 Q S 1 4 7400 Q R 1 4 7400 Q R 1 4 7400 Q * but note that the state after LL input condition is removed depends on which input signal goes high first. In other words. Q is the opposite of Q. except when both S and R are asserted. How would the ﬂip-ﬂop’s operation be different if it were made of positive-logic NORs? (You don’t need to build this circuit to answer the question. so try a few different sequences of input states to make sure you understand what it is doing. to see the entire state table you need to try each input state for each of the ﬂip-ﬂop’s two internal states. Build and test your ﬂip-ﬂop and record its state table (what its outputs do for each of the four possible input states). while a high input does nothing. rather than blindly using the NAND symbol just because the 7400 is called a NAND gate by its manufacturer. 11.2 are active low – a low input forces a particular output condition. it is just as good a symbol for 1 of a 7400 as the NAND symbol. . NORs). since it makes the circuit’s operation easier to see at a glance. And it is actually 4 better to use the OR symbol here. 11. set turns Q on while reset (clear) turns Q off. if both go high “simultaneously.” state is undefined Fig. Note on assertion-level logic notation The inputs in Fig. and it is a kind of hybrid between positive logic and negative logic – the idea is always to choose the gate symbol that best clariﬁes the operation of the circuit. Simple RS latch made of two-input NANDs with state table. The use of DeMorgan’s theorem in this way is called assertion-level logic notation. Since the OR gate with inverted inputs has the same truth table as the NAND gate.) Leave the RS latch in place for use later in this chapter.

) Note that. Sample timing diagram for a (positive-edge-triggered) 7474 D-type ﬂip-ﬂop. in addition to its ‘fancy’ D and clock inputs.” state is undefined Fig. 11.2. but is insensitive to D at all other times. then tie them to +5 V to make sure they are inactive. 11.2 D-type flip-flop In practice.4). Provide the clock signal using a momentary-contact breadboard ‘debounced push-button’ and the D input using a breadboard logic switch. 11.3). 7474 D-type ﬂip-ﬂop with state table. 11. if both go high “simultaneously. Next check out the clocked operation of the ’74 (illustrated in Fig. The most commonly used ﬂipﬂop is the clocked D-type. RS latches are seldom used. and display . OUT output state is not specified preceding the first rising clock edge output is stable between rising clock edges unpredictable output due to changing input coincident with clock edge minimum setup time not satisfied IN CLK Fig. which remembers the state of its D input at the time of a clock transition. the ’74 retains the simple reset and set inputs of the RS latch (which are active low since internally the ’74 is constructed from NANDs). which is sensitive to rising clock edges (low-to-high transitions). We will use the 7474 D-type ﬂip-ﬂop (Fig. Test the reset and set inputs and explain what they do.3. 11. it is positiveedge-triggered. (Later in this chapter you will encounter a negative-edgetriggered JK ﬂip-ﬂop.147 11 Flip-flops 4 10 S 2 S Q 5 12 D 1 2 D 1 2 Q 7474 Q R 9 7474 Q R 1 6 clk 3 clk 11 8 Vcc = pin 14 Gnd = pin 7 13 X = Don’t Care * but note that the state after LL input condition is removed depends on which input signal goes high first. in other words.4.

i. however.10).) What happens now when you apply clocks? Try clocking the toggling ﬂip-ﬂop using a digital square wave from the function generator.5 V for TTL and 2. apply a low level to it) and see whether you can clock in a high level applied at D. (Be sure to deassert set and reset. not the clock signal. a good technique for precise timing of digital signals is to set both scope channels to 1 or 2 V/division and use the scope’s verticalposition knobs to overlay both grounds below the center of the graticule. Disconnect the D input from the logic switch and connect the Q output to the D input to make a toggling ﬂip-ﬂop. 10.e. The following exercises use a negative-edge-triggered JK ﬂip-ﬂop with set and reset. When using the debounced push-buttons on the PB-503. for the following exercises we recommend using the 74112 JK ﬂip-ﬂop. Various chips are available (e.5 V for standard CMOS). Show that information presented on the D input is ignored except during clock transitions by changing the state of D while the clock is low or high. then. and use the scope to look at the input and output simultaneously. . Measure the ﬂip-ﬂop’s propagation delay..148 Hands-on electronics the Q and Q outputs using logic indicators. try asserting reset (i. As mentioned above. it can do everything a D can do plus more.g. 11. Be sure to review the data sheet to ensure that you have the correct pinout.5) is slightly more complicated than the D ﬂipﬂop. What happens? Now check whether the set and reset inputs take precedence over the clock and D inputs: for example.e. This is sometimes called a divide-by-two circuit – explain what this means. the time from the clock transition to the change of output voltage – what is it? Is it about what you would expect for chips of this family? How does it compare with the manufacturer’s speciﬁcations? Be sure to trigger the scope on Q. the 74HC112 or the 74LS76) – which you use will depend on what you have to hand. see Fig.. you can easily measure the time at which each signal crosses the midpoint between logic low and logic high (1. 11. be sure to add a pull-up resistor (as you did for the logic-level switches.3 JK flip-flop The JK ﬂip-ﬂop (Fig.

valid output states: high and low. The third state is thus referred to as the ‘high-Z ’ state. it is common for some logic ICs to have three output states. As in the case of the RS latch above. There are situations in which the designer wishes the output to be neither high nor low! Rather. This feature allows multiple outputs to be connected in parallel. and write down the JK state table. Now add an inverter (made from a NAND if you like) from the J to the K input to make a D ﬂip-ﬂop. check the four possible input states. hence. A common example of the use of tri-state outputs is the . 11. Tie set and reset high and explore the JK’s clocked operation. and only two. in such situations. Drive J from a level switch and write down the state table to verify that the circuit acts like a D ﬂip-ﬂop. Next remove the inverter and connect J and K together. What does this circuit do? How is it different from the toggling D ﬂip-ﬂop? 11. Nowadays. you need to try each input state for both possible internal states..4 Tri-state outputs The ICs that we have used so far have two. the output should become a high impedance. as long as all but one of them are in the high-Z state. Try it out and write down the state table. Pinout of the 74112 JK ﬂip-ﬂop.5. (Note that the 74112 senses its J and K inputs only at downward transitions of the clock. one wants to turn off the output completely.149 11 Flip-flops 4 3 10 S J 1 2 11 S J 1 2 Q 74112 Q R 15 5 Q 74112 Q R 14 9 clk 1 2 clk 6 13 12 7 K K Vcc = pin 16 Gnd = pin 8 Fig.) Driving the clock from a debounced push-button and J and K from logic switches. it is referred to as negative-edge-triggered. i.e.

OE. We can explore the third output state using the 74373 octal D-type transparent-latch IC. Verify that the Q output follows the D input for both input states. Begin by wiring the ’373 as shown in Fig. The eight output pins are driven by the IC when the output enable OE is held low.6. (b) Input and output connections for testing the tri-state output. Measure the propagation delay between the input and output. Set D high and observe the output voltage (voltage at the output pin) as you vary the potentiometer setting (there is no need to be excessively quantitative here). . (If logic-level switches are available. The outputs are latched (held constant) while LE is low.150 Hands-on electronics connection of multiple memory chips (say 4 Mbytes each) in parallel to build up a memory system (say 64 Mbytes). This type of ﬂip-ﬂop is called a transparent latch. Repeat your observation with D low. and they go into the high-Z state when OE is high.) Be sure LE is high and OE is low.g. 11. 11. Does the output logic level vary as the input changes? Explain your observations. Explain why the potentiometer has little effect on the output voltage. (a) +5 V +5 V +5 V VCC 8Q 8D 7D 7Q 6Q 6D 5D 5Q LE (b) 1 kΩ 74373 D OE 1Q 1D 2D 2Q 3Q 3D 4D 4Q GND Q 10 k pot 1 kΩ Fig. and D inputs in place of the resistor-SPST switch combinations shown. 1 k or larger). Set LE low.6. Choose any one of the D. Q pairs and apply a valid logic level to the D input. Unlabeled resistors merely need to be large enough to prevent excessive current ﬂow from +5 V to ground (e. use them for the LE. (a) Pinout and power connections for the 74373. The ’373 is often used to drive a data bus. Data are transferred from the input to the output while LE is high. The ’373 is equipped with a latch enable LE instead of a clock. and is equipped with three-state outputs to allow the bus to be driven in parallel by multiple data sources (‘drivers’).

i. Divide-by-four ripple counter. 11.7. First make the two-bit asynchronous or ‘ripple’ counter shown in Fig.151 11 Flip-flops Set LE high and OE high. Write down the +5 S J clk K +5 Q0 Q R J S Q1 Q R K Fig. Explain your observations.7. record the input state. Does the output behave differently for input high and input low? Now set the input either high or low. Does the ’373 remember the last input state correctly? In your own words and based on your observations. To see the binary counting pattern. that the output square wave changes state at 1 the frequency 4 of the input clock.5. (This circuit is asynchronous in that it does not have a common clock signal for all ﬂip-ﬂops. watch clock and Q0 . and verify that it divides by four.. explain the operation and features of the 74373. Write down its state diagram. Clock it with a square wave and look at the clock and the outputs on the scope. .e. always triggering on the slower waveform. Observe the output voltage as you vary the potentiometer. then Q0 and Q1 . 11.5 Flip-flop applications 11. independent of the state of OE. then set LE low. 11. Set OE low.) Clock it from a push-button while looking at the two Q outputs with logic indicators.1 Divide-by-four from JK flip-flops Ripple counter Cascading two toggling ﬂip-ﬂops makes a divide-by-four circuit. The input state at the moment that LE was set low should have been latched internally. otherwise known as a two-bit counter.

but if you are sending a clock pulse to a counter made of high-speed devices. Hook up an SPDT (single-pole-double-throw) switch as in Fig. Set the scope’s sweep rate in the vicinity of 0. You can see that. use ‘normal’ triggering. (Be sure to look at both edges of Q1 .8. This means that the switch closes and opens repeatedly for a period of order milliseconds until the contacts settle down. Play with the trigger threshold and sweep rate .) Synchronous counter Now conﬁgure the circuit in its synchronous form (Fig. Keep your synchronous counter (and RS latch) for the next exercise. there is usually an effect called ‘contact bounce’. Explain why not. timing diagram and label the states with their numeric values (0 through 3).5. 11. 11. This makes no difference if you are turning on a light. 11. Use the scope to conﬁrm that the ripple is gone. 11. Synchronous divide-by-four counter. and look at the output of the switch as you open and close the contact. Turn the scope’s sweep rate up until you can see the ‘ripple’: Q0 and Q1 don’t change at the same time. the counter can react to each bounce and therefore count a random number of times.152 Hands-on electronics +5 S J clk K +5 Q R K Q0 J S Q1 Q R Fig.1 to 1 ms/division. unlike D ﬂip-ﬂops.2 Contact bounce When a mechanical switch closes or opens.8). and make a timing diagram showing this. interpreting Q0 as the low-order bit and Q1 as the high-order bit of a 2-bit binary number. First use the scope to observe contact bounce directly.9(a). JK ﬂip-ﬂops are natural for building synchronous counters – explain.

(b) A NAND latch is used as a debouncer. to see if you can discern the bounces. Next.9. (a) Looking at contact bounce by driving a divide-by-four counter from a switch. Write down some typical sequences of states.) Connect the switch to the latch as shown. Make a sketch of the observed waveform.) 11. switch debouncing is one area in which RS latches continue to hold their own.5. and use the output from the latch to clock the counter. 11. 11. (Although D and JK ﬂip-ﬂops are used for most ﬂip-ﬂop applications. Connect the digital output of the function generator to the D input and connect the Q output to a logic indicator. clock your two-bit counter from the switch and see what happens. connect the output of the RS-latch debouncer to the clock input of a D-type ﬂip-ﬂop. Explain why this works. Set the function-generator .3 Electronic coin toss The RS-latch debouncer from the previous section can be combined with a D-type ﬂip-ﬂop to create an electronic coin-toss game.153 (a) 11 Flip-flops (b) +5 +5 1 4 7400 ÷4 ÷4 1 4 7400 +5 Fig. (The PB-503’s momentary-contact switches are already debounced by RS latches built into the unit. To build the circuit. How is the contact bounce affecting the sequence? RS latch as debouncer Now use the RS latch from the ﬁrst exercise as a switch debouncer (Fig. Verify that the counting sequence is now correct.9(b)).

Is the sequence random? Repeat another sequence of ten.154 Hands-on electronics frequency to about 10 Hz. How many tosses are required to determine if the system is truly random? If you were an unscrupulous game designer. Operate the switch and observe the output. You can assign the LED ‘on’ to be heads and ‘off’ to be tails. Record a sequence of ten coin tosses. how could you skew the ratio of heads to tails? . Write down the circuit diagram with pin numbers and explain how this circuit works.

the latter are particularly useful when a long pulse of stable and reproducible width is needed.g. 74121 (or similar) one-shot. As an example of the use of an address counter. ’122.12 Monostables. for every circuit you build. a schematic is essential when debugging subtle errors. Note The circuits in this lab are rather involved. and many of the details of their design are left for you to work out. 74189. Futhermore. etc. there are also available the family of timer chips (such as the 555). counters. Such designs often make use of pulses of various durations. In addition to monostables of a given logic family (such as the 74121. Monostable multivibrators are the usual solution. two 7490 and one 7493 counter. you will store and retrieve information in a small memory chip. Apparatus required Breadboard. to standardize a pulse from a push-button. oscilloscope. multiplexers. 74150 multiplexer. and RAM This chapter will introduce a variety of techniques that are important in sequential-logic design.). 7489. In this chapter you will also explore counters and their uses in timing and addressing. Be sure to write down the circuit’s schematic. e. with pin numbers. 7400 NAND. two TIL311 displays. You will ﬁnd the schematic especially useful should your circuit not work. or 74219 RAM chip. ’123. Sometimes a logic pulse of a given width needs to be formed in response to a particular input condition. A simple review of the schematic will often reveal the source of the problem. You will need to work them out in advance if you are to have any hope of completing the exercises in a timely fashion! 155 . assorted resistors and capacitors.

These circuits have only a single stable state.) r Bistable. 12-bit. r Astable. Counters are so useful that IC manufacturers provide 4-bit (and more) counters as a single chip. hence the name multivibrator. and will operate up to tens of seconds. They are thus stable in both of their allowed states. They are considered to be hybrid analog/digital chips in that the digital output is typically determined by the RC time constant of an external analog circuit connected to the chip.2 Counters In the last lab you wired up a divide-by-four circuit. Another term for monostable multivibrator is one-shot. (You used a 555 as an astable multivibrator in chapter 9. These circuits have no stable state but keep changing from one state to the other. They are very useful as clocks or oscillators. or greater range. They can be forced out of their stable state by a trigger pulse. These circuits can be induced to go from one state to another.. In digital designs. monostables such as the 74121 are preferred for pulse widths ranging from about 40 ns to 10 ms. r Monostable. longer. but they return to it after a very limited period of time. The primary use for monostables is to create pulses of known duration from triggering pulses of shorter.05%. since it is a device that ‘shoots’ once (i. Cascading means connecting multiple chips together (as in the multiple digits of a car’s odometer) so that each chip . That was of course a 2-bit binary counter. or variable duration.156 Hands-on electronics 12. Monostables are widely used to generate ‘gate’ signals for counter circuits.1 Multivibrators Multivibrator circuits fall into three general categories. 12. The more common term for bistable multivibrator is ﬂip-ﬂop.e. and can remain in either state permanently after the input signals have been removed. issues an output pulse) each time it receives an input signal. and for applications in which the pulse width must be stable to better than 0. but their pulse widths are not as predictable or stable. the 555 timer is the best choice for pulse widths ranging from milliseconds to hours. Among monostables. with carry-in and carryout connections that allow them to be ‘cascaded’ in multiple stages for 8-bit.

5 R0 R9 2 3 6 7 Fig. which are easier for humans to interpret than binary. with various arrangements of set.1. counters. . 12. Counters are available in both binary and decimal versions and in synchronous and asynchronous (‘ripple-through’) conﬁgurations.e.3 Experiments 12. R0 R9 QB QC QD . Some are actually bi-quinary counters. i. Note that decimal counters are also referred to as decade or BCD (‘binary-coded decimal’) counters. this makes them useful for driving decimal displays. synchronous counters issue a carry-out. the second produces a divide-by-ten sequence in +5: pin 5 GND: pin 10 14 7490 12 9 1 8 11 QA . 12.1) is a negative-edge-triggered bi-quinary ripple counter. and clock inputs. except that they roll over (and possibly issue carry-out) at 9 rather than at 15. possibly issuing a carry-out signal to the next stage. Four-bit binary counters count from 0–15 and then ‘roll over’ to 0 again.) Decimal counters work basically in the same way as binary counters.2 . Pinout of 7490 decade counter. and RAM counts when the preceding one ‘rolls over’ from its maximum count back to zero. It advances from one state to the next on the falling edge of its clock input. the high-order output bit from the preceding stage serves to clock the next stage. while for negative-edge-triggered asynchronous counters. a divide-by-ﬁve stage coupled to a divide-by-two. (More speciﬁcally.. reset. It consists of a 1-bit divide-by-two stage and a 3-bit divide-by-ﬁve stage that can be cascaded two different ways. . 12. multiplexers.1 Bi-quinary ripple counter The 7490 (Fig.3.157 12 Monostables. One way produces a decimal counting sequence (0 through 9).

e. clock it with a digital square wave. The data sheet can be obtained from the Texas Instruments website. (Also. DPL and DPR connect to LEDs for left and right decimal points. . 0010. 12.com/ . as explained in the following paragraphs. TIL311 numeric display The TIL311 is a handy (but expensive) hexadecimal display that combines. pins 6. and verify that the output is indeed a symmetrical (i.2. . First ﬁgure out how to conﬁgure a 7490 as a divide-by-ten. so that as successive clock pulses are applied it sequences through the states 0–9 in order (0000. Also. it 1 http://www. Write down the state table and sketch the timing diagram for the four output bits with respect to the clock input. each with its own driver circuit.) Next conﬁgure your 7490 as a decimal counter. 12. Note that. Pinout of TIL311 hex display: D0–D3 are data inputs.158 Hands-on electronics 1 +5 2 D1 3 D0 4 DPL 5 LE +5 14 D2 13 D3 12 DPR 10 TIL 311 7 GND BI 8 Fig.ti. and BI = high blanks the display. in a single package. 22 LEDs. LE = high latches the input data. which the high-order bit is a square wave at one-tenth the frequency of the input clock. the TIL311 has three notches on its package (rather than one). 0001. and a decoder that decides which LEDs to turn on for a given input state. .2). . Display the state of the counter with a TIL311 hexadecimal LED display (Fig. 1001). 9 and 11 are missing. high half the time and low half the time) square wave at one-tenth the input frequency. a latch that can store the four input bits.1 Although the TIL311 is a TTL device. don’t forget to write down your complete schematic with pin numbers. as indicated in Fig 12. unlike most chips.2.

The counter will count up the number of clock pulses. In brief. the predicted output pulse width is given by ln 2Rext Cext (74121) (12. 12.3).. i. with D3 being the highorder (23 ) bit. For the RC timing network.e. Ground BI (blanking input – when high the display is blank) and LE (latch enable – latches input when high). multiplexers. 12. To determine experimentally whether your one-shot is functioning properly.159 12 Monostables. Since you don’t want to display a decimal point. be sure to use a current-limiting resistor in series with the input pins. If you wish to experiment with the decimal place LEDs. which is proportional to the duration of the pulse from the one-shot. since CMOS logic levels satisfy the TTL input criteria. D3. and RAM will display correctly the output of a CMOS chip. (74HC123) Rext Cext 2 But the converse is not true: TTL logic levels do not satisfy the CMOS input criteria. D1. Clock your counter from a debounced switch and conﬁrm that it and the display both work. Save it for use in the following sections. 12.1) tw = K Rext Cext (74LS123) . connect the digital signals for the four bits to the pins labeled D0. . counters. Clock your circuit from a digital square wave at several hertz and verify that it works. See the TIL311 data sheet for additional information concerning these features.4 shows the pinout of the 74121 and 74123 monostable multivibrators.2 Monostable multivibrator The object of this exercise is to design a circuit that generates a pulse of about 500 s duration. Fig.3. use a conveniently sized resistor and capacitor – the timing rules vary by family and type. present the counter with a stream of clock pulses only while the one-shot is ﬁring (see Fig.2 To display a 4-bit hexadecimal number. so be sure to refer to the correct data sheet for your one-shot. leave the DPL (decimal place left) and DPR (decimal place right) pins open. D2. What are the state table and timing diagram for the four outputs? Try out the R0 and R9 inputs – what do they do? Add a second 7490 and TIL311 so that you can count from 0–99. use its output to gate the clock to your two-digit decimal counter.

3. Notice how the gated clock signal is simply the logical NAND of the gate and clock signals. Use a debounced switch to trigger the monostable.3. Trigger 5V Q Rext Cext 11 10 6 15 5V tW Rext Cext 14 13 A1 A2 3 4 Q B 5 A 1 B 1 2 Trigger 3 Q Q Trigger Q 74121 Clear 74123 4 Fig. Pinout of ’121 and ’123 one-shots with external RC timing network (see the data sheets for details). 12. there are either one or two A inputs (see Fig. 12. 12.160 Hands-on electronics Gated Clock Gate tW Clock Fig. There is always only one B input. The A and B inputs can be conﬁgured either to inhibit triggers. Design and build logic that produces a gated clock signal as shown in Fig. 12. (Calibrate your time scale by using the scope to measure carefully the period of your digital square-wave clock input. Timing diagram for a gated clock signal. Choose an appropriate input clock frequency for timing the duration of the one-shot pulse.) The clock frequency should be low enough that . Be sure to write down the schematic complete with pin numbers. (In the above equation.4. or to produce a trigger from a rising B input or falling A inputs edge.) The output pulse begins following a rising edge at the trigger input. Clock the two-digit counter with the gated clock signal.4). K is a parameter specifed on the 74LS123 data sheet. Depending on the speciﬁc chip used.

a substandard pulse might also be produced then. (root-mean-square) duration. Build the gated counter as described above. . Record the complete circuit diagram including all pin numbers. (Save your two-digit counter for use in the next section. a standard problem arises.s. 12. Why is your circuit insensitive to this problem? substandard clock pulses Gated Clock Gate Clock Fig. Since substandard clock pulses might fail to meet setup or hold requirements of ﬂip-ﬂops and counters. counters. Plot your results as a histogram and compute the mean and r. Include a push-button reset that zeros the counter.5).g. If the signal from the pushbutton arrives while a clock pulse is in progress.161 12 Monostables. When gating a clock is necessary.m. Similarly. one normally uses a pulse synchronization circuit such as the 74120 or a pair of cascaded ﬂip-ﬂops to ensure that signals used to gate clocks do not change state during the clock pulse. 12. but high enough that the width of the one-shot pulse can be measured accurately. and why? What clock frequency did you choose. Substandard outputs can result when gating clock signals. What pulse width is implied by the value of the counter. it is wise to avoid gating clocks whenever possible. and why? Repeat the measurement about ten times over a period of ﬁve minutes in order to determine the reproducibility and stability of the output pulse width. and RAM the counter does not go past 99.5.) Note on gating clocks When gating a clock with a signal that is independent of the clock (e. the push-button). Reset the counter and trigger the monostable. since the one-shot pulse might end during a clock pulse. multiplexers. a pulse of substandard width might be produced (see Fig.

if you want its output to be high during counter state i. 12.162 Hands-on electronics The 74123 ICs have several additional features that we haven’t explored here. ground data input i. For example. Record the truth table for this function.. It can also be used to implement logic functions. See the data sheets for details and operating rules.6) to the outputs of the low-order counter chip from the previous exercise (leave them connected to the hex display also). Note that the ’150 has an enable input that needs to be held low. and see if you are right – if not. each designated by a unique numeric address. under control of an input number in the range 0 to n − 1.4 RAM A random-access memory (RAM) is a chip containing a large number of ﬂip-ﬂops. conﬁgure the ’150 to identify which of the numbers from 0–9 are prime. The retrigger feature allows the output to persist longer than the time speciﬁed by Eq. in deciding when to open the hot-water valve in a washing machine). For example. Hard-wire the inputs appropriately and connect the ’150 output to a logic indicator. Finite-state machines are often useful in control applications (e. Since the ’150 is an inverting mux. by connecting each of the n inputs to low or high in a desired pattern.3. any desired 1-bit logic function of the input number can be produced. Which select input is high-order and which is low-order? What experiment can you do to ﬁnd out? Do it and ﬁnd out.1 through the application of additional trigger edges while the output pulse is in progress. 12. 12. a different input will be selected. Hook up the select inputs (A–D) of a ’150 16-to-1 multiplexer (Fig. You can also use a mux plus a counter to generate an arbitrary timingpulse sequence: on each clock cycle. Clock the counter from a debounced push-button. clear allows the output to be prematurely terminated.g. Each ﬂip-ﬂop can . It can thus be used to select among n different input signals. and the output will be either high or low depending on the state of the corresponding input. As an example of an arbitrary logic function. ﬁx it.3 Multiplexer and finite-state machine A multiplexer (or ‘mux’) is a device that connects one of n inputs to a single output. 12. This is an example of a ﬁnite-state machine – it repeatedly goes through a cycle of n internal states. the ’123 is a retriggerable monostable equipped with a clear input.3.

12 10 6 4 3 2 D I4 D I3 D I2 D I1 WE ME DO4 DO3 DO2 11 9 7 5 7489 DO1 +5: pin 16 GND: pin 8 A B C D 1 15 14 13 Fig. the 7489 (Fig. the ﬂip-ﬂops are organized into multi-bit words. be accessed by address for reading or writing. four data inputs (DI1 through DI4 ) for writing a value into the word being addressed. Pinout of 7489 16×4 RAM. 64 bits in a chip is nothing nowadays. Pinout of 74150 16-to-1 multiplexer. Each has four address bits (labeled A through D) for selecting words 0–15.6.163 12 Monostables. multiplexers. Frequently. 74189.7. 12. 12. counters. For example. with each word separately addressable. . Of course. and four data outputs (DO1 through DO4 ) for reading the word being addressed.7). 12. and RAM 9 16 17 18 19 20 21 22 23 1 2 3 4 5 6 7 8 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 Enable W 10 74150 +5: pin24 GND: pin12 A B C D 15 14 13 11 Fig. but it serves conveniently to illustrate the random-access-memory principle using a relatively simple chip. and 74219 are pin-compatible 64-bit RAM chips organized as sixteen words of 4 bits each.

Open-collector outputs RAM chips are designed for easy multiplexing with a minimum of additional components. Connect the data inputs and we to level switches. be sure to ground me. The chip accepts input data and puts out output data only when me = low. the word being addressed is read nondestructively.g. This causes the state of the inputs to be recorded in the word being addressed. to allow turning off the outputs of all but one chip. since to increase the total amount of memory available in a circuit one often wants to connect the outputs of multiple RAM chips together. but the ’93 counts from 0–15. Regardless of the state of we. thus eliminating the need for pull-up resistors and also improving the rise time when driving high capacitance. This means that the output transistors will not operate properly unless a pull-up resistor to +5 is provided for each one. If you apply some ingenuity. Use the address counter and the write-enable switch to program your memory to any desired sequence of hex digits (be sure to record what sequence you choose). not ten.. the word being addressed appears at the output. The 7490 and 7493 are pinout-compatible. Therefore.g. Then clock the address counter with a digital clock at a . you can spell out messages using the letters A–F plus I (1) and O (0) (e. before hooking it up replace your 7490 decimal counter with a 4-bit binary counter (e. Since we are not worrying about speed here. this is accomplished by making the data outputs open-collector (rather than the standard TTL ‘totem-pole’ output circuit). In the case of the TTL version of the 7489. the 7493). any convenient resistor in the range of a few hundred ohms to 10 k is suitable. The ‘master enable’ (me) signal is provided for use when the outputs of multiple chips are connected together. When the address bits change. FEED B0B A D10DE). Hook up the counter’s outputs to the address lines of the RAM and display the output data with your second TIL311. allowing all sixteen words of memory to be addressed. More modern memory chips use three-state outputs. Since the RAM chip contains sixteen words.164 Hands-on electronics To write a 4-bit word into the memory. we (write enable) is brought low. When we is high. the outputs settle to their new value after a propagation delay called the read access time. Clock the 4-bit binary counter from a debounced push-button and verify that it counts through all the hexadecimal (base 16) digits from 0 (binary 0000) to F (binary 1111).

counters. How could you use the ’150 mux to shorten the sequence to any desired fraction of the sixteen addresses? How could you use it to insert blank spaces between words? .165 12 Monostables. and RAM frequency of a couple of hertz and watch your message appear! Record the complete circuit diagram with pin numbers and explain how this circuit works. multiplexers.

.

converting between analog (voltage or current) and digital (binary-number) information. 74138. resistors. Please be sure to work through these circuits in advance.. capacitors. The reverse process is called analog-to-digital (A/D) conversion. complete schematic diagrams signiﬁcantly improve debugging efﬁciency.e. The original signal will then be re-created from the digitized data using a DAC080x D/A chip. four 7474. easy-to-use. 74191. you’ll use an ADC080x 8-bit successive-approximation A/D chip to digitize (i. 7400. Having learned the basic operating principles. As always. TIL311. i. 741 op amp. Apparatus required Breadboard. as well as how computers are used in laboratory research and process control. 311 comparator. The devices that carry out these conversions are called DACs and ADCs. The availability of highspeed. inexpensive digital⇒analog and analog⇒digital converter chips has dramatically changed the way audio and video information are recorded and processed. The process of converting digital information into voltages or currents whose magnitudes are proportional to the digitally encoded numbers is called digital-to-analog (D/A) conversion. ADC0804 (or similar).. This exercise will also allow you to explore the limitations of ADC and DAC operations. respectively. convert to digital) an arbitrary AC signal. after building a simple DAC from a digital counter and an op amp. oscilloscope.e. you will continue your exploration of analog/digital conversion by building a 4-bit tracking ADC. otherwise it is highly unlikely that you will successfully complete the exercises in a timely fashion! Carefully study the manufacturer’s data sheets which provide extensive details on operation and performance. 74112. In this chapter.13 Digital↔analog conversion In this chapter we will study simple techniques for generating and reading voltage or current levels. 167 . 7432. DAC0806 (or similar).

4R. with Q3 the MSB (most signiﬁcant bit) and Q0 the LSB (least signiﬁcant). 13. the output voltage will be proportional to this current.1(a)).168 Hands-on electronics 13. (b) output waveform resulting from input counting sequence.1(b)). If you double the resistance. To produce a desired output voltage. 2R. If a resistor R is connected from a voltage V to the inverting input of the op amp. Suppose you have four resistors with the resistances R. the noninverting input is grounded. which is tied to the output through a feedback resistor. Given a suitable feedback resistor such that the op amp does not saturate. half as much current will ﬂow. and ripple-clock and terminal-count outputs for use when cascading multiple stages. we get a 4-bit digital-to-analog converter. we can also increment or decrement the counter to get a voltage that changes in time in stepwise fashion (see Fig. and the inverting input. In addition it has four parallel-load inputs. This output can. be observed on an oscilloscope or other measurement device. 8R D0 4R D1 2R R1 0 Volts (0000) R = R1 VH = 5 V VL = 0 V Analog Out D2 R D3 _ op amp + (a) (b) −9. 13. and 8R. The corresponding current ﬂows will be in the proportion 8:4:2:1 (see Fig. we can load into the counter any desired value. and 8R (in order from MSB to LSB).4 Volts (1111) Fig. If we feed the counter outputs to the inverting input of an op amp through resistors R. The current into the feedback resistor will be proportional to the number that corresponds to the state of the counter. 2R. of course. A 74191 counter has four outputs Q3 through Q0 . (a) Simple D/A converter. acts as a ‘virtual ground’. 13. count-enable and count-direction (up/down) inputs.1 A simple D/A converter fabricated from familiar chips Recall that when an op amp is set up as an inverting ampliﬁer. . a current V/R will ﬂow. 4R.1.

To see a stable display of the waveform.50 k. clock it from a debounced switch. if we wanted to produce accurate analog output voltages. Despite the common misconception that modern electronics is strictly digital. .1b). Explain in your own words how this circuit works. Write down a complete circuit diagram with pin numbers.7 k resistor for Q2 . Next. The output should be a ﬁfteen-step staircase waveform (Fig. with each step having approximately the same height. Moreover. as described above. 13. for example) is only an approximation. 7. 5. Test it counting both up and down – you can control which way it counts using the ‘u/d’ input. To reduce the chances of hooking up the circuit incorrectly.3 k feedback resistor. Use a 2. we would need to take into account the inevitable small differences among the high and low levels of the counter outputs. and verify that it goes through all sixteen states in order. and 8 of the counter? Measure the four resistances and the high and low voltage levels of the four counter outputs (Q0 –Q3 ). and explain each DAC output voltage. you will build such a 4-bit DAC. a 10 k resistor for Q1 . hook up the counter outputs to the summing junction of the op amp through resistors.169 13 Digital↔analog conversion To demonstrate D/A conversion. We shall not worry about these reﬁnements. we would need precision resistors. What full-scale output voltage do you expect (i. For all practical purposes. a 4. analog electronics is still going strong.e. The digital representation of any waveform (music.0 k. Connect a 1 kHz digital signal to the clock input of the counter and view the analog output on an oscilloscope. and 20 k. for example 2. 5.2 k resistor to connect to Q3 . Of course. our everyday world is analog. To smooth out the discontinuities of digitized waveforms requires analog electronics. when the counter is at 15 in decimal or 1111 in binary)? What do you observe? Is the staircase rising or falling? Why is this? What simple change can you make to reverse the direction of the staircase? What are the output voltages corresponding to states 4. since it is our intention here merely to illustrate the basic idea of D/A conversion. Begin by setting up a 74191 counter and make sure it is working properly: hook up its outputs to a TIL311 display. and a 22 k resistor for Q0 . you can trigger the scope using the falling edge of the MSB. 10 k. and connect a 3. 6.

13. just as division is harder and slower than multiplication. the counter holds a digital approximation to the magnitude of the input. which you can vary between 0 V and −15 V. and taking the square-root harder and slower than squaring. a counter. since it has to count through all the intermediate values. Clock the counter at a few hertz and observe its state with the TIL311 as you vary the input voltage. and if you did it wrong the ﬁrst time. use some hysteresis by connecting a series 10 k resistor between the input voltage and the comparator noninverting input and 1 M between the comparator output and the noninverting input (see Fig. and a comparator. The slider controls the input voltage. ﬁx it. you stop counting when the comparator output switches states.170 Hands-on electronics 13. but. did you perhaps connect the comparator backwards? Explain. To stabilize the operation of the circuit. There are various ways of doing this.2). Given a DAC. Why is the output number never stable? How (if at all) does this affect the precision of the voltage measurement? Note that the tracking ADC is slow at following large input-voltage changes. At that point. Use a potentiometer to make the analog input voltage: connect one end to ground and the other to −15 V. . analog-to-digital conversion is harder and often slower than digital-to-analog.2 Tracking ADC To measure an analog signal you need to invert the process of D/A conversion. You can turn your 4-bit counter/DAC into such a tracking ADC by driving u/d from a comparator that compares the DAC output with the analog input voltage. Record the output numbers for a few different input voltages. a simple approach is to increment the counter (starting from zero) until the DAC output crosses the analog input. A simple variant of this circuit will follow (or ‘track’) changes in the input voltage. Using the comparator to compare the analog input to the DAC output. but it has good performance if the input voltage changes gradually. What do you observe? How should the comparator inputs be conﬁgured: which signal should go to the inverting and which to the noninverting input? Is the output number ‘homing in’ on the expected value? If not.

there are 2n possible output values. next try n/4. ﬁrst look at word n/2. then correct it to 2-bit accuracy. the logarithmic search has another name: successive approximation. you try each bit in turn.2 kΩ 1 kΩ _ +5 op amp + Analog In 10 kΩ comp 1 MΩ Fig. In the case of analog-to-digital conversion. 5n/8.3 080x ADC and DAC chips 13.) 13. or 7n/8 – and so on. 13. until you ﬁnd a word beyond the one you’re looking for).2.3 kΩ 4. This method will ﬁnd any word in at most log2 n steps.3. At the next step there are four possibilities: word n/8. if your word comes earlier. 3n/8. . A faster approach (and one used in many ADC chips) is based on the binary (or logarithmic) search algorithm.7 kΩ 2. but instead of trying each value in succession. (The polarity of the comparator inputs is left as an exercise for the reader. and so on. next try word 3n/4. If the word you want comes later in the alphabet.1 Successive-approximation ADC The technique just described is comparable to looking for a word in the dictionary by looking at each word one after the other until you ﬁnd the right one (or. starting from the MSB: ﬁrst generate a 1-bit approximation to the value. in which at each step you reduce the search range by half. For example. but if the DAC is far from the signal voltage it takes a long time to home in. If there are n output bits. Simple A/D converter. if you’re looking up a word in an n-word dictionary.171 13 Digital↔analog conversion U/D Clk +5 Load Enable 74191 22 kΩ QA QB QC QD 10 kΩ 3. then 3-bit. this is a ﬁne algorithm for tracking a slowly changing signal. As we’ve seen. if the word you’re looking for isn’t in the dictionary.

13. The digitized output thus measures the voltage difference Vin+ − Vin− .g. A few additional clock cycles are used during startup and after the conversion to latch the data on the output lines. as shown in Fig. The successive-approximation algorithm used in these chips requires sixty-four clock cycles to complete a conversion. . All of the input and output control signals are active-low. rd. 13. labeled cs.3. (b) The on-chip self-clocking conﬁguration. C Vin Vin(+) Vin(−) A GND Vref /2 D GND D4 D5 D6 D7 (MSB) ADC080X Fig. The self-clocking option uses an on-chip oscillator (with Schmitt-trigger timing input). The clock can originate from either an external or internal (‘self-clocking’) source. and the outputs have tri-state capability.3. (a) Pinout for the ADC080x series of A/D converters. and wr. The chips are general-purpose ADCs that can be used as stand-alone converters or interfaced with a computer or other logic system.1 RC R WR CLK IN DONE 4 CLK IN Internal Clock Osc. in combination with an external resistor and capacitor that determine the period. There are three input control lines. An A/D conversion is started (a) CS Inputs RD +5 V VCC CLK R D0 D1 D2 D3 Outputs (LSB) (b) 19 CLK R = TCLK ~ 1.172 Hands-on electronics ADC080x The ADC080x series of chips are inexpensive 8-bit successiveapproximation A/D converters. The logic inputs and outputs are compatible with both TTL and CMOS. ADC0804). They accept differential inputs for increased common-mode-noise rejection capability. The ‘x’ in ADC080x means that multiple versions of this IC exist (e. Note the locations of the most signiﬁcant bit (MSB) and the least signiﬁcant bit (LSB).

Adjust the input voltage and observe the digital output. 13.e. and how does the ADC respond to small excursions outside this range (not less than ground and not more than VCC )? With what precision does the ADC measure the input voltage? . start the digitization now!). called analog ground (a gnd) and digital ground (d gnd). Connecting done to wr (as shown in Fig. wr simpliﬁes to A/D start (i. Is the plot linear? What is the input range. when low. while the output 11111111 should correspond to a voltage near VCC . Connect pins 1. connect rd to ground. Connect pin 4 to ground using a 50 pF capacitor. 13. so you should ground Vin− . We shall use single-ended inputs. the analog and digital grounds are kept separate to reduce the analog noise introduced by rapidly changing digital signals.) Connect an ADC080x chip as shown in Fig.173 13 Digital↔analog conversion by bringing rd and wr low simultaneously. apply a variable voltage between +5 V and ground to pin 6. With cs tied to ground. We will ignore these effects and use a common ground for both pin 8 and pin 10. the output lines are active – therefore. A digital output of 00000000 should correspond to an input very close to zero volts. the ADC will repeatedly digitize the input voltage difference. The chip has two grounds. (A more complete explanation of the ADC080x’s features can be found in the manufacturer’s data sheets.3(a)) puts the ADC into a ‘free-run’ mode. Connect pin 20 to +5 V power.3(a). Once the digitization completes. 2. The range for the input voltage difference is determined by the input voltage reference – we are using VCC . 8. Once triggered. and 10 to ground. rd is equivalent to output-enable – rd high puts the outputs into their highimpedance state. What measured input voltage corresponds to the binary output 00000001? What measured input voltage corresponds to the binary output 11111110? Measure several other input-voltage values and plot the input voltage versus digital output. Using the 1 k pot. in which the completion of one conversion initiates the next. cs is used in microprocessorbased applications – for this exercise it should be connected to ground. so connect wr to a debounced push-button as well as to done. 7. To ensure free-run status following power-up. and add a connection to a push-button. Connect pins 11–18 to LED logic indicators. Connect pins 3 and 5. Place a 10 k resistor between pins 4 and 19.. the result is latched onto the output bus and a done pulse is issued (pin 5). and leave pin 9 unconnected. In many applications. wr may need to be brought low manually.

and 128R. these need to have better than 0. 13. 8R. Comment on your observations and measurements. The DAC0806 . If you need more accuracy. Our home-brew D/A converter above used 4 bits to distinguish sixteen different voltage levels.and 8-bit precision. the DAC0808. and explain in your own words how this circuit works. respectively.e. 64R. Replace the 50 pF capacitor with a 100 pF capacitor. 16R.5% tolerance. Write a complete circuit diagram with pin numbers.) What is the sampling rate? Does the conversion time depend on the input voltage? Taking into consideration the value of the external RC network. and less readily available. of which the DAC0806 is the least expensive – it is an 8-bit DAC with 6-bit precision. and explain the waveforms at pins 4 and 19.4. With the DAC0806 we can distinguish not sixteen but sixty-four different voltages (at least). you can use an external op amp. What happens if you remove the 100 pF capacitor completely (i. Save your ADC circuit for use in the next part. Replace the 50 pF capacitor. 2R. the expensive. and some small stray capacitance is common on breadboards such as the PB-503. do not use any external capacitor)? Estimate the value of the stray capacitance. Some D/A converter chips (e. Explain what happens to the waveforms. while the related DAC0807 and DAC0808 have 7.. 32R.. why not? (Hints: parallel capacitors add linearly.) Observe. as shown in Fig.) Chips of the DAC080x family output a current that is proportional to the value of the digital input. NE5018) include an op amp on the same chip. does the measured sampling rate agree with your expectations? If not. What exactly is the difference between the ‘top-of-the-line’ DAC0808 and the ‘budget’ DAC0806? Within the DAC0808 chip are eight resistors of nominal resistance R. you can use the DAC0807. (You may use whichever of these chips is available for this exercise. 4R. (Conversion time is the amount of time an ADC requires to digitize an input voltage.g. or one of the more sophisticated chips that are available. The DAC080x family does not! To convert the DAC’s output current to an analog voltage. The DAC080x D/A converters The DAC080x is a family of popular D/A converter chips.174 Hands-on electronics Estimate the conversion time by looking at pin 5 with the oscilloscope. sketch. For 8-bit accuracy.

but their tolerances are not as good. Comment on your results. 13. such that A0 is the MSB and A7 the LSB. such that the DAC0806 is not guaranteed to put out 256 distinguishable voltage levels.4. 4. Explain in your own words how this circuit works.175 13 Digital↔analog conversion −15V C = 0. 13. Pinout for the DAC080x series of D/A chips (with an output op amp added). Consult the manufacturer’s data sheet to see what else your D/A converter chip contains. to avoid inaccuracies due to voltage variations in logic levels on different input lines. Does it match the speciﬁed precision? Be sure to write down a complete circuit diagram with pin numbers.7 kΩ Vout Op Amp IQ + +5 V (LSB) Inputs A0 A1 A2 A3 A5 A4 DAC0806 Fig.1 µF N. Note that the bit order is the reverse of that used for the ADC080x. Connect the outputs of the ADC you built above to the inputs of the DAC.4. Wire up a DAC080x chip as shown in Fig. Measure the precision of your DAC. Comment on your observations and measurements. also has eight resistors.7 kΩ _ 4.C.7 kΩ Output (MSB) Inputs COMP GND VEE Vref(−) Vref(+) VCC A7 A6 4. but only sixty-four. These chips also contain circuitry to buffer the input signals and standardize their voltage levels. What is the output voltage that corresponds to the DAC digital input 00000001? What is the output voltage that corresponds to the digital input 11111110? Take a few more data points and plot output voltage versus digital input. This conﬁguration will give an output voltage between 0 and +5 V. .

the function-generator output is symmetric about ground. The most convenient AC input is a waveform from your function generator.5 Hz and try to measure the amplitude using the LED indicators. and comment on your observations. Measuring an AC input signal Replace the DC input with an AC input.5 V to the A/D input. . Warning: be careful not to exceed the ADC input voltage range! Input voltages larger than VCC or lower than ground may damage the ADC chip.5.5. Explain the relationship between sampling frequency and output frequency response. Apply a sine wave of 2 V amplitude centered at +2.176 Hands-on electronics +5 V R = 100 kΩ C = 100 µF C Vin V t t R _ + + R Op Amp To A/D input V Fig. However. 13. What happens to the output waveform as the input frequency approaches the sampling frequency? What happens when the input frequency exceeds the sampling frequency? Make several sketches of the input and output waveforms at various frequencies. Increase the input frequency (you may need to adjust the amplitude to stay within the range of the A/D input – why?). Method for producing a DC-shifted waveform. You can ‘freeze’ the A/D output using the push-button connected to pin 3. The waveform will thus be outside the ADC range 50% of the time! You can DC-shift the waveform as shown in Fig. Explain how this works! Compare the input and output waveforms. How well do they agree? Comment on your observations. Explain how the DC shift occurs and the signiﬁcance of the component values chosen. Set the frequency to ∼0. Suggest at least one other method for DC-shifting the input waveform. 13.

A/D and D/A conversions are merely approximations. The National Semiconductor LF398 is an SHA IC. You can also try adding nonlinear gain elements (say a logarithmic ampliﬁer) or ﬁltering to see how the sound is affected. 13. respectively. you would be able to store 32 s worth of sound.1 Digital recording You can convert your A/D–D/A circuit into a digital recording and audioprocessing system by adding an audio input. which you can make from 7493s or 74191s. Increasing the precision to 16-bit and sampling rate to 100 kHz increases the required space by a factor of 20. For example. buffered with suitable op amp circuits. The 32 k × 8 CY62256 memory chips require a 15-bit address counter.g. and the output is at pin 5.4. Can you ﬁgure out a way to program reverb? For faithful recording.177 13 Digital↔analog conversion Experiment with triangle and square waves of various frequencies. . Higher precision and higher sampling rates improve the approximation at the expense of increased cost and data size.4 Additional exercises 13. it is important that the input voltage be constant during the entire time of the conversion. To digitize and record a 30 s waveform with 8-bit precision and 10 kHz sampling rate requires 2. to 48 Mbit. on pins 3 and 8. and an audio output. As discussed above. Try recording a sound sample. Record and sketch a few input and output waveforms.4 Mbit of memory storage. An external capacitor (1000 pF) is connected between pin 6 and ground. at 1 kHz sampling rate. and comment on your results. then playing it back at various speeds. This capacitor is used to hold the analog data until the ADC has had an opportunity to process it. Be careful not to exceed the allowed input range. let’s say we’re recording the sound track for a TV commercial. Ground pin 7. This can be accomplished using a sample-and-hold ampliﬁer (SHA) to sample the input and hold it until the conversion is complete. What other control circuitry do you need? By varying the clock speed you can trade off ﬁdelity (sampling rate) for message length – e. a memory to store the digital data. There are separate analog and logic inputs. Pin 1 is connected to +15 V and pin 4 is connected to −15 V. You can use speakers for both input and output.

4. in order both to set and (possibly) reset each of the eight bits. you can build an 8-bit ADC using TTL or CMOS parts plus a DAC and comparator. you should ﬁrst spend some time understanding the timing cycle and drawing a complete timing diagram for the control logic (shown in Fig. Pay particular attention to the sequence and timing of signals at the beginning and end of the conversion cycle. The sixteen-state cycle is provided using a ’191 four-bit binary counter. In our successive-approximation circuit (Fig. The 311 comparator compares the DAC output with the analog input. the currently addressed bit is set to 0 in order to raise the DAC output voltage. 13.2 Successive-approximation ADC built from components To see ﬁrst-hand how the successive-approximation algorithm works. the circuit goes through a sixteen-state cycle. Timing and control logic If you choose to undertake this exercise. The ’138 1-of-8 decoder routes clock pulses (how many?) to each ﬂip-ﬂop in turn. and its output is connected to the D input of each ﬂip-ﬂop. starting from the MSB and ending with the LSB. Note that the DAC080x is a current-sinking DAC. what additional control logic is needed? Can you hear the difference it makes in ﬁdelity? How would you describe the difference. and see if you can ﬁgure out the answers to the following questions.7). and how would you explain it? 13. The successive-approximation algorithm consists of trying each bit in both the 0 and 1 states. 13. and its output thus becomes more negative as its digital input increases from 0 to 255. The connection shown thus provides negative feedback: if the DAC output is too negative.) What happens if you attempt to start a new conversion cycle while one is already in progress? Why is it important for the control ﬂip-ﬂops to be negative-edge . To accomplish this.6 and the lower left-hand corner of Fig. and if the DAC output is too positive. Each bit starts out at 0 and is then set to 1. the bit is set to 1 in order to lower it. In what state does the counter start out? In what state does it end? Exactly what function does each decoder enable perform? (Hint: each performs a slightly different function. the bit is set back to 0. otherwise it is left as a 1.7) the eight bits are stored in four ’74 dual D-type ﬂip-ﬂop chips.178 Hands-on electronics If you choose to add a sample-and-hold. 13. If the DAC output exceeds the analog input.

’ Verify that your timing diagram is correct.) Try to arrange the chips and connecting wires neatly. Measure the DC analog input voltage Vin with a digital voltmeter. In case a chip needs to be replaced.e. Attach the reset input to the other debounced pushbutton. Clock your control circuit with a digital square wave from the function generator. or why not? Begin by building and debugging the control logic by itself – but be sure to leave room for the additional chips! (If you prefer.6.179 13 Digital↔analog conversion E3 E2 E1 7 6 5 4 3 2 1 0 74138 A2 A1 A0 R J Start Convert K K Q J R Q LD Q3 Q2 Q1 Q0 74191 Q U/D +5 P 3 P2 P1 P0 E clk Done Fig. . Trigger the scope on the output of the second ﬂip-ﬂop and observe each of the other signals as you repeatedly issue ‘start convert.2)? Why. Complete ADC circuit Now add the rest of the circuit.3. You should be able to see the full timing sequence on the oscilloscope. you may choose to build and analyze the simpler control logic described below. Try several voltages over the full range and make a graph of your results. have a pulse of substandard width – see discussion in section 12. try to avoid overly tight wiring across the top of any chip. and provide the ‘start convert’ signal with a debounced push-button. 13. so that it is easy to see where each wire goes – some color-coding could be helpful. triggered: can the clock signal to the ’191 glitch (i. Use the breadboard logic indicators to display the data bits. Control logic for 8-bit successive-approximation ADC.

How good is your ADC? Is it linear? What is its zero offset (the number it puts out for Vin = 0).) Will this ADC always work correctly the ﬁrst time after power is turned on? Why. 8-bit successive-approximation ADC. 13.7. or why not? . and what is its slope constant (volts per output count)? What is its least count (the voltage change corresponding to one ADC count)? What is its full-scale voltage? Try raising the clock frequency.180 Hands-on electronics A0 A1 A2 A3 A4 A5 A6 A7 74112 74112 reset Fig. At what frequency does it stop working? Does this make sense? What do you think limits the conversion speed of this circuit? (Illustrate your answer with the relevant timing diagram.

thereby changing the time at which it crosses threshold. you can use the ‘diode trick’ to delay the clock relative to set. and then reset. by the clock edge if the analog output is too big. depending on the internal timing of the ﬂip-ﬂop (clock and set are changing simultaneously) – the question is whether the set signal goes away within the ﬂip-ﬂop soon enough so as not to override the clock edge. and circulate them around an 8-bit shift register. so that each ﬂip-ﬂop is ﬁrst set.) . This simple approach might or might not work.181 13 Digital↔analog conversion Simpler version of control logic You might want to build and analyze a simpler version of the control logic using parallel-output shift registers instead of the counter/decoder version discussed above. (The diode trick consists of adding a series diode to shift a digital signal in voltage. The idea is to produce eight bits. If it does not. of which seven are 1s and one is a 0. Connect each shift-register output both to clock and to set of a ﬂip-ﬂop.

.

06. D2 D3. A2 A1 A1.: A. 4. 1. 5 8 8 8 9 9 10 11 12 12. Principles of Electronic Instrumentation (Saunders.34 4. 8.17. 8. A8 A6 A6 A7. 2. 1. 1989). 9. 1. H. Simpson. 183 . D4 D5 D6 H.23–24. Simpson: Robert E.: Paul Horowitz and Winﬁeld Hill. 8.09 2. 8 thru 8. App. 1. 1. 6 2. D1.08–10 8. Barnaal: Dennis Barnaal. App. C 2. 1987). James Diefenderfer and Brian E.11–12 4. (5) 1.11. The Art of Electronics (2nd edition. A1 Suppl.Further reading The following table indicates the sections or chapters in four popular textbooks where you can ﬁnd additional background information for each chapter of our text.16–17 8.13. & H.14.02. & H.09. A 4 5 6 9. 10 10 10.10. 1.15–16 3 thru 3.16. 3.09. Introductory Electronics for Scientists and Engineers (2nd edition.01 1. 1989). 1994). 3 4. Cambridge University Press. Analog Electronics for Scientiﬁc Application and Digital Electronics for Scientiﬁc Application (reissued by Waveland Press. 8. 13 14 Barnaal A1. Holton.18.02–03.15–16. 12. 4. A4 A8 A8 A4.14.18–26.07–09.20 Simpson 1. Expt 1 2 3 4 5 6 7 8 9 10 11 12 13 D. 4.12–16.09–10.18–20 1. & H.32.25–28 2 thru 2. & H. 11 12 13 13 15 Key: D. 2.07. 1. 1 thru 1. A5. Prentice-Hall.19–20 4.14 4 thru 4. 9. 1.

.

185 .7 k 10 k 22 k 100 k 330 k 1M 10 M 1k a a b 1 4 W Number required 1 1 3 2 1 1 2 1 3 3 7 2 2 1 1 1 1 Capacitors 50 pF 100 pF 300 pF 0. or equivalent.2 k 3.47 F 1 F 100 Fb 1000 Fb Number required 1 1 1 1 1 3 1 1 1 1 1 2 W resistor.3 k 4. one digital multimeter with probes.0047 F 0. two oscilloscope probes with 10X attenuation.033 F 0. four 50–100 cm banana leads (two red and two black). one power tranformer (12.Appendix A Equipment and supplies General equipment you will need: one Global Specialties PB-503 powered breadboard.1 F 0. one Tektronix TDS 210 Dual Trace Oscilloscope. or similar. 33 68 100 330 560 820 1k 2.01 F 0.6 V each side of center tap). Analog components Resistors. 50 V capacitor.

two fat-pin adapter sockets. one diode bridge element. two 3. one 411 op amp. one 5. Digital components Component 7400 7404 7432 7474 7485 7486 7489 7490 7493 74112 74121 74138 74150 74191 74373 DAC0806 ADC0804 TIL311 Number required 1 1 1 4 1 1 1 2 1 1 1 1 1 1 1 1 1 2 . two VN0610N MOSFETs. transistors. three 2N3906 PNP transistors. three 2N3904 NPN transistors. analog IC’s four Si signal diodes. two VP0610L MOSFETs. one 311 comparator. two 2N5485 N-channel JFETs.186 Hands-on electronics Diodes.1 V 1 W Zener diode. one 555 timer. one red light-emitting diode. three 741 op amps. one 1 A Si rectiﬁer diode. Miscellaneous four alligator clips.3 V 1 W Zener diodes.

Several (e. However.net/ Electronix Express http://www.digikey. .com/ Jensen http://www.com/ Arrow Electronics http://www. so shop around! RadioShack http://www. Product information.com/ Digi-Key Corp http://www. We’ve included a few URLs to help get you started. the parts and supplies needed to complete the exercises within this book could be purchased from the companies below. availability. and pricing are easily found through a few quick web searches.tequipment. the publisher has no responsibility for the websites and can make no guarantee that a site will remain live or that the content is or will remain appropriate.1 At the time of going to press.com/ 1 The publisher has used its best endeavors to ensure that all URLs referred to in this book are correct and active at the time of going to press. Digi-Key Corp) even have links to product data sheets as part of their online catalog.newark. Pricing and availability may vary. Prices are reasonable and service is excellent. Equipment and supplies Suppliers of parts There are numerous companies selling electronic components and supplies.g.. Most allow customers to purchase small quantities directly over the web.com/ Newark Electronics http://www.arrow.187 Appendix A.radioshack.com/ Tequipment http://www.elexp.jensentools.

Appendix B Common abbreviations and circuit symbols Order-of-magnitude prefixes m = milli = 10−3 = micro = 10−6 n = nano = 10−9 p = pico = 10−12 f = femto = 10−15 k = kilo = 103 (or kilohm = 103 ) M = mega = 106 (or megohm = 106 G = giga = 109 T = tera = 1012 ) Mathematical symbols ∼ of order ≈ approximately equal to ≡ equals by deﬁnition change in ⇒ implies Electrical terms β = h FE = transistor current gain ω = angular frequency = ohm A = ampere AC = alternating current C = coulomb C = capacitance dB = decibel DC = direct current F = farad 188 .

Common abbreviations and circuit symbols f = frequency gm = transconductance H = henry Hz = hertz I = current L = inductance P = power Q = quality factor (of a bandpass ﬁlter) R = resistance V = volt VCC = most positive voltage in a circuit (positive supply voltage) VEE = most negative voltage in a circuit (negative supply voltage) X = reactance Z = impedance Electrical devices ADC C CMOS CRT DAC ECL FET JFET L MOSFET op amp Q R SPDT SPST TTL analog to digital converter symbol used in schematics for a capacitor complementary-MOSFET integrated-circuit family cathode-ray tube digital to analog converter emitter-coupled-logic integrated-circuit family ﬁeld-effect transistor junction FET symbol used in schematics for an inductor metal-oxide-semiconductor FET operational ampliﬁer symbol used in schematics for a transistor. can also refer to the latched output of a ﬂip-ﬂop or register symbol used in schematics for a resistor single-pole-double-throw switch single-pole-single-throw switch transistor–transistor-logic integrated-circuit family .189 Appendix B.

190 Hands-on electronics .

(C. V = IR+ 1 C t (C. they must have the same current ﬂowing through them. the charge Q stored on the capacitor is proportional to the voltage VC across it: Q = C VC .5) 1 C t = I0 R sin (ωt + φ) + I0 sin (ωt + φ) dt. which is out of phase with the sine by 90◦ . V . C. but out of phase with the voltage by 90◦ (since the derivative of the sine is the cosine.2) i.1) If the voltage across the capacitor is varying sinusoidally in time. Then.4.3) Kirchhoff’s voltage law tells us that. Since the resistor and capacitor are in series. the applied voltage must equal the sum of the voltage across the capacitor and that across the resistor: V = VR + VC .e.Appendix C RC circuits: frequency-domain analysis In many freshman-physics textbooks. the frequency-domain analysis of RC circuits is not explicitly treated. however. since the current I ﬂowing onto one plate of the capacitor is the time derivative of the stored charge. Then. At any moment of time. it is not necessarily in phase with the source voltage. Now. C. Here is a detailed derivation. it is not particularly difﬁcult.6) where we have made use of Eq. and we also have VC = Q/C = (1/C) these relations into Eq. (C. Substituting I dt t0 (C. t0 (C.1). Now consider a series RC circuit being driven by a sinusoidal AC voltage source (Fig. at any moment of time. we can write I = I0 sin (ωt + φ). however. we have chosen the zero of time to be a moment when the voltage across the source is zero.4) I dt. (C. allowing for an unknown phase difference between the current in the circuit and the voltage applied by the source. Suppose (for the sake of deﬁniteness) that V = V0 sin ωt. C.3. by Ohm’s law. and the derivative of the cosine is minus the sine). VR = I R. it follows that the charge must also vary sinusoidally. 191 . the current must also be a sinusoidal function.

Eq. C. C.10) I0 [cos φ − cos (ωt0 + φ)].6 using the substitution u = ωt + φ. thus I0 [cos (ωt + φ) − (cos φ − ω RC sin φ)] (C. ω RC (C. ωC (C.7) I0 [cos (ωt + φ) − cos (ωt0 + φ)].2: V (0) = 0 = I0 R sin φ − giving cos (ωt0 + φ) = cos φ − ω RC sin φ.14 can be simpliﬁed as V0 = I0 R cos φ + I0 sin φ. which clearly does not describe the behavior of the circuit.14) (C. can be determined by the condition V (0) = 0. which we assumed in writing Eq.1 thus φ = tan−1 1 . since sin ωt and cos ωt are independent functions of time.16) Eq. C.12) = I0 R[sin (ωt + φ) − sin φ] − ωC which clearly satisﬁes V (0) = 0. ωC (C.192 Hands-on electronics We can easily carry out the integration in Eq. C. (C.13) ωC Gathering and separating terms in sin ωt and cos ωt. and using V = V0 sin ωt.12 can be simpliﬁed using the trigonometric identities for sines and cosines of sums: V = I0 R sin (ωt + φ) − V = I0 R[sin ωt cos φ + cos ωt sin φ − sin φ] I0 [cos ωt cos φ − sin ωt sin φ − cos φ]. giving V = I0 R sin (ωt + φ) + = I0 R sin (ωt + φ) − I0 ωC ωt+φ sin u du ωt0 +φ (C. This can be satisﬁed for all times only if the constant is zero. .11) ωC I0 [cos (ωt + φ) − cos φ]. we obtain two equations: − V0 sin ωt = I0 R sin φ − I0 R cos φ + I0 cos φ = ωC I0 sin φ sin ωt ωC I0 R sin φ − I0 cos φ cos ωt.17) 1 Otherwise we could divide through by the constant to obtain cos ωt = 1. cos (ωt0 + φ).15) (C. C.9) Eq.8) ωC The constant of integration. (C. ωC (C. (C.15 states that a constant is equal to the same constant times a function of time.

resistor voltage. ωC This describes a right triangle with hypoteneuse of length V0 and sides of length I0 R and I0 /ωC (Fig.20) If we take the output as the capacitor voltage.1. we get a high-pass ﬁlter: Vout = I0 R = V0 1+ 2 1 ω RC .17. C. RC circuits: frequency-domain analysis Fig.193 Appendix C. which is a useful way of visualizing the relationship among the amplitudes of the source voltage. I0 ωC I0 R φ V0 φ I0 R cos φ I0 sin φ ωC Fig. we get a low-pass ﬁlter: Vout = I0 = ωC V0 (ω RC)2 + 1 .18) 1 2 ωC . C.19) If we take the output as the resistor voltage. (C. (C. The relationship is Pythagorean: V02 = (I0 R)2 + We thus have V0 I0 = R2 + I0 ωC 2 . C.21) . Right triangle represented by Eq. (C. Series RC circuit. and capacitor voltage. illustrating that V0 = I0 R cos φ + I0 sin φ.2. C.2). (C.

. . .2 R0 R9 1 QB QC QD . .2 R0 9 1 8 11 QB QC QD .5 R0 R9 16 17 18 19 20 21 22 23 1 2 3 4 5 6 7 8 9 R 13 4 3 1 2 K J S Q CLK Q R 15 74112 JK Flip-Flop 5 6 +5: pin 16 GND: pin 8 11 13 12 K J 10 S Q CLK Q R 14 9 7 +5: pin 5 GND: pin 10 14 2 3 7493 Binary Counter 12 QA .8 R0 D15 D14 D13 D12 D11 D10 D9 D8 D7 D6 D5 D4 D3 D2 D1 D0 Enable W 10 74150 Multiplexer +5: pin24 GND: pin12 A B C D 15 14 13 11 1 +5 2 D1 3 D0 4 DPL 5 LE +5 14 D2 13 D3 12 5 74191 Four-Bit Counter 7 6 2 3 QD QC QB QA 13 U/D RC 4 Enable 12 Max/Min 14 CLK Load DD DC DB DA 11 9 10 1 15 +5: pin 16 GND: pin 8 74121 Monostable +5: pin 14 GND: pin 7 3 4 B 5 7489 16 x 4 RAM Rext Cext 11 10 Q Trigger Q 6 1 12 10 6 4 3 2 D I4 D I3 D I2 D I1 DO4 DO3 DO2 DO1 +5: pin 16 GND: pin 8 11 9 7 5 DPR 10 A1 A2 WE ME TIL 311 7 GND BI 8 A B C D 1 15 14 13 194 .Appendix D Pinouts V+ 7 2N 4 390 2N 6 390 85 54 2N T E JF MO V+ 3 _ 6 2 8 7 1 4 V555 TIMER 1 GND V+ 8 0L 061 ET SF 2 _ 3 741 or 411 op amp 311 comparator 2 Trigger Discharge 7 3 Out 4 Reset Threshold 6 Control 5 + 4 V- + E B C E B C D S G S G D 1 2 3 4 5 6 7 GND +5 14 13 12 11 10 9 8 1 2 3 4 5 6 7 GND +5 14 13 12 11 10 9 8 1 2 3 4 5 6 7 GND +5 14 13 12 11 10 9 8 1 2 3 4 5 6 7 GND +5 14 13 12 11 10 9 8 1 2 3 4 5 6 7 GND +5 14 13 12 11 10 9 8 7400 Quad NAND 7408 Quad AND 7404 Hex INVERTER 7432 Quad OR 7486 Quad XOR 4 2 3 S D CLK R 1 Q Q 7474 D-Type Flip-Flop 5 6 12 11 +5: pin 14 GND: pin 7 D CLK 10 S Q 8 Q 9 +5: pin 5 GND: pin 10 14 2 3 6 7 7490 Bi-Quinary Counter 9 8 11 12 QA .

195 Appendix D. Pinouts .

.

feedback A design approach or situation in which an electronic signal communicates information from the output of an electronic device or circuit to its input.Glossary of basic electrical and electronic terms ampere Basic unit of current: 1 ampere = 1 A = 1 coulomb/second. charge Q. frequency domain AC circuit analysis approach that focuses on circuit response to sine waves vs.. charge A fundamental property of some elementary particles.602 × 10−19 coulombs. 197 . cathode The positive terminal cathode-ray tube A large vacuum tube in which the electron beam can be steered to create a visible pattern on a phosphorescent screen. A convenient analogy is the rate at which water ﬂows under a bridge or through a pipe. such as a diode or transistor junction. common Voltage reference point (0 V). Electrons have charge −1e. where e is the charge of the electron. and protons and holes have charge +1e. while negative feedback counteracts a change at the output. Capacitance C. compliance (usually of a current source): Range over which circuit performance is stable. where e = 1. Also called ground. farad Unit of capacitance: 1 farad = 1 F = 1 coulomb/volt.241 × 1018 e. a growing output with positive feedback grows even larger). their frequency. Positive feedback enhances a change at the output (i. (Due to Benjamin Franklin’s choice for the deﬁnition of positive charge. Positive current ﬂows from points of higher voltage to points of lower voltage. anode The negative terminal attenuation Decrease in voltage or current (also implies power reduction). angular frequency Rate of change of phase. current Rate at which charge ﬂows. commonly available capacitors range in size from a few picofarads to many thousands of microfarads. and voltage V are related by the equation Q = C V. dynamic resistance Effective resistance of a nonlinear element (typically a PN junction). coulomb Unit of charge: 1 coulomb = 6. Measured in radians per second: ω = 2π f . Deﬁned as the amount of charge that passes through a given surface (such as the cross-section of a wire) per unit time. this is opposite to the ﬂow of electrons. decibel Unit for specifying a voltage or power ratio on a logarithmic scale.) daraf Unit of inverse capacitance.e. The capacity of a capacitor is called the capacitance. capacitor Device used to store charge and energy. The farad is a large unit.

mho Unit of transconductance. volt Unit of electrostatic potential: 1 volt = 1 V = 1 joule/coulomb. ground Voltage reference point (0 V). jack Connector used to accept a plug. Kirchhoff’s voltage law The total voltage around any closed loop is zero. socket. time. quiescent Default voltage and/or current values when an input signal is absent. resistance Degree to which a device impedes the ﬂow of DC current. i. Also called common. Voltage is deﬁned and measured with respect to a common reference or ground point. gate A circuit that performs digital logic. Measured in ohms. time domain AC circuit analysis approach that focuses on circuit response to an arbitrary waveform vs. 1 ohm = 1 = 1 volt/ampere. inverse of an ohm.198 Hands-on electronics gain Increase in voltage or current (also implies power ampliﬁcation). The e ´venin equivalent A way to model complex circuits based on Th´ venin’s theorem. ohm Unit of resistance. impedance Degree to which a circuit element impedes the ﬂow of current. jack. which reduces most circuits to a single ideal voltage source in series with a single impedance. voltage gives the potential energy of the charge with respect to that reference. Z = R. When multiplied by the value of the charge. such as an AND gate or a NOR gate. VCC Most positive voltage in a circuit. nonreactive component of impedance.e. Kirchhoff’s current law The net current ﬂowing into or out of any point in a circuit is zero. . plug Connector that plugs into a socket or jack. includes both resistive and reactive components. socket Connector used to accept a plug. corresponding to their ±90◦ phase difference. VEE Most negative voltage in a circuit. voltage Electrostatic potential. resistance is a real quantity and reactance is imaginary. (For a nonreactive device. thus impedance is given by Z = R + i X . henry Unit of inductance.) slew rate Rate at which an output voltage changes. In the standard electrical-engineering notation. reactance Capacitive (X C ) and/or inductive (X L ) component of a circuit element’s impedance. Positive current ﬂows from points of higher voltage to points of lower voltage (from larger potential energy to lower potential energy).. hertz Unit of frequency: 1 hertz = 1Hz = 1 cycle/second. for a resistor. also the degree to which the device impedes the ﬂow of AC current.

125 741 op amp. 157 bi-quinary counter. 50 VCE . 86 exponential. 159 74138 decoder. 48 BCD counter. 52. 168 op amp. 162 74191 counter. 163 7489 pinout. 105 noninverting. 88 logarithmic. 39 ampliﬁer. 126. 108 ADC. 156 7400 IC series. 91 bandpass ﬁlter. 171 binary-coded decimal. 157 bipolar junction transistor. 114 555 timer. 114 311 pinout. binary. 168 7489 RAM. 57 difference. 141 bounce. 146 astable multivibrator. 157 binary search algorithm.Index β. 163 AC coupling. 48–50 h FE . 141 assertion-level logic. 76. 52 VCB . 156 asynchronous counter. 123 bandwidth. 2 breadboard LED indicators. 17. 50 common-emitter. 170 addition. 106 active ﬁlter. 126. 123 active differentiator. 140. binary. 57 VBE . 125. 171 tracking. 120. 35. 156. 127. 53. 43 acceptor. 157 attenuating probe. 167 analog-to-digital conversion. 151. 56 Boolean algebra. 47 bistable multivibrator. 123 active integrator. 141 alternating current. 85 741 pinout. 152 breadboard. 77. 26 buffer. 69 311 comparator. 126. 105 grounded-emitter. 10 attenuation. 124 rBE . contact. 51 −3 dB point. 79. 59 inverting. 141 binary counter. 18 analog. 102. Boolean. 86 74121 monostable. 125. 54 arithmetic. 103. 89 operational. 94 binary addition. 140. 167 analog information. 137 breadboard level switches. 178 74150 mux. ideal. 50. 95 differential. 146 assertion-level logic notation. 141 algebra. 54 2N3906 pinout. 10. 162 74150 pinout. 156. 13. 32 active bandpass ﬁlter. 26. 55 push–pull. 141 binary arithmetic. 87 base. 156 blocking capacitor. 62 199 . 138 breakpoint. 137. 157 bias current. 126. 77 attenuator. 54 2N5485 pinout. 159 74121 pinout. 15 ammeter. 85 amplitude. 118. 57 re . 26 2N3904 pinout. 48–50 Q. 107 active rectiﬁer. 89 op amp. 167 analog-to-digital converter. 143. 167 successive-approximation. 167 anode.

157 negative-edge-triggered. n. powering. 157 ripple. 95 differential ampliﬁer. 115 capacitive reactance. 142 compliance. 83 decimal. 16 mica. 157 bi-quinary. 43 DIP IC package. 156 74191. 167 counter. 59. 167 digital information. 113 magnitude. 167 DAC0806. 167 digital logic. 158 four-bit. 156. 59 current-source load. 157 BCD. 77 differential signal. 157 synchronous. 106 op amp. 70 op amp. 157. 141. 156. 168 asynchronous. 16 polarized. 78 common-mode gain. 162 DC coupling. 85 DC offset. multimeter. 131 diode test. 177 digital-to-analog conversion. 133 CMOS TTL. 114 voltage. 16 types of. 48 common-emitter ampliﬁer. 75. 102 active. 16 dielectric constant. 19 capacitor. 106 diode drop. 97 transistor. 157 cascaded. 157 cascaded counters. 16 paper. 79. 35. 71 contact bounce. 151. 123 diode characteristic. 146 diagram state. 97 collector. 86 differential gain. 16 difference ampliﬁer. 152. 34 diode constant.200 Index buffering. 56 ceramic. 20 capacitance parasitic. 147 DAC. 76. 157 decibel. 9 current mirror. 113 311. 153 debugging. 167 diode. propagation. 77 closed-loop. 31. 174 Darlington. 63. 133 CMRR. 106 digital. 125 CMOS ICs. 157 two-bit. 148 DeMorgan’s theorem. 157 decade. 151. 151. 78. 15. 152. 136 CMOS logic. 75. 156 cathode. binary-coded. 178 delay. 55 capacitance code. switch. 143 dielectric. 27. 167 digital-to-analog. 37 diode logic. 16 electrolytic. 78. 33. 60 zener. 26. 56 stray. 144 digital meter. 144 decade counter. 1 digital recording. 156 carry-out. 144 debugging digital logic. 19. 151. 16. 31 gate-channel. 102. 78 common-mode rejection ratio. 61 data selector. 156 decimal. 167 digital-to-analog converter. 178 74138. 96 comparator. 102. 109 CRT. 157 binary. 54 diode-bridge rectiﬁer. 156 crossover distortion. 167 digital-to-analog. 157 decimal counter. 54 clipping. 76. 87 debouncer. 177 D-type ﬂip-ﬂop. 16 carry-in. 89 CMOS. 178 family. 158 decoder. 57 common-mode. 75 differentiator. 79 current source FET. 143 timing. pin numbers. 167 converter analog-to-digital. 15. 157. 78 common-mode rejection. 72 CY62256 memory. 66 light-emitting. FET. 152 conversion analog-to-digital. 96. 174 DAC080x. 125 debugging. 20 blocking. 129 .

55 emitter resistance. 147 JK. 116. 58 crossover. 137. 85 ideal ammeter. 146. 107 internal state. summing. 1 donor. 167 input bias current. 164 hexadecimal display. 117 FET. 85 linear. 63. 169 latch. 59 half-power frequency. 137 level switches. 107 op amp. 171 . 137 TIL311. 145. 142 XOR. 77 gate current. 151 divide-by-ten circuit. 45 measuring. 141 NAND. 67 ﬁlter. 105 logarithmic search algorithm. 68 logarithmic ampliﬁer. 78 gain. 123 bandpass. TIL311. op amp. 52 equality tester. 15. 37. 141 exponential ampliﬁer. 143. differential. 52 ECL. 34 four-bit counter. 126 feedback. quality (Q). 19 inductor. 153 LED. 15. 25. 75. 28 low-pass. 48 emitter follower. breadboard. 15. 148 follower. 125 electrolytic. common-mode. 168 ground clip. 28 ﬁnite-state machine. 72 Ebers–Moll transistor model. 105 factor. 66 gate exclusive-OR. 19 inductive reactance. 79. 60 LED indicators. RS. 138 LF398 SHA. 65 JK ﬂip-ﬂop. virtual. 52. 149. 94 forward-bias. 143. FET. 24. see LED linear region. 103. 157 frequency domain. 148 DMM. 103 active. 148 junction diode. 109 DVM. 93. 167 digital. 45 output. 85 linear. 60. 143. 62. 113. 94 input impedance. 141. 71. 141 exclusive-OR gate. 1 dynamic resistance. 57 of diode. 149 inverting ampliﬁer. 158.201 Index display logic-level. 90 ground. 123 high-pass. 45 input offset voltage. 39 impedance input. 93. 177 light-emitting diode. 103. voltage. breadboard. 156 D-type. 70 FET saturation. 32 junction. 19 information analog. push–pull. 126 hysteresis. 78. 109 divide-by-four circuit. 158 high (logic level). 85 integrator. 162 inverter. 15. 52 FET source. 158 divide-by-two circuit. 117 positive. 123 active. 2. 168 JFET. 13 gain. 88. 32 driver. 16 electrolytic capacitor. 46 indicators. 170 IC digital. 91 integrated circuit digital. 39 ideal op amp. 36 ideal voltmeter. 58. 158 distortion. 141 gate-channel diode. 10 grounded-emitter ampliﬁer. 137 inductance. 88. 26 hexadecimal. 88. 37 of emitter. breadboard. 101 function generator. 65 FET current source. 140 OR. 79 negative. 124 false. 87 ideal rectiﬁer. 148 toggling. 66 golden rules. 162 ﬂip-ﬂop. LED. 168 op amp. 16 emitter.

85 differentiator. 147 potentiometer. 75. 65 MOSFET logic. 47 PNP transistor. 114. noise. 156 bistable. 120. 10 propagation delay. 115 pin numbers. 48 offset voltage. 136 probe. 149. 79. 10 oscilloscope. 10 probe compensation adjustment. 162 7489. 23 power supplies. 108 signal-processing. 156. DIP IC package. tri-state. 127 positive. 157 noise margin. 113 positive logic. 126 noninverting ampliﬁer. 159 74150. 25 pickup. 162 NAND. 101 op-amp inverting ampliﬁer. 159 monostable. 88 negative logic. 127 positive-edge triggering. 136 powering TTL ICs. 62 push–pull driver. 32. 89 NPN transistor. 16 phase shift. 56 passband. 143. universal. op amp. 137 low (logic level). 156. 120. 126 one-shot. 159 op amp. 66 pinch-off voltage. 17 permittivity. 127 sequential. 147. 89 rectiﬁer. 102 golden rules. 140 negative feedback. 126 logic-level displays. 156. 148 pull-up resistor. 1 multimeter diode test. 54 multiplexer. 45 measurement of. digital. 129 pinch-off. 157 negative-edge triggering. 156. 48 MKS. 117 relaxation. 159 mux. see output. 146 digital. 162 word-addressable. 143. 10 attenuating. 164 output. 127. FET. 162 magnitude comparator.202 Index logic assertion-level. square-wave. see output. 16 momentary-contact switch. 14 measurement. 153 monostable. 148 measurement of. 124 peak-to-peak voltage. state. 162 negative. three-state parasitic capacitance. 74121. 56 parasitic oscillation. open-collector. 117 sine–cosine. 88. 122 oscilloscope. 159 MOSFET. totem-pole. 133. 168 open-collector output. 159 monostable multivibrator. 140 logic levels. 140 NAND gate. 46 output. 1 mho. 12 output impedance. 103 inverting ampliﬁer. 90 ideal. 164 open-loop. 6 one (logic level). 68 pinout 74121. 143 synchronous. 125 diode. 13 triggering. see op amp OR. 156. 143. three-state output. 164 output. 163 PN junction. 142 OR gate. 125. 144 logic function. 40 powering CMOS ICs. 87 integrator. 6. 113 operational ampliﬁer. 133 multimeter. 3-state. 162 multiplexer logic. 126 TTL. 164 output. 115 parasitic. 48 positive feedback. 156. 79. three-state. 131 multiplexer. 159 astable. 126 machine. 142 margin. 162 multivibrator. 156 monostable. 56 oscillator. 78 Ohm’s law. 22. 8–10 cursors. 109 . 88 noninverting ampliﬁer. 126 memory random-access. 127 negative-edge-triggered counter. 164 push–pull buffer. 91. 163 meter. 142 oscillation.

33. 146 table. 51 sine–cosine oscillator. 43 half-wave. 52 saturation drain current. see output. common-mode. 15. 116 search algorithm binary.203 Index quality factor (Q). 156. 10 register. 152 level. equality. 40 ideal. state. 68 transistor. 157 synchronous logic. 150 tri-state output. 164 three-terminal voltage regulators. 140. simple model. 67 saturation voltage. 60 saturation bipolar-transistor. 146 . digital. 108 full-wave. op amp. 143. 47 transistor. 141. 34 ripple counter. 48 transistor. 61 stray capacitance. 24 time domain. 171 summing junction. 36 rectiﬁer. 51 transistor action. three-state trigger. 19 inductive. 181 regulator. Schmitt. 143 state machine. 149. 144 table. 48 transconductance FET. 151. 122 slew rate. 171 sequential logic. 45 threshold voltage. 126 tri-state. 143 timing network. 11 shunt resistor. 170 transconductance ampliﬁer. 148 totem-pole output. 93. 149. 36. 71 SPDT switch. 8 SHA. 124 quiescent voltage. 146 tester. 43 RS latch. see RAM RC circuit. 116 triggering negative-edge. 143. 68 transition speed. 48 transformer. 40 active. 157 ripple voltage. 96 relaxation oscillator. 126 state. shift. 91 Schmitt trigger. 118. 143. 37 resistor pull-up. 177 saturated switch. ﬁeld-effect. 143. 138 switch debouncer. junction. 52 reverse-bias. 54 toggling ﬂip-ﬂop. 171 logarithmic. 177 shift register. transition. 147 true (logic level). 33. 145. 108 reference lead. 140. 133. 68 gm . 36 op amp. 68 saturation current. 153 SPDT. 59 transistor model. DeMorgan’s. 181 short circuit. 143 series. truth. 87 source follower. 37 stored charge. 57 RAM. 40 rejection ratio. 149. 48 transistor current source. 52. 115 successive-approximation ADC. 7. 101 simple transistor model. 117 resistance dynamic. 164 shunt. 141 theorem. 164 tracking ADC. 169 switch. 152. 162 7489. 104 signal processing. FET. 59. 158 time constant. 141. 114. 146. 37. 141. 146 Th´ venin equivalent circuit. 104 reverse saturation current. 6. breadboard. 162 state diagram. 163 random-access memory. RC. 159 reactance. 65 transistor. 555 timing diagram. 147. 57 static. 5. 45 e three-state output. 137. Ebers–Moll. 152 speed. 146 static resistance. 162 state table. 68 saturation region. 153 sample-and-hold. 15 RC timing network. internal. 159 TO-92 case. 52 transistor saturation. 102 timer. 113 TIL311 display. 60 transistor. 152 momentary-contact. 157 positive-edge. 31. 126 truth table. 19 capacitive. 153 synchronous counter. 40 transistor. 19 recording. FET. 177 rectiﬁcation.

126 two-bit counter. 168 voltage peak-to-peak. 17 quiescent. 44 voltmeter. 125 TTL families. 126 . 141 Zener diode. 141 XOR gate. 23. 57 threshold. 128. 15. 136 TTL logic levels. 151. 22. 123 zero. 128. 43 voltage follower. 35. 26 voltage droop. powering. ideal.204 Index TTL. 39 word addressing. 140 virtual ground. 93. 156 universal logic function. 133 TTL ICs. 163 XOR. 113 voltage comparator. 113 voltage-divider. 133 TTL history. 94 voltage regulation.

- Introductory Electronics
- Electronics for Technician Engineers
- Analog Electronics
- The Giant Book of Electronics Projects
- Electronics Repair Manual
- Digital Electrincs
- Electronics
- Consumer Electronics Troubleshooting and Repair Handbook
- Projects in Electronics
- Practical Electronics
- Useful Electronic Circuits
- Electronics
- Engineer's Mini Notebook Vol III Electronic Sensor Circuits Amp Projects
- Electronics Guide
- IntroToPracticalElectronicsMicrocontrollersAndSoftwareDesign.pdf
- Troubleshooting Electronics
- Basic Electronics
- Popular Electronics Electronic Hobbyists Handbook 1989
- Power Electronics
- electronics
- Power Electronic-EE IIT Kharagpur
- All Time Favorite Electronic Projects
- Electronic Circuit Design
- EDC k Lal Kishore
- A Designers Guide to Instrumentation Amplifiers 2ed
- Electronics
- 121 Electronic Projects (BPB)
- Troubleshooting-Analog-Circuits
- Platt Make Electronics

Skip carousel

- UT Dallas Syllabus for ce3311.001.10s taught by Bhaskar Banerjee (bxb073000)
- UT Dallas Syllabus for ee3311.501.10s taught by Ricardo Saad (rsaad)
- Super Sensitive Intruder Alarm
- Design and Analysisi of Widw Swing Folded-Cascode OTA
- UT Dallas Syllabus for ee6326.501 05f taught by Jin Liu (jinliu)
- Applied Electronics
- Heart beat detector using infrared pulse sensor
- Linear Integrated Circuits
- UT Dallas Syllabus for ee3311.001.09f taught by Murat Torlak (torlak)
- Applied Electronics
- UT Dallas Syllabus for ee3311.002.07f taught by Gil Lee (gslee)
- UT Dallas Syllabus for ee3311.501 06f taught by Ricardo Saad (rsaad)
- UT Dallas Syllabus for ee3311.001.07s taught by Gil Lee (gslee)
- lb6 sp05b
- AUTOMATIC HEADLIGHT FOR SOLAR ASSISTED ELECTRIC BICYCLE
- Desining and FFT Analysis of Sigma Delta Converter using Spice

Skip carousel

- Design of Low Noise Amplifier at 3.4GHz for Ranging Transponder (IRNSS)
- DCT- SLM Technique for Peak-To-Average Power Ratio Reduction in OFCDM System
- Power Electronics III
- Design Of High Power Amplifier At 3.4 GHz For Satellite Transponder (IRNSS)
- UT Dallas Syllabus for ee3111.004.08s taught by Eric Vogel (exv061000)
- Brainwave Controlled Robot
- UT Dallas Syllabus for ee6326.001.07s taught by Jin Liu (jinliu)
- UT Dallas Syllabus for ee3111.002.09s taught by Gil Lee (gslee)
- WWII Radar & Comm Equipment
- Housing Works, Inc. v. Bernard Kerik, Commissioner of the New York City Police Department and the City of New York, 283 F.3d 471, 2d Cir. (2002)
- UT Dallas Syllabus for ee6326.501 06f taught by Jin Liu (jinliu)
- Applied Electronics
- Applied Electronics
- Approach of Thought Control using Electro-Oculogram signal for different Eyeball positions
- Detrola Radio & Television Corp. v. Hazeltine Corp., 313 U.S. 259 (1941)
- UT Dallas Syllabus for ee3111.5u2.08u taught by Balkan Kecicioglu (bxk022000)
- Standard Coil Products Co., Inc. v. General Electric Company, 306 F.2d 319, 2d Cir. (1962)
- UT Dallas Syllabus for ee3111.501.07f taught by Randall Lehmann (rel041000)
- Gain Analysis of Fiber Raman Amplifier at 870nm band.
- Design of Low Power, High Speed 3-Bit Pipelined ADC
- UT Dallas Syllabus for ee3111.004.09s taught by Gil Lee (gslee)
- Design & Analysis of CMOS Telescopic Operational Transconductance Amplifier (OTA) with its Process Parameters
- Embedded Based Colour Recognition
- Docket No. 01-7245 August Term, 2000, 2001 March 7, 2d Cir. (2002)
- UT Dallas Syllabus for ce3111.002.09s taught by Gil Lee (gslee)
- UT Dallas Syllabus for ee3111.501.07s taught by Gil Lee (gslee)
- UT Dallas Syllabus for ee3311.501.10f taught by Bhaskar Banerjee (bxb073000)
- UT Dallas Syllabus for eerf6355.501.11f taught by Ricardo Saad (rsaad)
- UT Dallas Syllabus for ee3111.001.09s taught by Gil Lee (gslee)
- tmp1C8F

Sign up to vote on this title

UsefulNot usefulClose Dialog## Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

Close Dialog## This title now requires a credit

Use one of your book credits to continue reading from where you left off, or restart the preview.

Loading