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.
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

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

(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

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.

Sign up to vote on this title
UsefulNot useful