Fourth Edition, last update November 01, 2007
2
Lessons In Electric Circuits, Volume IV – Digital
By Tony R. Kuphaldt
Fourth Edition, last update November 01, 2007
i
c©20002011, Tony R. Kuphaldt
This book is published under the terms and conditions of the Design Science License. These
terms and conditions allow for free copying, distribution, and/or modification of this document
by the general public. The full Design Science License text is included in the last chapter.
As an open and collaboratively developed text, this book is distributed in the hope that
it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the Design Science
License for more details.
Available in its entirety as part of the Open Book Project collection at:
openbookproject.net/electricCircuits
PRINTING HISTORY
• First Edition: Printed in June of 2000. PlainASCII illustrations for universal computer
readability.
• Second Edition: Printed in September of 2000. Illustrations reworked in standard graphic
(eps and jpeg) format. Source files translated to Texinfo format for easy online and printed
publication.
• Third Edition: Printed in February 2001. Source files translated to SubML format.
SubML is a simple markup language designed to easily convert to other markups like
LATEX, HTML, or DocBook using nothing but searchandreplace substitutions.
• Fourth Edition: Printed in March 2002. Additions and improvements to 3rd edition.
ii
Contents
1 NUMERATION SYSTEMS 1
1.1 Numbers and symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Systems of numeration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Decimal versus binary numeration . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Octal and hexadecimal numeration . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Octal and hexadecimal to decimal conversion . . . . . . . . . . . . . . . . . . . . . 12
1.6 Conversion from decimal numeration . . . . . . . . . . . . . . . . . . . . . . . . . 13
2 BINARY ARITHMETIC 19
2.1 Numbers versus numeration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Binary addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Negative binary numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Subtraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Overflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6 Bit groupings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3 LOGIC GATES 29
3.1 Digital signals and gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 The NOT gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3 The ”buffer” gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.4 Multipleinput gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.5 TTL NAND and AND gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.6 TTL NOR and OR gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.7 CMOS gate circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.8 Specialoutput gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.9 Gate universality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.10 Logic signal voltage levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.11 DIP gate packaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.12 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
4 SWITCHES 103
4.1 Switch types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.2 Switch contact design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4.3 Contact ”normal” state and make/break sequence . . . . . . . . . . . . . . . . . . 111
iii
iv CONTENTS
4.4 Contact ”bounce” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
5 ELECTROMECHANICAL RELAYS 119
5.1 Relay construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
5.2 Contactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.3 Timedelay relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.4 Protective relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.5 Solidstate relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
6 LADDER LOGIC 135
6.1 ”Ladder” diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.2 Digital logic functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
6.3 Permissive and interlock circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.4 Motor control circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
6.5 Failsafe design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.6 Programmable logic controllers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
7 BOOLEAN ALGEBRA 173
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
7.2 Boolean arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
7.3 Boolean algebraic identities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.4 Boolean algebraic properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.5 Boolean rules for simplification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
7.6 Circuit simplification examples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
7.7 The ExclusiveOR function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
7.8 DeMorgan’s Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
7.9 Converting truth tables into Boolean expressions . . . . . . . . . . . . . . . . . . 200
8 KARNAUGH MAPPING 219
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
8.2 Venn diagrams and sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
8.3 Boolean Relationships on Venn Diagrams . . . . . . . . . . . . . . . . . . . . . . . 223
8.4 Making a Venn diagram look like a Karnaugh map . . . . . . . . . . . . . . . . . 228
8.5 Karnaugh maps, truth tables, and Boolean expressions . . . . . . . . . . . . . . . 231
8.6 Logic simplification with Karnaugh maps . . . . . . . . . . . . . . . . . . . . . . . 238
8.7 Larger 4variable Karnaugh maps . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
8.8 Minterm vs maxterm solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
8.9 Σ (sum) and Π (product) notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.10 Don’t care cells in the Karnaugh map . . . . . . . . . . . . . . . . . . . . . . . . . 262
8.11 Larger 5 & 6variable Karnaugh maps . . . . . . . . . . . . . . . . . . . . . . . . 265
9 COMBINATIONAL LOGIC FUNCTIONS 273
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
9.2 A HalfAdder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
9.3 A FullAdder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
CONTENTS v
9.4 Decoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
9.5 Encoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
9.6 Demultiplexers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
9.7 Multiplexers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
9.8 Using multiple combinational circuits . . . . . . . . . . . . . . . . . . . . . . . . . 294
10 MULTIVIBRATORS 299
10.1 Digital logic with feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
10.2 The SR latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
10.3 The gated SR latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
10.4 The D latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
10.5 Edgetriggered latches: FlipFlops . . . . . . . . . . . . . . . . . . . . . . . . . . . 310
10.6 The JK flipflop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
10.7 Asynchronous flipflop inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
10.8 Monostable multivibrators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
11 SEQUENTIAL CIRCUITSCOUNTERS 323
11.1 Binary count sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
11.2 Asynchronous counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
11.3 Synchronous counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
11.4 Counter modulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
11.5 Finite State Machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
12 SHIFT REGISTERS 349
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
12.2 Serialin/serialout shift register . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
12.3 Parallelin, serialout shift register . . . . . . . . . . . . . . . . . . . . . . . . . . . 361
12.4 Serialin, parallelout shift register . . . . . . . . . . . . . . . . . . . . . . . . . . 372
12.5 Parallelin, parallelout, universal shift register . . . . . . . . . . . . . . . . . . . 381
12.6 Ring counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
12.7 references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
13 DIGITALANALOG CONVERSION 407
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
13.2 The R/2nR DAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
13.3 The R/2R DAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
13.4 Flash ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
13.5 Digital ramp ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
13.6 Successive approximation ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
13.7 Tracking ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
13.8 Slope (integrating) ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
13.9 DeltaSigma (∆Σ) ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
13.10Practical considerations of ADC circuits . . . . . . . . . . . . . . . . . . . . . . . . 427
vi CONTENTS
14 DIGITAL COMMUNICATION 433
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
14.2 Networks and busses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
14.3 Data flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
14.4 Electrical signal types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
14.5 Optical data communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
14.6 Network topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
14.7 Network protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.8 Practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
15 DIGITAL STORAGE (MEMORY) 455
15.1 Why digital? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
15.2 Digital memory terms and concepts . . . . . . . . . . . . . . . . . . . . . . . . . . 456
15.3 Modern nonmechanical memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . 458
15.4 Historical, nonmechanical memory technologies . . . . . . . . . . . . . . . . . . . 460
15.5 Readonly memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
15.6 Memory with moving parts: ”Drives” . . . . . . . . . . . . . . . . . . . . . . . . . 467
16 PRINCIPLES OF DIGITAL COMPUTING 471
16.1 A binary adder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
16.2 Lookup tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472
16.3 Finitestate machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
16.4 Microprocessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481
16.5 Microprocessor programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
A1 ABOUT THIS BOOK 487
A2 CONTRIBUTOR LIST 493
A3 DESIGN SCIENCE LICENSE 497
INDEX 500
Chapter 1
NUMERATION SYSTEMS
Contents
1.1 Numbers and symbols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Systems of numeration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Decimal versus binary numeration . . . . . . . . . . . . . . . . . . . . . . . . 8
1.4 Octal and hexadecimal numeration . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Octal and hexadecimal to decimal conversion . . . . . . . . . . . . . . . . . 12
1.6 Conversion from decimal numeration . . . . . . . . . . . . . . . . . . . . . . 13
”There are three types of people: those who can count, and those who can’t.”
Anonymous
1.1 Numbers and symbols
The expression of numerical quantities is something we tend to take for granted. This is both
a good and a bad thing in the study of electronics. It is good, in that we’re accustomed to
the use and manipulation of numbers for the many calculations used in analyzing electronic
circuits. On the other hand, the particular system of notation we’ve been taught from grade
school onward is not the system used internally in modern electronic computing devices, and
learning any different system of notation requires some reexamination of deeply ingrained
assumptions.
First, we have to distinguish the difference between numbers and the symbols we use to
represent numbers. A number is a mathematical quantity, usually correlated in electronics to
a physical quantity such as voltage, current, or resistance. There are many different types of
numbers. Here are just a few types, for example:
WHOLE NUMBERS:
1, 2, 3, 4, 5, 6, 7, 8, 9 . . .
1
2 CHAPTER 1. NUMERATION SYSTEMS
INTEGERS:
4, 3, 2, 1, 0, 1, 2, 3, 4 . . .
IRRATIONAL NUMBERS:
pi (approx. 3.1415927), e (approx. 2.718281828),
square root of any prime
REAL NUMBERS:
(All onedimensional numerical values, negative and positive,
including zero, whole, integer, and irrational numbers)
COMPLEX NUMBERS:
3  j4 , 34.5 6 20o
Different types of numbers find different application in the physical world. Whole numbers
work well for counting discrete objects, such as the number of resistors in a circuit. Integers
are needed when negative equivalents of whole numbers are required. Irrational numbers are
numbers that cannot be exactly expressed as the ratio of two integers, and the ratio of a perfect
circle’s circumference to its diameter (pi) is a good physical example of this. The noninteger
quantities of voltage, current, and resistance that we’re used to dealing with in DC circuits can
be expressed as real numbers, in either fractional or decimal form. For AC circuit analysis,
however, real numbers fail to capture the dual essence of magnitude and phase angle, and so
we turn to the use of complex numbers in either rectangular or polar form.
If we are to use numbers to understand processes in the physical world, make scientific
predictions, or balance our checkbooks, we must have a way of symbolically denoting them.
In other words, we may know how much money we have in our checking account, but to keep
record of it we need to have some system worked out to symbolize that quantity on paper, or in
some other kind of form for recordkeeping and tracking. There are two basic ways we can do
this: analog and digital. With analog representation, the quantity is symbolized in a way that
is infinitely divisible. With digital representation, the quantity is symbolized in a way that is
discretely packaged.
You’re probably already familiar with an analog representation of money, and didn’t realize
it for what it was. Have you ever seen a fundraising poster made with a picture of a ther
mometer on it, where the height of the red column indicated the amount of money collected for
the cause? The more money collected, the taller the column of red ink on the poster.
1.1. NUMBERS AND SYMBOLS 3
$50,000
$0
$10,000
$20,000
$30,000
$40,000
An analog representation
of a numerical quantity
This is an example of an analog representation of a number. There is no real limit to how
finely divided the height of that column can be made to symbolize the amount of money in the
account. Changing the height of that column is something that can be done without changing
the essential nature of what it is. Length is a physical quantity that can be divided as small
as you would like, with no practical limit. The slide rule is a mechanical device that uses the
very same physical quantity – length – to represent numbers, and to help perform arithmetical
operations with two or more numbers at a time. It, too, is an analog device.
On the other hand, a digital representation of that same monetary figure, written with
standard symbols (sometimes called ciphers), looks like this:
$35,955.38
Unlike the ”thermometer” poster with its red column, those symbolic characters above can
not be finely divided: that particular combination of ciphers stand for one quantity and one
quantity only. If more money is added to the account (+ $40.12), different symbols must be
used to represent the new balance ($35,995.50), or at least the same symbols arranged in dif
ferent patterns. This is an example of digital representation. The counterpart to the slide rule
(analog) is also a digital device: the abacus, with beads that are moved back and forth on rods
to symbolize numerical quantities:
4 CHAPTER 1. NUMERATION SYSTEMS
Numerical quantities are represented by
the positioning of the slide.
Slide
Slide rule (an analog device)
Numerical quantities are represented by
Abacus (a digital device)
the discrete positions of the beads.
Let’s contrast these two methods of numerical representation:
ANALOG DIGITAL

Intuitively understood  Requires training to interpret
Infinitely divisible  Discrete
Prone to errors of precision  Absolute precision
Interpretation of numerical symbols is something we tend to take for granted, because it
has been taught to us for many years. However, if you were to try to communicate a quantity
of something to a person ignorant of decimal numerals, that person could still understand the
simple thermometer chart!
The infinitely divisible vs. discrete and precision comparisons are really flipsides of the
same coin. The fact that digital representation is composed of individual, discrete symbols
(decimal digits and abacus beads) necessarily means that it will be able to symbolize quantities
in precise steps. On the other hand, an analog representation (such as a slide rule’s length)
is not composed of individual steps, but rather a continuous range of motion. The ability
for a slide rule to characterize a numerical quantity to infinite resolution is a tradeoff for
imprecision. If a slide rule is bumped, an error will be introduced into the representation of
1.1. NUMBERS AND SYMBOLS 5
the number that was ”entered” into it. However, an abacus must be bumped much harder
before its beads are completely dislodged from their places (sufficient to represent a different
number).
Please don’t misunderstand this difference in precision by thinking that digital represen
tation is necessarily more accurate than analog. Just because a clock is digital doesn’t mean
that it will always read time more accurately than an analog clock, it just means that the
interpretation of its display is less ambiguous.
Divisibility of analog versus digital representation can be further illuminated by talking
about the representation of irrational numbers. Numbers such as pi are called irrational, be
cause they cannot be exactly expressed as the fraction of integers, or whole numbers. Although
you might have learned in the past that the fraction 22/7 can be used for pi in calculations, this
is just an approximation. The actual number ”pi” cannot be exactly expressed by any finite, or
limited, number of decimal places. The digits of pi go on forever:
3.1415926535897932384 . . . . .
It is possible, at least theoretically, to set a slide rule (or even a thermometer column) so
as to perfectly represent the number pi, because analog symbols have no minimum limit to the
degree that they can be increased or decreased. If my slide rule shows a figure of 3.141593
instead of 3.141592654, I can bump the slide just a bit more (or less) to get it closer yet. How
ever, with digital representation, such as with an abacus, I would need additional rods (place
holders, or digits) to represent pi to further degrees of precision. An abacus with 10 rods sim
ply cannot represent any more than 10 digits worth of the number pi, no matter how I set the
beads. To perfectly represent pi, an abacus would have to have an infinite number of beads
and rods! The tradeoff, of course, is the practical limitation to adjusting, and reading, analog
symbols. Practically speaking, one cannot read a slide rule’s scale to the 10th digit of precision,
because the marks on the scale are too coarse and human vision is too limited. An abacus, on
the other hand, can be set and read with no interpretational errors at all.
Furthermore, analog symbols require some kind of standard by which they can be compared
for precise interpretation. Slide rules have markings printed along the length of the slides to
translate length into standard quantities. Even the thermometer chart has numerals written
along its height to show how much money (in dollars) the red column represents for any given
amount of height. Imagine if we all tried to communicate simple numbers to each other by
spacing our hands apart varying distances. The number 1 might be signified by holding our
hands 1 inch apart, the number 2 with 2 inches, and so on. If someone held their hands 17
inches apart to represent the number 17, would everyone around them be able to immediately
and accurately interpret that distance as 17? Probably not. Some would guess short (15 or 16)
and some would guess long (18 or 19). Of course, fishermen who brag about their catches don’t
mind overestimations in quantity!
Perhaps this is why people have generally settled upon digital symbols for representing
numbers, especially whole numbers and integers, which find the most application in everyday
life. Using the fingers on our hands, we have a ready means of symbolizing integers from 0 to
10. We can make hash marks on paper, wood, or stone to represent the same quantities quite
easily:
6 CHAPTER 1. NUMERATION SYSTEMS
5 + 5 + 3 = 13
For large numbers, though, the ”hash mark” numeration system is too inefficient.
1.2 Systems of numeration
The Romans devised a system that was a substantial improvement over hash marks, because it
used a variety of symbols (or ciphers) to represent increasingly large quantities. The notation
for 1 is the capital letter I. The notation for 5 is the capital letter V. Other ciphers possess
increasing values:
X = 10
L = 50
C = 100
D = 500
M = 1000
If a cipher is accompanied by another cipher of equal or lesser value to the immediate right
of it, with no ciphers greater than that other cipher to the right of that other cipher, that
other cipher’s value is added to the total quantity. Thus, VIII symbolizes the number 8, and
CLVII symbolizes the number 157. On the other hand, if a cipher is accompanied by another
cipher of lesser value to the immediate left, that other cipher’s value is subtracted from the
first. Therefore, IV symbolizes the number 4 (V minus I), and CM symbolizes the number 900
(M minus C). You might have noticed that ending credit sequences for most motion pictures
contain a notice for the date of production, in Roman numerals. For the year 1987, it would
read: MCMLXXXVII. Let’s break this numeral down into its constituent parts, from left to right:
M = 1000
+
CM = 900
+
L = 50
+
XXX = 30
+
V = 5
+
II = 2
Aren’t you glad we don’t use this system of numeration? Large numbers are very difficult to
denote this way, and the left vs. right / subtraction vs. addition of values can be very confusing,
too. Another major problem with this system is that there is no provision for representing
the number zero or negative numbers, both very important concepts in mathematics. Roman
1.2. SYSTEMS OF NUMERATION 7
culture, however, was more pragmatic with respect to mathematics than most, choosing only
to develop their numeration system as far as it was necessary for use in daily life.
We owe one of the most important ideas in numeration to the ancient Babylonians, who
were the first (as far as we know) to develop the concept of cipher position, or place value, in
representing larger numbers. Instead of inventing new ciphers to represent larger numbers,
as the Romans did, they reused the same ciphers, placing them in different positions from
right to left. Our own decimal numeration system uses this concept, with only ten ciphers (0,
1, 2, 3, 4, 5, 6, 7, 8, and 9) used in ”weighted” positions to represent very large and very small
numbers.
Each cipher represents an integer quantity, and each place from right to left in the notation
represents a multiplying constant, or weight, for each integer quantity. For example, if we
see the decimal notation ”1206”, we known that this may be broken down into its constituent
weightproducts as such:
1206 = 1000 + 200 + 6
1206 = (1 x 1000) + (2 x 100) + (0 x 10) + (6 x 1)
Each cipher is called a digit in the decimal numeration system, and each weight, or place
value, is ten times that of the one to the immediate right. So, we have a ones place, a tens
place, a hundreds place, a thousands place, and so on, working from right to left.
Right about now, you’re probably wondering why I’m laboring to describe the obvious. Who
needs to be told how decimal numeration works, after you’ve studied math as advanced as
algebra and trigonometry? The reason is to better understand other numeration systems, by
first knowing the how’s and why’s of the one you’re already used to.
The decimal numeration system uses ten ciphers, and placeweights that are multiples of
ten. What if we made a numeration system with the same strategy of weighted places, except
with fewer or more ciphers?
The binary numeration system is such a system. Instead of ten different cipher symbols,
with each weight constant being ten times the one before it, we only have two cipher symbols,
and each weight constant is twice as much as the one before it. The two allowable cipher
symbols for the binary system of numeration are ”1” and ”0,” and these ciphers are arranged
righttoleft in doubling values of weight. The rightmost place is the ones place, just as with
decimal notation. Proceeding to the left, we have the twos place, the fours place, the eights
place, the sixteens place, and so on. For example, the following binary number can be expressed,
just like the decimal number 1206, as a sum of each cipher value times its respective weight
constant:
11010 = 2 + 8 + 16 = 26
11010 = (1 x 16) + (1 x 8) + (0 x 4) + (1 x 2) + (0 x 1)
This can get quite confusing, as I’ve written a number with binary numeration (11010),
and then shown its place values and total in standard, decimal numeration form (16 + 8 + 2
= 26). In the above example, we’re mixing two different kinds of numerical notation. To avoid
unnecessary confusion, we have to denote which form of numeration we’re using when we write
(or type!). Typically, this is done in subscript form, with a ”2” for binary and a ”10” for decimal,
so the binary number 110102 is equal to the decimal number 2610.
8 CHAPTER 1. NUMERATION SYSTEMS
The subscripts are not mathematical operation symbols like superscripts (exponents) are.
All they do is indicate what system of numeration we’re using when we write these symbols for
other people to read. If you see ”310”, all this means is the number three written using decimal
numeration. However, if you see ”310”, this means something completely different: three to
the tenth power (59,049). As usual, if no subscript is shown, the cipher(s) are assumed to be
representing a decimal number.
Commonly, the number of cipher types (and therefore, the placevalue multiplier) used in a
numeration system is called that system’s base. Binary is referred to as ”base two” numeration,
and decimal as ”base ten.” Additionally, we refer to each cipher position in binary as a bit rather
than the familiar word digit used in the decimal system.
Now, why would anyone use binary numeration? The decimal system, with its ten ciphers,
makes a lot of sense, being that we have ten fingers on which to count between our two hands.
(It is interesting that some ancient central American cultures used numeration systems with a
base of twenty. Presumably, they used both fingers and toes to count!!). But the primary reason
that the binary numeration system is used in modern electronic computers is because of the
ease of representing two cipher states (0 and 1) electronically. With relatively simple circuitry,
we can perform mathematical operations on binary numbers by representing each bit of the
numbers by a circuit which is either on (current) or off (no current). Just like the abacus
with each rod representing another decimal digit, we simply add more circuits to give us more
bits to symbolize larger numbers. Binary numeration also lends itself well to the storage and
retrieval of numerical information: on magnetic tape (spots of iron oxide on the tape either
being magnetized for a binary ”1” or demagnetized for a binary ”0”), optical disks (a laser
burned pit in the aluminum foil representing a binary ”1” and an unburned spot representing
a binary ”0”), or a variety of other media types.
Before we go on to learning exactly how all this is done in digital circuitry, we need to
become more familiar with binary and other associated systems of numeration.
1.3 Decimal versus binary numeration
Let’s count from zero to twenty using four different kinds of numeration systems: hash marks,
Roman numerals, decimal, and binary:
System: Hash Marks Roman Decimal Binary
    
Zero n/a n/a 0 0
One  I 1 1
Two  II 2 10
Three  III 3 11
Four  IV 4 100
Five // V 5 101
Six //  VI 6 110
Seven //  VII 7 111
Eight //  VIII 8 1000
Nine //  IX 9 1001
Ten // // X 10 1010
1.3. DECIMAL VERSUS BINARY NUMERATION 9
Eleven // //  XI 11 1011
Twelve // //  XII 12 1100
Thirteen // //  XIII 13 1101
Fourteen // //  XIV 14 1110
Fifteen // // // XV 15 1111
Sixteen // // //  XVI 16 10000
Seventeen // // //  XVII 17 10001
Eighteen // // //  XVIII 18 10010
Nineteen // // //  XIX 19 10011
Twenty // // // // XX 20 10100
Neither hash marks nor the Roman system are very practical for symbolizing large num
bers. Obviously, placeweighted systems such as decimal and binary are more efficient for the
task. Notice, though, how much shorter decimal notation is over binary notation, for the same
number of quantities. What takes five bits in binary notation only takes two digits in decimal
notation.
This raises an interesting question regarding different numeration systems: how large of
a number can be represented with a limited number of cipher positions, or places? With the
crude hashmark system, the number of places IS the largest number that can be represented,
since one hash mark ”place” is required for every integer step. For placeweighted systems of
numeration, however, the answer is found by taking base of the numeration system (10 for
decimal, 2 for binary) and raising it to the power of the number of places. For example, 5 digits
in a decimal numeration system can represent 100,000 different integer number values, from
0 to 99,999 (10 to the 5th power = 100,000). 8 bits in a binary numeration system can repre
sent 256 different integer number values, from 0 to 11111111 (binary), or 0 to 255 (decimal),
because 2 to the 8th power equals 256. With each additional place position to the number field,
the capacity for representing numbers increases by a factor of the base (10 for decimal, 2 for
binary).
An interesting footnote for this topic is the one of the first electronic digital computers,
the Eniac. The designers of the Eniac chose to represent numbers in decimal form, digitally,
using a series of circuits called ”ring counters” instead of just going with the binary numeration
system, in an effort to minimize the number of circuits required to represent and calculate very
large numbers. This approach turned out to be counterproductive, and virtually all digital
computers since then have been purely binary in design.
To convert a number in binary numeration to its equivalent in decimal form, all you have to
do is calculate the sum of all the products of bits with their respective placeweight constants.
To illustrate:
Convert 110011012 to decimal form:
bits = 1 1 0 0 1 1 0 1
.        
weight = 1 6 3 1 8 4 2 1
(in decimal 2 4 2 6
notation) 8
10 CHAPTER 1. NUMERATION SYSTEMS
The bit on the far right side is called the Least Significant Bit (LSB), because it stands in
the place of the lowest weight (the one’s place). The bit on the far left side is called the Most
Significant Bit (MSB), because it stands in the place of the highest weight (the one hundred
twentyeight’s place). Remember, a bit value of ”1” means that the respective place weight gets
added to the total value, and a bit value of ”0” means that the respective place weight does not
get added to the total value. With the above example, we have:
12810 + 6410 + 810 + 410 + 110 = 20510
If we encounter a binary number with a dot (.), called a ”binary point” instead of a decimal
point, we follow the same procedure, realizing that each place weight to the right of the point is
onehalf the value of the one to the left of it (just as each place weight to the right of a decimal
point is onetenth the weight of the one to the left of it). For example:
Convert 101.0112 to decimal form:
.
bits = 1 0 1 . 0 1 1
.       
weight = 4 2 1 1 1 1
(in decimal / / /
notation) 2 4 8
410 + 110 + 0.2510 + 0.12510 = 5.37510
1.4 Octal and hexadecimal numeration
Because binary numeration requires so many bits to represent relatively small numbers com
pared to the economy of the decimal system, analyzing the numerical states inside of digital
electronic circuitry can be a tedious task. Computer programmers who design sequences of
number codes instructing a computer what to do would have a very difficult task if they were
forced to work with nothing but long strings of 1’s and 0’s, the ”native language” of any digital
circuit. To make it easier for human engineers, technicians, and programmers to ”speak” this
language of the digital world, other systems of placeweighted numeration have been made
which are very easy to convert to and from binary.
One of those numeration systems is called octal, because it is a placeweighted system with
a base of eight. Valid ciphers include the symbols 0, 1, 2, 3, 4, 5, 6, and 7. Each place weight
differs from the one next to it by a factor of eight.
Another system is called hexadecimal, because it is a placeweighted system with a base of
sixteen. Valid ciphers include the normal decimal symbols 0, 1, 2, 3, 4, 5, 6, 7, 8, and 9, plus
six alphabetical characters A, B, C, D, E, and F, to make a total of sixteen. As you might have
guessed already, each place weight differs from the one before it by a factor of sixteen.
Let’s count again from zero to twenty using decimal, binary, octal, and hexadecimal to
contrast these systems of numeration:
Number Decimal Binary Octal Hexadecimal
1.4. OCTAL AND HEXADECIMAL NUMERATION 11
    
Zero 0 0 0 0
One 1 1 1 1
Two 2 10 2 2
Three 3 11 3 3
Four 4 100 4 4
Five 5 101 5 5
Six 6 110 6 6
Seven 7 111 7 7
Eight 8 1000 10 8
Nine 9 1001 11 9
Ten 10 1010 12 A
Eleven 11 1011 13 B
Twelve 12 1100 14 C
Thirteen 13 1101 15 D
Fourteen 14 1110 16 E
Fifteen 15 1111 17 F
Sixteen 16 10000 20 10
Seventeen 17 10001 21 11
Eighteen 18 10010 22 12
Nineteen 19 10011 23 13
Twenty 20 10100 24 14
Octal and hexadecimal numeration systems would be pointless if not for their ability to be
easily converted to and from binary notation. Their primary purpose in being is to serve as a
”shorthand” method of denoting a number represented electronically in binary form. Because
the bases of octal (eight) and hexadecimal (sixteen) are even multiples of binary’s base (two),
binary bits can be grouped together and directly converted to or from their respective octal or
hexadecimal digits. With octal, the binary bits are grouped in three’s (because 23 = 8), and
with hexadecimal, the binary bits are grouped in four’s (because 24 = 16):
BINARY TO OCTAL CONVERSION
Convert 10110111.12 to octal:
.
. implied zero implied zeros
.  
. 010 110 111 100
Convert each group of bits ### ### ### . ###
to its octal equivalent: 2 6 7 4
.
Answer: 10110111.12 = 267.48
We had to group the bits in three’s, from the binary point left, and from the binary point
right, adding (implied) zeros as necessary to make complete 3bit groups. Each octal digit was
translated from the 3bit binary groups. BinarytoHexadecimal conversion is much the same:
BINARY TO HEXADECIMAL CONVERSION
12 CHAPTER 1. NUMERATION SYSTEMS
Convert 10110111.12 to hexadecimal:
.
. implied zeros
. 
. 1011 0111 1000
Convert each group of bits   . 
to its hexadecimal equivalent: B 7 8
.
Answer: 10110111.12 = B7.816
Here we had to group the bits in four’s, from the binary point left, and from the binary point
right, adding (implied) zeros as necessary to make complete 4bit groups:
Likewise, the conversion from either octal or hexadecimal to binary is done by taking each
octal or hexadecimal digit and converting it to its equivalent binary (3 or 4 bit) group, then
putting all the binary bit groups together.
Incidentally, hexadecimal notation is more popular, because binary bit groupings in digital
equipment are commonly multiples of eight (8, 16, 32, 64, and 128 bit), which are also multiples
of 4. Octal, being based on binary bit groups of 3, doesn’t work out evenly with those common
bit group sizings.
1.5 Octal and hexadecimal to decimal conversion
Although the prime intent of octal and hexadecimal numeration systems is for the ”shorthand”
representation of binary numbers in digital electronics, we sometimes have the need to convert
from either of those systems to decimal form. Of course, we could simply convert the hexadeci
mal or octal format to binary, then convert from binary to decimal, since we already know how
to do both, but we can also convert directly.
Because octal is a baseeight numeration system, each placeweight value differs from ei
ther adjacent place by a factor of eight. For example, the octal number 245.37 can be broken
down into place values as such:
octal
digits = 2 4 5 . 3 7
.      
weight = 6 8 1 1 1
(in decimal 4 / /
notation) 8 6
. 4
The decimal value of each octal placeweight times its respective cipher multiplier can be
determined as follows:
(2 x 6410) + (4 x 810) + (5 x 110) + (3 x 0.12510) +
(7 x 0.01562510) = 165.48437510
1.6. CONVERSION FROM DECIMAL NUMERATION 13
The technique for converting hexadecimal notation to decimal is the same, except that each
successive placeweight changes by a factor of sixteen. Simply denote each digit’s weight,
multiply each hexadecimal digit value by its respective weight (in decimal form), then add up
all the decimal values to get a total. For example, the hexadecimal number 30F.A916 can be
converted like this:
hexadecimal
digits = 3 0 F . A 9
.      
weight = 2 1 1 1 1
(in decimal 5 6 / /
notation) 6 1 2
. 6 5
. 6
(3 x 25610) + (0 x 1610) + (15 x 110) + (10 x 0.062510) +
(9 x 0.0039062510) = 783.6601562510
These basic techniques may be used to convert a numerical notation of any base into decimal
form, if you know the value of that numeration system’s base.
1.6 Conversion from decimal numeration
Because octal and hexadecimal numeration systems have bases that are multiples of binary
(base 2), conversion back and forth between either hexadecimal or octal and binary is very
easy. Also, because we are so familiar with the decimal system, converting binary, octal, or
hexadecimal to decimal form is relatively easy (simply add up the products of cipher values
and placeweights). However, conversion from decimal to any of these ”strange” numeration
systems is a different matter.
The method which will probably make the most sense is the ”trialandfit” method, where
you try to ”fit” the binary, octal, or hexadecimal notation to the desired value as represented
in decimal form. For example, let’s say that I wanted to represent the decimal value of 87 in
binary form. Let’s start by drawing a binary number field, complete with placeweight values:
.
.        
weight = 1 6 3 1 8 4 2 1
(in decimal 2 4 2 6
notation) 8
Well, we know that we won’t have a ”1” bit in the 128’s place, because that would immedi
ately give us a value greater than 87. However, since the next weight to the right (64) is less
than 87, we know that we must have a ”1” there.
. 1
14 CHAPTER 1. NUMERATION SYSTEMS
.        Decimal value so far = 6410
weight = 6 3 1 8 4 2 1
(in decimal 4 2 6
notation)
If we were to make the next place to the right a ”1” as well, our total value would be 6410
+ 3210, or 9610. This is greater than 8710, so we know that this bit must be a ”0”. If we make
the next (16’s) place bit equal to ”1,” this brings our total value to 6410 + 1610, or 8010, which is
closer to our desired value (8710) without exceeding it:
. 1 0 1
.        Decimal value so far = 8010
weight = 6 3 1 8 4 2 1
(in decimal 4 2 6
notation)
By continuing in this progression, setting each lesserweight bit as we need to come up to
our desired total value without exceeding it, we will eventually arrive at the correct figure:
. 1 0 1 0 1 1 1
.        Decimal value so far = 8710
weight = 6 3 1 8 4 2 1
(in decimal 4 2 6
notation)
This trialandfit strategy will work with octal and hexadecimal conversions, too. Let’s take
the same decimal figure, 8710, and convert it to octal numeration:
.
.   
weight = 6 8 1
(in decimal 4
notation)
If we put a cipher of ”1” in the 64’s place, we would have a total value of 6410 (less than
8710). If we put a cipher of ”2” in the 64’s place, we would have a total value of 12810 (greater
than 8710). This tells us that our octal numeration must start with a ”1” in the 64’s place:
. 1
.    Decimal value so far = 6410
weight = 6 8 1
(in decimal 4
notation)
Now, we need to experiment with cipher values in the 8’s place to try and get a total (deci
mal) value as close to 87 as possible without exceeding it. Trying the first few cipher options,
we get:
1.6. CONVERSION FROM DECIMAL NUMERATION 15
"1" = 6410 + 810 = 7210
"2" = 6410 + 1610 = 8010
"3" = 6410 + 2410 = 8810
A cipher value of ”3” in the 8’s place would put us over the desired total of 8710, so ”2” it is!
. 1 2
.    Decimal value so far = 8010
weight = 6 8 1
(in decimal 4
notation)
Now, all we need to make a total of 87 is a cipher of ”7” in the 1’s place:
. 1 2 7
.    Decimal value so far = 8710
weight = 6 8 1
(in decimal 4
notation)
Of course, if you were paying attention during the last section on octal/binary conversions,
you will realize that we can take the binary representation of (decimal) 8710, which we previ
ously determined to be 10101112, and easily convert from that to octal to check our work:
. Implied zeros
. 
. 001 010 111 Binary
.   
. 1 2 7 Octal
.
Answer: 10101112 = 1278
Can we do decimaltohexadecimal conversion the same way? Sure, but who would want
to? This method is simple to understand, but laborious to carry out. There is another way to do
these conversions, which is essentially the same (mathematically), but easier to accomplish.
This other method uses repeated cycles of division (using decimal notation) to break the
decimal numeration down into multiples of binary, octal, or hexadecimal placeweight values.
In the first cycle of division, we take the original decimal number and divide it by the base of
the numeration system that we’re converting to (binary=2 octal=8, hex=16). Then, we take the
wholenumber portion of division result (quotient) and divide it by the base value again, and
so on, until we end up with a quotient of less than 1. The binary, octal, or hexadecimal digits
are determined by the ”remainders” left over by each division step. Let’s see how this works
for binary, with the decimal example of 8710:
. 87 Divide 87 by 2, to get a quotient of 43.5
.  = 43.5 Division "remainder" = 1, or the < 1 portion
16 CHAPTER 1. NUMERATION SYSTEMS
. 2 of the quotient times the divisor (0.5 x 2)
.
. 43 Take the wholenumber portion of 43.5 (43)
.  = 21.5 and divide it by 2 to get 21.5, or 21 with
. 2 a remainder of 1
.
. 21 And so on . . . remainder = 1 (0.5 x 2)
.  = 10.5
. 2
.
. 10 And so on . . . remainder = 0
.  = 5.0
. 2
.
. 5 And so on . . . remainder = 1 (0.5 x 2)
.  = 2.5
. 2
.
. 2 And so on . . . remainder = 0
.  = 1.0
. 2
.
. 1 . . . until we get a quotient of less than 1
.  = 0.5 remainder = 1 (0.5 x 2)
. 2
The binary bits are assembled from the remainders of the successive division steps, begin
ning with the LSB and proceeding to the MSB. In this case, we arrive at a binary notation of
10101112. When we divide by 2, we will always get a quotient ending with either ”.0” or ”.5”,
i.e. a remainder of either 0 or 1. As was said before, this repeatdivision technique for con
version will work for numeration systems other than binary. If we were to perform successive
divisions using a different number, such as 8 for conversion to octal, we will necessarily get
remainders between 0 and 7. Let’s try this with the same decimal number, 8710:
. 87 Divide 87 by 8, to get a quotient of 10.875
.  = 10.875 Division "remainder" = 7, or the < 1 portion
. 8 of the quotient times the divisor (.875 x 8)
.
. 10
.  = 1.25 Remainder = 2
. 8
.
. 1
.  = 0.125 Quotient is less than 1, so we’ll stop here.
. 8 Remainder = 1
.
1.6. CONVERSION FROM DECIMAL NUMERATION 17
. RESULT: 8710 = 1278
We can use a similar technique for converting numeration systems dealing with quantities
less than 1, as well. For converting a decimal number less than 1 into binary, octal, or hexadec
imal, we use repeated multiplication, taking the integer portion of the product in each step as
the next digit of our converted number. Let’s use the decimal number 0.812510 as an example,
converting to binary:
. 0.8125 x 2 = 1.625 Integer portion of product = 1
.
. 0.625 x 2 = 1.25 Take < 1 portion of product and remultiply
. Integer portion of product = 1
.
. 0.25 x 2 = 0.5 Integer portion of product = 0
.
. 0.5 x 2 = 1.0 Integer portion of product = 1
. Stop when product is a pure integer
. (ends with .0)
.
. RESULT: 0.812510 = 0.11012
As with the repeatdivision process for integers, each step gives us the next digit (or bit)
further away from the ”point.” With integer (division), we worked from the LSB to the MSB
(righttoleft), but with repeated multiplication, we worked from the left to the right. To convert
a decimal number greater than 1, with a ¡ 1 component, we must use both techniques, one at a
time. Take the decimal example of 54.4062510, converting to binary:
REPEATED DIVISION FOR THE INTEGER PORTION:
.
. 54
.  = 27.0 Remainder = 0
. 2
.
. 27
.  = 13.5 Remainder = 1 (0.5 x 2)
. 2
.
. 13
.  = 6.5 Remainder = 1 (0.5 x 2)
. 2
.
. 6
.  = 3.0 Remainder = 0
. 2
.
. 3
18 CHAPTER 1. NUMERATION SYSTEMS
.  = 1.5 Remainder = 1 (0.5 x 2)
. 2
.
. 1
.  = 0.5 Remainder = 1 (0.5 x 2)
. 2
.
PARTIAL ANSWER: 5410 = 1101102
REPEATED MULTIPLICATION FOR THE < 1 PORTION:
.
. 0.40625 x 2 = 0.8125 Integer portion of product = 0
.
. 0.8125 x 2 = 1.625 Integer portion of product = 1
.
. 0.625 x 2 = 1.25 Integer portion of product = 1
.
. 0.25 x 2 = 0.5 Integer portion of product = 0
.
. 0.5 x 2 = 1.0 Integer portion of product = 1
.
. PARTIAL ANSWER: 0.4062510 = 0.011012
.
. COMPLETE ANSWER: 5410 + 0.4062510 = 54.4062510
.
. 1101102 + 0.011012 = 110110.011012
Chapter 2
BINARY ARITHMETIC
Contents
2.1 Numbers versus numeration . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2 Binary addition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.3 Negative binary numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.4 Subtraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
2.5 Overflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.6 Bit groupings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
2.1 Numbers versus numeration
It is imperative to understand that the type of numeration system used to represent numbers
has no impact upon the outcome of any arithmetical function (addition, subtraction, multipli
cation, division, roots, powers, or logarithms). A number is a number is a number; one plus
one will always equal two (so long as we’re dealing with real numbers), no matter how you
symbolize one, one, and two. A prime number in decimal form is still prime if its shown in
binary form, or octal, or hexadecimal. pi is still the ratio between the circumference and diam
eter of a circle, no matter what symbol(s) you use to denote its value. The essential functions
and interrelations of mathematics are unaffected by the particular system of symbols we might
choose to represent quantities. This distinction between numbers and systems of numeration
is critical to understand.
The essential distinction between the two is much like that between an object and the
spoken word(s) we associate with it. A house is still a house regardless of whether we call it
by its English name house or its Spanish name casa. The first is the actual thing, while the
second is merely the symbol for the thing.
That being said, performing a simple arithmetic operation such as addition (longhand) in
binary form can be confusing to a person accustomed to working with decimal numeration
only. In this lesson, we’ll explore the techniques used to perform simple arithmetic functions
19
20 CHAPTER 2. BINARY ARITHMETIC
on binary numbers, since these techniques will be employed in the design of electronic circuits
to do the same. You might take longhand addition and subtraction for granted, having used
a calculator for so long, but deep inside that calculator’s circuitry all those operations are
performed ”longhand,” using binary numeration. To understand how that’s accomplished, we
need to review to the basics of arithmetic.
2.2 Binary addition
Adding binary numbers is a very simple task, and very similar to the longhand addition of
decimal numbers. As with decimal numbers, you start by adding the bits (digits) one column,
or place weight, at a time, from right to left. Unlike decimal addition, there is little to memorize
in the way of rules for the addition of binary bits:
0 + 0 = 0
1 + 0 = 1
0 + 1 = 1
1 + 1 = 10
1 + 1 + 1 = 11
Just as with decimal addition, when the sum in one column is a twobit (twodigit) number,
the least significant figure is written as part of the total sum and the most significant figure is
”carried” to the next left column. Consider the following examples:
. 11 1 < Carry bits > 11
. 1001101 1001001 1000111
. + 0010010 + 0011001 + 0010110
.   
. 1011111 1100010 1011101
The addition problem on the left did not require any bits to be carried, since the sum of bits
in each column was either 1 or 0, not 10 or 11. In the other two problems, there definitely were
bits to be carried, but the process of addition is still quite simple.
As we’ll see later, there are ways that electronic circuits can be built to perform this very
task of addition, by representing each bit of each binary number as a voltage signal (either
”high,” for a 1; or ”low” for a 0). This is the very foundation of all the arithmetic which modern
digital computers perform.
2.3 Negative binary numbers
With addition being easily accomplished, we can perform the operation of subtraction with the
same technique simply by making one of the numbers negative. For example, the subtraction
problem of 7  5 is essentially the same as the addition problem 7 + (5). Since we already know
how to represent positive numbers in binary, all we need to know now is how to represent their
negative counterparts and we’ll be able to subtract.
2.3. NEGATIVE BINARY NUMBERS 21
Usually we represent a negative decimal number by placing a minus sign directly to the left
of the most significant digit, just as in the example above, with 5. However, the whole purpose
of using binary notation is for constructing on/off circuits that can represent bit values in
terms of voltage (2 alternative values: either ”high” or ”low”). In this context, we don’t have
the luxury of a third symbol such as a ”minus” sign, since these circuits can only be on or off
(two possible states). One solution is to reserve a bit (circuit) that does nothing but represent
the mathematical sign:
. 1012 = 510 (positive)
.
. Extra bit, representing sign (0=positive, 1=negative)
. 
. 01012 = 510 (positive)
.
. Extra bit, representing sign (0=positive, 1=negative)
. 
. 11012 = 510 (negative)
As you can see, we have to be careful when we start using bits for any purpose other than
standard placeweighted values. Otherwise, 11012 could be misinterpreted as the number
thirteen when in fact we mean to represent negative five. To keep things straight here, we must
first decide how many bits are going to be needed to represent the largest numbers we’ll be
dealing with, and then be sure not to exceed that bit field length in our arithmetic operations.
For the above example, I’ve limited myself to the representation of numbers from negative
seven (11112) to positive seven (01112), and no more, by making the fourth bit the ”sign” bit.
Only by first establishing these limits can I avoid confusion of a negative number with a larger,
positive number.
Representing negative five as 11012 is an example of the signmagnitude system of nega
tive binary numeration. By using the leftmost bit as a sign indicator and not a placeweighted
value, I am sacrificing the ”pure” form of binary notation for something that gives me a prac
tical advantage: the representation of negative numbers. The leftmost bit is read as the sign,
either positive or negative, and the remaining bits are interpreted according to the standard
binary notation: left to right, place weights in multiples of two.
As simple as the signmagnitude approach is, it is not very practical for arithmetic purposes.
For instance, how do I add a negative five (11012) to any other number, using the standard
technique for binary addition? I’d have to invent a new way of doing addition in order for it
to work, and if I do that, I might as well just do the job with longhand subtraction; there’s no
arithmetical advantage to using negative numbers to perform subtraction through addition if
we have to do it with signmagnitude numeration, and that was our goal!
There’s another method for representing negative numbers which works with our familiar
technique of longhand addition, and also happens to make more sense from a placeweighted
numeration point of view, called complementation. With this strategy, we assign the leftmost
bit to serve a special purpose, just as we did with the signmagnitude approach, defining our
number limits just as before. However, this time, the leftmost bit is more than just a sign bit;
rather, it possesses a negative placeweight value. For example, a value of negative five would
be represented as such:
22 CHAPTER 2. BINARY ARITHMETIC
Extra bit, place weight = negative eight
. 
. 10112 = 510 (negative)
.
. (1 x 810) + (0 x 410) + (1 x 210) + (1 x 110) = 510
With the right three bits being able to represent a magnitude from zero through seven, and
the leftmost bit representing either zero or negative eight, we can successfully represent any
integer number from negative seven (10012 = 810 + 110 = 710) to positive seven (01112 = 010 +
710 = 710).
Representing positive numbers in this scheme (with the fourth bit designated as the neg
ative weight) is no different from that of ordinary binary notation. However, representing
negative numbers is not quite as straightforward:
zero 0000
positive one 0001 negative one 1111
positive two 0010 negative two 1110
positive three 0011 negative three 1101
positive four 0100 negative four 1100
positive five 0101 negative five 1011
positive six 0110 negative six 1010
positive seven 0111 negative seven 1001
. negative eight 1000
Note that the negative binary numbers in the right column, being the sum of the right three
bits’ total plus the negative eight of the leftmost bit, don’t ”count” in the same progression as
the positive binary numbers in the left column. Rather, the right three bits have to be set at
the proper value to equal the desired (negative) total when summed with the negative eight
place value of the leftmost bit.
Those right three bits are referred to as the two’s complement of the corresponding positive
number. Consider the following comparison:
positive number two’s complement
 
001 111
010 110
011 101
100 100
101 011
110 010
111 001
In this case, with the negative weight bit being the fourth bit (place value of negative eight),
the two’s complement for any positive number will be whatever value is needed to add to
negative eight to make that positive value’s negative equivalent. Thankfully, there’s an easy
way to figure out the two’s complement for any binary number: simply invert all the bits of that
2.4. SUBTRACTION 23
number, changing all 1’s to 0’s and vice versa (to arrive at what is called the one’s complement)
and then add one! For example, to obtain the two’s complement of five (1012), we would first
invert all the bits to obtain 0102 (the ”one’s complement”), then add one to obtain 0112, or 510
in threebit, two’s complement form.
Interestingly enough, generating the two’s complement of a binary number works the same
if you manipulate all the bits, including the leftmost (sign) bit at the same time as the mag
nitude bits. Let’s try this with the former example, converting a positive five to a negative
five, but performing the complementation process on all four bits. We must be sure to include
the 0 (positive) sign bit on the original number, five (01012). First, inverting all bits to obtain
the one’s complement: 10102. Then, adding one, we obtain the final answer: 10112, or 510
expressed in fourbit, two’s complement form.
It is critically important to remember that the place of the negativeweight bit must be al
ready determined before any two’s complement conversions can be done. If our binary numera
tion field were such that the eighth bit was designated as the negativeweight bit (100000002),
we’d have to determine the two’s complement based on all seven of the other bits. Here, the
two’s complement of five (00001012) would be 11110112. A positive five in this system would
be represented as 000001012, and a negative five as 111110112.
2.4 Subtraction
We can subtract one binary number from another by using the standard techniques adapted
for decimal numbers (subtraction of each bit pair, right to left, ”borrowing” as needed from bits
to the left). However, if we can leverage the already familiar (and easier) technique of binary
addition to subtract, that would be better. As we just learned, we can represent negative binary
numbers by using the ”two’s complement” method and a negative placeweight bit. Here, we’ll
use those negative binary numbers to subtract through addition. Here’s a sample problem:
Subtraction: 710  510 Addition equivalent: 710 + (510)
If all we need to do is represent seven and negative five in binary (two’s complemented)
form, all we need is three bits plus the negativeweight bit:
positive seven = 01112
negative five = 10112
Now, let’s add them together:
. 1111 < Carry bits
. 0111
. + 1011
. 
. 10010
. 
. Discard extra bit
.
24 CHAPTER 2. BINARY ARITHMETIC
. Answer = 00102
Since we’ve already defined our number bit field as three bits plus the negativeweight bit,
the fifth bit in the answer (1) will be discarded to give us a result of 00102, or positive two,
which is the correct answer.
Another way to understand why we discard that extra bit is to remember that the leftmost
bit of the lower number possesses a negative weight, in this case equal to negative eight.
When we add these two binary numbers together, what we’re actually doing with the MSBs is
subtracting the lower number’s MSB from the upper number’s MSB. In subtraction, one never
”carries” a digit or bit on to the next left placeweight.
Let’s try another example, this time with larger numbers. If we want to add 2510 to 1810,
we must first decide how large our binary bit field must be. To represent the largest (absolute
value) number in our problem, which is twentyfive, we need at least five bits, plus a sixth bit
for the negativeweight bit. Let’s start by representing positive twentyfive, then finding the
two’s complement and putting it all together into one numeration:
+2510 = 0110012 (showing all six bits)
One’s complement of 110012 = 1001102
One’s complement + 1 = two’s complement = 1001112
2510 = 1001112
Essentially, we’re representing negative twentyfive by using the negativeweight (sixth)
bit with a value of negative thirtytwo, plus positive seven (binary 1112).
Now, let’s represent positive eighteen in binary form, showing all six bits:
. 1810 = 0100102
.
. Now, let’s add them together and see what we get:
.
. 11 < Carry bits
. 100111
. + 010010
. 
. 111001
Since there were no ”extra” bits on the left, there are no bits to discard. The leftmost bit
on the answer is a 1, which means that the answer is negative, in two’s complement form,
as it should be. Converting the answer to decimal form by summing all the bits times their
respective weight values, we get:
(1 x 3210) + (1 x 1610) + (1 x 810) + (1 x 110) = 710
Indeed 710 is the proper sum of 2510 and 1810.
2.5. OVERFLOW 25
2.5 Overflow
One caveat with signed binary numbers is that of overflow, where the answer to an addition or
subtraction problem exceeds the magnitude which can be represented with the alloted number
of bits. Remember that the place of the sign bit is fixed from the beginning of the problem.
With the last example problem, we used five binary bits to represent the magnitude of the
number, and the leftmost (sixth) bit as the negativeweight, or sign, bit. With five bits to
represent magnitude, we have a representation range of 25, or thirtytwo integer steps from 0
to maximum. This means that we can represent a number as high as +3110 (0111112), or as
low as 3210 (1000002). If we set up an addition problem with two binary numbers, the sixth
bit used for sign, and the result either exceeds +3110 or is less than 3210, our answer will be
incorrect. Let’s try adding 1710 and 1910 to see how this overflow condition works for excessive
positive numbers:
. 1710 = 100012 1910 = 100112
.
. 1 11 < Carry bits
. (Showing sign bits) 010001
. + 010011
. 
. 100100
The answer (1001002), interpreted with the sixth bit as the 3210 place, is actually equal
to 2810, not +3610 as we should get with +1710 and +1910 added together! Obviously, this is
not correct. What went wrong? The answer lies in the restrictions of the sixbit number field
within which we’re working Since the magnitude of the true and proper sum (3610) exceeds the
allowable limit for our designated bit field, we have an overflow error. Simply put, six places
doesn’t give enough bits to represent the correct sum, so whatever figure we obtain using the
strategy of discarding the leftmost ”carry” bit will be incorrect.
A similar error will occur if we add two negative numbers together to produce a sum that
is too low for our sixbit binary field. Let’s try adding 1710 and 1910 together to see how this
works (or doesn’t work, as the case may be!):
. 1710 = 1011112 1910 = 1011012
.
. 1 1111 < Carry bits
. (Showing sign bits) 101111
. + 101101
. 
. 1011100
. 
. Discard extra bit
.
FINAL ANSWER: 0111002 = +2810
The (incorrect) answer is a positive twentyeight. The fact that the real sum of negative sev
enteen and negative nineteen was too low to be properly represented with a five bit magnitude
26 CHAPTER 2. BINARY ARITHMETIC
field and a sixth sign bit is the root cause of this difficulty.
Let’s try these two problems again, except this time using the seventh bit for a sign bit, and
allowing the use of 6 bits for representing the magnitude:
. 1710 + 1910 (1710) + (1910)
.
. 1 11 11 1111
. 0010001 1101111
. + 0010011 + 1101101
.  
. 01001002 110111002
. 
. Discard extra bit
.
. ANSWERS: 01001002 = +3610
. 10111002 = 3610
By using bit fields sufficiently large to handle the magnitude of the sums, we arrive at the
correct answers.
In these sample problems we’ve been able to detect overflow errors by performing the ad
dition problems in decimal form and comparing the results with the binary answers. For ex
ample, when adding +1710 and +1910 together, we knew that the answer was supposed to be
+3610, so when the binary sum checked out to be 2810, we knew that something had to be
wrong. Although this is a valid way of detecting overflow, it is not very efficient. After all, the
whole idea of complementation is to be able to reliably add binary numbers together and not
have to doublecheck the result by adding the same numbers together in decimal form! This is
especially true for the purpose of building electronic circuits to add binary quantities together:
the circuit has to be able to check itself for overflow without the supervision of a human being
who already knows what the correct answer is.
What we need is a simple errordetection method that doesn’t require any additional arith
metic. Perhaps the most elegant solution is to check for the sign of the sum and compare
it against the signs of the numbers added. Obviously, two positive numbers added together
should give a positive result, and two negative numbers added together should give a negative
result. Notice that whenever we had a condition of overflow in the example problems, the sign
of the sum was always opposite of the two added numbers: +1710 plus +1910 giving 2810, or
1710 plus 1910 giving +2810. By checking the signs alone we are able to tell that something is
wrong.
But what about cases where a positive number is added to a negative number? What sign
should the sum be in order to be correct. Or, more precisely, what sign of sum would neces
sarily indicate an overflow error? The answer to this is equally elegant: there will never be an
overflow error when two numbers of opposite signs are added together! The reason for this is
apparent when the nature of overflow is considered. Overflow occurs when the magnitude of a
number exceeds the range allowed by the size of the bit field. The sum of two identicallysigned
numbers may very well exceed the range of the bit field of those two numbers, and so in this
case overflow is a possibility. However, if a positive number is added to a negative number, the
2.6. BIT GROUPINGS 27
sum will always be closer to zero than either of the two added numbers: its magnitude must be
less than the magnitude of either original number, and so overflow is impossible.
Fortunately, this technique of overflow detection is easily implemented in electronic cir
cuitry, and it is a standard feature in digital adder circuits: a subject for a later chapter.
2.6 Bit groupings
The singular reason for learning and using the binary numeration system in electronics is to
understand how to design, build, and troubleshoot circuits that represent and process numeri
cal quantities in digital form. Since the bivalent (twovalued) system of binary bit numeration
lends itself so easily to representation by ”on” and ”off” transistor states (saturation and cutoff,
respectively), it makes sense to design and build circuits leveraging this principle to perform
binary calculations.
If we were to build a circuit to represent a binary number, we would have to allocate enough
transistor circuits to represent as many bits as we desire. In other words, in designing a digital
circuit, we must first decide how many bits (maximum) we would like to be able to represent,
since each bit requires one on/off circuit to represent it. This is analogous to designing an
abacus to digitally represent decimal numbers: we must decide how many digits we wish to
handle in this primitive ”calculator” device, for each digit requires a separate rod with its own
beads.
A 10rod abacus
Each rod represents
a single decimal digit
A tenrod abacus would be able to represent a tendigit decimal number, or a maxmium
value of 9,999,999,999. If we wished to represent a larger number on this abacus, we would be
unable to, unless additional rods could be added to it.
In digital, electronic computer design, it is common to design the system for a common ”bit
width:” a maximum number of bits allocated to represent numerical quantities. Early digital
computers handled bits in groups of four or eight. More modern systems handle numbers in
clusters of 32 bits or more. To more conveniently express the ”bit width” of such clusters in a
digital computer, specific labels were applied to the more common groupings.
28 CHAPTER 2. BINARY ARITHMETIC
Eight bits, grouped together to form a single binary quantity, is known as a byte. Four bits,
grouped together as one binary number, is known by the humorous title of nibble, often spelled
as nybble.
A multitude of terms have followed byte and nibble for labeling specfiic groupings of binary
bits. Most of the terms shown here are informal, and have not been made ”authoritative” by
any standards group or other sanctioning body. However, their inclusion into this chapter is
warranted by their occasional appearance in technical literature, as well as the levity they add
to an otherwise dry subject:
• Bit: A single, bivalent unit of binary notation. Equivalent to a decimal ”digit.”
• Crumb, Tydbit, or Tayste: Two bits.
• Nibble, or Nybble: Four bits.
• Nickle: Five bits.
• Byte: Eight bits.
• Deckle: Ten bits.
• Playte: Sixteen bits.
• Dynner: Thirtytwo bits.
• Word: (system dependent).
The most ambiguous term by far is word, referring to the standard bitgrouping within a
particular digital system. For a computer system using a 32 bitwide ”data path,” a ”word”
would mean 32 bits. If the system used 16 bits as the standard grouping for binary quantities,
a ”word” would mean 16 bits. The terms playte and dynner, by contrast, always refer to 16 and
32 bits, respectively, regardless of the system context in which they are used.
Context dependence is likewise true for derivative terms of word, such as double word
and longword (both meaning twice the standard bitwidth), halfword (half the standard bit
width), and quad (meaning four times the standard bitwidth). One humorous addition to this
somewhat boring collection of wordderivatives is the term chawmp, which means the same as
halfword. For example, a chawmp would be 16 bits in the context of a 32bit digital system,
and 18 bits in the context of a 36bit system. Also, the term gawble is sometimes synonymous
with word.
Definitions for bit grouping terms were taken from Eric S. Raymond’s ”Jargon Lexicon,” an
indexed collection of terms – both common and obscure – germane to the world of computer
programming.
Chapter 3
LOGIC GATES
Contents
3.1 Digital signals and gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
3.2 The NOT gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
3.3 The ”buffer” gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.4 Multipleinput gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3.4.1 The AND gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
3.4.2 The NAND gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
3.4.3 The OR gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
3.4.4 The NOR gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
3.4.5 The NegativeAND gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
3.4.6 The NegativeOR gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
3.4.7 The ExclusiveOR gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
3.4.8 The ExclusiveNOR gate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
3.5 TTL NAND and AND gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.6 TTL NOR and OR gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.7 CMOS gate circuitry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
3.8 Specialoutput gates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.9 Gate universality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
3.9.1 Constructing the NOT function . . . . . . . . . . . . . . . . . . . . . . . . 85
3.9.2 Constructing the ”buffer” function . . . . . . . . . . . . . . . . . . . . . . 86
3.9.3 Constructing the AND function . . . . . . . . . . . . . . . . . . . . . . . . 86
3.9.4 Constructing the NAND function . . . . . . . . . . . . . . . . . . . . . . . 87
3.9.5 Constructing the OR function . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.9.6 Constructing the NOR function . . . . . . . . . . . . . . . . . . . . . . . . 89
3.10 Logic signal voltage levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
3.11 DIP gate packaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.12 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
29
30 CHAPTER 3. LOGIC GATES
3.1 Digital signals and gates
While the binary numeration system is an interesting mathematical abstraction, we haven’t
yet seen its practical application to electronics. This chapter is devoted to just that: practically
applying the concept of binary bits to circuits. What makes binary numeration so important to
the application of digital electronics is the ease in which bits may be represented in physical
terms. Because a binary bit can only have one of two different values, either 0 or 1, any
physical medium capable of switching between two saturated states may be used to represent a
bit. Consequently, any physical system capable of representing binary bits is able to represent
numerical quantities, and potentially has the ability to manipulate those numbers. This is the
basic concept underlying digital computing.
Electronic circuits are physical systems that lend themselves well to the representation of
binary numbers. Transistors, when operated at their bias limits, may be in one of two differ
ent states: either cutoff (no controlled current) or saturation (maximum controlled current). If
a transistor circuit is designed to maximize the probability of falling into either one of these
states (and not operating in the linear, or active, mode), it can serve as a physical representa
tion of a binary bit. A voltage signal measured at the output of such a circuit may also serve as
a representation of a single bit, a low voltage representing a binary ”0” and a (relatively) high
voltage representing a binary ”1.” Note the following transistor circuit:
5 V
Vin = 5 V
Vout ≈ 0 V
0 V = "low" logic level (0)
"high" input
5 V = "high" logic level (1)
"low" output
Transistor in saturation
In this circuit, the transistor is in a state of saturation by virtue of the applied input voltage
(5 volts) through the twoposition switch. Because its saturated, the transistor drops very
little voltage between collector and emitter, resulting in an output voltage of (practically) 0
volts. If we were using this circuit to represent binary bits, we would say that the input signal
is a binary ”1” and that the output signal is a binary ”0.” Any voltage close to full supply
voltage (measured in reference to ground, of course) is considered a ”1” and a lack of voltage is
considered a ”0.” Alternative terms for these voltage levels are high (same as a binary ”1”) and
low (same as a binary ”0”). A general term for the representation of a binary bit by a circuit
voltage is logic level.
3.1. DIGITAL SIGNALS AND GATES 31
Moving the switch to the other position, we apply a binary ”0” to the input and receive a
binary ”1” at the output:
5 V
0 V = "low" logic level (0)
Vin = 0 V
Vout = 5 V
"low" input "high" output
5 V = "high" logic level (1)
Transistor in cutoff
What we’ve created here with a single transistor is a circuit generally known as a logic
gate, or simply gate. A gate is a special type of amplifier circuit designed to accept and gener
ate voltage signals corresponding to binary 1’s and 0’s. As such, gates are not intended to be
used for amplifying analog signals (voltage signals between 0 and full voltage). Used together,
multiple gates may be applied to the task of binary number storage (memory circuits) or ma
nipulation (computing circuits), each gate’s output representing one bit of a multibit binary
number. Just how this is done is a subject for a later chapter. Right now it is important to focus
on the operation of individual gates.
The gate shown here with the single transistor is known as an inverter, or NOT gate, be
cause it outputs the exact opposite digital signal as what is input. For convenience, gate circuits
are generally represented by their own symbols rather than by their constituent transistors
and resistors. The following is the symbol for an inverter:
Inverter, or NOT gate
Input Output
An alternative symbol for an inverter is shown here:
Input Output
Notice the triangular shape of the gate symbol, much like that of an operational amplifier.
As was stated before, gate circuits actually are amplifiers. The small circle, or ”bubble” shown
on either the input or output terminal is standard for representing the inversion function. As
you might suspect, if we were to remove the bubble from the gate symbol, leaving only a trian
gle, the resulting symbol would no longer indicate inversion, but merely direct amplification.
32 CHAPTER 3. LOGIC GATES
Such a symbol and such a gate actually do exist, and it is called a buffer, the subject of the next
section.
Like an operational amplifier symbol, input and output connections are shown as single
wires, the implied reference point for each voltage signal being ”ground.” In digital gate cir
cuits, ground is almost always the negative connection of a single voltage source (power sup
ply). Dual, or ”split,” power supplies are seldom used in gate circuitry. Because gate circuits are
amplifiers, they require a source of power to operate. Like operational amplifiers, the power
supply connections for digital gates are often omitted from the symbol for simplicity’s sake. If
we were to show all the necessary connections needed for operating this gate, the schematic
would look something like this:
5 V
Vin Vout
Vcc
Ground
Power supply conductors are rarely shown in gate circuit schematics, even if the power
supply connections at each gate are. Minimizing lines in our schematic, we get this:
VccVcc
”Vcc” stands for the constant voltage supplied to the collector of a bipolar junction transistor
circuit, in reference to ground. Those points in a gate circuit marked by the label ”Vcc” are all
connected to the same point, and that point is the positive terminal of a DC voltage source,
usually 5 volts.
As we will see in other sections of this chapter, there are quite a few different types of logic
gates, most of which have multiple input terminals for accepting more than one signal. The
output of any gate is dependent on the state of its input(s) and its logical function.
One common way to express the particular function of a gate circuit is called a truth table.
Truth tables show all combinations of input conditions in terms of logic level states (either
”high” or ”low,” ”1” or ”0,” for each input terminal of the gate), along with the corresponding
output logic level, either ”high” or ”low.” For the inverter, or NOT, circuit just illustrated, the
truth table is very simple indeed:
3.2. THE NOT GATE 33
OutputInput
0 1
1 0
NOT gate truth table
Input Output
Truth tables for more complex gates are, of course, larger than the one shown for the NOT
gate. A gate’s truth table must have as many rows as there are possibilities for unique input
combinations. For a singleinput gate like the NOT gate, there are only two possibilities, 0 and
1. For a two input gate, there are four possibilities (00, 01, 10, and 11), and thus four rows to
the corresponding truth table. For a threeinput gate, there are eight possibilities (000, 001,
010, 011, 100, 101, 110, and 111), and thus a truth table with eight rows are needed. The
mathematically inclined will realize that the number of truth table rows needed for a gate is
equal to 2 raised to the power of the number of input terminals.
• REVIEW:
• In digital circuits, binary bit values of 0 and 1 are represented by voltage signals mea
sured in reference to a common circuit point called ground. An absence of voltage repre
sents a binary ”0” and the presence of full DC supply voltage represents a binary ”1.”
• A logic gate, or simply gate, is a special form of amplifier circuit designed to input and
output logic level voltages (voltages intended to represent binary bits). Gate circuits are
most commonly represented in a schematic by their own unique symbols rather than by
their constituent transistors and resistors.
• Just as with operational amplifiers, the power supply connections to gates are often omit
ted in schematic diagrams for the sake of simplicity.
• A truth table is a standard way of representing the input/output relationships of a gate
circuit, listing all the possible input logic level combinations with their respective output
logic levels.
3.2 The NOT gate
The singletransistor inverter circuit illustrated earlier is actually too crude to be of practical
use as a gate. Real inverter circuits contain more than one transistor to maximize voltage gain
(so as to ensure that the final output transistor is either in full cutoff or full saturation), and
other components designed to reduce the chance of accidental damage.
Shown here is a schematic diagram for a real inverter circuit, complete with all necessary
components for efficient and reliable operation:
34 CHAPTER 3. LOGIC GATES
Output
Input
Vcc
Practical inverter (NOT) circuit
Q1 Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
This circuit is composed exclusively of resistors and bipolar transistors. Bear in mind that
other circuit designs are capable of performing the NOT gate function, including designs sub
stituting fieldeffect transistors for bipolar (discussed later in this chapter).
Let’s analyze this circuit for the condition where the input is ”high,” or in a binary ”1” state.
We can simulate this by showing the input terminal connected to Vcc through a switch:
Output
Input
Vcc
Q1 Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
+

5 V
0 V
In this case, diode D1 will be reversebiased, and therefore not conduct any current. In fact,
3.2. THE NOT GATE 35
the only purpose for having D1 in the circuit is to prevent transistor damage in the case of a
negative voltage being impressed on the input (a voltage that is negative, rather than positive,
with respect to ground). With no voltage between the base and emitter of transistor Q1, we
would expect no current through it, either. However, as strange as it may seem, transistor Q1
is not being used as is customary for a transistor. In reality, Q1 is being used in this circuit
as nothing more than a backtoback pair of diodes. The following schematic shows the real
function of Q1:
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
+

5 V
0 V
"Q1"
The purpose of these diodes is to ”steer” current to or away from the base of transistor
Q2, depending on the logic level of the input. Exactly how these two diodes are able to ”steer”
current isn’t exactly obvious at first inspection, so a short example may be necessary for under
standing. Suppose we had the following diode/resistor circuit, representing the baseemitter
junctions of transistors Q2 and Q4 as single diodes, stripping away all other portions of the cir
cuit so that we can concentrate on the current ”steered” through the two backtoback diodes:
36 CHAPTER 3. LOGIC GATES
Vcc
R3
"Q1"
Q2(BE)
Q4(BE)
5 V
R1
With the input switch in the ”up” position (connected to Vcc), it should be obvious that there
will be no current through the left steering diode of Q1, because there isn’t any voltage in
the switchdiodeR1switch loop to motivate electrons to flow. However, there will be current
through the right steering diode of Q1, as well as through Q2’s baseemitter diode junction and
Q4’s baseemitter diode junction:
Vcc
R3
"Q1"
Q2(BE)
Q4(BE)
5 V
R1
This tells us that in the real gate circuit, transistors Q2 and Q4 will have base current,
which will turn them on to conduct collector current. The total voltage dropped between the
base of Q1 (the node joining the two backtoback steering diodes) and ground will be about 2.1
volts, equal to the combined voltage drops of three PN junctions: the right steering diode, Q2’s
baseemitter diode, and Q4’s baseemitter diode.
Now, let’s move the input switch to the ”down” position and see what happens:
3.2. THE NOT GATE 37
Vcc
R3
"Q1"
Q2(BE)
Q4(BE)
5 V
R1
If we were to measure current in this circuit, we would find that all of the current goes
through the left steering diode of Q1 and none of it through the right diode. Why is this? It
still appears as though there is a complete path for current through Q4’s diode, Q2’s diode, the
right diode of the pair, and R1, so why will there be no current through that path?
Remember that PN junction diodes are very nonlinear devices: they do not even begin to
conduct current until the forward voltage applied across them reaches a certain minimum
quantity, approximately 0.7 volts for silicon and 0.3 volts for germanium. And then when
they begin to conduct current, they will not drop substantially more than 0.7 volts. When the
switch in this circuit is in the ”down” position, the left diode of the steering diode pair is fully
conducting, and so it drops about 0.7 volts across it and no more.
Vcc
R3
"Q1"
Q2(BE)
Q4(BE)
5 V
R1
0.7 V
+

Recall that with the switch in the ”up” position (transistors Q2 and Q4 conducting), there
was about 2.1 volts dropped between those same two points (Q1’s base and ground), which also
happens to be the minimum voltage necessary to forwardbias three seriesconnected silicon
PN junctions into a state of conduction. The 0.7 volts provided by the left diode’s forward volt
age drop is simply insufficient to allow any electron flow through the series string of the right
diode, Q2’s diode, and the R3//Q4 diode parallel subcircuit, and so no electrons flow through
38 CHAPTER 3. LOGIC GATES
that path. With no current through the bases of either transistor Q2 or Q4, neither one will be
able to conduct collector current: transistors Q2 and Q4 will both be in a state of cutoff.
Consequently, this circuit configuration allows 100 percent switching of Q2 base current
(and therefore control over the rest of the gate circuit, including voltage at the output) by
diversion of current through the left steering diode.
In the case of our example gate circuit, the input is held ”high” by the switch (connected
to Vcc), making the left steering diode (zero voltage dropped across it). However, the right
steering diode is conducting current through the base of Q2, through resistor R1:
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
+

5 V
0 V
"Q1"
With base current provided, transistor Q2 will be turned ”on.” More specifically, it will be
saturated by virtue of the morethanadequate current allowed by R1 through the base. With
Q2 saturated, resistor R3 will be dropping enough voltage to forwardbias the baseemitter
junction of transistor Q4, thus saturating it as well:
3.2. THE NOT GATE 39
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
+

5 V
0 V
"Q1"
≈ 0 V
≈ 0.7 V
With Q4 saturated, the output terminal will be almost directly shorted to ground, leaving
the output terminal at a voltage (in reference to ground) of almost 0 volts, or a binary ”0” (”low”)
logic level. Due to the presence of diode D2, there will not be enough voltage between the base
of Q3 and its emitter to turn it on, so it remains in cutoff.
Let’s see now what happens if we reverse the input’s logic level to a binary ”0” by actuating
the input switch:
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
"Q1"
5 V
0 V
+

≈ 5 V
40 CHAPTER 3. LOGIC GATES
Now there will be current through the left steering diode of Q1 and no current through the
right steering diode. This eliminates current through the base of Q2, thus turning it off. With
Q2 off, there is no longer a path for Q4 base current, so Q4 goes into cutoff as well. Q3, on the
other hand, now has sufficient voltage dropped between its base and ground to forwardbias
its baseemitter junction and saturate it, thus raising the output terminal voltage to a ”high”
state. In actuality, the output voltage will be somewhere around 4 volts depending on the
degree of saturation and any load current, but still high enough to be considered a ”high” (1)
logic level.
With this, our simulation of the inverter circuit is complete: a ”1” in gives a ”0” out, and
vice versa.
The astute observer will note that this inverter circuit’s input will assume a ”high” state of
left floating (not connected to either Vcc or ground). With the input terminal left unconnected,
there will be no current through the left steering diode of Q1, leaving all of R1’s current to go
through Q2’s base, thus saturating Q2 and driving the circuit output to a ”low” state:
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc = 5 volts
+

5 V
0 V
"Q1"
≈ 0 V
≈ 0.7 V
(floating)
The tendency for such a circuit to assume a high input state if left floating is one shared
by all gate circuits based on this type of design, known as TransistortoTransistor Logic, or
TTL. This characteristic may be taken advantage of in simplifying the design of a gate’s output
circuitry, knowing that the outputs of gates typically drive the inputs of other gates. If the
input of a TTL gate circuit assumes a high state when floating, then the output of any gate
driving a TTL input need only provide a path to ground for a low state and be floating for
a high state. This concept may require further elaboration for full understanding, so I will
explore it in detail here.
A gate circuit as we have just analyzed has the ability to handle output current in two
directions: in and out. Technically, this is known as sourcing and sinking current, respectively.
When the gate output is high, there is continuity from the output terminal to Vcc through the
top output transistor (Q3), allowing electrons to flow from ground, through a load, into the
3.2. THE NOT GATE 41
gate’s output terminal, through the emitter of Q3, and eventually up to the Vcc power terminal
(positive side of the DC power supply):
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
"Q1"
5 V
0 V
+

≈ 5 V Load
Inverter gate sourcing current
To simplify this concept, we may show the output of a gate circuit as being a doublethrow
switch, capable of connecting the output terminal either to Vcc or ground, depending on its
state. For a gate outputting a ”high” logic level, the combination of Q3 saturated and Q4 cutoff
is analogous to a doublethrow switch in the ”Vcc” position, providing a path for current through
a grounded load:
Input Output
Vcc
...
Load
Simplified gate circuit sourcing current
Please note that this twoposition switch shown inside the gate symbol is representative of
transistors Q3 and Q4 alternately connecting the output terminal to Vcc or ground, not of the
switch previously shown sending an input signal to the gate!
Conversely, when a gate circuit is outputting a ”low” logic level to a load, it is analogous
to the doublethrow switch being set in the ”ground” position. Current will then be going the
42 CHAPTER 3. LOGIC GATES
other way if the load resistance connects to Vcc: from ground, through the emitter of Q4, out
the output terminal, through the load resistance, and back to Vcc. In this condition, the gate is
said to be sinking current:
Output
Input
Vcc
Q2
Q3
Q4
D1
D2
R1 R2
R3
R4
Vcc
Vcc = 5 volts
+

5 V
0 V
"Q1"
≈ 0.7 V
Vcc
Inverter gate sinking current
Load
Input
Output
Vcc
...
Load
Simplified gate circuit sinking current
Vcc
The combination of Q3 and Q4 working as a ”pushpull” transistor pair (otherwise known
as a totem pole output) has the ability to either source current (draw in current to Vcc) or sink
current (output current from ground) to a load. However, a standard TTL gate input never
needs current to be sourced, only sunk. That is, since a TTL gate input naturally assumes a
high state if left floating, any gate output driving a TTL input need only sink current to provide
a ”0” or ”low” input, and need not source current to provide a ”1” or a ”high” logic level at the
input of the receiving gate:
3.2. THE NOT GATE 43
Input
Vcc
... ...
TTL
gate
A direct connection to Vcc is not
necessary to drive the TTL gate
input high!
Input
Vcc
... ...
TTL
gate
An output that "floats" when high
is sufficient.
Input
Vcc
... ...
TTL
gate
Vcc
Any gate driving a TTL
input must sink some
current in the low state.
This means we have the option of simplifying the output stage of a gate circuit so as to
eliminate Q3 altogether. The result is known as an opencollector output:
44 CHAPTER 3. LOGIC GATES
Output
Input
Vcc
D1
R1 R2
R3
Inverter circuit with opencollector output
To designate opencollector output circuitry within a standard gate symbol, a special marker
is used. Shown here is the symbol for an inverter gate with opencollector output:
Inverter with open
collector output
Please keep in mind that the ”high” default condition of a floating gate input is only true
for TTL circuitry, and not necessarily for other types, especially for logic gates constructed of
fieldeffect transistors.
• REVIEW:
• An inverter, or NOT, gate is one that outputs the opposite state as what is input. That is,
a ”low” input (0) gives a ”high” output (1), and vice versa.
• Gate circuits constructed of resistors and bipolar transistors as illustrated in this section
are called TTL. TTL is an acronym standing for TransistortoTransistor Logic. There are
other design methodologies used in gate circuits, some which use fieldeffect transistors
rather than bipolar transistors.
• A gate is said to be sourcing current when it provides a path for current between the
output terminal and the positive side of the DC power supply (Vcc). In other words, it is
connecting the output terminal to the power source (+V).
• A gate is said to be sinking current when it provides a path for current between the output
terminal and ground. In other words, it is grounding (sinking) the output terminal.
3.3. THE ”BUFFER” GATE 45
• Gate circuits with totem pole output stages are able to both source and sink current. Gate
circuits with opencollector output stages are only able to sink current, and not source
current. Opencollector gates are practical when used to drive TTL gate inputs because
TTL inputs don’t require current sourcing.
3.3 The ”buffer” gate
If we were to connect two inverter gates together so that the output of one fed into the input
of another, the two inversion functions would ”cancel” each other out so that there would be no
inversion from input to final output:
0
1
0
0 inverted into a 1
Logic state reinverted
to original status
Double inversion
While this may seem like a pointless thing to do, it does have practical application. Remem
ber that gate circuits are signal amplifiers, regardless of what logic function they may perform.
A weak signal source (one that is not capable of sourcing or sinking very much current to a
load) may be boosted by means of two inverters like the pair shown in the previous illustra
tion. The logic level is unchanged, but the full currentsourcing or sinking capabilities of the
final inverter are available to drive a load resistance if needed.
For this purpose, a special logic gate called a buffer is manufactured to perform the same
function as two inverters. Its symbol is simply a triangle, with no inverting ”bubble” on the
output terminal:
Input Output
"Buffer" gate
OutputInput
0
11
0
The internal schematic diagram for a typical opencollector buffer is not much different
from that of a simple inverter: only one more commonemitter transistor stage is added to
reinvert the output signal.
46 CHAPTER 3. LOGIC GATES
OutputInput
Vcc
D1
R1 R2
R3
Buffer circuit with opencollector output
Q1 Q2
Q3
Q4
R4
Inverter Inverter
Let’s analyze this circuit for two conditions: an input logic level of ”1” and an input logic
level of ”0.” First, a ”high” (1) input:
Output
Input
Vcc
D1
R1 R2
R3
Q1 Q2
Q3
Q4
R4
Vcc
As before with the inverter circuit, the ”high” input causes no conduction through the left
steering diode of Q1 (emittertobase PN junction). All of R1’s current goes through the base of
transistor Q2, saturating it:
3.3. THE ”BUFFER” GATE 47
Output
Input
Vcc
D1
R1 R2
R3
Q1 Q2
Q3
Q4
R4
Vcc
Having Q2 saturated causes Q3 to be saturated as well, resulting in very little voltage
dropped between the base and emitter of the final output transistor Q4. Thus, Q4 will be in
cutoff mode, conducting no current. The output terminal will be floating (neither connected to
ground nor Vcc), and this will be equivalent to a ”high” state on the input of the next TTL gate
that this one feeds in to. Thus, a ”high” input gives a ”high” output.
With a ”low” input signal (input terminal grounded), the analysis looks something like this:
Output
Input
Vcc
D1
R1 R2
R3
Q1 Q2
Q3
Q4
R4
Vcc
All of R1’s current is now diverted through the input switch, thus eliminating base current
through Q2. This forces transistor Q2 into cutoff so that no base current goes through Q3
either. With Q3 cutoff as well, Q4 is will be saturated by the current through resistor R4, thus
connecting the output terminal to ground, making it a ”low” logic level. Thus, a ”low” input
48 CHAPTER 3. LOGIC GATES
gives a ”low” output.
The schematic diagram for a buffer circuit with totem pole output transistors is a bit more
complex, but the basic principles, and certainly the truth table, are the same as for the open
collector circuit:
Output
Input
Vcc
D1
R1 R2
R3
Q1 Q2
Q3
Buffer circuit with totem pole output
Q4
Q5
Q6
D2
D3
R4
R5
Inverter Inverter
• REVIEW:
• Two inverter, or NOT, gates connected in ”series” so as to invert, then reinvert, a binary
bit perform the function of a buffer. Buffer gates merely serve the purpose of signal
amplification: taking a ”weak” signal source that isn’t capable of sourcing or sinking
much current, and boosting the current capacity of the signal so as to be able to drive a
load.
• Buffer circuits are symbolized by a triangle symbol with no inverter ”bubble.”
• Buffers, like inverters, may be made in opencollector output or totem pole output forms.
3.4 Multipleinput gates
Inverters and buffers exhaust the possibilities for singleinput gate circuits. What more can
be done with a single logic signal but to buffer it or invert it? To explore more logic gate
possibilities, we must add more input terminals to the circuit(s).
Adding more input terminals to a logic gate increases the number of input state possibil
ities. With a singleinput gate such as the inverter or buffer, there can only be two possible
input states: either the input is ”high” (1) or it is ”low” (0). As was mentioned previously in
3.4. MULTIPLEINPUT GATES 49
this chapter, a two input gate has four possibilities (00, 01, 10, and 11). A threeinput gate has
eight possibilities (000, 001, 010, 011, 100, 101, 110, and 111) for input states. The number of
possible input states is equal to two to the power of the number of inputs:
Number of possible input states = 2n
Where,
n = Number of inputs
This increase in the number of possible input states obviously allows for more complex gate
behavior. Now, instead of merely inverting or amplifying (buffering) a single ”high” or ”low”
logic level, the output of the gate will be determined by whatever combination of 1’s and 0’s is
present at the input terminals.
Since so many combinations are possible with just a few input terminals, there are many
different types of multipleinput gates, unlike singleinput gates which can only be inverters
or buffers. Each basic gate type will be presented in this section, showing its standard symbol,
truth table, and practical operation. The actual TTL circuitry of these different gates will be
explored in subsequent sections.
3.4.1 The AND gate
One of the easiest multipleinput gates to understand is the AND gate, socalled because the
output of this gate will be ”high” (1) if and only if all inputs (first input and the second input
and . . .) are ”high” (1). If any input(s) are ”low” (0), the output is guaranteed to be in a ”low”
state as well.
InputA
InputB
Output
2input AND gate
Output
3input AND gate
InputA
InputB
InputC
In case you might have been wondering, AND gates are made with more than three inputs,
but this is less common than the simple twoinput variety.
A twoinput AND gate’s truth table looks like this:
50 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
2input AND gate
A B Output
00
0 1
01
1 1
0
0
0
1
What this truth table means in practical terms is shown in the following sequence of illus
trations, with the 2input AND gate subjected to all possibilities of input logic levels. An LED
(LightEmitting Diode) provides visual indication of the output logic level:
InputA
InputB
Output
Vcc
Vcc
0
0 0
InputA =
InputB =
Output =
0
0
0 (no light)
InputA
InputB
Output
Vcc
Vcc
0
0
InputA =
InputB =
Output =
0
0 (no light)
1
1
3.4. MULTIPLEINPUT GATES 51
InputA
InputB
Output
Vcc
Vcc
0 0
InputA =
InputB =
Output =
0
0 (no light)
1
1
InputA
InputB
Output
Vcc
Vcc
InputA =
InputB =
Output =
1
1
1 1
1
1 (light!)
It is only with all inputs raised to ”high” logic levels that the AND gate’s output goes ”high,”
thus energizing the LED for only one out of the four input combination states.
3.4.2 The NAND gate
A variation on the idea of the AND gate is called the NAND gate. The word ”NAND” is a
verbal contraction of the words NOT and AND. Essentially, a NAND gate behaves the same as
an AND gate with a NOT (inverter) gate connected to the output terminal. To symbolize this
output signal inversion, the NAND gate symbol has a bubble on the output line. The truth
table for a NAND gate is as one might expect, exactly opposite as that of an AND gate:
52 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
1
1
2input NAND gate
InputA
InputB
Output
Equivalent gate circuit
As with AND gates, NAND gates are made with more than two inputs. In such cases, the
same general principle applies: the output will be ”low” (0) if and only if all inputs are ”high”
(1). If any input is ”low” (0), the output will go ”high” (1).
3.4.3 The OR gate
Our next gate to investigate is the OR gate, socalled because the output of this gate will be
”high” (1) if any of the inputs (first input or the second input or . . .) are ”high” (1). The output
of an OR gate goes ”low” (0) if and only if all inputs are ”low” (0).
InputA
InputB
Output Output
InputA
InputB
InputC
2input OR gate 3input OR gate
A twoinput OR gate’s truth table looks like this:
3.4. MULTIPLEINPUT GATES 53
InputA
InputB
Output
A B Output
00
0 1
01
1 1
0
1
2input OR gate
1
1
The following sequence of illustrations demonstrates the OR gate’s function, with the 2
inputs experiencing all possible logic levels. An LED (LightEmitting Diode) provides visual
indication of the gate’s output logic level:
InputA
InputB
Output
Vcc
Vcc
0
0 0
InputA =
InputB =
Output =
0
0
0 (no light)
InputA
InputB
Output
Vcc
Vcc
0
InputA =
InputB =
Output =
0
1
1
1
1 (light!)
54 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
Vcc
Vcc
0
InputA =
InputB =
Output =
0
1
1
1
1 (light!)
InputA
InputB
Output
Vcc
Vcc
InputA =
InputB =
Output =
1
1
1
1 (light!)
1
1
A condition of any input being raised to a ”high” logic level makes the OR gate’s output go
”high,” thus energizing the LED for three out of the four input combination states.
3.4.4 The NOR gate
As you might have suspected, the NOR gate is an OR gate with its output inverted, just like a
NAND gate is an AND gate with an inverted output.
3.4. MULTIPLEINPUT GATES 55
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
InputA
InputB
Output
Equivalent gate circuit
2input NOR gate
0
0
NOR gates, like all the other multipleinput gates seen thus far, can be manufactured with
more than two inputs. Still, the same logical principle applies: the output goes ”low” (0) if any
of the inputs are made ”high” (1). The output is ”high” (1) only when all inputs are ”low” (0).
3.4.5 The NegativeAND gate
A NegativeAND gate functions the same as an AND gate with all its inputs inverted (con
nected through NOT gates). In keeping with standard gate symbol convention, these inverted
inputs are signified by bubbles. Contrary to most peoples’ first instinct, the logical behavior of
a NegativeAND gate is not the same as a NAND gate. Its truth table, actually, is identical to
a NOR gate:
56 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
InputA
InputB
Output
2input NegativeAND gate
0
0
InputA
InputB
Output
Equivalent gate circuits
3.4.6 The NegativeOR gate
Following the same pattern, a NegativeOR gate functions the same as an OR gate with all its
inputs inverted. In keeping with standard gate symbol convention, these inverted inputs are
signified by bubbles. The behavior and truth table of a NegativeOR gate is the same as for a
NAND gate:
3.4. MULTIPLEINPUT GATES 57
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
InputA
InputB
Output
InputA
InputB
Output
Equivalent gate circuits
2input NegativeOR gate
1
1
3.4.7 The ExclusiveOR gate
The last six gate types are all fairly direct variations on three basic functions: AND, OR, and
NOT. The ExclusiveOR gate, however, is something quite different.
ExclusiveOR gates output a ”high” (1) logic level if the inputs are at different logic levels,
either 0 and 1 or 1 and 0. Conversely, they output a ”low” (0) logic level if the inputs are at
the same logic levels. The ExclusiveOR (sometimes called XOR) gate has both a symbol and a
truth table pattern that is unique:
58 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
1
ExclusiveOR gate
0
There are equivalent circuits for an ExclusiveOR gate made up of AND, OR, and NOT
gates, just as there were for NAND, NOR, and the negativeinput gates. A rather direct ap
proach to simulating an ExclusiveOR gate is to start with a regular OR gate, then add addi
tional gates to inhibit the output from going ”high” (1) when both inputs are ”high” (1):
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
1
0
ExclusiveOR equivalent circuit
In this circuit, the final AND gate acts as a buffer for the output of the OR gate whenever
the NAND gate’s output is high, which it is for the first three input state combinations (00, 01,
and 10). However, when both inputs are ”high” (1), the NAND gate outputs a ”low” (0) logic
level, which forces the final AND gate to produce a ”low” (0) output.
Another equivalent circuit for the ExclusiveOR gate uses a strategy of two AND gates with
inverters, set up to generate ”high” (1) outputs for input conditions 01 and 10. A final OR gate
then allows either of the AND gates’ ”high” outputs to create a final ”high” output:
3.4. MULTIPLEINPUT GATES 59
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
1
0
ExclusiveOR equivalent circuit
ExclusiveOR gates are very useful for circuits where two or more binary numbers are to be
compared bitforbit, and also for error detection (parity check) and code conversion (binary to
Grey and vice versa).
3.4.8 The ExclusiveNOR gate
Finally, our last gate for analysis is the ExclusiveNOR gate, otherwise known as the XNOR
gate. It is equivalent to an ExclusiveOR gate with an inverted output. The truth table for this
gate is exactly opposite as for the ExclusiveOR gate:
60 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1
1
InputA
InputB
Output
Equivalent gate circuit
0
0
ExclusiveNOR gate
1
As indicated by the truth table, the purpose of an ExclusiveNOR gate is to output a ”high”
(1) logic level whenever both inputs are at the same logic levels (either 00 or 11).
• REVIEW:
• Rule for an AND gate: output is ”high” only if first input and second input are both ”high.”
• Rule for an OR gate: output is ”high” if input A or input B are ”high.”
• Rule for a NAND gate: output is not ”high” if both the first input and the second input
are ”high.”
• Rule for a NOR gate: output is not ”high” if either the first input or the second input are
”high.”
• A NegativeAND gate behaves like a NOR gate.
• A NegativeOR gate behaves like a NAND gate.
• Rule for an ExclusiveOR gate: output is ”high” if the input logic levels are different.
• Rule for an ExclusiveNOR gate: output is ”high” if the input logic levels are the same.
3.5 TTL NAND and AND gates
Suppose we altered our basic opencollector inverter circuit, adding a second input terminal
just like the first:
3.5. TTL NAND AND AND GATES 61
Output
Vcc
D1
R1 R2
R3
D2
"Q1"
Q2
Q3
A twoinput inverter circuit
InputA
InputB
This schematic illustrates a real circuit, but it isn’t called a ”twoinput inverter.” Through
analysis we will discover what this circuit’s logic function is and correspondingly what it should
be designated as.
Just as in the case of the inverter and buffer, the ”steering” diode cluster marked ”Q1” is
actually formed like a transistor, even though it isn’t used in any amplifying capacity. Unfor
tunately, a simple NPN transistor structure is inadequate to simulate the three PN junctions
necessary in this diode network, so a different transistor (and symbol) is needed. This transis
tor has one collector, one base, and two emitters, and in the circuit it looks like this:
Output
Vcc
D1
R1 R2
R3
D2
Q2
Q3
Q1InputAInputB
62 CHAPTER 3. LOGIC GATES
In the singleinput (inverter) circuit, grounding the input resulted in an output that as
sumed the ”high” (1) state. In the case of the opencollector output configuration, this ”high”
state was simply ”floating.” Allowing the input to float (or be connected to Vcc) resulted in the
output becoming grounded, which is the ”low” or 0 state. Thus, a 1 in resulted in a 0 out, and
vice versa.
Since this circuit bears so much resemblance to the simple inverter circuit, the only differ
ence being a second input terminal connected in the same way to the base of transistor Q2, we
can say that each of the inputs will have the same effect on the output. Namely, if either of the
inputs are grounded, transistor Q2 will be forced into a condition of cutoff, thus turning Q3 off
and floating the output (output goes ”high”). The following series of illustrations shows this for
three input states (00, 01, and 10):
Output
Vcc
D1
R1 R2
R3
D2
"Q1"
Q2
Q3
InputA
InputB
Vcc
0
0 Cutoff
Cutoff
1
InputA =
InputB =
Output =
0
0
1
Output
Vcc
D1
R1 R2
R3
D2
"Q1"
Q2
Q3
InputA
InputB
Vcc
0 Cutoff
Cutoff
1
InputA =
InputB =
Output =
0
1
1
1
3.5. TTL NAND AND AND GATES 63
Output
Vcc
D1
R1 R2
R3
D2
"Q1"
Q2
Q3
InputA
InputB
Vcc
0
Cutoff
Cutoff
1
InputA =
InputB =
Output =
0
1
1
1
In any case where there is a grounded (”low”) input, the output is guaranteed to be floating
(”high”). Conversely, the only time the output will ever go ”low” is if transistor Q3 turns on,
which means transistor Q2 must be turned on (saturated), which means neither input can
be diverting R1 current away from the base of Q2. The only condition that will satisfy this
requirement is when both inputs are ”high” (1):
Output
Vcc
D1
R1 R2
R3
D2
"Q1"
Q2
Q3
InputA
InputB
Vcc
InputA =
InputB =
Output = 0
1
1
1
Saturation
Saturation
1
0
Collecting and tabulating these results into a truth table, we see that the pattern matches
that of the NAND gate:
64 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
1
1
NAND gate
In the earlier section on NAND gates, this type of gate was created by taking an AND gate
and increasing its complexity by adding an inverter (NOT gate) to the output. However, when
we examine this circuit, we see that the NAND function is actually the simplest, most natural
mode of operation for this TTL design. To create an AND function using TTL circuitry, we need
to increase the complexity of this circuit by adding an inverter stage to the output, just like we
had to add an additional transistor stage to the TTL inverter circuit to turn it into a buffer:
Output
Vcc
D1
R1 R2
R3
Q1 Q2
Q3
Q4
R4
Inverter
AND gate with opencollector output
NAND gate
D2
InputA
InputB
The truth table and equivalent gate circuit (an invertedoutput NAND gate) are shown
here:
3.6. TTL NOR AND OR GATES 65
InputA
InputB
Output
A B Output
00
0 1
01
1 1 1
AND gate
Equivalent circuit
InputA
InputB
Output
0
0
0
Of course, both NAND and AND gate circuits may be designed with totempole output
stages rather than opencollector. I am opting to show the opencollector versions for the sake
of simplicity.
• REVIEW:
• A TTL NAND gate can be made by taking a TTL inverter circuit and adding another
input.
• An AND gate may be created by adding an inverter stage to the output of the NAND gate
circuit.
3.6 TTL NOR and OR gates
Let’s examine the following TTL circuit and analyze its operation:
66 CHAPTER 3. LOGIC GATES
Output
Vcc
D1
R1 R2 R3
D2
Q2
Q1InputA
InputB
Q3 Q4
Q5
R4
Transistors Q1 and Q2 are both arranged in the same manner that we’ve seen for transistor
Q1 in all the other TTL circuits. Rather than functioning as amplifiers, Q1 and Q2 are both
being used as twodiode ”steering” networks. We may replace Q1 and Q2 with diode sets to
help illustrate:
Output
Vcc
D1
R1 R2 R3
D2
InputA
InputB
Q3 Q4
Q5
R4
"Q1"
"Q2"
If input A is left floating (or connected to Vcc), current will go through the base of transistor
Q3, saturating it. If input A is grounded, that current is diverted away from Q3’s base through
the left steering diode of ”Q1,” thus forcing Q3 into cutoff. The same can be said for input B and
transistor Q4: the logic level of input B determines Q4’s conduction: either saturated or cutoff.
Notice how transistors Q3 and Q4 are paralleled at their collector and emitter terminals.
In essence, these two transistors are acting as paralleled switches, allowing current through
3.6. TTL NOR AND OR GATES 67
resistors R3 and R4 according to the logic levels of inputs A and B. If any input is at a ”high” (1)
level, then at least one of the two transistors (Q3 and/or Q4) will be saturated, allowing current
through resistors R3 and R4, and turning on the final output transistor Q5 for a ”low” (0) logic
level output. The only way the output of this circuit can ever assume a ”high” (1) state is if
both Q3 and Q4 are cutoff, which means both inputs would have to be grounded, or ”low” (0).
This circuit’s truth table, then, is equivalent to that of the NOR gate:
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
1
NOR gate
0
0
In order to turn this NOR gate circuit into an OR gate, we would have to invert the output
logic level with another transistor stage, just like we did with the NANDtoAND gate example:
Output
Vcc
D1
R1 R2 R3
D2
Q2
Q1InputA
InputB
Q3 Q4
Q5
R4
R5
Q6
OR gate with opencollector output
InverterNOR gate
The truth table and equivalent gate circuit (an invertedoutput NOR gate) are shown here:
68 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 1
Equivalent circuit
InputA
InputB
Output
0
OR gate
1
1
Of course, totempole output stages are also possible in both NOR and OR TTL logic circuits.
• REVIEW:
• An OR gate may be created by adding an inverter stage to the output of the NOR gate
circuit.
3.7 CMOS gate circuitry
Up until this point, our analysis of transistor logic circuits has been limited to the TTL design
paradigm, whereby bipolar transistors are used, and the general strategy of floating inputs
being equivalent to ”high” (connected to Vcc) inputs – and correspondingly, the allowance of
”opencollector” output stages – is maintained. This, however, is not the only way we can build
logic gates.
Fieldeffect transistors, particularly the insulatedgate variety, may be used in the design
of gate circuits. Being voltagecontrolled rather than currentcontrolled devices, IGFETs tend
to allow very simple circuit designs. Take for instance, the following inverter circuit built using
P and Nchannel IGFETs:
3.7. CMOS GATE CIRCUITRY 69
Vdd (+5 volts)
Inverter circuit using IGFETs
Input Output
Notice the ”Vdd” label on the positive power supply terminal. This label follows the same
convention as ”Vcc” in TTL circuits: it stands for the constant voltage applied to the drain of a
field effect transistor, in reference to ground.
Let’s connect this gate circuit to a power source and input switch, and examine its opera
tion. Please note that these IGFET transistors are Etype (Enhancementmode), and so are
normallyoff devices. It takes an applied voltage between gate and drain (actually, between
gate and substrate) of the correct polarity to bias them on.
Input Output
0 V
5 V
+

Saturated
Cutoff
5 V
+

5 V
+

Input = "low" (0)
Output = "high" (1)
The upper transistor is a Pchannel IGFET. When the channel (substrate) is made more
positive than the gate (gate negative in reference to the substrate), the channel is enhanced and
current is allowed between source and drain. So, in the above illustration, the top transistor is
turned on.
The lower transistor, having zero voltage between gate and substrate (source), is in its
normal mode: off. Thus, the action of these two transistors are such that the output terminal
of the gate circuit has a solid connection to Vdd and a very high resistance connection to ground.
This makes the output ”high” (1) for the ”low” (0) state of the input.
Next, we’ll move the input switch to its other position and see what happens:
70 CHAPTER 3. LOGIC GATES
Input Output
0 V
5 V
+

Saturated
Cutoff
5 V
+

0 V
Input = "high" (1)
Output = "low" (0)
Now the lower transistor (Nchannel) is saturated because it has sufficient voltage of the
correct polarity applied between gate and substrate (channel) to turn it on (positive on gate,
negative on the channel). The upper transistor, having zero voltage applied between its gate
and substrate, is in its normal mode: off. Thus, the output of this gate circuit is now ”low” (0).
Clearly, this circuit exhibits the behavior of an inverter, or NOT gate.
Using fieldeffect transistors instead of bipolar transistors has greatly simplified the design
of the inverter gate. Note that the output of this gate never floats as is the case with the
simplest TTL circuit: it has a natural ”totempole” configuration, capable of both sourcing and
sinking load current. Key to this gate circuit’s elegant design is the complementary use of both
P and Nchannel IGFETs. Since IGFETs are more commonly known as MOSFETs (Metal
OxideSemiconductor Field Effect Transistor), and this circuit uses both P and Nchannel
transistors together, the general classification given to gate circuits like this one is CMOS:
Complementary Metal Oxide Semiconductor.
CMOS circuits aren’t plagued by the inherent nonlinearities of the fieldeffect transistors,
because as digital circuits their transistors always operate in either the saturated or cutoff
modes and never in the active mode. Their inputs are, however, sensitive to high voltages
generated by electrostatic (static electricity) sources, and may even be activated into ”high” (1)
or ”low” (0) states by spurious voltage sources if left floating. For this reason, it is inadvisable
to allow a CMOS logic gate input to float under any circumstances. Please note that this is
very different from the behavior of a TTL gate where a floating input was safely interpreted as
a ”high” (1) logic level.
This may cause a problem if the input to a CMOS logic gate is driven by a singlethrow
switch, where one state has the input solidly connected to either Vdd or ground and the other
state has the input floating (not connected to anything):
3.7. CMOS GATE CIRCUITRY 71
Input Output...
CMOS gate
When switch is closed, the gate sees a
definite "low" (0) input. However, when
switch is open, the input logic level will
be uncertain because it’s floating.
Also, this problem arises if a CMOS gate input is being driven by an opencollector TTL
gate. Because such a TTL gate’s output floats when it goes ”high” (1), the CMOS gate input
will be left in an uncertain state:
Input Output
Vcc
...
TTL gate
...
Input
CMOS gateOpencollector
Vdd
When the opencollector TTL gate’s output
is "high" (1), the CMOS gate’s input will be
left floating and in an uncertain logic state.
Fortunately, there is an easy solution to this dilemma, one that is used frequently in CMOS
logic circuitry. Whenever a singlethrow switch (or any other sort of gate output incapable of
both sourcing and sinking current) is being used to drive a CMOS input, a resistor connected to
either Vdd or groundmay be used to provide a stable logic level for the state in which the driving
device’s output is floating. This resistor’s value is not critical: 10 kΩ is usually sufficient. When
used to provide a ”high” (1) logic level in the event of a floating signal source, this resistor is
known as a pullup resistor:
72 CHAPTER 3. LOGIC GATES
Input Output
...
CMOS gate
When switch is closed, the gate sees a
Vdd
Rpullup
definite "low" (0) input. When the switch
is open, Rpullup will provide the connection
to Vdd needed to secure a reliable "high"
logic level for the CMOS gate input.
When such a resistor is used to provide a ”low” (0) logic level in the event of a floating
signal source, it is known as a pulldown resistor. Again, the value for a pulldown resistor is
not critical:
Input
Output...
CMOS gate
When switch is closed, the gate sees a
logic level for the CMOS gate input.
Vdd
Rpulldown
is open, Rpulldown will provide the connection
to ground needed to secure a reliable "low"
definite "high" (1) input. When the switch
Because opencollector TTL outputs always sink, never source, current, pullup resistors are
necessary when interfacing such an output to a CMOS gate input:
Vcc
...
TTL gate
...
CMOS gateOpencollector
Vdd
Vdd
Rpullup
3.7. CMOS GATE CIRCUITRY 73
Although the CMOS gates used in the preceding examples were all inverters (singleinput),
the same principle of pullup and pulldown resistors applies to multipleinput CMOS gates. Of
course, a separate pullup or pulldown resistor will be required for each gate input:
InputA
InputB
InputC
Vdd
Pullup resistors for a 3input
CMOS AND gate
This brings us to the next question: how do we design multipleinput CMOS gates such as
AND, NAND, OR, and NOR? Not surprisingly, the answer(s) to this question reveal a simplicity
of design much like that of the CMOS inverter over its TTL equivalent.
For example, here is the schematic diagram for a CMOS NAND gate:
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
CMOS NAND gate
Notice how transistors Q1 and Q3 resemble the seriesconnected complementary pair from
the inverter circuit. Both are controlled by the same input signal (input A), the upper transistor
turning off and the lower transistor turning on when the input is ”high” (1), and vice versa.
Notice also how transistors Q2 and Q4 are similarly controlled by the same input signal (input
B), and how they will also exhibit the same on/off behavior for the same input logic levels. The
74 CHAPTER 3. LOGIC GATES
upper transistors of both pairs (Q1 and Q2) have their source and drain terminals paralleled,
while the lower transistors (Q3 and Q4) are seriesconnected. What this means is that the
output will go ”high” (1) if either top transistor saturates, and will go ”low” (0) only if both
lower transistors saturate. The following sequence of illustrations shows the behavior of this
NAND gate for all four possibilities of input logic levels (00, 01, 10, and 11):
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
Vdd
ON ON
OFF
OFF
1
0
0
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
Vdd
ON
OFF
OFF
1
0
1
ON
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
Vdd
ON
OFF
1
0
1 ON
OFF
3.7. CMOS GATE CIRCUITRY 75
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
Vdd
1 ON
OFF
1
ON
OFF
0
As with the TTL NAND gate, the CMOS NAND gate circuit may be used as the starting
point for the creation of an AND gate. All that needs to be added is another stage of transistors
to invert the output signal:
Vdd
Output
InputA
InputB
Q1 Q2
Q3
Q4
CMOS AND gate
InverterNAND gate
Q5
Q6
A CMOS NOR gate circuit uses four MOSFETs just like the NAND gate, except that its
transistors are differently arranged. Instead of two paralleled sourcing (upper) transistors
connected to Vdd and two seriesconnected sinking (lower) transistors connected to ground, the
NOR gate uses two seriesconnected sourcing transistors and two parallelconnected sinking
transistors like this:
76 CHAPTER 3. LOGIC GATES
Vdd
Output
InputA
InputB
Q1
Q2
Q3 Q4
CMOS NOR gate
As with the NAND gate, transistors Q1 and Q3 work as a complementary pair, as do tran
sistors Q2 and Q4. Each pair is controlled by a single input signal. If either input A or input B
are ”high” (1), at least one of the lower transistors (Q3 or Q4) will be saturated, thus making
the output ”low” (0). Only in the event of both inputs being ”low” (0) will both lower transistors
be in cutoff mode and both upper transistors be saturated, the conditions necessary for the
output to go ”high” (1). This behavior, of course, defines the NOR logic function.
The OR function may be built up from the basic NOR gate with the addition of an inverter
stage on the output:
3.7. CMOS GATE CIRCUITRY 77
Vdd
Output
InputA
InputB
Q1
Q2
Q3 Q4
Q5
Q6
InverterNOR gate
CMOS OR gate
Since it appears that any gate possible to construct using TTL technology can be duplicated
in CMOS, why do these two ”families” of logic design still coexist? The answer is that both TTL
and CMOS have their own unique advantages.
First and foremost on the list of comparisons between TTL and CMOS is the issue of power
consumption. In this measure of performance, CMOS is the unchallenged victor. Because
the complementary P and Nchannel MOSFET pairs of a CMOS gate circuit are (ideally)
never conducting at the same time, there is little or no current drawn by the circuit from the
Vdd power supply except for what current is necessary to source current to a load. TTL, on
the other hand, cannot function without some current drawn at all times, due to the biasing
requirements of the bipolar transistors from which it is made.
There is a caveat to this advantage, though. While the power dissipation of a TTL gate
remains rather constant regardless of its operating state(s), a CMOS gate dissipates more
power as the frequency of its input signal(s) rises. If a CMOS gate is operated in a static
(unchanging) condition, it dissipates zero power (ideally). However, CMOS gate circuits draw
transient current during every output state switch from ”low” to ”high” and vice versa. So,
the more often a CMOS gate switches modes, the more often it will draw current from the Vdd
supply, hence greater power dissipation at greater frequencies.
A CMOS gate also draws much less current from a driving gate output than a TTL gate
because MOSFETs are voltagecontrolled, not currentcontrolled, devices. This means that
one gate can drive many more CMOS inputs than TTL inputs. The measure of how many gate
inputs a single gate output can drive is called fanout.
Another advantage that CMOS gate designs enjoy over TTL is a much wider allowable
range of power supply voltages. Whereas TTL gates are restricted to power supply (Vcc) volt
78 CHAPTER 3. LOGIC GATES
ages between 4.75 and 5.25 volts, CMOS gates are typically able to operate on any voltage
between 3 and 15 volts! The reason behind this disparity in power supply voltages is the
respective bias requirements of MOSFET versus bipolar junction transistors. MOSFETs are
controlled exclusively by gate voltage (with respect to substrate), whereas BJTs are current
controlled devices. TTL gate circuit resistances are precisely calculated for proper bias currents
assuming a 5 volt regulated power supply. Any significant variations in that power supply volt
age will result in the transistor bias currents being incorrect, which then results in unreliable
(unpredictable) operation. The only effect that variations in power supply voltage have on a
CMOS gate is the voltage definition of a ”high” (1) state. For a CMOS gate operating at 15 volts
of power supply voltage (Vdd), an input signal must be close to 15 volts in order to be considered
”high” (1). The voltage threshold for a ”low” (0) signal remains the same: near 0 volts.
One decided disadvantage of CMOS is slow speed, as compared to TTL. The input capaci
tances of a CMOS gate are much, much greater than that of a comparable TTL gate – owing
to the use of MOSFETs rather than BJTs – and so a CMOS gate will be slower to respond to a
signal transition (lowtohigh or vice versa) than a TTL gate, all other factors being equal. The
RC time constant formed by circuit resistances and the input capacitance of the gate tend to
impede the fast rise and falltimes of a digital logic level, thereby degrading highfrequency
performance.
A strategy for minimizing this inherent disadvantage of CMOS gate circuitry is to ”buffer”
the output signal with additional transistor stages, to increase the overall voltage gain of the
device. This provides a fastertransitioning output voltage (hightolow or lowtohigh) for an
input voltage slowly changing from one logic state to another. Consider this example, of an
”unbuffered” NOR gate versus a ”buffered,” or Bseries, NOR gate:
3.7. CMOS GATE CIRCUITRY 79
Vdd
Output
InputA
InputB
Q1
Q2
Q3 Q4
"Unbuffered" NOR gate
Vdd
Output
InputA
InputB
Q1
Q2
Q3 Q4
"Bseries" (buffered) NOR gate
In essence, the Bseries design enhancement adds two inverters to the output of a simple
NOR circuit. This serves no purpose as far as digital logic is concerned, since two cascaded
inverters simply cancel:
80 CHAPTER 3. LOGIC GATES
(same as)
(same as)
However, adding these inverter stages to the circuit does serve the purpose of increasing
overall voltage gain, making the output more sensitive to changes in input state, working to
overcome the inherent slowness caused by CMOS gate input capacitance.
• REVIEW:
• CMOS logic gates are made of IGFET (MOSFET) transistors rather than bipolar junction
transistors.
• CMOS gate inputs are sensitive to static electricity. They may be damaged by high volt
ages, and they may assume any logic level if left floating.
• Pullup and pulldown resistors are used to prevent a CMOS gate input from floating if
being driven by a signal source capable only of sourcing or sinking current.
• CMOS gates dissipate far less power than equivalent TTL gates, but their power dissi
pation increases with signal frequency, whereas the power dissipation of a TTL gate is
approximately constant over a wide range of operating conditions.
• CMOS gate inputs draw far less current than TTL inputs, because MOSFETs are voltage
controlled, not currentcontrolled, devices.
• CMOS gates are able to operate on a much wider range of power supply voltages than
TTL: typically 3 to 15 volts versus 4.75 to 5.25 volts for TTL.
• CMOS gates tend to have a much lower maximum operating frequency than TTL gates
due to input capacitances caused by the MOSFET gates.
3.8. SPECIALOUTPUT GATES 81
• Bseries CMOS gates have ”buffered” outputs to increase voltage gain from input to out
put, resulting in faster output response to input signal changes. This helps overcome the
inherent slowness of CMOS gates due to MOSFET input capacitance and the RC time
constant thereby engendered.
3.8 Specialoutput gates
It is sometimes desirable to have a logic gate that provides both inverted and noninverted
outputs. For example, a singleinput gate that is both a buffer and an inverter, with a sep
arate output terminal for each function. Or, a twoinput gate that provides both the AND
and the NAND functions in a single circuit. Such gates do exist and they are referred to as
complementary output gates.
The general symbology for such a gate is the basic gate figure with a bar and two output
lines protruding from it. An array of complementary gate symbols is shown in the following
illustration:
Complementary buffer
Complementary AND gate
Complementary OR gate
Complementary XOR gate
Complementary gates are especially useful in ”crowded” circuits where there may not be
enough physical room to mount the additional integrated circuit chips necessary to provide
both inverted and noninverted outputs using standard gates and additional inverters. They
are also useful in applications where a complementary output is necessary from a gate, but the
addition of an inverter would introduce an unwanted time lag in the inverted output relative
to the noninverted output. The internal circuitry of complemented gates is such that both
82 CHAPTER 3. LOGIC GATES
inverted and noninverted outputs change state at almost exactly the same time:
...
... ...
...
Time delay introduced
by the inverter
Complemented gate Standard gate with inverter added
Another type of special gate output is called tristate, because it has the ability to provide
three different output modes: current sinking (”low” logic level), current sourcing (”high”),
and floating (”highZ,” or highimpedance). Tristate outputs are usually found as an optional
feature on buffer gates. Such gates require an extra input terminal to control the ”highZ”
mode, and this input is usually called the enable.
Input Output...
+V
Tristate buffer gate
.
.
.
Enable
With the enable input held ”high” (1), the buffer acts like an ordinary buffer with a totem
pole output stage: it is capable of both sourcing and sinking current. However, the output
terminal floats (goes into ”highZ” mode) if ever the enable input is grounded (”low”), regardless
of the data signal’s logic level. In other words, making the enable input terminal ”low” (0)
effectively disconnects the gate from whatever its output is wired to so that it can no longer
have any effect.
Tristate buffers are marked in schematic diagrams by a triangle character within the gate
3.8. SPECIALOUTPUT GATES 83
symbol like this:
Input Output
Tristate buffer symbol
A B Output
00
0 1
01
1 1
HighZ
HighZ
0
1
Truth table
(A)
Enable (B)
Tristate buffers are also made with inverted enable inputs. Such a gate acts normal when
the enable input is ”low” (0) and goes into highZ output mode when the enable input is ”high”
(1):
Input Output
A B Output
00
0 1
01
1 1
HighZ
HighZ
0
1
Truth table
Tristate buffer with
inverted enable input
(A)
Enable (B)
One special type of gate known as the bilateral switch uses gatecontrolled MOSFET tran
sistors acting as on/off switches to switch electrical signals, analog or digital. The ”on” resis
tance of such a switch is in the range of several hundred ohms, the ”off” resistance being in the
range of several hundred megaohms.
84 CHAPTER 3. LOGIC GATES
Bilateral switches appear in schematics as SPST (SinglePole, SingleThrow) switches in
side of rectangular boxes, with a control terminal on one of the box’s long sides:
Control
In/Out In/Out
CMOS bilateral switch
A bilateral switch might be best envisioned as a solidstate (semiconductor) version of an
electromechanical relay: a signalactuated switch contact that may be used to conduct virtu
ally any type of electric signal. Of course, being solidstate, the bilateral switch has none of
the undesirable characteristics of electromechanical relays, such as contact ”bouncing,” arcing,
slow speed, or susceptibility to mechanical vibration. Conversely, though, they are rather lim
ited in their currentcarrying ability. Additionally, the signal conducted by the ”contact” must
not exceed the power supply ”rail” voltages powering the bilateral switch circuit.
Four bilateral switches are packaged inside the popular model ”4066” integrated circuit:
14 13 12 11 10 9 8
7654321
Vdd
GND
Quad CMOS bilateral switch
4066
• REVIEW:
• Complementary gates provide both inverted and noninverted output signals, in such a
way that neither one is delayed with respect to the other.
• Tristate gates provide three different output states: high, low, and floating (HighZ). Such
gates are commanded into their highimpedance output modes by a separate input ter
minal called the enable.
• Bilateral switches are MOSFET circuits providing on/off switching for a variety of electri
cal signal types (analog and digital), controlled by logic level voltage signals. In essence,
they are solidstate relays with very low currenthandling ability.
3.9. GATE UNIVERSALITY 85
3.9 Gate universality
NAND and NOR gates possess a special property: they are universal. That is, given enough
gates, either type of gate is able to mimic the operation of any other gate type. For example,
it is possible to build a circuit exhibiting the OR function using three interconnected NAND
gates. The ability for a single gate type to be able to mimic any other gate type is one enjoyed
only by the NAND and the NOR. In fact, digital control systems have been designed around
nothing but either NAND or NOR gates, all the necessary logic functions being derived from
collections of interconnected NANDs or NORs.
As proof of this property, this section will be divided into subsections showing how all the
basic gate types may be formed using only NANDs or only NORs.
3.9.1 Constructing the NOT function
OutputInput
0 1
1 0
Input Output
Input
Output
Input
Output
. . . or . . .
Input
Output Output
+V
Input
As you can see, there are two ways to use a NAND gate as an inverter, and two ways to use
a NOR gate as an inverter. Either method works, although connecting TTL inputs together
increases the amount of current loading to the driving gate. For CMOS gates, common input
terminals decreases the switching speed of the gate due to increased input capacitance.
Inverters are the fundamental tool for transforming one type of logic function into another,
and so there will be many inverters shown in the illustrations to follow. In those diagrams, I
will only show one method of inversion, and that will be where the unused NAND gate input
is connected to +V (either Vcc or Vdd, depending on whether the circuit is TTL or CMOS) and
where the unused input for the NOR gate is connected to ground. Bear in mind that the other
inversion method (connecting both NAND or NOR inputs together) works just as well from a
86 CHAPTER 3. LOGIC GATES
logical (1’s and 0’s) point of view, but is undesirable from the practical perspectives of increased
current loading for TTL and increased input capacitance for CMOS.
3.9.2 Constructing the ”buffer” function
Being that it is quite easy to employ NAND and NOR gates to perform the inverter (NOT)
function, it stands to reason that two such stages of gates will result in a buffer function,
where the output is the same logical state as the input.
OutputInput
0
11
0
Input Output
Input
Output
+V
+V
Input
Output
3.9.3 Constructing the AND function
To make the AND function from NAND gates, all that is needed is an inverter (NOT) stage on
the output of a NAND gate. This extra inversion ”cancels out” the first N in NAND, leaving
the AND function. It takes a little more work to wrestle the same functionality out of NOR
gates, but it can be done by inverting (”NOT”) all of the inputs to a NOR gate.
3.9. GATE UNIVERSALITY 87
InputA
InputB
Output
2input AND gate
A B Output
00
0 1
01
1 1
0
0
0
1
+V
OutputInputA
InputB
Output
InputA
InputB
3.9.4 Constructing the NAND function
It would be pointless to show you how to ”construct” the NAND function using a NAND gate,
since there is nothing to do. To make a NOR gate perform the NAND function, we must invert
all inputs to the NOR gate as well as the NOR gate’s output. For a twoinput gate, this requires
three more NOR gates connected as inverters.
88 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1 0
Output
InputA
InputB
2input NAND gate
1
1
1
3.9.5 Constructing the OR function
Inverting the output of a NOR gate (with another NOR gate connected as an inverter) results
in the OR function. The NAND gate, on the other hand, requires inversion of all inputs to
mimic the OR function, just as we needed to invert all inputs of a NOR gate to obtain the
AND function. Remember that inversion of all inputs to a gate results in changing that gate’s
essential function from AND to OR (or vice versa), plus an inverted output. Thus, with all
inputs inverted, a NAND behaves as an OR, a NOR behaves as an AND, an AND behaves as
a NOR, and an OR behaves as a NAND. In Boolean algebra, this transformation is referred to
as DeMorgan’s Theorem, covered in more detail in a later chapter of this book.
3.9. GATE UNIVERSALITY 89
InputA
InputB
Output
A B Output
00
0 1
01
1 1
0
1
2input OR gate
1
1
Output
+V
+V
InputA
InputB
InputA
InputB Output
3.9.6 Constructing the NOR function
Much the same as the procedure for making a NOR gate behave as a NAND, we must invert
all inputs and the output to make a NAND gate function as a NOR.
90 CHAPTER 3. LOGIC GATES
InputA
InputB
Output
A B Output
00
0 1
01
1 1
1
Output
+V
+V
InputA
InputB
+V
2input NOR gate
0
0
0
• REVIEW:
• NAND and NOR gates are universal: that is, they have the ability to mimic any type of
gate, if interconnected in sufficient numbers.
3.10 Logic signal voltage levels
Logic gate circuits are designed to input and output only two types of signals: ”high” (1) and
”low” (0), as represented by a variable voltage: full power supply voltage for a ”high” state and
zero voltage for a ”low” state. In a perfect world, all logic circuit signals would exist at these
extreme voltage limits, and never deviate from them (i.e., less than full voltage for a ”high,”
or more than zero voltage for a ”low”). However, in reality, logic signal voltage levels rarely
attain these perfect limits due to stray voltage drops in the transistor circuitry, and so we must
understand the signal level limitations of gate circuits as they try to interpret signal voltages
lying somewhere between full supply voltage and zero.
TTL gates operate on a nominal power supply voltage of 5 volts, +/ 0.25 volts. Ideally, a TTL
”high” signal would be 5.00 volts exactly, and a TTL ”low” signal 0.00 volts exactly. However,
real TTL gate circuits cannot output such perfect voltage levels, and are designed to accept
”high” and ”low” signals deviating substantially from these ideal values. ”Acceptable” input
signal voltages range from 0 volts to 0.8 volts for a ”low” logic state, and 2 volts to 5 volts for
a ”high” logic state. ”Acceptable” output signal voltages (voltage levels guaranteed by the gate
manufacturer over a specified range of load conditions) range from 0 volts to 0.5 volts for a
”low” logic state, and 2.7 volts to 5 volts for a ”high” logic state:
3.10. LOGIC SIGNAL VOLTAGE LEVELS 91
Low
High
5 V
2 V
0.8 V
0 V
Acceptable TTL gate
input signal levels
5 V
0 V
High
Low
Acceptable TTL gate
output signal levels
Vcc = 5 V
2.7 V
0.5 V
If a voltage signal ranging between 0.8 volts and 2 volts were to be sent into the input
of a TTL gate, there would be no certain response from the gate. Such a signal would be
considered uncertain, and no logic gate manufacturer would guarantee how their gate circuit
would interpret such a signal.
As you can see, the tolerable ranges for output signal levels are narrower than for input
signal levels, to ensure that any TTL gate outputting a digital signal into the input of another
TTL gate will transmit voltages acceptable to the receiving gate. The difference between the
tolerable output and input ranges is called the noise margin of the gate. For TTL gates, the
lowlevel noise margin is the difference between 0.8 volts and 0.5 volts (0.3 volts), while the
highlevel noise margin is the difference between 2.7 volts and 2 volts (0.7 volts). Simply put,
the noise margin is the peak amount of spurious or ”noise” voltage that may be superimposed
on a weak gate output voltage signal before the receiving gate might interpret it wrongly:
Low
High
5 V
2 V
0.8 V
0 V
Acceptable TTL gate
input signal levels
5 V
0 V
High
Low
Acceptable TTL gate
output signal levels
2.7 V
0.5 V
highlevel noise margin
lowlevel noise margin
CMOS gate circuits have input and output signal specifications that are quite different
from TTL. For a CMOS gate operating at a power supply voltage of 5 volts, the acceptable
input signal voltages range from 0 volts to 1.5 volts for a ”low” logic state, and 3.5 volts to 5
volts for a ”high” logic state. ”Acceptable” output signal voltages (voltage levels guaranteed by
the gate manufacturer over a specified range of load conditions) range from 0 volts to 0.05 volts
for a ”low” logic state, and 4.95 volts to 5 volts for a ”high” logic state:
92 CHAPTER 3. LOGIC GATES
Low
High
5 V
0 V
input signal levels
5 V
0 V
High
Low
output signal levels
Acceptable CMOS gate Acceptable CMOS gate
1.5 V
3.5 V
0.05 V
4.95 V
Vdd = 5 V
It should be obvious from these figures that CMOS gate circuits have far greater noise
margins than TTL: 1.45 volts for CMOS lowlevel and highlevel margins, versus a maximum
of 0.7 volts for TTL. In other words, CMOS circuits can tolerate over twice the amount of
superimposed ”noise” voltage on their input lines before signal interpretation errors will result.
CMOS noise margins widen even further with higher operating voltages. Unlike TTL,
which is restricted to a power supply voltage of 5 volts, CMOS may be powered by voltages
as high as 15 volts (some CMOS circuits as high as 18 volts). Shown here are the acceptable
”high” and ”low” states, for both input and output, of CMOS integrated circuits operating at 10
volts and 15 volts, respectively:
Low
High
0 V
input signal levels
0 V
High
Low
output signal levels
Acceptable CMOS gate Acceptable CMOS gate
0.05 V
Vdd = 10 V
10 V 10 V
9.95 V
3 V
7 V
3.10. LOGIC SIGNAL VOLTAGE LEVELS 93
Low
High
0 V
input signal levels
0 V
High
Low
output signal levels
Acceptable CMOS gate Acceptable CMOS gate
0.05 V
Vdd = 15 V
15 V
14.95 V15 V
11 V
4 V
The margins for acceptable ”high” and ”low” signals may be greater than what is shown in
the previous illustrations. What is shown represents ”worstcase” input signal performance,
based on manufacturer’s specifications. In practice, it may be found that a gate circuit will
tolerate ”high” signals of considerably less voltage and ”low” signals of considerably greater
voltage than those specified here.
Conversely, the extremely small output margins shown – guaranteeing output states for
”high” and ”low” signals to within 0.05 volts of the power supply ”rails” – are optimistic. Such
”solid” output voltage levels will be true only for conditions of minimum loading. If the gate is
sourcing or sinking substantial current to a load, the output voltage will not be able to maintain
these optimum levels, due to internal channel resistance of the gate’s final output MOSFETs.
Within the ”uncertain” range for any gate input, there will be some point of demarcation
dividing the gate’s actual ”low” input signal range from its actual ”high” input signal range.
That is, somewhere between the lowest ”high” signal voltage level and the highest ”low” signal
voltage level guaranteed by the gate manufacturer, there is a threshold voltage at which the
gate will actually switch its interpretation of a signal from ”low” or ”high” or vice versa. For
most gate circuits, this unspecified voltage is a single point:
94 CHAPTER 3. LOGIC GATES
0 V
5 V Vdd = 5 V
Vin
Vout
Typical response of a logic gate
to a variable (analog) input voltage
Time
threshold
In the presence of AC ”noise” voltage superimposed on the DC input signal, a single thresh
old point at which the gate alters its interpretation of logic level will result in an erratic output:
0 V
5 V Vdd = 5 V
Vin
Vout
Time
threshold
Slowlychanging DC signal with
AC noise superimposed
If this scenario looks familiar to you, its because you remember a similar problem with
(analog) voltage comparator opamp circuits. With a single threshold point at which an input
causes the output to switch between ”high” and ”low” states, the presence of significant noise
will cause erratic changes in the output:
3.10. LOGIC SIGNAL VOLTAGE LEVELS 95
−
+
+V
V
VoutVin
AC input
voltage
Square wave
output voltage
DC reference
voltage
Vref
The solution to this problem is a bit of positive feedback introduced into the amplifier circuit.
With an opamp, this is done by connecting the output back around to the noninverting (+)
input through a resistor. In a gate circuit, this entails redesigning the internal gate circuitry,
establishing the feedback inside the gate package rather than through external connections. A
gate so designed is called a Schmitt trigger. Schmitt triggers interpret varying input voltages
according to two threshold voltages: a positivegoing threshold (VT+), and a negativegoing
threshold (VT−):
0 V
5 V Vdd = 5 V
Vin
Vout
Time
VT+
VT
Schmitt trigger response to a
"noisy" input signal
Schmitt trigger gates are distinguished in schematic diagrams by the small ”hysteresis”
symbol drawn within them, reminiscent of the BH curve for a ferromagnetic material. Hys
teresis engendered by positive feedback within the gate circuitry adds an additional level of
noise immunity to the gate’s performance. Schmitt trigger gates are frequently used in ap
plications where noise is expected on the input signal line(s), and/or where an erratic output
would be very detrimental to system performance.
96 CHAPTER 3. LOGIC GATES
The differing voltage level requirements of TTL and CMOS technology present problems
when the two types of gates are used in the same system. Although operating CMOS gates on
the same 5.00 volt power supply voltage required by the TTL gates is no problem, TTL output
voltage levels will not be compatible with CMOS input voltage requirements.
Take for instance a TTL NAND gate outputting a signal into the input of a CMOS inverter
gate. Both gates are powered by the same 5.00 volt supply (Vcc). If the TTL gate outputs a
”low” signal (guaranteed to be between 0 volts and 0.5 volts), it will be properly interpreted by
the CMOS gate’s input as a ”low” (expecting a voltage between 0 volts and 1.5 volts):
TTL CMOS
Vcc Vdd
5 V
+

. . .
. . .
0 V
0.5 V
5 V
TTL
output
CMOS
input
5 V
0 V
1.5 V
TTL output falls within
acceptable limits for
CMOS input
"low"
However, if the TTL gate outputs a ”high” signal (guaranteed to be between 5 volts and 2.7
volts), it might not be properly interpreted by the CMOS gate’s input as a ”high” (expecting a
voltage between 5 volts and 3.5 volts):
3.10. LOGIC SIGNAL VOLTAGE LEVELS 97
TTL CMOS
Vcc Vdd
5 V
+

. . .
. . .
0 V
5 V
TTL
output
CMOS
input
5 V
0 V
acceptable limits for
CMOS input
"high"
2.7 V
3.5 V
TTL output falls outside of
Given this mismatch, it is entirely possible for the TTL gate to output a valid ”high” signal
(valid, that is, according to the standards for TTL) that lies within the ”uncertain” range for
the CMOS input, and may be (falsely) interpreted as a ”low” by the receiving gate. An easy
”fix” for this problem is to augment the TTL gate’s ”high” signal voltage level by means of a
pullup resistor:
TTL CMOS
Vcc Vdd
5 V
+

. . .
. . .
0 V
5 V
TTL
output CMOS
input
5 V
0 V
3.5 V
Rpullup
TTL "high" output voltage
assisted by Rpullup
Something more than this, though, is required to interface a TTL output with a CMOS
98 CHAPTER 3. LOGIC GATES
input, if the receiving CMOS gate is powered by a greater power supply voltage:
TTL CMOS
Vcc Vdd
5 V
+

. . .
. . .
+

10 V
5 V
0 V
10 V
0 V
TTL
output
CMOS
input
3 V
7 V
2.7 V
0.5 V
The TTL "high" signal will
definitely not fall within the
CMOS gate’s acceptable limits
There will be no problem with the CMOS gate interpreting the TTL gate’s ”low” output,
of course, but a ”high” signal from the TTL gate is another matter entirely. The guaranteed
output voltage range of 2.7 volts to 5 volts from the TTL gate output is nowhere near the CMOS
gate’s acceptable range of 7 volts to 10 volts for a ”high” signal. If we use an opencollector TTL
gate instead of a totempole output gate, though, a pullup resistor to the 10 volt Vdd supply rail
will raise the TTL gate’s ”high” output voltage to the full power supply voltage supplying the
CMOS gate. Since an opencollector gate can only sink current, not source current, the ”high”
state voltage level is entirely determined by the power supply to which the pullup resistor is
attached, thus neatly solving the mismatch problem:
3.10. LOGIC SIGNAL VOLTAGE LEVELS 99
TTL CMOS
Vcc Vdd
5 V
+

. . .
. . .
+

10 V
0 V
10 V
0 V
TTL
output
CMOS
input
3 V
7 V
0.5 V
Rpullup
10 V
Now, both "low" and "high"
TTL signals are acceptable
to the CMOS gate input
(opencollector)
Due to the excellent output voltage characteristics of CMOS gates, there is typically no
problem connecting a CMOS output to a TTL input. The only significant issue is the current
loading presented by the TTL inputs, since the CMOS output must sink current for each of the
TTL inputs while in the ”low” state.
When the CMOS gate in question is powered by a voltage source in excess of 5 volts (Vcc),
though, a problem will result. The ”high” output state of the CMOS gate, being greater than 5
volts, will exceed the TTL gate’s acceptable input limits for a ”high” signal. A solution to this
problem is to create an ”opencollector” inverter circuit using a discrete NPN transistor, and
use it to interface the two gates together:
TTLCMOS
VccVdd
5 V
+

. . .
. . .
+

10 V
Rpullup
The ”Rpullup” resistor is optional, since TTL inputs automatically assume a ”high” state
when left floating, which is what will happen when the CMOS gate output is ”low” and the
transistor cuts off. Of course, one very important consequence of implementing this solution is
the logical inversion created by the transistor: when the CMOS gate outputs a ”low” signal, the
TTL gate sees a ”high” input; and when the CMOS gate outputs a ”high” signal, the transistor
saturates and the TTL gate sees a ”low” input. So long as this inversion is accounted for in the
logical scheme of the system, all will be well.
100 CHAPTER 3. LOGIC GATES
3.11 DIP gate packaging
Digital logic gate circuits are manufactured as integrated circuits: all the constituent transis
tors and resistors built on a single piece of semiconductor material. The engineer, technician,
or hobbyist using small numbers of gates will likely find what he or she needs enclosed in a
DIP (Dual Inline Package) housing. DIPenclosed integrated circuits are available with even
numbers of pins, located at 0.100 inch intervals from each other for standard circuit board
layout compatibility. Pin counts of 8, 14, 16, 18, and 24 are common for DIP ”chips.”
Part numbers given to these DIP packages specify what type of gates are enclosed, and how
many. These part numbers are industry standards, meaning that a ”74LS02” manufactured by
Motorola will be identical in function to a ”74LS02” manufactured by Fairchild or by any other
manufacturer. Letter codes prepended to the part number are unique to the manufacturer,
and are not industrystandard codes. For instance, a SN74LS02 is a quad 2input TTL NOR
gate manufactured by Motorola, while a DM74LS02 is the exact same circuit manufactured by
Fairchild.
Logic circuit part numbers beginning with ”74” are commercialgrade TTL. If the part num
ber begins with the number ”54”, the chip is a militarygrade unit: having a greater operat
ing temperature range, and typically more robust in regard to allowable power supply and
signal voltage levels. The letters ”LS” immediately following the 74/54 prefix indicate ”Low
power Schottky” circuitry, using Schottkybarrier diodes and transistors throughout, to de
crease power dissipation. NonSchottky gate circuits consume more power, but are able to
operate at higher frequencies due to their faster switching times.
A few of the more common TTL ”DIP” circuit packages are shown here for reference:
3.11. DIP GATE PACKAGING 101
14 13 12 11 10 9 8
7654321
Vcc
GND
Quad NAND gate
5400/7400
Vcc
GND
14 13 12 11 10 9 8
7654321
5402/7402
Quad NOR gate
14 13 12 11 10 9 8
7654321
Vcc
GND
5408/7408
Quad AND gate
14 13 12 11 10 9 8
7654321
Vcc
GND
5432/7432
Quad OR gate
14 13 12 11 10 9 8
7654321
Vcc
GND
5486/7486
Quad XOR gate
14 13 12 11 10 9 8
7654321
Vcc
GND
5404/7404
Hex inverter
102 CHAPTER 3. LOGIC GATES
14 13 12 11 10 9 8
7654321
GND
Quad NAND gate
GND
14 13 12 11 10 9 8
7654321
Quad NOR gate
14 13 12 11 10 9 8
7654321
GND
Quad AND gate
14 13 12 11 10 9 8
7654321
GND
Quad OR gate
14 13 12 11 10 9 8
7654321
GND
Quad XOR gate
14 13 12 11 10 9 8
7654321
GND
Hex inverter
4011
Vdd Vdd
Vdd Vdd
Vdd Vdd
4001
4081 4071
4070 4069
3.12 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
JanWillem Rensman (May 2, 2002): Suggested the inclusion of Schmitt triggers and gate
hysteresis to this chapter.
Chapter 4
SWITCHES
Contents
4.1 Switch types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
4.2 Switch contact design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
4.3 Contact ”normal” state and make/break sequence . . . . . . . . . . . . . . 111
4.4 Contact ”bounce” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
4.1 Switch types
An electrical switch is any device used to interrupt the flow of electrons in a circuit. Switches
are essentially binary devices: they are either completely on (”closed”) or completely off (”open”).
There are many different types of switches, and we will explore some of these types in this
chapter.
Though it may seem strange to cover this elementary electrical topic at such a late stage
in this book series, I do so because the chapters that follow explore an older realm of digital
technology based on mechanical switch contacts rather than solidstate gate circuits, and a
thorough understanding of switch types is necessary for the undertaking. Learning the func
tion of switchbased circuits at the same time that you learn about solidstate logic gates makes
both topics easier to grasp, and sets the stage for an enhanced learning experience in Boolean
algebra, the mathematics behind digital logic circuits.
The simplest type of switch is one where two electrical conductors are brought in contact
with each other by the motion of an actuating mechanism. Other switches are more complex,
containing electronic circuits able to turn on or off depending on some physical stimulus (such
as light or magnetic field) sensed. In any case, the final output of any switch will be (at least) a
pair of wireconnection terminals that will either be connected together by the switch’s internal
contact mechanism (”closed”), or not connected together (”open”).
Any switch designed to be operated by a person is generally called a hand switch, and they
are manufactured in several varieties:
103
104 CHAPTER 4. SWITCHES
Toggle switch
Toggle switches are actuated by a lever angled in one of two or more positions. The common
light switch used in household wiring is an example of a toggle switch. Most toggle switches
will come to rest in any of their lever positions, while others have an internal spring mecha
nism returning the lever to a certain normal position, allowing for what is called ”momentary”
operation.
Pushbutton switch
Pushbutton switches are twoposition devices actuated with a button that is pressed and
released. Most pushbutton switches have an internal spring mechanism returning the button
to its ”out,” or ”unpressed,” position, for momentary operation. Some pushbutton switches
will latch alternately on or off with every push of the button. Other pushbutton switches will
stay in their ”in,” or ”pressed,” position until the button is pulled back out. This last type of
pushbutton switches usually have a mushroomshaped button for easy pushpull action.
Selector switch
Selector switches are actuated with a rotary knob or lever of some sort to select one of two
or more positions. Like the toggle switch, selector switches can either rest in any of their
positions or contain springreturn mechanisms for momentary operation.
Joystick switch
A joystick switch is actuated by a lever free to move in more than one axis of motion. One or
more of several switch contact mechanisms are actuated depending on which way the lever is
pushed, and sometimes by how far it is pushed. The circleanddot notation on the switch sym
bol represents the direction of joystick lever motion required to actuate the contact. Joystick
hand switches are commonly used for crane and robot control.
Some switches are specifically designed to be operated by the motion of a machine rather
than by the hand of a human operator. These motionoperated switches are commonly called
limit switches, because they are often used to limit the motion of a machine by turning off the
actuating power to a component if it moves too far. As with hand switches, limit switches come
in several varieties:
4.1. SWITCH TYPES 105
Lever actuator limit switch
These limit switches closely resemble rugged toggle or selector hand switches fitted with
a lever pushed by the machine part. Often, the levers are tipped with a small roller bearing,
preventing the lever from being worn off by repeated contact with the machine part.
prox
Proximity switch
Proximity switches sense the approach of a metallic machine part either by a magnetic or
highfrequency electromagnetic field. Simple proximity switches use a permanent magnet to
actuate a sealed switch mechanism whenever the machine part gets close (typically 1 inch or
less). More complex proximity switches work like a metal detector, energizing a coil of wire
with a highfrequency current, and electronically monitoring the magnitude of that current.
If a metallic part (not necessarily magnetic) gets close enough to the coil, the current will
increase, and trip the monitoring circuit. The symbol shown here for the proximity switch is
of the electronic variety, as indicated by the diamondshaped box surrounding the switch. A
nonelectronic proximity switch would use the same symbol as the leveractuated limit switch.
Another form of proximity switch is the optical switch, comprised of a light source and
photocell. Machine position is detected by either the interruption or reflection of a light beam.
Optical switches are also useful in safety applications, where beams of light can be used to
detect personnel entry into a dangerous area.
In many industrial processes, it is necessary to monitor various physical quantities with
switches. Such switches can be used to sound alarms, indicating that a process variable has
exceeded normal parameters, or they can be used to shut down processes or equipment if those
variables have reached dangerous or destructive levels. There are many different types of
process switches:
Speed switch
These switches sense the rotary speed of a shaft either by a centrifugal weight mechanism
mounted on the shaft, or by some kind of noncontact detection of shaft motion such as optical
or magnetic.
Pressure switch
106 CHAPTER 4. SWITCHES
Gas or liquid pressure can be used to actuate a switch mechanism if that pressure is applied
to a piston, diaphragm, or bellows, which converts pressure to mechanical force.
Temperature switch
An inexpensive temperaturesensing mechanism is the ”bimetallic strip:” a thin strip of two
metals, joined backtoback, each metal having a different rate of thermal expansion. When
the strip heats or cools, differing rates of thermal expansion between the two metals causes
it to bend. The bending of the strip can then be used to actuate a switch contact mechanism.
Other temperature switches use a brass bulb filled with either a liquid or gas, with a tiny
tube connecting the bulb to a pressuresensing switch. As the bulb is heated, the gas or liquid
expands, generating a pressure increase which then actuates the switch mechanism.
Liquid level switch
A floating object can be used to actuate a switch mechanism when the liquid level in an tank
rises past a certain point. If the liquid is electrically conductive, the liquid itself can be used as
a conductor to bridge between two metal probes inserted into the tank at the required depth.
The conductivity technique is usually implemented with a special design of relay triggered by
a small amount of current through the conductive liquid. In most cases it is impractical and
dangerous to switch the full load current of the circuit through a liquid.
Level switches can also be designed to detect the level of solid materials such as wood chips,
grain, coal, or animal feed in a storage silo, bin, or hopper. A common design for this application
is a small paddle wheel, inserted into the bin at the desired height, which is slowly turned by a
small electric motor. When the solid material fills the bin to that height, the material prevents
the paddle wheel from turning. The torque response of the small motor than trips the switch
mechanism. Another design uses a ”tuning fork” shaped metal prong, inserted into the bin
from the outside at the desired height. The fork is vibrated at its resonant frequency by an
electronic circuit and magnet/electromagnet coil assembly. When the bin fills to that height,
the solid material dampens the vibration of the fork, the change in vibration amplitude and/or
frequency detected by the electronic circuit.
Liquid flow switch
Inserted into a pipe, a flow switch will detect any gas or liquid flow rate in excess of a
certain threshold, usually with a small paddle or vane which is pushed by the flow. Other flow
switches are constructed as differential pressure switches, measuring the pressure drop across
a restriction built into the pipe.
4.1. SWITCH TYPES 107
Another type of level switch, suitable for liquid or solid material detection, is the nu
clear switch. Composed of a radioactive source material and a radiation detector, the two are
mounted across the diameter of a storage vessel for either solid or liquid material. Any height
of material beyond the level of the source/detector arrangement will attenuate the strength
of radiation reaching the detector. This decrease in radiation at the detector can be used to
trigger a relay mechanism to provide a switch contact for measurement, alarm point, or even
control of the vessel level.
source
source
detector
detector
Nuclear level switch
(for solid or liquid material)
Both source and detector are outside of the vessel, with no intrusion at all except the radi
ation flux itself. The radioactive sources used are fairly weak and pose no immediate health
threat to operations or maintenance personnel.
As usual, there is usually more than one way to implement a switch to monitor a physical
process or serve as an operator control. There is usually no single ”perfect” switch for any
application, although some obviously exhibit certain advantages over others. Switches must
be intelligently matched to the task for efficient and reliable operation.
• REVIEW:
• A switch is an electrical device, usually electromechanical, used to control continuity
between two points.
• Hand switches are actuated by human touch.
• Limit switches are actuated by machine motion.
• Process switches are actuated by changes in some physical process (temperature, level,
flow, etc.).
108 CHAPTER 4. SWITCHES
4.2 Switch contact design
A switch can be constructed with any mechanism bringing two conductors into contact with
each other in a controlled manner. This can be as simple as allowing two copper wires to
touch each other by the motion of a lever, or by directly pushing two metal strips into contact.
However, a good switch design must be rugged and reliable, and avoid presenting the operator
with the possibility of electric shock. Therefore, industrial switch designs are rarely this crude.
The conductive parts in a switch used to make and break the electrical connection are
called contacts. Contacts are typically made of silver or silvercadmium alloy, whose conductive
properties are not significantly compromised by surface corrosion or oxidation. Gold contacts
exhibit the best corrosion resistance, but are limited in currentcarrying capacity and may
”cold weld” if brought together with high mechanical force. Whatever the choice of metal, the
switch contacts are guided by a mechanism ensuring square and even contact, for maximum
reliability and minimum resistance.
Contacts such as these can be constructed to handle extremely large amounts of electric cur
rent, up to thousands of amps in some cases. The limiting factors for switch contact ampacity
are as follows:
• Heat generated by current through metal contacts (while closed).
• Sparking caused when contacts are opened or closed.
• The voltage across open switch contacts (potential of current jumping across the gap).
One major disadvantage of standard switch contacts is the exposure of the contacts to the
surrounding atmosphere. In a nice, clean, controlroom environment, this is generally not a
problem. However, most industrial environments are not this benign. The presence of corrosive
chemicals in the air can cause contacts to deteriorate and fail prematurely. Even more trouble
some is the possibility of regular contact sparking causing flammable or explosive chemicals to
ignite.
When such environmental concerns exist, other types of contacts can be considered for
small switches. These other types of contacts are sealed from contact with the outside air, and
therefore do not suffer the same exposure problems that standard contacts do.
A common type of sealedcontact switch is the mercury switch. Mercury is a metallic ele
ment, liquid at room temperature. Being a metal, it possesses excellent conductive properties.
Being a liquid, it can be brought into contact with metal probes (to close a circuit) inside of
a sealed chamber simply by tilting the chamber so that the probes are on the bottom. Many
industrial switches use small glass tubes containing mercury which are tilted one way to close
the contact, and tilted another way to open. Aside from the problems of tube breakage and
spilling mercury (which is a toxic material), and susceptibility to vibration, these devices are
an excellent alternative to openair switch contacts wherever environmental exposure prob
lems are a concern.
Here, a mercury switch (often called a tilt switch) is shown in the open position, where the
mercury is out of contact with the two metal contacts at the other end of the glass bulb:
4.2. SWITCH CONTACT DESIGN 109
Here, the same switch is shown in the closed position. Gravity now holds the liquid mercury
in contact with the two metal contacts, providing electrical continuity from one to the other:
Mercury switch contacts are impractical to build in large sizes, and so you will typically
find such contacts rated at no more than a few amps, and no more than 120 volts. There are
exceptions, of course, but these are common limits.
Another sealedcontact type of switch is the magnetic reed switch. Like the mercury switch,
a reed switch’s contacts are located inside a sealed tube. Unlike the mercury switch which uses
liquid metal as the contact medium, the reed switch is simply a pair of very thin, magnetic,
metal strips (hence the name ”reed”) which are brought into contact with each other by apply
110 CHAPTER 4. SWITCHES
ing a strong magnetic field outside the sealed tube. The source of the magnetic field in this
type of switch is usually a permanent magnet, moved closer to or further away from the tube
by the actuating mechanism. Due to the small size of the reeds, this type of contact is typically
rated at lower currents and voltages than the average mercury switch. However, reed switches
typically handle vibration better than mercury contacts, because there is no liquid inside the
tube to splash around.
It is common to find generalpurpose switch contact voltage and current ratings to be
greater on any given switch or relay if the electric power being switched is AC instead of DC.
The reason for this is the selfextinguishing tendency of an alternatingcurrent arc across an
air gap. Because 60 Hz power line current actually stops and reverses direction 120 times per
second, there are many opportunities for the ionized air of an arc to lose enough temperature
to stop conducting current, to the point where the arc will not restart on the next voltage peak.
DC, on the other hand, is a continuous, uninterrupted flow of electrons which tends to main
tain an arc across an air gap much better. Therefore, switch contacts of any kind incur more
wear when switching a given value of direct current than for the same value of alternating cur
rent. The problem of switching DC is exaggerated when the load has a significant amount of
inductance, as there will be very high voltages generated across the switch’s contacts when the
circuit is opened (the inductor doing its best to maintain circuit current at the same magnitude
as when the switch was closed).
With both AC and DC, contact arcing can be minimized with the addition of a ”snubber”
circuit (a capacitor and resistor wired in series) in parallel with the contact, like this:
R C
"Snubber"
A sudden rise in voltage across the switch contact caused by the contact opening will be
tempered by the capacitor’s charging action (the capacitor opposing the increase in voltage by
drawing current). The resistor limits the amount of current that the capacitor will discharge
through the contact when it closes again. If the resistor were not there, the capacitor might
actually make the arcing during contact closure worse than the arcing during contact open
ing without a capacitor! While this addition to the circuit helps mitigate contact arcing, it is
not without disadvantage: a prime consideration is the possibility of a failed (shorted) capaci
tor/resistor combination providing a path for electrons to flow through the circuit at all times,
even when the contact is open and current is not desired. The risk of this failure, and the
severity of the resulting consequences must be considered against the increased contact wear
(and inevitable contact failure) without the snubber circuit.
The use of snubbers in DC switch circuits is nothing new: automobile manufacturers have
been doing this for years on engine ignition systems, minimizing the arcing across the switch
contact ”points” in the distributor with a small capacitor called a condenser. As any mechanic
can tell you, the service life of the distributor’s ”points” is directly related to how well the
condenser is functioning.
With all this discussion concerning the reduction of switch contact arcing, one might be
4.3. CONTACT ”NORMAL” STATE AND MAKE/BREAK SEQUENCE 111
led to think that less current is always better for a mechanical switch. This, however, is not
necessarily so. It has been found that a small amount of periodic arcing can actually be good
for the switch contacts, because it keeps the contact faces free from small amounts of dirt and
corrosion. If a mechanical switch contact is operated with too little current, the contacts will
tend to accumulate excessive resistance and may fail prematurely! This minimum amount
of electric current necessary to keep a mechanical switch contact in good health is called the
wetting current.
Normally, a switch’s wetting current rating is far below its maximum current rating, and
well below its normal operating current load in a properly designed system. However, there are
applications where a mechanical switch contact may be required to routinely handle currents
below normal wetting current limits (for instance, if a mechanical selector switch needs to open
or close a digital logic or analog electronic circuit where the current value is extremely small).
In these applications, is it highly recommended that goldplated switch contacts be specified.
Gold is a ”noble” metal and does not corrode as other metals will. Such contacts have extremely
low wetting current requirements as a result. Normal silver or copper alloy contacts will not
provide reliable operation if used in such lowcurrent service!
• REVIEW:
• The parts of a switch responsible for making and breaking electrical continuity are called
the ”contacts.” Usually made of corrosionresistant metal alloy, contacts are made to touch
each other by a mechanism which helps maintain proper alignment and spacing.
• Mercury switches use a slug of liquid mercury metal as a moving contact. Sealed in a
glass tube, the mercury contact’s spark is sealed from the outside environment, making
this type of switch ideally suited for atmospheres potentially harboring explosive vapors.
• Reed switches are another type of sealedcontact device, contact being made by two thin
metal ”reeds” inside a glass tube, brought together by the influence of an external mag
netic field.
• Switch contacts suffer greater duress switching DC than AC. This is primarily due to the
selfextinguishing nature of an AC arc.
• A resistorcapacitor network called a ”snubber” can be connected in parallel with a switch
contact to reduce contact arcing.
• Wetting current is the minimum amount of electric current necessary for a switch contact
to carry in order for it to be selfcleaning. Normally this value is far below the switch’s
maximum current rating.
4.3 Contact ”normal” state and make/break sequence
Any kind of switch contact can be designed so that the contacts ”close” (establish continu
ity) when actuated, or ”open” (interrupt continuity) when actuated. For switches that have a
springreturn mechanism in them, the direction that the spring returns it to with no applied
force is called the normal position. Therefore, contacts that are open in this position are called
normally open and contacts that are closed in this position are called normally closed.
112 CHAPTER 4. SWITCHES
For process switches, the normal position, or state, is that which the switch is in when there
is no process influence on it. An easy way to figure out the normal condition of a process switch
is to consider the state of the switch as it sits on a storage shelf, uninstalled. Here are some
examples of ”normal” process switch conditions:
• Speed switch: Shaft not turning
• Pressure switch: Zero applied pressure
• Temperature switch: Ambient (room) temperature
• Level switch: Empty tank or bin
• Flow switch: Zero liquid flow
It is important to differentiate between a switch’s ”normal” condition and its ”normal” use
in an operating process. Consider the example of a liquid flow switch that serves as a lowflow
alarm in a cooling water system. The normal, or properlyoperating, condition of the cooling
water system is to have fairly constant coolant flow going through this pipe. If we want the
flow switch’s contact to close in the event of a loss of coolant flow (to complete an electric
circuit which activates an alarm siren, for example), we would want to use a flow switch with
normallyclosed rather than normallyopen contacts. When there’s adequate flow through the
pipe, the switch’s contacts are forced open; when the flow rate drops to an abnormally low level,
the contacts return to their normal (closed) state. This is confusing if you think of ”normal” as
being the regular state of the process, so be sure to always think of a switch’s ”normal” state
as that which its in as it sits on a shelf.
The schematic symbology for switches vary according to the switch’s purpose and actuation.
A normallyopen switch contact is drawn in such a way as to signify an open connection, ready
to close when actuated. Conversely, a normallyclosed switch is drawn as a closed connection
which will be opened when actuated. Note the following symbols:
Pushbutton switch
Normallyopen Normallyclosed
There is also a generic symbology for any switch contact, using a pair of vertical lines
to represent the contact points in a switch. Normallyopen contacts are designated by the
lines not touching, while normallyclosed contacts are designated with a diagonal line bridging
between the two lines. Compare the two:
Normallyopen Normallyclosed
Generic switch contact designation
4.3. CONTACT ”NORMAL” STATE AND MAKE/BREAK SEQUENCE 113
The switch on the left will close when actuated, and will be open while in the ”normal”
(unactuated) position. The switch on the right will open when actuated, and is closed in the
”normal” (unactuated) position. If switches are designated with these generic symbols, the
type of switch usually will be noted in text immediately beside the symbol. Please note that
the symbol on the left is not to be confused with that of a capacitor. If a capacitor needs to be
represented in a control logic schematic, it will be shown like this:
Capacitor
In standard electronic symbology, the figure shown above is reserved for polaritysensitive
capacitors. In control logic symbology, this capacitor symbol is used for any type of capacitor,
even when the capacitor is not polarity sensitive, so as to clearly distinguish it from a normally
open switch contact.
With multipleposition selector switches, another design factor must be considered: that is,
the sequence of breaking old connections and making new connections as the switch is moved
from position to position, the moving contact touching several stationary contacts in sequence.
common
1
2
3
4
5
The selector switch shown above switches a common contact lever to one of five different
positions, to contact wires numbered 1 through 5. The most common configuration of a multi
position switch like this is one where the contact with one position is broken before the contact
with the next position is made. This configuration is called breakbeforemake. To give an
example, if the switch were set at position number 3 and slowly turned clockwise, the contact
lever would move off of the number 3 position, opening that circuit, move to a position between
number 3 and number 4 (both circuit paths open), and then touch position number 4, closing
that circuit.
There are applications where it is unacceptable to completely open the circuit attached to
the ”common” wire at any point in time. For such an application, a makebeforebreak switch
design can be built, in which the movable contact lever actually bridges between two positions
of contact (between number 3 and number 4, in the above scenario) as it travels between
positions. The compromise here is that the circuit must be able to tolerate switch closures
between adjacent position contacts (1 and 2, 2 and 3, 3 and 4, 4 and 5) as the selector knob is
turned from position to position. Such a switch is shown here:
114 CHAPTER 4. SWITCHES
common
1
2
3
4
5
When movable contact(s) can be brought into one of several positions with stationary con
tacts, those positions are sometimes called throws. The number of movable contacts is some
times called poles. Both selector switches shown above with one moving contact and five sta
tionary contacts would be designated as ”singlepole, fivethrow” switches.
If two identical singlepole, fivethrow switches were mechanically ganged together so that
they were actuated by the same mechanism, the whole assembly would be called a ”double
pole, fivethrow” switch:
Doublepole, 5throw switch
assembly
Here are a few common switch configurations and their abbreviated designations:
Singlepole, singlethrow
(SPST)
Doublepole, singlethrow
(DPST)
4.3. CONTACT ”NORMAL” STATE AND MAKE/BREAK SEQUENCE 115
Singlepole, doublethrow
(SPDT)
Doublepole, doublethrow
(DPDT)
Fourpole, doublethrow
(4PDT)
• REVIEW:
• The normal state of a switch is that where it is unactuated. For process switches, this is
the condition its in when sitting on a shelf, uninstalled.
• A switch that is open when unactuated is called normallyopen. A switch that is closed
when unactuated is called normallyclosed. Sometimes the terms ”normallyopen” and
”normallyclosed” are abbreviated N.O. and N.C., respectively.
• The generic symbology for N.O. and N.C. switch contacts is as follows:
Normallyopen Normallyclosed
Generic switch contact designation
•
• Multiposition switches can be either breakbeforemake (most common) or makebefore
break.
• The ”poles” of a switch refers to the number of moving contacts, while the ”throws” of a
switch refers to the number of stationary contacts per moving contact.
116 CHAPTER 4. SWITCHES
4.4 Contact ”bounce”
When a switch is actuated and contacts touch one another under the force of actuation, they
are supposed to establish continuity in a single, crisp moment. Unfortunately, though, switches
do not exactly achieve this goal. Due to the mass of the moving contact and any elasticity
inherent in the mechanism and/or contact materials, contacts will ”bounce” upon closure for a
period of milliseconds before coming to a full rest and providing unbroken contact. In many
applications, switch bounce is of no consequence: it matters little if a switch controlling an
incandescent lamp ”bounces” for a few cycles every time it is actuated. Since the lamp’s warm
up time greatly exceeds the bounce period, no irregularity in lamp operation will result.
However, if the switch is used to send a signal to an electronic amplifier or some other
circuit with a fast response time, contact bounce may produce very noticeable and undesired
effects:
Switch
actuated
A closer look at the oscilloscope display reveals a rather ugly set of makes and breaks when
the switch is actuated a single time:
Closeup view of oscilloscope display:
Switch is actuated
Contacts bouncing
If, for example, this switch is used to provide a ”clock” signal to a digital counter circuit, so
that each actuation of the pushbutton switch is supposed to increment the counter by a value
of 1, what will happen instead is the counter will increment by several counts each time the
switch is actuated. Since mechanical switches often interface with digital electronic circuits
in modern systems, switch contact bounce is a frequent design consideration. Somehow, the
4.4. CONTACT ”BOUNCE” 117
”chattering” produced by bouncing contacts must be eliminated so that the receiving circuit
sees a clean, crisp off/on transition:
Switch is actuated
"Bounceless" switch operation
Switch contacts may be debounced several different ways. The most direct means is to
address the problem at its source: the switch itself. Here are some suggestions for designing
switch mechanisms for minimum bounce:
• Reduce the kinetic energy of the moving contact. This will reduce the force of impact as
it comes to rest on the stationary contact, thus minimizing bounce.
• Use ”buffer springs” on the stationary contact(s) so that they are free to recoil and gently
absorb the force of impact from the moving contact.
• Design the switch for ”wiping” or ”sliding” contact rather than direct impact. ”Knife”
switch designs use sliding contacts.
• Dampen the switch mechanism’s movement using an air or oil ”shock absorber” mecha
nism.
• Use sets of contacts in parallel with each other, each slightly different in mass or contact
gap, so that when one is rebounding off the stationary contact, at least one of the others
will still be in firm contact.
• ”Wet” the contacts with liquid mercury in a sealed environment. After initial contact is
made, the surface tension of the mercury will maintain circuit continuity even though
the moving contact may bounce off the stationary contact several times.
Each one of these suggestions sacrifices some aspect of switch performance for limited
bounce, and so it is impractical to design all switches with limited contact bounce in mind.
Alterations made to reduce the kinetic energy of the contact may result in a small opencontact
gap or a slowmoving contact, which limits the amount of voltage the switch may handle and
the amount of current it may interrupt. Sliding contacts, while nonbouncing, still produce
”noise” (irregular current caused by irregular contact resistance when moving), and suffer
from more mechanical wear than normal contacts.
Multiple, parallel contacts give less bounce, but only at greater switch complexity and cost.
Using mercury to ”wet” the contacts is a very effective means of bounce mitigation, but it is
unfortunately limited to switch contacts of low ampacity. Also, mercurywetted contacts are
118 CHAPTER 4. SWITCHES
usually limited in mounting position, as gravity may cause the contacts to ”bridge” accidently
if oriented the wrong way.
If redesigning the switch mechanism is not an option, mechanical switch contacts may
be debounced externally, using other circuit components to condition the signal. A lowpass
filter circuit attached to the output of the switch, for example, will reduce the voltage/current
fluctuations generated by contact bounce:
Switch
actuated
C
Switch contacts may be debounced electronically, using hysteretic transistor circuits (cir
cuits that ”latch” in either a high or a low state) with builtin time delays (called ”oneshot”
circuits), or two inputs controlled by a doublethrow switch. These hysteretic circuits, called
multivibrators, are discussed in detail in a later chapter.
Chapter 5
ELECTROMECHANICAL
RELAYS
Contents
5.1 Relay construction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
5.2 Contactors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
5.3 Timedelay relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
5.4 Protective relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
5.5 Solidstate relays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
5.1 Relay construction
An electric current through a conductor will produce a magnetic field at right angles to the
direction of electron flow. If that conductor is wrapped into a coil shape, the magnetic field
produced will be oriented along the length of the coil. The greater the current, the greater the
strength of the magnetic field, all other factors being equal:
current current
magnetic field
Inductors react against changes in current because of the energy stored in this magnetic
field. When we construct a transformer from two inductor coils around a common iron core,
we use this field to transfer energy from one coil to the other. However, there are simpler and
more direct uses for electromagnetic fields than the applications we’ve seen with inductors and
transformers. The magnetic field produced by a coil of currentcarrying wire can be used to
119
120 CHAPTER 5. ELECTROMECHANICAL RELAYS
exert a mechanical force on any magnetic object, just as we can use a permanent magnet to
attract magnetic objects, except that this magnet (formed by the coil) can be turned on or off
by switching the current on or off through the coil.
If we place a magnetic object near such a coil for the purpose of making that object move
when we energize the coil with electric current, we have what is called a solenoid. The movable
magnetic object is called an armature, and most armatures can be moved with either direct
current (DC) or alternating current (AC) energizing the coil. The polarity of the magnetic
field is irrelevant for the purpose of attracting an iron armature. Solenoids can be used to
electrically open door latches, open or shut valves, move robotic limbs, and even actuate electric
switch mechanisms. However, if a solenoid is used to actuate a set of switch contacts, we have
a device so useful it deserves its own name: the relay.
Relays are extremely useful when we have a need to control a large amount of current
and/or voltage with a small electrical signal. The relay coil which produces the magnetic field
may only consume fractions of a watt of power, while the contacts closed or opened by that
magnetic field may be able to conduct hundreds of times that amount of power to a load. In
effect, a relay acts as a binary (on or off) amplifier.
Just as with transistors, the relay’s ability to control one electrical signal with another finds
application in the construction of logic functions. This topic will be covered in greater detail in
another lesson. For now, the relay’s ”amplifying” ability will be explored.
relay
VAC
Load
12
VDC480
In the above schematic, the relay’s coil is energized by the lowvoltage (12 VDC) source,
while the singlepole, singlethrow (SPST) contact interrupts the highvoltage (480 VAC) cir
cuit. It is quite likely that the current required to energize the relay coil will be hundreds of
times less than the current rating of the contact. Typical relay coil currents are well below 1
amp, while typical contact ratings for industrial relays are at least 10 amps.
One relay coil/armature assembly may be used to actuate more than one set of contacts.
Those contacts may be normallyopen, normallyclosed, or any combination of the two. As with
switches, the ”normal” state of a relay’s contacts is that state when the coil is deenergized,
just as you would find the relay sitting on a shelf, not connected to any circuit.
Relay contacts may be openair pads of metal alloy, mercury tubes, or even magnetic reeds,
just as with other types of switches. The choice of contacts in a relay depends on the same
factors which dictate contact choice in other types of switches. Openair contacts are the best
for highcurrent applications, but their tendency to corrode and spark may cause problems in
some industrial environments. Mercury and reed contacts are sparkless and won’t corrode, but
5.1. RELAY CONSTRUCTION 121
they tend to be limited in currentcarrying capacity.
Shown here are three small relays (about two inches in height, each), installed on a panel
as part of an electrical control system at a municipal water treatment plant:
The relay units shown here are called ”octalbase,” because they plug into matching sockets,
the electrical connections secured via eight metal pins on the relay bottom. The screw terminal
connections you see in the photograph where wires connect to the relays are actually part of
the socket assembly, into which each relay is plugged. This type of construction facilitates easy
removal and replacement of the relay(s) in the event of failure.
Aside from the ability to allow a relatively small electric signal to switch a relatively large
electric signal, relays also offer electrical isolation between coil and contact circuits. This
means that the coil circuit and contact circuit(s) are electrically insulated from one another.
One circuit may be DC and the other AC (such as in the example circuit shown earlier), and/or
they may be at completely different voltage levels, across the connections or from connections
to ground.
While relays are essentially binary devices, either being completely on or completely off,
there are operating conditions where their state may be indeterminate, just as with semi
conductor logic gates. In order for a relay to positively ”pull in” the armature to actuate the
contact(s), there must be a certain minimum amount of current through the coil. This mini
mum amount is called the pullin current, and it is analogous to the minimum input voltage
that a logic gate requires to guarantee a ”high” state (typically 2 Volts for TTL, 3.5 Volts for
CMOS). Once the armature is pulled closer to the coil’s center, however, it takes less magnetic
field flux (less coil current) to hold it there. Therefore, the coil current must drop below a value
significantly lower than the pullin current before the armature ”drops out” to its springloaded
position and the contacts resume their normal state. This current level is called the dropout
current, and it is analogous to the maximum input voltage that a logic gate input will allow to
guarantee a ”low” state (typically 0.8 Volts for TTL, 1.5 Volts for CMOS).
The hysteresis, or difference between pullin and dropout currents, results in operation
that is similar to a Schmitt trigger logic gate. Pullin and dropout currents (and voltages)
vary widely from relay to relay, and are specified by the manufacturer.
• REVIEW:
122 CHAPTER 5. ELECTROMECHANICAL RELAYS
• A solenoid is a device that produces mechanical motion from the energization of an elec
tromagnet coil. The movable portion of a solenoid is called an armature.
• A relay is a solenoid set up to actuate switch contacts when its coil is energized.
• Pullin current is the minimum amount of coil current needed to actuate a solenoid or
relay from its ”normal” (deenergized) position.
• Dropout current is the maximum coil current below which an energized relay will return
to its ”normal” state.
5.2 Contactors
When a relay is used to switch a large amount of electrical power through its contacts, it is
designated by a special name: contactor. Contactors typically have multiple contacts, and
those contacts are usually (but not always) normallyopen, so that power to the load is shut off
when the coil is deenergized. Perhaps the most common industrial use for contactors is the
control of electric motors.
motor
A
B
C
3phase
AC power
relay
120 VAC
coil
The top three contacts switch the respective phases of the incoming 3phase AC power, typ
ically at least 480 Volts for motors 1 horsepower or greater. The lowest contact is an ”auxiliary”
contact which has a current rating much lower than that of the large motor power contacts, but
is actuated by the same armature as the power contacts. The auxiliary contact is often used
in a relay logic circuit, or for some other part of the motor control scheme, typically switching
120 Volt AC power instead of the motor voltage. One contactor may have several auxiliary
contacts, either normallyopen or normallyclosed, if required.
The three ”opposedquestionmark” shaped devices in series with each phase going to the
motor are called overload heaters. Each ”heater” element is a lowresistance strip of metal
intended to heat up as the motor draws current. If the temperature of any of these heater ele
ments reaches a critical point (equivalent to a moderate overloading of the motor), a normally
closed switch contact (not shown in the diagram) will spring open. This normallyclosed contact
is usually connected in series with the relay coil, so that when it opens the relay will automat
ically deenergize, thereby shutting off power to the motor. We will see more of this overload
protection wiring in the next chapter. Overload heaters are intended to provide overcurrent
5.2. CONTACTORS 123
protection for large electric motors, unlike circuit breakers and fuses which serve the primary
purpose of providing overcurrent protection for power conductors.
Overload heater function is often misunderstood. They are not fuses; that is, it is not
their function to burn open and directly break the circuit as a fuse is designed to do. Rather,
overload heaters are designed to thermally mimic the heating characteristic of the particular
electric motor to be protected. All motors have thermal characteristics, including the amount
of heat energy generated by resistive dissipation (I2R), the thermal transfer characteristics of
heat ”conducted” to the cooling medium through the metal frame of the motor, the physical
mass and specific heat of the materials constituting the motor, etc. These characteristics are
mimicked by the overload heater on a miniature scale: when the motor heats up toward its
critical temperature, so will the heater toward its critical temperature, ideally at the same rate
and approach curve. Thus, the overload contact, in sensing heater temperature with a thermo
mechanical mechanism, will sense an analogue of the real motor. If the overload contact trips
due to excessive heater temperature, it will be an indication that the real motor has reached
its critical temperature (or, would have done so in a short while). After tripping, the heaters
are supposed to cool down at the same rate and approach curve as the real motor, so that they
indicate an accurate proportion of the motor’s thermal condition, and will not allow power to
be reapplied until the motor is truly ready for startup again.
Shown here is a contactor for a threephase electric motor, installed on a panel as part of
an electrical control system at a municipal water treatment plant:
124 CHAPTER 5. ELECTROMECHANICAL RELAYS
Threephase, 480 volt AC power comes in to the three normallyopen contacts at the top of
the contactor via screw terminals labeled ”L1,” ”L2,” and ”L3” (The ”L2” terminal is hidden be
hind a squareshaped ”snubber” circuit connected across the contactor’s coil terminals). Power
to the motor exits the overload heater assembly at the bottom of this device via screw terminals
labeled ”T1,” ”T2,” and ”T3.”
The overload heater units themselves are black, squareshaped blocks with the label ”W34,”
indicating a particular thermal response for a certain horsepower and temperature rating of
electric motor. If an electric motor of differing power and/or temperature ratings were to be
substituted for the one presently in service, the overload heater units would have to be replaced
with units having a thermal response suitable for the new motor. The motor manufacturer can
provide information on the appropriate heater units to use.
A white pushbutton located between the ”T1” and ”T2” line heaters serves as a way to
manually reset the normallyclosed switch contact back to its normal state after having been
5.2. CONTACTORS 125
tripped by excessive heater temperature. Wire connections to the ”overload” switch contact
may be seen at the lowerright of the photograph, near a label reading ”NC” (normallyclosed).
On this particular overload unit, a small ”window” with the label ”Tripped” indicates a tripped
condition by means of a colored flag. In this photograph, there is no ”tripped” condition, and
the indicator appears clear.
As a footnote, heater elements may be used as a crude current shunt resistor for determin
ing whether or not a motor is drawing current when the contactor is closed. There may be
times when you’re working on a motor control circuit, where the contactor is located far away
from the motor itself. How do you know if the motor is consuming power when the contac
tor coil is energized and the armature has been pulled in? If the motor’s windings are burnt
open, you could be sending voltage to the motor through the contactor contacts, but still have
zero current, and thus no motion from the motor shaft. If a clampon ammeter isn’t available
to measure line current, you can take your multimeter and measure millivoltage across each
heater element: if the current is zero, the voltage across the heater will be zero (unless the
heater element itself is open, in which case the voltage across it will be large); if there is cur
rent going to the motor through that phase of the contactor, you will read a definite millivoltage
across that heater:
motor
A
B
C
3phase
AC power
relay
120 VAC
coil
V Ω
COMA
mV
This is an especially useful trick to use for troubleshooting 3phase AC motors, to see if
one phase winding is burnt open or disconnected, which will result in a rapidly destructive
condition known as ”singlephasing.” If one of the lines carrying power to the motor is open,
it will not have any current through it (as indicated by a 0.00 mV reading across its heater),
although the other two lines will (as indicated by small amounts of voltage dropped across the
respective heaters).
• REVIEW:
• A contactor is a large relay, usually used to switch current to an electric motor or other
highpower load.
• Large electric motors can be protected from overcurrent damage through the use of over
load heaters and overload contacts. If the seriesconnected heaters get too hot from exces
126 CHAPTER 5. ELECTROMECHANICAL RELAYS
sive current, the normallyclosed overload contact will open, deenergizing the contactor
sending power to the motor.
5.3 Timedelay relays
Some relays are constructed with a kind of ”shock absorber” mechanism attached to the arma
ture which prevents immediate, full motion when the coil is either energized or deenergized.
This addition gives the relay the property of timedelay actuation. Timedelay relays can be
constructed to delay armature motion on coil energization, deenergization, or both.
Timedelay relay contacts must be specified not only as either normallyopen or normally
closed, but whether the delay operates in the direction of closing or in the direction of opening.
The following is a description of the four basic types of timedelay relay contacts.
First we have the normallyopen, timedclosed (NOTC) contact. This type of contact is nor
mally open when the coil is unpowered (deenergized). The contact is closed by the application
of power to the relay coil, but only after the coil has been continuously powered for the speci
fied amount of time. In other words, the direction of the contact’s motion (either to close or to
open) is identical to a regular NO contact, but there is a delay in closing direction. Because the
delay occurs in the direction of coil energization, this type of contact is alternatively known as
a normallyopen, ondelay:
5 sec.
Closes 5 seconds after coil energization
Opens immediately upon coil deenergization
Normallyopen, timedclosed
The following is a timing diagram of this relay contact’s operation:
NOTC
5 sec.
Coil
power
Contact
status
Time
5
seconds
on
off
closed
open
5.3. TIMEDELAY RELAYS 127
Next we have the normallyopen, timedopen (NOTO) contact. Like the NOTC contact,
this type of contact is normally open when the coil is unpowered (deenergized), and closed
by the application of power to the relay coil. However, unlike the NOTC contact, the timing
action occurs upon deenergization of the coil rather than upon energization. Because the delay
occurs in the direction of coil deenergization, this type of contact is alternatively known as a
normallyopen, off delay:
5 sec.
Normallyopen, timedopen
Closes immediately upon coil energization
Opens 5 seconds after coil deenergization
The following is a timing diagram of this relay contact’s operation:
5 sec.
Coil
power
Contact
status
Time
5
seconds
on
off
closed
open
NOTO
Next we have the normallyclosed, timedopen (NCTO) contact. This type of contact is
normally closed when the coil is unpowered (deenergized). The contact is opened with the
application of power to the relay coil, but only after the coil has been continuously powered
for the specified amount of time. In other words, the direction of the contact’s motion (either
to close or to open) is identical to a regular NC contact, but there is a delay in the opening
direction. Because the delay occurs in the direction of coil energization, this type of contact is
alternatively known as a normallyclosed, ondelay:
128 CHAPTER 5. ELECTROMECHANICAL RELAYS
5 sec.
Normallyclosed, timedopen
Opens 5 seconds after coil energization
Closes immediately upon coil deenergization
The following is a timing diagram of this relay contact’s operation:
5 sec.
Coil
power
Contact
status
Time
5
seconds
on
off
closed
open
NCTO
Finally we have the normallyclosed, timedclosed (NCTC) contact. Like the NCTO contact,
this type of contact is normally closed when the coil is unpowered (deenergized), and opened
by the application of power to the relay coil. However, unlike the NCTO contact, the timing
action occurs upon deenergization of the coil rather than upon energization. Because the delay
occurs in the direction of coil deenergization, this type of contact is alternatively known as a
normallyclosed, off delay:
5 sec.
Normallyclosed, timedclosed
Opens immediately upon coil energization
Closes 5 seconds after coil deenergization
The following is a timing diagram of this relay contact’s operation:
5.3. TIMEDELAY RELAYS 129
5 sec.
Coil
power
Contact
status
Time
5
seconds
on
off
closed
open
NCTC
Timedelay relays are very important for use in industrial control logic circuits. Some ex
amples of their use include:
• Flashing light control (time on, time off): two timedelay relays are used in conjunction
with one another to provide a constantfrequency on/off pulsing of contacts for sending
intermittent power to a lamp.
• Engine autostart control: Engines that are used to power emergency generators are often
equipped with ”autostart” controls that allow for automatic startup if the main electric
power fails. To properly start a large engine, certain auxiliary devices must be started
first and allowed some brief time to stabilize (fuel pumps, prelubrication oil pumps)
before the engine’s starter motor is energized. Timedelay relays help sequence these
events for proper startup of the engine.
• Furnace safety purge control: Before a combustiontype furnace can be safely lit, the
air fan must be run for a specified amount of time to ”purge” the furnace chamber of
any potentially flammable or explosive vapors. A timedelay relay provides the furnace
control logic with this necessary time element.
• Motor softstart delay control: Instead of starting large electric motors by switching full
power from a dead stop condition, reduced voltage can be switched for a ”softer” start and
less inrush current. After a prescribed time delay (provided by a timedelay relay), full
power is applied.
• Conveyor belt sequence delay: when multiple conveyor belts are arranged to transport
material, the conveyor belts must be started in reverse sequence (the last one first and
the first one last) so that material doesn’t get piled on to a stopped or slowmoving con
veyor. In order to get large belts up to full speed, some time may be needed (especially if
softstart motor controls are used). For this reason, there is usually a timedelay circuit
arranged on each conveyor to give it adequate time to attain full belt speed before the
next conveyor belt feeding it is started.
130 CHAPTER 5. ELECTROMECHANICAL RELAYS
The older, mechanical timedelay relays used pneumatic dashpots or fluidfilled piston/cylinder
arrangements to provide the ”shock absorbing” needed to delay the motion of the armature.
Newer designs of timedelay relays use electronic circuits with resistorcapacitor (RC) net
works to generate a time delay, then energize a normal (instantaneous) electromechanical
relay coil with the electronic circuit’s output. The electronictimer relays are more versatile
than the older, mechanical models, and less prone to failure. Many models provide advanced
timer features such as ”oneshot” (one measured output pulse for every transition of the input
from deenergized to energized), ”recycle” (repeated on/off output cycles for as long as the input
connection is energized) and ”watchdog” (changes state if the input signal does not repeatedly
cycle on and off).
Coil
power
Contact
status
Time
on
off
closed
open
time
"Oneshot" normallyopen relay contact
Coil
power
Contact
status
Time
on
off
closed
open
"Recycle" normallyopen relay contact
5.3. TIMEDELAY RELAYS 131
Coil
power
Contact
status
Time
on
off
closed
open
"Watchdog" relay contact
time
The ”watchdog” timer is especially useful for monitoring of computer systems. If a computer
is being used to control a critical process, it is usually recommended to have an automatic alarm
to detect computer ”lockup” (an abnormal halting of program execution due to any number of
causes). An easy way to set up such a monitoring system is to have the computer regularly en
ergize and deenergize the coil of a watchdog timer relay (similar to the output of the ”recycle”
timer). If the computer execution halts for any reason, the signal it outputs to the watchdog
relay coil will stop cycling and freeze in one or the other state. A short time thereafter, the
watchdog relay will ”time out” and signal a problem.
• REVIEW:
• Time delay relays are built in these four basic modes of contact operation:
• 1: Normallyopen, timedclosed. Abbreviated ”NOTC”, these relays open immediately
upon coil deenergization and close only if the coil is continuously energized for the time
duration period. Also called normallyopen, ondelay relays.
• 2: Normallyopen, timedopen. Abbreviated ”NOTO”, these relays close immediately
upon coil energization and open after the coil has been deenergized for the time duration
period. Also called normallyopen, off delay relays.
• 3: Normallyclosed, timedopen. Abbreviated ”NCTO”, these relays close immediately
upon coil deenergization and open only if the coil is continuously energized for the time
duration period. Also called normallyclosed, ondelay relays.
• 4: Normallyclosed, timedclosed. Abbreviated ”NCTC”, these relays open immediately
upon coil energization and close after the coil has been deenergized for the time duration
period. Also called normallyclosed, off delay relays.
• Oneshot timers provide a single contact pulse of specified duration for each coil energiza
tion (transition from coil off to coil on).
• Recycle timers provide a repeating sequence of onoff contact pulses as long as the coil is
maintained in an energized state.
• Watchdog timers actuate their contacts only if the coil fails to be continuously sequenced
on and off (energized and deenergized) at a minimum frequency.
132 CHAPTER 5. ELECTROMECHANICAL RELAYS
5.4 Protective relays
A special type of relay is one which monitors the current, voltage, frequency, or any other type
of electric power measurement either from a generating source or to a load for the purpose of
triggering a circuit breaker to open in the event of an abnormal condition. These relays are
referred to in the electrical power industry as protective relays.
The circuit breakers which are used to switch large quantities of electric power on and
off are actually electromechanical relays, themselves. Unlike the circuit breakers found in
residential and commercial use which determine when to trip (open) by means of a bimetallic
strip inside that bends when it gets too hot from overcurrent, large industrial circuit breakers
must be ”told” by an external device when to open. Such breakers have two electromagnetic
coils inside: one to close the breaker contacts and one to open them. The ”trip” coil can be
energized by one or more protective relays, as well as by hand switches, connected to switch
125 Volt DC power. DC power is used because it allows for a battery bank to supply close/trip
power to the breaker control circuits in the event of a complete (AC) power failure.
Protective relays can monitor large AC currents by means of current transformers (CT’s),
which encircle the currentcarrying conductors exiting a large circuit breaker, transformer,
generator, or other device. Current transformers step down the monitored current to a sec
ondary (output) range of 0 to 5 amps AC to power the protective relay. The current relay uses
this 05 amp signal to power its internal mechanism, closing a contact to switch 125 Volt DC
power to the breaker’s trip coil if the monitored current becomes excessive.
Likewise, (protective) voltage relays can monitor high AC voltages by means of voltage, or
potential, transformers (PT’s) which step down the monitored voltage to a secondary range of
0 to 120 Volts AC, typically. Like (protective) current relays, this voltage signal powers the
internal mechanism of the relay, closing a contact to switch 125 Volt DC power to the breaker’s
trip coil is the monitored voltage becomes excessive.
There are many types of protective relays, some with highly specialized functions. Not all
monitor voltage or current, either. They all, however, share the common feature of outputting
a contact closure signal which can be used to switch power to a breaker trip coil, close coil,
or operator alarm panel. Most protective relay functions have been categorized into an ANSI
standard number code. Here are a few examples from that code list:
ANSI protective relay designation numbers
12 = Overspeed
24 = Overexcitation
25 = Syncrocheck
27 = Bus/Line undervoltage
32 = Reverse power (antimotoring)
38 = Stator overtemp (RTD)
39 = Bearing vibration
40 = Loss of excitation
46 = Negative sequence undercurrent (phase current imbalance)
47 = Negative sequence undervoltage (phase voltage imbalance)
49 = Bearing overtemp (RTD)
50 = Instantaneous overcurrent
51 = Time overcurrent
5.5. SOLIDSTATE RELAYS 133
51V = Time overcurrent  voltage restrained
55 = Power factor
59 = Bus overvoltage
60FL = Voltage transformer fuse failure
67 = Phase/Ground directional current
79 = Autoreclose
81 = Bus over/underfrequency
• REVIEW:
• Large electric circuit breakers do not contain within themselves the necessary mecha
nisms to automatically trip (open) in the event of overcurrent conditions. They must be
”told” to trip by external devices.
• Protective relays are devices built to automatically trigger the actuation coils of large
electric circuit breakers under certain conditions.
5.5 Solidstate relays
As versatile as electromechanical relays can be, they do suffer many limitations. They can be
expensive to build, have a limited contact cycle life, take up a lot of room, and switch slowly,
compared to modern semiconductor devices. These limitations are especially true for large
power contactor relays. To address these limitations, many relay manufacturers offer ”solid
state” relays, which use an SCR, TRIAC, or transistor output instead of mechanical contacts to
switch the controlled power. The output device (SCR, TRIAC, or transistor) is opticallycoupled
to an LED light source inside the relay. The relay is turned on by energizing this LED, usually
with lowvoltage DC power. This optical isolation between input to output rivals the best that
electromechanical relays can offer.
Load
LED OptoTRIAC
Solidstate relay
Being solidstate devices, there are no moving parts to wear out, and they are able to switch
on and off much faster than any mechanical relay armature can move. There is no sparking
between contacts, and no problems with contact corrosion. However, solidstate relays are
still too expensive to build in very high current ratings, and so electromechanical contactors
continue to dominate that application in industry today.
One significant advantage of a solidstate SCR or TRIAC relay over an electromechanical
device is its natural tendency to open the AC circuit only at a point of zero load current. Be
cause SCR’s and TRIAC’s are thyristors, their inherent hysteresis maintains circuit continuity
after the LED is deenergized until the AC current falls below a threshold value (the holding
134 CHAPTER 5. ELECTROMECHANICAL RELAYS
current). In practical terms what this means is the circuit will never be interrupted in the
middle of a sine wave peak. Such untimely interruptions in a circuit containing substantial
inductance would normally produce large voltage spikes due to the sudden magnetic field col
lapse around the inductance. This will not happen in a circuit broken by an SCR or TRIAC.
This feature is called zerocrossover switching.
One disadvantage of solid state relays is their tendency to fail ”shorted” on their outputs,
while electromechanical relay contacts tend to fail ”open.” In either case, it is possible for a
relay to fail in the other mode, but these are the most common failures. Because a ”failopen”
state is generally considered safer than a ”failclosed” state, electromechanical relays are still
favored over their solidstate counterparts in many applications.
Chapter 6
LADDER LOGIC
Contents
6.1 ”Ladder” diagrams . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6.2 Digital logic functions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
6.3 Permissive and interlock circuits . . . . . . . . . . . . . . . . . . . . . . . . . 144
6.4 Motor control circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
6.5 Failsafe design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
6.6 Programmable logic controllers . . . . . . . . . . . . . . . . . . . . . . . . . 154
6.7 Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
6.1 ”Ladder” diagrams
Ladder diagrams are specialized schematics commonly used to document industrial control
logic systems. They are called ”ladder” diagrams because they resemble a ladder, with two
vertical rails (supply power) and as many ”rungs” (horizontal lines) as there are control circuits
to represent. If we wanted to draw a simple ladder diagram showing a lamp that is controlled
by a hand switch, it would look like this:
L1 L2
1
Switch Lamp
The ”L1” and ”L2” designations refer to the two poles of a 120 VAC supply, unless otherwise
noted. L1 is the ”hot” conductor, and L2 is the grounded (”neutral”) conductor. These designa
tions have nothing to do with inductors, just to make things confusing. The actual transformer
or generator supplying power to this circuit is omitted for simplicity. In reality, the circuit looks
something like this:
135
136 CHAPTER 6. LADDER LOGIC
L1 L2
1
Switch Lamp
To 480 volt AC
power source (typical)
stepdown "control power"
transformer
120 VAC
fuse fuse
fuse
Typically in industrial relay logic circuits, but not always, the operating voltage for the
switch contacts and relay coils will be 120 volts AC. Lower voltage AC and even DC systems
are sometimes built and documented according to ”ladder” diagrams:
L1 L2
1
Switch Lamp
fuse 24 VDC
So long as the switch contacts and relay coils are all adequately rated, it really doesn’t
matter what level of voltage is chosen for the system to operate with.
Note the number ”1” on the wire between the switch and the lamp. In the real world, that
wire would be labeled with that number, using heatshrink or adhesive tags, wherever it was
convenient to identify. Wires leading to the switch would be labeled ”L1” and ”1,” respectively.
Wires leading to the lamp would be labeled ”1” and ”L2,” respectively. These wire numbers
make assembly and maintenance very easy. Each conductor has its own unique wire number
for the control system that its used in. Wire numbers do not change at any junction or node,
even if wire size, color, or length changes going into or out of a connection point. Of course, it is
preferable to maintain consistent wire colors, but this is not always practical. What matters is
that any one, electrically continuous point in a control circuit possesses the same wire number.
Take this circuit section, for example, with wire #25 as a single, electrically continuous point
threading to many different devices:
6.1. ”LADDER” DIAGRAMS 137
25
25
25
25
25
25
25
25
In ladder diagrams, the load device (lamp, relay coil, solenoid coil, etc.) is almost always
drawn at the righthand side of the rung. While it doesn’t matter electrically where the re
lay coil is located within the rung, it does matter which end of the ladder’s power supply is
grounded, for reliable operation.
Take for instance this circuit:
L1 L2
1
Switch Lamp
120 VAC
Here, the lamp (load) is located on the righthand side of the rung, and so is the ground
connection for the power source. This is no accident or coincidence; rather, it is a purposeful
element of good design practice. Suppose that wire #1 were to accidently come in contact with
ground, the insulation of that wire having been rubbed off so that the bare conductor came in
contact with grounded, metal conduit. Our circuit would now function like this:
138 CHAPTER 6. LADDER LOGIC
L1 L2
1
Switch Lamp
120 VAC
accidental ground
Lamp cannot light!
Fuse will blow
if switch is
closed!
With both sides of the lamp connected to ground, the lamp will be ”shorted out” and un
able to receive power to light up. If the switch were to close, there would be a shortcircuit,
immediately blowing the fuse.
However, consider what would happen to the circuit with the same fault (wire #1 coming
in contact with ground), except this time we’ll swap the positions of switch and fuse (L2 is still
grounded):
L1 L2
1
SwitchLamp
120 VAC
accidental ground
Lamp is energized! Switch has no
effect!
This time the accidental grounding of wire #1 will force power to the lamp while the switch
will have no effect. It is much safer to have a system that blows a fuse in the event of a
ground fault than to have a system that uncontrollably energizes lamps, relays, or solenoids
6.2. DIGITAL LOGIC FUNCTIONS 139
in the event of the same fault. For this reason, the load(s) must always be located nearest the
grounded power conductor in the ladder diagram.
• REVIEW:
• Ladder diagrams (sometimes called ”ladder logic”) are a type of electrical notation and
symbology frequently used to illustrate how electromechanical switches and relays are
interconnected.
• The two vertical lines are called ”rails” and attach to opposite poles of a power supply,
usually 120 volts AC. L1 designates the ”hot” AC wire and L2 the ”neutral” (grounded)
conductor.
• Horizontal lines in a ladder diagram are called ”rungs,” each one representing a unique
parallel circuit branch between the poles of the power supply.
• Typically, wires in control systems are marked with numbers and/or letters for identifica
tion. The rule is, all permanently connected (electrically common) points must bear the
same label.
6.2 Digital logic functions
We can construct simply logic functions for our hypothetical lamp circuit, using multiple con
tacts, and document these circuits quite easily and understandably with additional rungs to
our original ”ladder.” If we use standard binary notation for the status of the switches and
lamp (0 for unactuated or deenergized; 1 for actuated or energized), a truth table can be made
to show how the logic works:
L1 L2
B
A
A
B
A B Output
00
0 1
01
1 1
0
1
1
1
1
140 CHAPTER 6. LADDER LOGIC
Now, the lamp will come on if either contact A or contact B is actuated, because all it takes
for the lamp to be energized is to have at least one path for current from wire L1 to wire 1.
What we have is a simple OR logic function, implemented with nothing more than contacts
and a lamp.
We canmimic the AND logic function by wiring the two contacts in series instead of parallel:
L1 L2
BA
A
B
A B Output
00
0 1
01
1 1
0
1
1 2
0
0
Now, the lamp energizes only if contact A and contact B are simultaneously actuated. A
path exists for current from wire L1 to the lamp (wire 2) if and only if both switch contacts are
closed.
The logical inversion, or NOT, function can be performed on a contact input simply by using
a normallyclosed contact instead of a normallyopen contact:
L1 L2
A
A
1
A Output
0 1
1 0
Now, the lamp energizes if the contact is not actuated, and deenergizes when the contact
is actuated.
If we take our OR function and invert each ”input” through the use of normallyclosed
contacts, we will end up with a NAND function. In a special branch of mathematics known
as Boolean algebra, this effect of gate function identity changing with the inversion of input
signals is described by DeMorgan’s Theorem, a subject to be explored in more detail in a later
chapter.
6.2. DIGITAL LOGIC FUNCTIONS 141
L1 L2
B
A
A
B
A B Output
00
0 1
01
1 1 0
1
1
1
1
or
A
B
The lamp will be energized if either contact is unactuated. It will go out only if both contacts
are actuated simultaneously.
Likewise, if we take our AND function and invert each ”input” through the use of normally
closed contacts, we will end up with a NOR function:
L1 L2
BA
A
B
A B Output
00
0 1
01
1 1 0
1
1
or
A
B
0
0
2
A pattern quickly reveals itself when ladder circuits are compared with their logic gate
counterparts:
142 CHAPTER 6. LADDER LOGIC
• Parallel contacts are equivalent to an OR gate.
• Series contacts are equivalent to an AND gate.
• Normallyclosed contacts are equivalent to a NOT gate (inverter).
We can build combinational logic functions by grouping contacts in seriesparallel arrange
ments, as well. In the following example, we have an ExclusiveOR function built from a
combination of AND, OR, and inverter (NOT) gates:
L1 L2
BA
A
B
A B Output
00
0 1
01
1 1 0
1
1
or
A
B
0
2
3
A B
2
1
The top rung (NC contact A in series with NO contact B) is the equivalent of the top
NOT/AND gate combination. The bottom rung (NO contact A in series with NC contact B)
is the equivalent of the bottom NOT/AND gate combination. The parallel connection between
the two rungs at wire number 2 forms the equivalent of the OR gate, in allowing either rung 1
or rung 2 to energize the lamp.
To make the ExclusiveOR function, we had to use two contacts per input: one for direct
input and the other for ”inverted” input. The two ”A” contacts are physically actuated by the
same mechanism, as are the two ”B” contacts. The common association between contacts is
denoted by the label of the contact. There is no limit to how many contacts per switch can be
represented in a ladder diagram, as each new contact on any switch or relay (either normally
open or normallyclosed) used in the diagram is simply marked with the same label.
Sometimes, multiple contacts on a single switch (or relay) are designated by a compound
labels, such as ”A1” and ”A2” instead of two ”A” labels. This may be especially useful if
6.2. DIGITAL LOGIC FUNCTIONS 143
you want to specifically designate which set of contacts on each switch or relay is being used
for which part of a circuit. For simplicity’s sake, I’ll refrain from such elaborate labeling in
this lesson. If you see a common label for multiple contacts, you know those contacts are all
actuated by the same mechanism.
If we wish to invert the output of any switchgenerated logic function, we must use a relay
with a normallyclosed contact. For instance, if we want to energize a load based on the inverse,
or NOT, of a normallyopen contact, we could do this:
L1 L2
A
1
A Output
0 1
1 0
A CR1
CR1
CR1
0
1
2
We will call the relay, ”control relay 1,” or CR1. When the coil of CR1 (symbolized with the
pair of parentheses on the first rung) is energized, the contact on the second rung opens, thus
deenergizing the lamp. From switch A to the coil of CR1, the logic function is noninverted.
The normallyclosed contact actuated by relay coil CR1 provides a logical inverter function to
drive the lamp opposite that of the switch’s actuation status.
Applying this inversion strategy to one of our invertedinput functions created earlier, such
as the ORtoNAND, we can invert the output with a relay to create a noninverted function:
144 CHAPTER 6. LADDER LOGIC
L1 L2
B
A
A
B
A B Output
00
0 1
01
1 1
1
or
A
B
CR1
CR1
2
0
0
0
1
From the switches to the coil of CR1, the logical function is that of a NAND gate. CR1’s
normallyclosed contact provides one final inversion to turn the NAND function into an AND
function.
• REVIEW:
• Parallel contacts are logically equivalent to an OR gate.
• Series contacts are logically equivalent to an AND gate.
• Normally closed (N.C.) contacts are logically equivalent to a NOT gate.
• A relay must be used to invert the output of a logic gate function, while simple normally
closed switch contacts are sufficient to represent inverted gate inputs.
6.3 Permissive and interlock circuits
A practical application of switch and relay logic is in control systems where several process
conditions have to be met before a piece of equipment is allowed to start. A good example of
this is burner control for large combustion furnaces. In order for the burners in a large furnace
to be started safely, the control system requests ”permission” from several process switches,
including high and low fuel pressure, air fan flow check, exhaust stack damper position, access
6.3. PERMISSIVE AND INTERLOCK CIRCUITS 145
door position, etc. Each process condition is called a permissive, and each permissive switch
contact is wired in series, so that if any one of them detects an unsafe condition, the circuit will
be opened:
L1 L2
low fuel
pressure
high fuel
pressure
minimum
air flow open
damper CR1
green
red
CR1
CR1
Green light = conditions met: safe to start
Red light = conditions not met: unsafe to start
If all permissive conditions are met, CR1 will energize and the green lamp will be lit. In
real life, more than just a green lamp would be energized: usually a control relay or fuel valve
solenoid would be placed in that rung of the circuit to be energized when all the permissive
contacts were ”good:” that is, all closed. If any one of the permissive conditions are not met,
the series string of switch contacts will be broken, CR2 will deenergize, and the red lamp will
light.
Note that the high fuel pressure contact is normallyclosed. This is because we want the
switch contact to open if the fuel pressure gets too high. Since the ”normal” condition of any
pressure switch is when zero (low) pressure is being applied to it, and we want this switch to
open with excessive (high) pressure, we must choose a switch that is closed in its normal state.
Another practical application of relay logic is in control systems where we want to ensure
two incompatible events cannot occur at the same time. An example of this is in reversible mo
tor control, where two motor contactors are wired to switch polarity (or phase sequence) to an
electric motor, and we don’t want the forward and reverse contactors energized simultaneously:
146 CHAPTER 6. LADDER LOGIC
motor
3phase
AC
power
M1
M2
M1 = forward
M2 = reverse
A
B
C
1
2
3
When contactor M1 is energized, the 3 phases (A, B, and C) are connected directly to termi
nals 1, 2, and 3 of the motor, respectively. However, when contactor M2 is energized, phases A
and B are reversed, A going to motor terminal 2 and B going to motor terminal 1. This reversal
of phase wires results in the motor spinning the opposite direction. Let’s examine the control
circuit for these two contactors:
L1 L2
forward M1 OL
M2reverse
1
2
3
Take note of the normallyclosed ”OL” contact, which is the thermal overload contact acti
vated by the ”heater” elements wired in series with each phase of the AC motor. If the heaters
get too hot, the contact will change from its normal (closed) state to being open, which will
prevent either contactor from energizing.
This control system will work fine, so long as no one pushes both buttons at the same time.
If someone were to do that, phases A and B would be shortcircuited together by virtue of the
fact that contactor M1 sends phases A and B straight to the motor and contactor M2 reverses
them; phase A would be shorted to phase B and vice versa. Obviously, this is a bad control
system design!
To prevent this occurrence from happening, we can design the circuit so that the energiza
tion of one contactor prevents the energization of the other. This is called interlocking, and it
is accomplished through the use of auxiliary contacts on each contactor, as such:
6.4. MOTOR CONTROL CIRCUITS 147
L1 L2
forward M1 OL
M2reverse
1
2
3
M1
M2
4
5
Now, when M1 is energized, the normallyclosed auxiliary contact on the second rung will be
open, thus preventing M2 from being energized, even if the ”Reverse” pushbutton is actuated.
Likewise, M1’s energization is prevented when M2 is energized. Note, as well, how additional
wire numbers (4 and 5) were added to reflect the wiring changes.
It should be noted that this is not the only way to interlock contactors to prevent a short
circuit condition. Some contactors come equipped with the option of a mechanical interlock: a
lever joining the armatures of two contactors together so that they are physically prevented
from simultaneous closure. For additional safety, electrical interlocks may still be used, and
due to the simplicity of the circuit there is no good reason not to employ them in addition to
mechanical interlocks.
• REVIEW:
• Switch contacts installed in a rung of ladder logic designed to interrupt a circuit if cer
tain physical conditions are not met are called permissive contacts, because the system
requires permission from these inputs to activate.
• Switch contacts designed to prevent a control system from taking two incompatible ac
tions at once (such as powering an electric motor forward and backward simultaneously)
are called interlocks.
6.4 Motor control circuits
The interlock contacts installed in the previous section’s motor control circuit work fine, but
the motor will run only as long as each pushbutton switch is held down. If we wanted to keep
the motor running even after the operator takes his or her hand off the control switch(es),
we could change the circuit in a couple of different ways: we could replace the pushbutton
switches with toggle switches, or we could add some more relay logic to ”latch” the control
circuit with a single, momentary actuation of either switch. Let’s see how the second approach
is implemented, since it is commonly used in industry:
148 CHAPTER 6. LADDER LOGIC
L1 L2
forward M1 OL
M2reverse
1
2
3
M1
M2
4
5
M1
M2
When the ”Forward” pushbutton is actuated, M1 will energize, closing the normallyopen
auxiliary contact in parallel with that switch. When the pushbutton is released, the closed M1
auxiliary contact will maintain current to the coil of M1, thus latching the ”Forward” circuit in
the ”on” state. The same sort of thing will happen when the ”Reverse” pushbutton is pressed.
These parallel auxiliary contacts are sometimes referred to as sealin contacts, the word ”seal”
meaning essentially the same thing as the word latch.
However, this creates a new problem: how to stop the motor! As the circuit exists right now,
the motor will run either forward or backward once the corresponding pushbutton switch is
pressed, and will continue to run as long as there is power. To stop either circuit (forward or
backward), we require some means for the operator to interrupt power to the motor contactors.
We’ll call this new switch, Stop:
L1 L2
forward M1 OL
M2reverse
1
2
3
M1
M2
4
5
M1
M2
stop
6
6
6
6
6.4. MOTOR CONTROL CIRCUITS 149
Now, if either forward or reverse circuits are latched, they may be ”unlatched” by momen
tarily pressing the ”Stop” pushbutton, which will open either forward or reverse circuit, de
energizing the energized contactor, and returning the sealin contact to its normal (open) state.
The ”Stop” switch, having normallyclosed contacts, will conduct power to either forward or re
verse circuits when released.
So far, so good. Let’s consider another practical aspect of our motor control scheme before we
quit adding to it. If our hypothetical motor turned a mechanical load with a lot of momentum,
such as a large air fan, the motor might continue to coast for a substantial amount of time
after the stop button had been pressed. This could be problematic if an operator were to try
to reverse the motor direction without waiting for the fan to stop turning. If the fan was
still coasting forward and the ”Reverse” pushbutton was pressed, the motor would struggle to
overcome that inertia of the large fan as it tried to begin turning in reverse, drawing excessive
current and potentially reducing the life of the motor, drive mechanisms, and fan. What we
might like to have is some kind of a timedelay function in this motor control system to prevent
such a premature startup from happening.
Let’s begin by adding a couple of timedelay relay coils, one in parallel with each motor
contactor coil. If we use contacts that delay returning to their normal state, these relays will
provide us a ”memory” of which direction the motor was last powered to turn. What we want
each timedelay contact to do is to open the startingswitch leg of the opposite rotation circuit
for several seconds, while the fan coasts to a halt.
L1 L2
forward M1 OL
M2reverse
1
2
3
M1
M2
4
5
M1
M2
stop
6
6
6
6
TD1
TD2
TD2
TD1
7
8
If the motor has been running in the forward direction, both M1 and TD1 will have been
energized. This being the case, the normallyclosed, timedclosed contact of TD1 between wires
8 and 5 will have immediately opened the moment TD1 was energized. When the stop button
is pressed, contact TD1 waits for the specified amount of time before returning to its normally
closed state, thus holding the reverse pushbutton circuit open for the duration so M2 can’t
be energized. When TD1 times out, the contact will close and the circuit will allow M2 to be
energized, if the reverse pushbutton is pressed. In like manner, TD2 will prevent the ”Forward”
pushbutton from energizing M1 until the prescribed time delay after M2 (and TD2) have been
150 CHAPTER 6. LADDER LOGIC
deenergized.
The careful observer will notice that the timeinterlocking functions of TD1 and TD2 ren
der the M1 and M2 interlocking contacts redundant. We can get rid of auxiliary contacts M1
and M2 for interlocks and just use TD1 and TD2’s contacts, since they immediately open when
their respective relay coils are energized, thus ”locking out” one contactor if the other is ener
gized. Each time delay relay will serve a dual purpose: preventing the other contactor from
energizing while the motor is running, and preventing the same contactor from energizing un
til a prescribed time after motor shutdown. The resulting circuit has the advantage of being
simpler than the previous example:
L1 L2
forward M1 OL
M2reverse
1
2
34
5
M1
M2
stop
6
6
6
6
TD1
TD2
TD2
TD1
• REVIEW:
• Motor contactor (or ”starter”) coils are typically designated by the letter ”M” in ladder
logic diagrams.
• Continuous motor operation with a momentary ”start” switch is possible if a normally
open ”sealin” contact from the contactor is connected in parallel with the start switch, so
that once the contactor is energized it maintains power to itself and keeps itself ”latched”
on.
• Time delay relays are commonly used in large motor control circuits to prevent the motor
from being started (or reversed) until a certain amount of time has elapsed from an event.
6.5 Failsafe design
Logic circuits, whether comprised of electromechanical relays or solidstate gates, can be built
in many different ways to perform the same functions. There is usually no one ”correct” way to
design a complex logic circuit, but there are usually ways that are better than others.
6.5. FAILSAFE DESIGN 151
In control systems, safety is (or at least should be) an important design priority. If there
are multiple ways in which a digital control circuit can be designed to perform a task, and one
of those ways happens to hold certain advantages in safety over the others, then that design is
the better one to choose.
Let’s take a look at a simple system and consider how it might be implemented in relay
logic. Suppose that a large laboratory or industrial building is to be equipped with a fire alarm
system, activated by any one of several latching switches installed throughout the facility.
The system should work so that the alarm siren will energize if any one of the switches is
actuated. At first glance it seems as though the relay logic should be incredibly simple: just
use normallyopen switch contacts and connect them all in parallel with each other:
L1 L2
sirenswitch 1
switch 2
switch 3
switch 4
Essentially, this is the OR logic function implemented with four switch inputs. We could
expand this circuit to include any number of switch inputs, each new switch being added to the
parallel network, but I’ll limit it to four in this example to keep things simple. At any rate, it
is an elementary system and there seems to be little possibility of trouble.
Except in the event of a wiring failure, that is. The nature of electric circuits is such that
”open” failures (open switch contacts, broken wire connections, open relay coils, blown fuses,
etc.) are statistically more likely to occur than any other type of failure. With that in mind, it
makes sense to engineer a circuit to be as tolerant as possible to such a failure. Let’s suppose
that a wire connection for Switch #2 were to fail open:
152 CHAPTER 6. LADDER LOGIC
L1 L2
sirenswitch 1
switch 2
switch 3
switch 4
open wire connection!
If this failure were to occur, the result would be that Switch #2 would no longer energize the
siren if actuated. This, obviously, is not good in a fire alarm system. Unless the system were
regularly tested (a good idea anyway), no one would know there was a problem until someone
tried to use that switch in an emergency.
What if the system were reengineered so as to sound the alarm in the event of an open
failure? That way, a failure in the wiring would result in a false alarm, a scenario much more
preferable than that of having a switch silently fail and not function when needed. In order
to achieve this design goal, we would have to rewire the switches so that an open contact
sounded the alarm, rather than a closed contact. That being the case, the switches will have to
be normallyclosed and in series with each other, powering a relay coil which then activates a
normallyclosed contact for the siren:
L1 L2
switch 1
switch 2
switch 3
switch 4
CR1
CR1 siren
When all switches are unactuated (the regular operating state of this system), relay CR1
will be energized, thus keeping contact CR1 open, preventing the siren from being powered.
However, if any of the switches are actuated, relay CR1 will deenergize, closing contact CR1
and sounding the alarm. Also, if there is a break in the wiring anywhere in the top rung of the
circuit, the alarm will sound. When it is discovered that the alarm is false, the workers in the
facility will know that something failed in the alarm system and that it needs to be repaired.
Granted, the circuit is more complex than it was before the addition of the control relay, and
the system could still fail in the ”silent” mode with a broken connection in the bottom rung,
6.5. FAILSAFE DESIGN 153
but its still a safer design than the original circuit, and thus preferable from the standpoint of
safety.
This design of circuit is referred to as failsafe, due to its intended design to default to
the safest mode in the event of a common failure such as a broken connection in the switch
wiring. Failsafe design always starts with an assumption as to the most likely kind of wiring or
component failure, and then tries to configure things so that such a failure will cause the circuit
to act in the safest way, the ”safest way” being determined by the physical characteristics of
the process.
Take for example an electricallyactuated (solenoid) valve for turning on cooling water to
a machine. Energizing the solenoid coil will move an armature which then either opens or
closes the valve mechanism, depending on what kind of valve we specify. A spring will return
the valve to its ”normal” position when the solenoid is deenergized. We already know that
an open failure in the wiring or solenoid coil is more likely than a short or any other type of
failure, so we should design this system to be in its safest mode with the solenoid deenergized.
If its cooling water we’re controlling with this valve, chances are it is safer to have the
cooling water turn on in the event of a failure than to shut off, the consequences of a machine
running without coolant usually being severe. This means we should specify a valve that turns
on (opens up) when deenergized and turns off (closes down) when energized. This may seem
”backwards” to have the valve set up this way, but it will make for a safer system in the end.
One interesting application of failsafe design is in the power generation and distribution
industry, where large circuit breakers need to be opened and closed by electrical control sig
nals from protective relays. If a 50/51 relay (instantaneous and time overcurrent) is going to
command a circuit breaker to trip (open) in the event of excessive current, should we design it
so that the relay closes a switch contact to send a ”trip” signal to the breaker, or opens a switch
contact to interrupt a regularly ”on” signal to initiate a breaker trip? We know that an open
connection will be the most likely to occur, but what is the safest state of the system: breaker
open or breaker closed?
At first, it would seem that it would be safer to have a large circuit breaker trip (open up
and shut off power) in the event of an open fault in the protective relay control circuit, just
like we had the fire alarm system default to an alarm state with any switch or wiring failure.
However, things are not so simple in the world of high power. To have a large circuit breaker
indiscriminately trip open is no small matter, especially when customers are depending on
the continued supply of electric power to supply hospitals, telecommunications systems, water
treatment systems, and other important infrastructures. For this reason, power system engi
neers have generally agreed to design protective relay circuits to output a closed contact signal
(power applied) to open large circuit breakers, meaning that any open failure in the control
wiring will go unnoticed, simply leaving the breaker in the status quo position.
Is this an ideal situation? Of course not. If a protective relay detects an overcurrent condi
tion while the control wiring is failed open, it will not be able to trip open the circuit breaker.
Like the first fire alarm system design, the ”silent” failure will be evident only when the system
is needed. However, to engineer the control circuitry the other way – so that any open failure
would immediately shut the circuit breaker off, potentially blacking out large potions of the
power grid – really isn’t a better alternative.
An entire book could be written on the principles and practices of good failsafe system
design. At least here, you know a couple of the fundamentals: that wiring tends to fail open
more often than shorted, and that an electrical control system’s (open) failure mode should be
154 CHAPTER 6. LADDER LOGIC
such that it indicates and/or actuates the reallife process in the safest alternative mode. These
fundamental principles extend to nonelectrical systems as well: identify the most common
mode of failure, then engineer the system so that the probable failure mode places the system
in the safest condition.
• REVIEW:
• The goal of failsafe design is to make a control system as tolerant as possible to likely
wiring or component failures.
• The most common type of wiring and component failure is an ”open” circuit, or broken
connection. Therefore, a failsafe system should be designed to default to its safest mode
of operation in the case of an open circuit.
6.6 Programmable logic controllers
Before the advent of solidstate logic circuits, logical control systems were designed and built
exclusively around electromechanical relays. Relays are far from obsolete in modern design,
but have been replaced in many of their former roles as logiclevel control devices, relegated
most often to those applications demanding high current and/or high voltage switching.
Systems and processes requiring ”on/off” control abound in modern commerce and industry,
but such control systems are rarely built from either electromechanical relays or discrete logic
gates. Instead, digital computers fill the need, which may be programmed to do a variety of
logical functions.
In the late 1960’s an American company named Bedford Associates released a computing
device they called the MODICON. As an acronym, it meant Modular Digital Controller, and
later became the name of a company division devoted to the design, manufacture, and sale
of these specialpurpose control computers. Other engineering firms developed their own ver
sions of this device, and it eventually came to be known in nonproprietary terms as a PLC, or
Programmable Logic Controller. The purpose of a PLC was to directly replace electromechan
ical relays as logic elements, substituting instead a solidstate digital computer with a stored
program, able to emulate the interconnection of many relays to perform certain logical tasks.
A PLC has many ”input” terminals, through which it interprets ”high” and ”low” logical
states from sensors and switches. It also has many output terminals, through which it out
puts ”high” and ”low” signals to power lights, solenoids, contactors, small motors, and other
devices lending themselves to on/off control. In an effort to make PLCs easy to program, their
programming language was designed to resemble ladder logic diagrams. Thus, an industrial
electrician or electrical engineer accustomed to reading ladder logic schematics would feel com
fortable programming a PLC to perform the same control functions.
PLCs are industrial computers, and as such their input and output signals are typically 120
volts AC, just like the electromechanical control relays they were designed to replace. Although
some PLCs have the ability to input and output lowlevel DC voltage signals of the magnitude
used in logic gate circuits, this is the exception and not the rule.
Signal connection and programming standards vary somewhat between different models of
PLC, but they are similar enough to allow a ”generic” introduction to PLC programming here.
The following illustration shows a simple PLC, as it might appear from a front view. Two screw
6.6. PROGRAMMABLE LOGIC CONTROLLERS 155
terminals provide connection to 120 volts AC for powering the PLC’s internal circuitry, labeled
L1 and L2. Six screw terminals on the lefthand side provide connection to input devices, each
terminal representing a different input ”channel” with its own ”X” label. The lowerleft screw
terminal is a ”Common” connection, which is generally connected to L2 (neutral) of the 120
VAC power source.
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
Inside the PLC housing, connected between each input terminal and the Common terminal,
is an optoisolator device (LightEmitting Diode) that provides an electrically isolated ”high”
logic signal to the computer’s circuitry (a phototransistor interprets the LED’s light) when
there is 120 VAC power applied between the respective input terminal and the Common termi
nal. An indicating LED on the front panel of the PLC gives visual indication of an ”energized”
input:
PLC
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
portCommon Source
L1 L2
X2
X3
X4
X5
X6
V Ω
COMA
120 V
Input X1 energized
Output signals are generated by the PLC’s computer circuitry activating a switching device
(transistor, TRIAC, or even an electromechanical relay), connecting the ”Source” terminal to
156 CHAPTER 6. LADDER LOGIC
any of the ”Y” labeled output terminals. The ”Source” terminal, correspondingly, is usually
connected to the L1 side of the 120 VAC power source. As with each input, an indicating LED
on the front panel of the PLC gives visual indication of an ”energized” output:
PLC
L1 L2
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X2
X3
X4
X5
X6
V Ω
COMA
120 V
X1
Output Y1 energized
In this way, the PLC is able to interface with realworld devices such as switches and
solenoids.
The actual logic of the control system is established inside the PLC by means of a computer
program. This program dictates which output gets energized under which input conditions.
Although the program itself appears to be a ladder logic diagram, with switch and relay sym
bols, there are no actual switch contacts or relay coils operating inside the PLC to create the
logical relationships between input and output. These are imaginary contacts and coils, if
you will. The program is entered and viewed via a personal computer connected to the PLC’s
programming port.
Consider the following circuit and PLC program:
6.6. PROGRAMMABLE LOGIC CONTROLLERS 157
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
Personal
computer
Programming
cableX1 Y1
display
When the pushbutton switch is unactuated (unpressed), no power is sent to the X1 input of
the PLC. Following the program, which shows a normallyopen X1 contact in series with a Y1
coil, no ”power” will be sent to the Y1 coil. Thus, the PLC’s Y1 output remains deenergized,
and the indicator lamp connected to it remains dark.
If the pushbutton switch is pressed, however, power will be sent to the PLC’s X1 input. Any
and all X1 contacts appearing in the program will assume the actuated (nonnormal) state,
as though they were relay contacts actuated by the energizing of a relay coil named ”X1”. In
this case, energizing the X1 input will cause the normallyopen X1 contact will ”close,” sending
”power” to the Y1 coil. When the Y1 coil of the program ”energizes,” the real Y1 output will
become energized, lighting up the lamp connected to it:
158 CHAPTER 6. LADDER LOGIC
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
Personal
computer
Programming
cableX1 Y1
display
switch actuated
lamp
lights!
contact and coil symbols
turn color, indicating "power"
conducted through "circuit"
It must be understood that the X1 contact, Y1 coil, connecting wires, and ”power” appearing
in the personal computer’s display are all virtual. They do not exist as real electrical compo
nents. They exist as commands in a computer program – a piece of software only – that just
happens to resemble a real relay schematic diagram.
Equally important to understand is that the personal computer used to display and edit
the PLC’s program is not necessary for the PLC’s continued operation. Once a program has
been loaded to the PLC from the personal computer, the personal computer may be unplugged
from the PLC, and the PLC will continue to follow the programmed commands. I include the
personal computer display in these illustrations for your sake only, in aiding to understand the
relationship between reallife conditions (switch closure and lamp status) and the program’s
status (”power” through virtual contacts and virtual coils).
The true power and versatility of a PLC is revealed when we want to alter the behavior
of a control system. Since the PLC is a programmable device, we can alter its behavior by
changing the commands we give it, without having to reconfigure the electrical components
connected to it. For example, suppose we wanted to make this switchandlamp circuit function
in an inverted fashion: push the button to make the lamp turn off, and release it to make it
turn on. The ”hardware” solution would require that a normallyclosed pushbutton switch be
6.6. PROGRAMMABLE LOGIC CONTROLLERS 159
substituted for the normallyopen switch currently in place. The ”software” solution is much
easier: just alter the program so that contact X1 is normallyclosed rather than normallyopen.
In the following illustration, we have the altered system shown in the state where the
pushbutton is unactuated (not being pressed):
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
Personal
computer
Programming
cableX1 Y1
display
lamp
lights!
colored components show
"power" conducted through
the "circuit" with contact X1
in its "normal" status
In this next illustration, the switch is shown actuated (pressed):
160 CHAPTER 6. LADDER LOGIC
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
Personal
computer
Programming
cableX1 Y1
display
switch actuated
lamp is
dark
uncolored components
show no "continuity" in
the "circuit"
One of the advantages of implementing logical control in software rather than in hardware
is that input signals can be reused as many times in the program as is necessary. For example,
take the following circuit and program, designed to energize the lamp if at least two of the three
pushbutton switches are simultaneously actuated:
6.6. PROGRAMMABLE LOGIC CONTROLLERS 161
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
X2 X3
X1 X3
To build an equivalent circuit using electromechanical relays, three relays with two normally
open contacts each would have to be used, to provide two contacts per input switch. Using a
PLC, however, we can program as many contacts as we wish for each ”X” input without adding
additional hardware, since each input and each output is nothing more than a single bit in the
PLC’s digital memory (either 0 or 1), and can be recalled as many times as necessary.
Furthermore, since each output in the PLC is nothing more than a bit in its memory as
well, we can assign contacts in a PLC program ”actuated” by an output (Y) status. Take for
instance this next system, a motor startstop control circuit:
162 CHAPTER 6. LADDER LOGIC
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
The pushbutton switch connected to input X1 serves as the ”Start” switch, while the switch
connected to input X2 serves as the ”Stop.” Another contact in the program, named Y1, uses
the output coil status as a sealin contact, directly, so that the motor contactor will continue to
be energized after the ”Start” pushbutton switch is released. You can see the normallyclosed
contact X2 appear in a colored block, showing that it is in a closed (”electrically conducting”)
state.
If we were to press the ”Start” button, input X1 would energize, thus ”closing” the X1 contact
in the program, sending ”power” to the Y1 ”coil,” energizing the Y1 output and applying 120
volt AC power to the real motor contactor coil. The parallel Y1 contact will also ”close,” thus
latching the ”circuit” in an energized state:
6.6. PROGRAMMABLE LOGIC CONTROLLERS 163
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
(actuated)
(energized)
Now, if we release the ”Start” pushbutton, the normallyopen X1 ”contact” will return to its
”open” state, but the motor will continue to run because the Y1 sealin ”contact” continues to
provide ”continuity” to ”power” coil Y1, thus keeping the Y1 output energized:
164 CHAPTER 6. LADDER LOGIC
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
(energized)
(released)
To stop the motor, we must momentarily press the ”Stop” pushbutton, which will energize
the X2 input and ”open” the normallyclosed ”contact,” breaking continuity to the Y1 ”coil:”
6.6. PROGRAMMABLE LOGIC CONTROLLERS 165
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
(actuated)
When the ”Stop” pushbutton is released, input X2 will deenergize, returning ”contact” X2 to
its normal, ”closed” state. The motor, however, will not start again until the ”Start” pushbutton
is actuated, because the ”sealin” of Y1 has been lost:
166 CHAPTER 6. LADDER LOGIC
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
(released)
An important point to make here is that failsafe design is just as important in PLC
controlled systems as it is in electromechanical relaycontrolled systems. One should always
consider the effects of failed (open) wiring on the device or devices being controlled. In this
motor control circuit example, we have a problem: if the input wiring for X2 (the ”Stop” switch)
were to fail open, there would be no way to stop the motor!
The solution to this problem is a reversal of logic between the X2 ”contact” inside the PLC
program and the actual ”Stop” pushbutton switch:
6.6. PROGRAMMABLE LOGIC CONTROLLERS 167
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1 Y1X2
Y1
Motor
start
Motor
stop
M1
Motor
contactor
When the normallyclosed ”Stop” pushbutton switch is unactuated (not pressed), the PLC’s
X2 input will be energized, thus ”closing” the X2 ”contact” inside the program. This allows
the motor to be started when input X1 is energized, and allows it to continue to run when the
”Start” pushbutton is no longer pressed. When the ”Stop” pushbutton is actuated, input X2
will deenergize, thus ”opening” the X2 ”contact” inside the PLC program and shutting off the
motor. So, we see there is no operational difference between this new design and the previous
design.
However, if the input wiring on input X2 were to fail open, X2 input would deenergize in
the same manner as when the ”Stop” pushbutton is pressed. The result, then, for a wiring
failure on the X2 input is that the motor will immediately shut off. This is a safer design
than the one previously shown, where a ”Stop” switch wiring failure would have resulted in an
inability to turn off the motor.
In addition to input (X) and output (Y) program elements, PLCs provide ”internal” coils and
contacts with no intrinsic connection to the outside world. These are used much the same as
”control relays” (CR1, CR2, etc.) are used in standard relay circuits: to provide logic signal
inversion when necessary.
To demonstrate how one of these ”internal” relays might be used, consider the following
168 CHAPTER 6. LADDER LOGIC
example circuit and program, designed to emulate the function of a threeinput NAND gate.
Since PLC program elements are typically designed by single letters, I will call the internal
control relay ”C1” rather than ”CR1” as would be customary in a relay control circuit:
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1
Y1
X2 X3 C1
C1
lamp
lights!
In this circuit, the lamp will remain lit so long as any of the pushbuttons remain unactuated
(unpressed). To make the lamp turn off, we will have to actuate (press) all three switches, like
this:
6.6. PROGRAMMABLE LOGIC CONTROLLERS 169
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
X1
Y1
X2 X3 C1
C1
All three switches actuated
lamp is
dark
This section on programmable logic controllers illustrates just a small sample of their ca
pabilities. As computers, PLCs can perform timing functions (for the equivalent of timedelay
relays), drum sequencing, and other advanced functions with far greater accuracy and reliabil
ity than what is possible using electromechanical logic devices. Most PLCs have the capacity
for far more than six inputs and six outputs. The following photograph shows several input
and output modules of a single AllenBradley PLC.
170 CHAPTER 6. LADDER LOGIC
With each module having sixteen ”points” of either input or output, this PLC has the ability
to monitor and control dozens of devices. Fit into a control cabinet, a PLC takes up little room,
especially considering the equivalent space that would be needed by electromechanical relays
to perform the same functions:
One advantage of PLCs that simply cannot be duplicated by electromechanical relays is
remote monitoring and control via digital computer networks. Because a PLC is nothing more
than a specialpurpose digital computer, it has the ability to communicate with other comput
ers rather easily. The following photograph shows a personal computer displaying a graphic
image of a real liquidlevel process (a pumping, or ”lift,” station for a municipal wastewater
treatment system) controlled by a PLC. The actual pumping station is located miles away from
6.7. CONTRIBUTORS 171
the personal computer display:
6.7 Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
Roger Hollingsworth (May 2003): Suggested a way to make the PLC motor control circuit
failsafe.
172 CHAPTER 6. LADDER LOGIC
Chapter 7
BOOLEAN ALGEBRA
Contents
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
7.2 Boolean arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
7.3 Boolean algebraic identities . . . . . . . . . . . . . . . . . . . . . . . . . . . . 178
7.4 Boolean algebraic properties . . . . . . . . . . . . . . . . . . . . . . . . . . . 181
7.5 Boolean rules for simplification . . . . . . . . . . . . . . . . . . . . . . . . . . 184
7.6 Circuit simplification examples . . . . . . . . . . . . . . . . . . . . . . . . . . 187
7.7 The ExclusiveOR function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
7.8 DeMorgan’s Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
7.9 Converting truth tables into Boolean expressions . . . . . . . . . . . . . . 200
0 + 0 = 0
0 + 1 = 1
1 + 0 = 1
1 + 1 = 1
Rules of addition for Boolean quantities
”Gee Toto, I don’t think we’re in Kansas anymore!”
Dorothy, in The Wizard of Oz
7.1 Introduction
Mathematical rules are based on the defining limits we place on the particular numerical
quantities dealt with. When we say that 1 + 1 = 2 or 3 + 4 = 7, we are implying the use of
integer quantities: the same types of numbers we all learned to count in elementary education.
173
174 CHAPTER 7. BOOLEAN ALGEBRA
What most people assume to be selfevident rules of arithmetic – valid at all times and for all
purposes – actually depend on what we define a number to be.
For instance, when calculating quantities in AC circuits, we find that the ”real” number
quantities which served us so well in DC circuit analysis are inadequate for the task of repre
senting AC quantities. We know that voltages add when connected in series, but we also know
that it is possible to connect a 3volt AC source in series with a 4volt AC source and end up
with 5 volts total voltage (3 + 4 = 5)! Does this mean the inviolable and selfevident rules of
arithmetic have been violated? No, it just means that the rules of ”real” numbers do not apply
to the kinds of quantities encountered in AC circuits, where every variable has both a magni
tude and a phase. Consequently, we must use a different kind of numerical quantity, or object,
for AC circuits (complex numbers, rather than real numbers), and along with this different
system of numbers comes a different set of rules telling us how they relate to one another.
An expression such as ”3 + 4 = 5” is nonsense within the scope and definition of real num
bers, but it fits nicely within the scope and definition of complex numbers (think of a right
triangle with opposite and adjacent sides of 3 and 4, with a hypotenuse of 5). Because complex
numbers are twodimensional, they are able to ”add” with one another trigonometrically as
singledimension ”real” numbers cannot.
Logic is much like mathematics in this respect: the socalled ”Laws” of logic depend on how
we define what a proposition is. The Greek philosopher Aristotle founded a system of logic
based on only two types of propositions: true and false. His bivalent (twomode) definition of
truth led to the four foundational laws of logic: the Law of Identity (A is A); the Law of Non
contradiction (A is not nonA); the Law of the Excluded Middle (either A or nonA); and the Law
of Rational Inference. These socalled Laws function within the scope of logic where a propo
sition is limited to one of two possible values, but may not apply in cases where propositions
can hold values other than ”true” or ”false.” In fact, much work has been done and continues
to be done on ”multivalued,” or fuzzy logic, where propositions may be true or false to a limited
degree. In such a system of logic, ”Laws” such as the Law of the Excluded Middle simply do
not apply, because they are founded on the assumption of bivalence. Likewise, many premises
which would violate the Law of Noncontradiction in Aristotelian logic have validity in ”fuzzy”
logic. Again, the defining limits of propositional values determine the Laws describing their
functions and relations.
The English mathematician George Boole (18151864) sought to give symbolic form to Aris
totle’s system of logic. Boole wrote a treatise on the subject in 1854, titled An Investigation of
the Laws of Thought, on Which Are Founded the Mathematical Theories of Logic and Probabil
ities, which codified several rules of relationship between mathematical quantities limited to
one of two possible values: true or false, 1 or 0. His mathematical system became known as
Boolean algebra.
All arithmetic operations performed with Boolean quantities have but one of two possible
outcomes: either 1 or 0. There is no such thing as ”2” or ”1” or ”1/2” in the Boolean world. It is a
world in which all other possibilities are invalid by fiat. As one might guess, this is not the kind
of math you want to use when balancing a checkbook or calculating current through a resistor.
However, Claude Shannon of MIT fame recognized how Boolean algebra could be applied to
onandoff circuits, where all signals are characterized as either ”high” (1) or ”low” (0). His
1938 thesis, titled A Symbolic Analysis of Relay and Switching Circuits, put Boole’s theoretical
work to use in a way Boole never could have imagined, giving us a powerful mathematical tool
for designing and analyzing digital circuits.
7.2. BOOLEAN ARITHMETIC 175
In this chapter, you will find a lot of similarities between Boolean algebra and ”normal” al
gebra, the kind of algebra involving socalled real numbers. Just bear in mind that the system
of numbers defining Boolean algebra is severely limited in terms of scope, and that there can
only be one of two possible values for any Boolean variable: 1 or 0. Consequently, the ”Laws”
of Boolean algebra often differ from the ”Laws” of realnumber algebra, making possible such
statements as 1 + 1 = 1, which would normally be considered absurd. Once you comprehend
the premise of all quantities in Boolean algebra being limited to the two possibilities of 1 and
0, and the general philosophical principle of Laws depending on quantitative definitions, the
”nonsense” of Boolean algebra disappears.
It should be clearly understood that Boolean numbers are not the same as binary numbers.
Whereas Boolean numbers represent an entirely different system of mathematics from real
numbers, binary is nothing more than an alternative notation for real numbers. The two are
often confused because both Boolean math and binary notation use the same two ciphers: 1
and 0. The difference is that Boolean quantities are restricted to a single bit (either 1 or 0),
whereas binary numbers may be composed of many bits adding up in placeweighted form to
a value of any finite size. The binary number 100112 (”nineteen”) has no more place in the
Boolean world than the decimal number 210 (”two”) or the octal number 328 (”twentysix”).
7.2 Boolean arithmetic
Let us begin our exploration of Boolean algebra by adding numbers together:
0 + 0 = 0
0 + 1 = 1
1 + 0 = 1
1 + 1 = 1
The first three sums make perfect sense to anyone familiar with elementary addition. The
last sum, though, is quite possibly responsible for more confusion than any other single state
ment in digital electronics, because it seems to run contrary to the basic principles of math
ematics. Well, it does contradict principles of addition for real numbers, but not for Boolean
numbers. Remember that in the world of Boolean algebra, there are only two possible values
for any quantity and for any arithmetic operation: 1 or 0. There is no such thing as ”2” within
the scope of Boolean values. Since the sum ”1 + 1” certainly isn’t 0, it must be 1 by process of
elimination.
It does not matter how many or few terms we add together, either. Consider the following
sums:
0 + 1 + 1 = 1
0 + 1 + 1 + 1 = 1
1 + 0 + 1 + 1 + 1 = 1
1 + 1 + 1 = 1
Take a close look at the twoterm sums in the first set of equations. Does that pattern look
familiar to you? It should! It is the same pattern of 1’s and 0’s as seen in the truth table for an
176 CHAPTER 7. BOOLEAN ALGEBRA
OR gate. In other words, Boolean addition corresponds to the logical function of an ”OR” gate,
as well as to parallel switch contacts:
0 + 0 = 0
0
0
0
0 0
0
0
0
0 + 1 = 1
1
1
1
1
0
0
1
1 1
1
1 + 0 = 1
1
1 1
1
1 + 1 = 1
1
1
There is no such thing as subtraction in the realm of Boolean mathematics. Subtraction
implies the existence of negative numbers: 5  3 is the same thing as 5 + (3), and in Boolean
algebra negative quantities are forbidden. There is no such thing as division in Boolean mathe
matics, either, since division is really nothing more than compounded subtraction, in the same
way that multiplication is compounded addition.
Multiplication is valid in Boolean algebra, and thankfully it is the same as in realnumber
algebra: anything multiplied by 0 is 0, and anything multiplied by 1 remains unchanged:
0 × 0 = 0
0 × 1 = 0
1 × 0 = 0
1 × 1 = 1
This set of equations should also look familiar to you: it is the same pattern found in the
7.2. BOOLEAN ARITHMETIC 177
truth table for an AND gate. In other words, Boolean multiplication corresponds to the logical
function of an ”AND” gate, as well as to series switch contacts:
0 × 0 = 0
0
0
0
0 0 0
0
0
0 0
0 × 1 = 0
1
1
0 0
0 01 1
1 × 0 = 0
1 1
1
1
1 × 1 = 1
1 1
Like ”normal” algebra, Boolean algebra uses alphabetical letters to denote variables. Un
like ”normal” algebra, though, Boolean variables are always CAPITAL letters, never lower
case. Because they are allowed to possess only one of two possible values, either 1 or 0, each
and every variable has a complement: the opposite of its value. For example, if variable ”A” has
a value of 0, then the complement of A has a value of 1. Boolean notation uses a bar above the
variable character to denote complementation, like this:
If: A=0
Then: A=1
If:
Then:
A=1
A=0
In written form, the complement of ”A” denoted as ”Anot” or ”Abar”. Sometimes a ”prime”
symbol is used to represent complementation. For example, A’ would be the complement of A,
much the same as using a prime symbol to denote differentiation in calculus rather than the
fractional notation d/dt. Usually, though, the ”bar” symbol finds more widespread use than the
”prime” symbol, for reasons that will become more apparent later in this chapter.
Boolean complementation finds equivalency in the form of the NOT gate, or a normally
closed switch or relay contact:
178 CHAPTER 7. BOOLEAN ALGEBRA
1
If:
Then: A=1
A=0
0
A A 0 1
A A
1
If:
Then:
A=1
A=0
0
A A 01
A A
The basic definition of Boolean quantities has led to the simple rules of addition and mul
tiplication, and has excluded both subtraction and division as valid arithmetic operations. We
have a symbology for denoting Boolean variables, and their complements. In the next section
we will proceed to develop Boolean identities.
• REVIEW:
• Boolean addition is equivalent to the OR logic function, as well as parallel switch con
tacts.
• Boolean multiplication is equivalent to the AND logic function, as well as series switch
contacts.
• Boolean complementation is equivalent to the NOT logic function, as well as normally
closed relay contacts.
7.3 Boolean algebraic identities
In mathematics, an identity is a statement true for all possible values of its variable or vari
ables. The algebraic identity of x + 0 = x tells us that anything (x) added to zero equals the
original ”anything,” no matter what value that ”anything” (x) may be. Like ordinary algebra,
Boolean algebra has its own unique identities based on the bivalent states of Boolean variables.
The first Boolean identity is that the sum of anything and zero is the same as the original
”anything.” This identity is no different from its realnumber algebraic equivalent:
0
A
0
A + 0 = A
A
A A
7.3. BOOLEAN ALGEBRAIC IDENTITIES 179
No matter what the value of A, the output will always be the same: when A=1, the output
will also be 1; when A=0, the output will also be 0.
The next identity is most definitely different from any seen in normal algebra. Here we
discover that the sum of anything and one is one:
A
A
A + 1 = 1
1
1
1
1
No matter what the value of A, the sum of A and 1 will always be 1. In a sense, the ”1”
signal overrides the effect of A on the logic circuit, leaving the output fixed at a logic level of 1.
Next, we examine the effect of adding A and A together, which is the same as connecting
both inputs of an OR gate to each other and activating them with the same signal:
A
A
A
A + A = A
A
A
In realnumber algebra, the sum of two identical variables is twice the original variable’s
value (x + x = 2x), but remember that there is no concept of ”2” in the world of Boolean math,
only 1 and 0, so we cannot say that A + A = 2A. Thus, when we add a Boolean quantity to itself,
the sum is equal to the original quantity: 0 + 0 = 0, and 1 + 1 = 1.
Introducing the uniquely Boolean concept of complementation into an additive identity, we
find an interesting effect. Since there must be one ”1” value between any variable and its
complement, and since the sum of any Boolean quantity and 1 is 1, the sum of a variable and
its complement must be 1:
A
A
1
A
A + A = 1
1A
A
Just as there are four Boolean additive identities (A+0, A+1, A+A, and A+A’), so there are
also four multiplicative identities: Ax0, Ax1, AxA, and AxA’. Of these, the first two are no
different from their equivalent expressions in regular algebra:
180 CHAPTER 7. BOOLEAN ALGEBRA
A
A
0A = 0
0
0
0 0
A
A
1A = A
1
A
A1
The third multiplicative identity expresses the result of a Boolean quantity multiplied by
itself. In normal algebra, the product of a variable and itself is the square of that variable (3
x 3 = 32 = 9). However, the concept of ”square” implies a quantity of 2, which has no meaning
in Boolean algebra, so we cannot say that A x A = A2. Instead, we find that the product of a
Boolean quantity and itself is the original quantity, since 0 x 0 = 0 and 1 x 1 = 1:
A
A
A
A
AA = A
A
The fourth multiplicative identity has no equivalent in regular algebra because it uses the
complement of a variable, a concept unique to Boolean mathematics. Since there must be
one ”0” value between any variable and its complement, and since the product of any Boolean
quantity and 0 is 0, the product of a variable and its complement must be 0:
A
A A
AA = 0
0
A
A
0
To summarize, then, we have four basic Boolean identities for addition and four for multi
plication:
A + 0 = A
A + 1 = 1
A + A = A
A + A = 1
0A = 0
1A = A
AA = A
AA = 0
Additive Multiplicative
Basic Boolean algebraic identities
Another identity having to do with complementation is that of the double complement: a
variable inverted twice. Complementing a variable twice (or any even number of times) results
in the original Boolean value. This is analogous to negating (multiplying by 1) in realnumber
7.4. BOOLEAN ALGEBRAIC PROPERTIES 181
algebra: an even number of negations cancel to leave the original value:
A
AA
A = A
A
A
(same)
A
CR1
CR1 CR2
CR2
A
A
(same)
7.4 Boolean algebraic properties
Another type of mathematical identity, called a ”property” or a ”law,” describes how differing
variables relate to each other in a system of numbers. One of these properties is known as
the commutative property, and it applies equally to addition and multiplication. In essence,
the commutative property tells us we can reverse the order of variables that are either added
together or multiplied together without changing the truth of the expression:
Commutative property of addition
A + B = B + A
A
B
A
B
(same)
A
B
A
B
(same)
182 CHAPTER 7. BOOLEAN ALGEBRA
A
B
A
B
(same)
A B
AB
(same)
Commutative property of multiplication
AB = BA
Along with the commutative properties of addition and multiplication, we have the associa
tive property, again applying equally well to addition and multiplication. This property tells
us we can associate groups of added or multiplied variables together with parentheses without
altering the truth of the equations.
A
B
A
B
Associative property of addition
A + (B + C) = (A + B) + C
C
A
B
C
(same)
C
A
B
C
(same)
7.4. BOOLEAN ALGEBRAIC PROPERTIES 183
A
B
A
B
C
A
B
C
(same)
C
Associative property of multiplication
A(BC) = (AB)C
A B
C
(same)
Lastly, we have the distributive property, illustrating how to expand a Boolean expression
formed by the product of a sum, and in reverse shows us how terms may be factored out of
Boolean sumsofproducts:
A
B
A B
C
A
B
C
(same)
C
A B
C
(same)
Distributive property
A(B + C) = AB + AC
A A
To summarize, here are the three basic properties: commutative, associative, and distribu
tive.
184 CHAPTER 7. BOOLEAN ALGEBRA
Additive Multiplicative
Basic Boolean algebraic properties
A + (B + C) = (A + B) + C A(BC) = (AB)C
A + B = B + A AB = BA
A(B + C) = AB + AC
7.5 Boolean rules for simplification
Boolean algebra finds its most practical use in the simplification of logic circuits. If we translate
a logic circuit’s function into symbolic (Boolean) form, and apply certain algebraic rules to the
resulting equation to reduce the number of terms and/or arithmetic operations, the simplified
equation may be translated back into circuit form for a logic circuit performing the same func
tion with fewer components. If equivalent function may be achieved with fewer components,
the result will be increased reliability and decreased cost of manufacture.
To this end, there are several rules of Boolean algebra presented in this section for use in
reducing expressions to their simplest forms. The identities and properties already reviewed
in this chapter are very useful in Boolean simplification, and for the most part bear similarity
to many identities and properties of ”normal” algebra. However, the rules shown in this section
are all unique to Boolean mathematics.
A
B
(same) A
A B
(same)
A + AB = A
A
AB
A + AB
A
A
A
This rule may be proven symbolically by factoring an ”A” out of the two terms, then applying
the rules of A + 1 = 1 and 1A = A to achieve the final result:
7.5. BOOLEAN RULES FOR SIMPLIFICATION 185
A + AB
A(1 + B)
Applying identity A + 1 = 1
A(1)
Applying identity 1A = A
A
Factoring A out of both terms
Please note how the rule A + 1 = 1 was used to reduce the (B + 1) term to 1. When a rule
like ”A + 1 = 1” is expressed using the letter ”A”, it doesn’t mean it only applies to expressions
containing ”A”. What the ”A” stands for in a rule like A + 1 = 1 is any Boolean variable or
collection of variables. This is perhaps the most difficult concept for new students to master in
Boolean simplification: applying standardized identities, properties, and rules to expressions
not in standard form.
For instance, the Boolean expression ABC + 1 also reduces to 1 by means of the ”A + 1 =
1” identity. In this case, we recognize that the ”A” term in the identity’s standard form can
represent the entire ”ABC” term in the original expression.
The next rule looks similar to the first one shown in this section, but is actually quite
different and requires a more clever proof:
B (same)
A
A
B
(same)
A
AB
A + AB
A
A
A + AB = A + B
A
A + B
B
186 CHAPTER 7. BOOLEAN ALGEBRA
A + AB
Applying the previous rule to expand A term
A + AB = A
A + AB + AB
A + B(A + A)
Applying identity A + A = 1
A + B(1)
Applying identity 1A = A
A + B
Factoring B out of 2nd and 3rd terms
Note how the last rule (A + AB = A) is used to ”unsimplify” the first ”A” term in the expres
sion, changing the ”A” into an ”A + AB”. While this may seem like a backward step, it certainly
helped to reduce the expression to something simpler! Sometimes in mathematics we must
take ”backward” steps to achieve the most elegant solution. Knowing when to take such a step
and when not to is part of the artform of algebra, just as a victory in a game of chess almost
always requires calculated sacrifices.
Another rule involves the simplification of a productofsums expression:
A
A
B
(same)
A
(A + B)(A + C) = A + BC
B
A
C
A+B
A+C
A
B
C
BC
(A+B)(A+C)
A + BC
(same)
A
C
B C
And, the corresponding proof:
7.6. CIRCUIT SIMPLIFICATION EXAMPLES 187
(A + B)(A + C)
Distributing terms
AA + AC + AB + BC
Applying identity AA = A
A + AC + AB + BC
Applying rule A + AB = A
to the A + AC term
A + AB + BC
Applying rule A + AB = A
to the A + AB term
A + BC
To summarize, here are the three new rules of Boolean simplification expounded in this
section:
Useful Boolean rules for simplification
A + AB = A
A + AB = A + B
(A + B)(A + C) = A + BC
7.6 Circuit simplification examples
Let’s begin with a semiconductor gate circuit in need of simplification. The ”A,” ”B,” and ”C”
input signals are assumed to be provided from switches, sensors, or perhaps other gate circuits.
Where these signals originate is of no concern in the task of gate reduction.
A
B
C
Q
Our first step in simplification must be to write a Boolean expression for this circuit. This
task is easily performed step by step if we start by writing subexpressions at the output of each
gate, corresponding to the respective input signals for each gate. Remember that OR gates are
equivalent to Boolean addition, while AND gates are equivalent to Boolean multiplication. For
example, I’ll write subexpressions at the outputs of the first three gates:
188 CHAPTER 7. BOOLEAN ALGEBRA
A
B
C
Q
AB
B+C
BC
. . . then another subexpression for the next gate:
A
B
C
Q
AB
B+C
BC
BC(B+C)
Finally, the output (”Q”) is seen to be equal to the expression AB + BC(B + C):
A
B
C
AB
B+C
BC
BC(B+C)
Q = AB + BC(B+C)
Now that we have a Boolean expression to work with, we need to apply the rules of Boolean
algebra to reduce the expression to its simplest form (simplest defined as requiring the fewest
gates to implement):
7.6. CIRCUIT SIMPLIFICATION EXAMPLES 189
Distributing terms
Applying identity AA = A
AB + BBC + BCC
AB + BC + BC
Applying identity A + A = A
to 2nd and 3rd terms
to 2nd and 3rd terms
AB + BC
Factoring B out of terms
B(A + C)
AB + BC(B + C)
The final expression, B(A + C), is much simpler than the original, yet performs the same
function. If you would like to verify this, you may generate a truth table for both expressions
and determine Q’s status (the circuits’ output) for all eight logicstate combinations of A, B,
and C, for both circuits. The two truth tables should be identical.
Now, we must generate a schematic diagram from this Boolean expression. To do this,
evaluate the expression, following proper mathematical order of operations (multiplication be
fore addition, operations inside parentheses before anything else), and draw gates for each
step. Remember again that OR gates are equivalent to Boolean addition, while AND gates are
equivalent to Boolean multiplication. In this case, we would begin with the subexpression ”A
+ C”, which is an OR gate:
A
C
A+C
The next step in evaluating the expression ”B(A + C)” is to multiply (AND gate) the signal
B by the output of the previous gate (A + C):
A
C
A+C
B
Q = B(A+C)
Obviously, this circuit is much simpler than the original, having only two logic gates instead
of five. Such component reduction results in higher operating speed (less delay time from input
signal transition to output signal transition), less power consumption, less cost, and greater
reliability.
Electromechanical relay circuits, typically being slower, consuming more electrical power to
operate, costing more, and having a shorter average life than their semiconductor counterparts,
benefit dramatically from Boolean simplification. Let’s consider an example circuit:
190 CHAPTER 7. BOOLEAN ALGEBRA
L1 L2
QA
B A
C
A C
As before, our first step in reducing this circuit to its simplest form must be to develop a
Boolean expression from the schematic. The easiest way I’ve found to do this is to follow the
same steps I’d normally follow to reduce a seriesparallel resistor network to a single, total
resistance. For example, examine the following resistor network with its resistors arranged
in the same connection pattern as the relay contacts in the former circuit, and corresponding
total resistance formula:
R1
R2 R3
R4
R5 R6
Rtotal
Rtotal = R1 // [(R3//R4)  R2] // (R5  R6)
Remember that parallel contacts are equivalent to Boolean addition, while series contacts
are equivalent to Boolean multiplication. Write a Boolean expression for this relay contact cir
cuit, following the same order of precedence that you would follow in reducing a seriesparallel
resistor network to a total resistance. It may be helpful to write a Boolean subexpression to
the left of each ladder ”rung,” to help organize your expressionwriting:
7.6. CIRCUIT SIMPLIFICATION EXAMPLES 191
L1 L2
QA
B A
C
A C
Q = A + B(A+C) + AC
A
B(A+C)
AC
Now that we have a Boolean expression to work with, we need to apply the rules of Boolean
algebra to reduce the expression to its simplest form (simplest defined as requiring the fewest
relay contacts to implement):
Distributing terms
A + B(A + C) + AC
A + AB + BC + AC
Applying rule A + AB = A
to 1st and 2nd terms
A + BC + AC
Applying rule A + AB = A
to 1st and 3rd terms
A + BC
The more mathematically inclined should be able to see that the two steps employing the
rule ”A + AB = A” may be combined into a single step, the rule being expandable to: ”A + AB +
AC + AD + . . . = A”
Distributing terms
A + B(A + C) + AC
A + AB + BC + AC
A + BC
Applying (expanded) rule A + AB = A
to 1st, 2nd, and 4th terms
As you can see, the reduced circuit is much simpler than the original, yet performs the same
logical function:
192 CHAPTER 7. BOOLEAN ALGEBRA
L1 L2
QA
CB
A + BCA
BC
• REVIEW:
• To convert a gate circuit to a Boolean expression, label each gate output with a Boolean
subexpression corresponding to the gates’ input signals, until a final expression is reached
at the last gate.
• To convert a Boolean expression to a gate circuit, evaluate the expression using standard
order of operations: multiplication before addition, and operations within parentheses
before anything else.
• To convert a ladder logic circuit to a Boolean expression, label each rung with a Boolean
subexpression corresponding to the contacts’ input signals, until a final expression is
reached at the last coil or light. To determine proper order of evaluation, treat the con
tacts as though they were resistors, and as if you were determining total resistance of the
seriesparallel network formed by them. In other words, look for contacts that are either
directly in series or directly in parallel with each other first, then ”collapse” them into
equivalent Boolean subexpressions before proceeding to other contacts.
• To convert a Boolean expression to a ladder logic circuit, evaluate the expression us
ing standard order of operations: multiplication before addition, and operations within
parentheses before anything else.
7.7 The ExclusiveOR function
One element conspicuously missing from the set of Boolean operations is that of Exclusive
OR. Whereas the OR function is equivalent to Boolean addition, the AND function to Boolean
multiplication, and the NOT function (inverter) to Boolean complementation, there is no direct
Boolean equivalent for ExclusiveOR. This hasn’t stopped people from developing a symbol to
represent it, though:
A
B
A ⊕ B
This symbol is seldom used in Boolean expressions because the identities, laws, and rules
of simplification involving addition, multiplication, and complementation do not apply to it.
However, there is a way to represent the ExclusiveOR function in terms of OR and AND, as
has been shown in previous chapters: AB’ + A’B
7.8. DEMORGAN’S THEOREMS 193
A
B
A ⊕ B
A
B
. . . is equivalent to . . .
AB
AB
AB + AB
A ⊕ B = AB + AB
As a Boolean equivalency, this rule may be helpful in simplifying some Boolean expressions.
Any expression following the AB’ + A’B form (two AND gates and an OR gate) may be replaced
by a single ExclusiveOR gate.
7.8 DeMorgan’s Theorems
A mathematician named DeMorgan developed a pair of important rules regarding group com
plementation in Boolean algebra. By group complementation, I’m referring to the complement
of a group of terms, represented by a long bar over more than one variable.
You should recall from the chapter on logic gates that inverting all inputs to a gate reverses
that gate’s essential function from AND to OR, or vice versa, and also inverts the output. So,
an OR gate with all inputs inverted (a NegativeOR gate) behaves the same as a NAND gate,
and an AND gate with all inputs inverted (a NegativeAND gate) behaves the same as a NOR
gate. DeMorgan’s theorems state the same equivalence in ”backward” form: that inverting the
output of any gate results in the same function as the opposite type of gate (AND vs. OR) with
inverted inputs:
194 CHAPTER 7. BOOLEAN ALGEBRA
A
B
AB
AB
. . . is equivalent to . . .
A
B
A
B
A + B
AB = A + B
A long bar extending over the term AB acts as a grouping symbol, and as such is entirely
different from the product of A and B independently inverted. In other words, (AB)’ is not
equal to A’B’. Because the ”prime” symbol (’) cannot be stretched over two variables like a bar
can, we are forced to use parentheses to make it apply to the whole term AB in the previous
sentence. A bar, however, acts as its own grouping symbol when stretched over more than one
variable. This has profound impact on how Boolean expressions are evaluated and reduced, as
we shall see.
DeMorgan’s theorem may be thought of in terms of breaking a long bar symbol. When
a long bar is broken, the operation directly underneath the break changes from addition to
multiplication, or vice versa, and the broken bar pieces remain over the individual variables.
To illustrate:
AB
break!
A + B
A + B
break!
AB
NAND to NegativeOR NOR to NegativeAND
DeMorgan’s Theorems
When multiple ”layers” of bars exist in an expression, you may only break one bar at a time,
and it is generally easier to begin simplification by breaking the longest (uppermost) bar first.
To illustrate, let’s take the expression (A + (BC)’)’ and reduce it using DeMorgan’s Theorems:
7.8. DEMORGAN’S THEOREMS 195
B
C
BC
A
A
A + BC
Following the advice of breaking the longest (uppermost) bar first, I’ll begin by breaking the
bar covering the entire expression as a first step:
A + BC
Breaking longest bar
A BC
Applying identity A = A
to BC
ABC
(addition changes to multiplication)
As a result, the original circuit is reduced to a threeinput AND gate with the A input
inverted:
A
B
C
A
ABC
You should never break more than one bar in a single step, as illustrated here:
A + BC
Applying identity A = A
Breaking long bar between A and B;
Breaking both bars between B and C
A B + C
Incorrect step!
to B and C
AB + CIncorrect answer:
As tempting as it may be to conserve steps and break more than one bar at a time, it often
leads to an incorrect result, so don’t do it!
It is possible to properly reduce this expression by breaking the short bar first, rather than
the long bar first:
196 CHAPTER 7. BOOLEAN ALGEBRA
A + BC
Breaking shortest bar
(multiplication changes to addition)
A + (B + C)
Applying associative property
to remove parentheses
A + B + C
Breaking long bar in two places,
between 1st and 2nd terms;
between 2nd and 3rd terms
A B C
Applying identity A = A
ABC
to B and C
The end result is the same, but more steps are required compared to using the first method,
where the longest bar was broken first. Note how in the third step we broke the long bar in two
places. This is a legitimate mathematical operation, and not the same as breaking two bars in
one step! The prohibition against breaking more than one bar in one step is not a prohibition
against breaking a bar in more than one place. Breaking in more than one place in a single
step is okay; breaking more than one bar in a single step is not.
You might be wondering why parentheses were placed around the subexpression B’ + C’,
considering the fact that I just removed them in the next step. I did this to emphasize an
important but easily neglected aspect of DeMorgan’s theorem. Since a long bar functions as
a grouping symbol, the variables formerly grouped by a broken bar must remain grouped lest
proper precedence (order of operation) be lost. In this example, it really wouldn’t matter if I
forgot to put parentheses in after breaking the short bar, but in other cases it might. Consider
this example, starting with a different expression:
AB + CD
Breaking bar in middle
(AB)(CD)
Breaking both bars in middle
(A + B)(C + D)
Notice the grouping maintained
with parentheses
Correct answer:
7.8. DEMORGAN’S THEOREMS 197
AB + CD
Breaking bar in middle
Breaking both bars in middle
AB CD
A + BC + DIncorrect answer:
Parentheses omitted
As you can see, maintaining the grouping implied by the complementation bars for this
expression is crucial to obtaining the correct answer.
Let’s apply the principles of DeMorgan’s theorems to the simplification of a gate circuit:
Q
A
B
C
As always, our first step in simplifying this circuit must be to generate an equivalent
Boolean expression. We can do this by placing a subexpression label at the output of each
gate, as the inputs become known. Here’s the first step in this process:
Q
A
B
C BC
B
Next, we can label the outputs of the first NOR gate and the NAND gate. When dealing
with invertedoutput gates, I find it easier to write an expression for the gate’s output without
the final inversion, with an arrow pointing to just before the inversion bubble. Then, at the
wire leading out of the gate (after the bubble), I write the full, complemented expression. This
helps ensure I don’t forget a complementing bar in the subexpression, by forcing myself to
split the expressionwriting task into two steps:
198 CHAPTER 7. BOOLEAN ALGEBRA
Q
A
B
C BC
B
A
A+BC
A+BC
AB
AB
Finally, we write an expression (or pair of expressions) for the last NOR gate:
A
B
C BC
B
A
A+BC
A+BC
AB
AB A+BC + AB
Q = A+BC + AB
Now, we reduce this expression using the identities, properties, rules, and theorems (De
Morgan’s) of Boolean algebra:
7.8. DEMORGAN’S THEOREMS 199
Applying identity A = A
A + BC + AB
Breaking longest bar
(A + BC) (AB)
wherever double bars of
equal length are found
(A + BC)(AB)
Distributive property
AAB + BCAB
Applying identity AA = A
to left term; applying identity
AA = 0 to B and B in right
term
AB + 0
Applying identity A + 0 = A
AB
The equivalent gate circuit for this muchsimplified expression is as follows:
A
B
Q = AB
• REVIEW
• DeMorgan’s Theorems describe the equivalence between gates with inverted inputs and
gates with inverted outputs. Simply put, a NAND gate is equivalent to a NegativeOR
gate, and a NOR gate is equivalent to a NegativeAND gate.
• When ”breaking” a complementation bar in a Boolean expression, the operation directly
underneath the break (addition or multiplication) reverses, and the broken bar pieces
remain over the respective terms.
• It is often easier to approach a problem by breaking the longest (uppermost) bar before
breaking any bars under it. You must never attempt to break two bars in one step!
• Complementation bars function as grouping symbols. Therefore, when a bar is broken,
the terms underneath it must remain grouped. Parentheses may be placed around these
grouped terms as a help to avoid changing precedence.
200 CHAPTER 7. BOOLEAN ALGEBRA
7.9 Converting truth tables into Boolean expressions
In designing digital circuits, the designer often begins with a truth table describing what the
circuit should do. The design task is largely to determine what type of circuit will perform the
function described in the truth table. While some people seem to have a natural ability to look
at a truth table and immediately envision the necessary logic gate or relay logic circuitry for
the task, there are procedural techniques available for the rest of us. Here, Boolean algebra
proves its utility in a most dramatic way.
To illustrate this procedural method, we should begin with a realistic design problem. Sup
pose we were given the task of designing a flame detection circuit for a toxic waste incinerator.
The intense heat of the fire is intended to neutralize the toxicity of the waste introduced into
the incinerator. Such combustionbased techniques are commonly used to neutralize medical
waste, which may be infected with deadly viruses or bacteria:
flame
Toxic waste
inlet
Fuel
inletExhaust
Toxic waste incinerator
So long as a flame is maintained in the incinerator, it is safe to inject waste into it to be
neutralized. If the flame were to be extinguished, however, it would be unsafe to continue to
inject waste into the combustion chamber, as it would exit the exhaust unneutralized, and
pose a health threat to anyone in close proximity to the exhaust. What we need in this system
is a sure way of detecting the presence of a flame, and permitting waste to be injected only if a
flame is ”proven” by the flame detection system.
Several different flamedetection technologies exist: optical (detection of light), thermal
(detection of high temperature), and electrical conduction (detection of ionized particles in the
flame path), each one with its unique advantages and disadvantages. Suppose that due to the
high degree of hazard involved with potentially passing unneutralized waste out the exhaust
of this incinerator, it is decided that the flame detection system be made redundant (multiple
sensors), so that failure of a single sensor does not lead to an emission of toxins out the exhaust.
Each sensor comes equipped with a normallyopen contact (open if no flame, closed if flame
detected) which we will use to activate the inputs of a logic system:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 201
flame
Toxic waste
inlet
Fuel
inletExhaust
Toxic waste incinerator
sensor sensor sensor
Logic system
(shuts off waste valve
if no flame detected)
Waste shutoff
valve
A B C
Our task, now, is to design the circuitry of the logic system to open the waste valve if and
only if there is good flame proven by the sensors. First, though, we must decide what the logical
behavior of this control system should be. Do we want the valve to be opened if only one out of
the three sensors detects flame? Probably not, because this would defeat the purpose of having
multiple sensors. If any one of the sensors were to fail in such a way as to falsely indicate the
presence of flame when there was none, a logic system based on the principle of ”any one out
of three sensors showing flame” would give the same output that a singlesensor system would
with the same failure. A far better solution would be to design the system so that the valve is
commanded to open if and only if all three sensors detect a good flame. This way, any single,
failed sensor falsely showing flame could not keep the valve in the open position; rather, it
would require all three sensors to be failed in the same manner – a highly improbable scenario
– for this dangerous condition to occur.
Thus, our truth table would look like this:
202 CHAPTER 7. BOOLEAN ALGEBRA
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
It does not require much insight to realize that this functionality could be generated with a
threeinput AND gate: the output of the circuit will be ”high” if and only if input A AND input
B AND input C are all ”high:”
flame
Toxic waste
inlet
Fuel
inletExhaust
Toxic waste incinerator
sensor sensor sensor
Waste shutoff
valve
A B C
Vdd
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 203
If using relay circuitry, we could create this AND function by wiring three relay contacts in
series, or simply by wiring the three sensor contacts in series, so that the only way electrical
power could be sent to open the waste valve is if all three sensors indicate flame:
flame
Toxic waste
inlet
Fuel
inletExhaust
Toxic waste incinerator
sensor sensor sensor
Waste shutoff
valve
A B C
L1
While this design strategy maximizes safety, it makes the system very susceptible to sensor
failures of the opposite kind. Suppose that one of the three sensors were to fail in such a way
that it indicated no flame when there really was a good flame in the incinerator’s combustion
chamber. That single failure would shut off the waste valve unnecessarily, resulting in lost
production time and wasted fuel (feeding a fire that wasn’t being used to incinerate waste).
It would be nice to have a logic system that allowed for this kind of failure without shutting
the system down unnecessarily, yet still provide sensor redundancy so as to maintain safety
in the event that any single sensor failed ”high” (showing flame at all times, whether or not
there was one to detect). A strategy that would meet both needs would be a ”two out of three”
sensor logic, whereby the waste valve is opened if at least two out of the three sensors show
good flame. The truth table for such a system would look like this:
204 CHAPTER 7. BOOLEAN ALGEBRA
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Here, it is not necessarily obvious what kind of logic circuit would satisfy the truth ta
ble. However, a simple method for designing such a circuit is found in a standard form of
Boolean expression called the SumOfProducts, or SOP, form. As you might suspect, a Sum
OfProducts Boolean expression is literally a set of Boolean terms added (summed) together,
each term being a multiplicative (product) combination of Boolean variables. An example of
an SOP expression would be something like this: ABC + BC + DF, the sum of products ”ABC,”
”BC,” and ”DF.”
SumOfProducts expressions are easy to generate from truth tables. All we have to do
is examine the truth table for any rows where the output is ”high” (1), and write a Boolean
product term that would equal a value of 1 given those input conditions. For instance, in the
fourth row down in the truth table for our twooutofthree logic system, where A=0, B=1, and
C=1, the product term would be A’BC, since that term would have a value of 1 if and only if
A=0, B=1, and C=1:
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
1
1
1 ABC = 1
Three other rows of the truth table have an output value of 1, so those rows also need
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 205
Boolean product expressions to represent them:
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
1
1
1 ABC = 1
ABC = 1
ABC = 1
ABC = 1
Finally, we join these four Boolean product expressions together by addition, to create a
single Boolean expression describing the truth table as a whole:
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
1
1
1 ABC = 1
ABC = 1
ABC = 1
ABC = 1
Output = ABC + ABC + ABC + ABC
Now that we have a Boolean SumOfProducts expression for the truth table’s function, we
can easily design a logic gate or relay logic circuit based on that expression:
206 CHAPTER 7. BOOLEAN ALGEBRA
Output = ABC + ABC + ABC + ABC
A B C
ABC
ABC
ABC
ABC
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 207
Output = ABC + ABC + ABC + ABC
A
B
C
ABC
ABC
ABC
L1 L2
CR1
CR2
CR3
Output
ABC
CR1 CR2 CR3
CR1 CR2 CR3
CR1 CR2 CR3
CR1 CR2 CR3
Unfortunately, both of these circuits are quite complex, and could benefit from simplifica
tion. Using Boolean algebra techniques, the expression may be significantly simplified:
208 CHAPTER 7. BOOLEAN ALGEBRA
ABC + ABC + ABC + ABC
Factoring BC out of 1st and 4th terms
BC(A + A) + ABC + ABC
Applying identity A + A = 1
BC(1) + ABC + ABC
Applying identity 1A = A
BC + ABC + ABC
Factoring B out of 1st and 3rd terms
B(C + AC) + ABC
Applying rule A + AB = A + B to
the C + AC term
B(C + A) + ABC
Distributing terms
BC + AB + ABC
Factoring A out of 2nd and 3rd terms
BC + A(B + BC)
Applying rule A + AB = A + B to
the B + BC term
BC + A(B + C)
Distributing terms
BC + AB + AC
or
AB + BC + AC
Simplified result
As a result of the simplification, we can now build much simpler logic circuits performing
the same function, in either gate or relay form:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 209
A B C
Output = AB + BC + AC
AB
BC
AC
A
B
C
L1 L2
CR1
CR2
CR3
OutputCR1 CR2
CR2 CR3
CR1 CR3
AB
BC
AC
Output = AB + BC + AC
Either one of these circuits will adequately perform the task of operating the incinerator
waste valve based on a flame verification from two out of the three flame sensors. At minimum,
this is what we need to have a safe incinerator system. We can, however, extend the function
ality of the system by adding to it logic circuitry designed to detect if any one of the sensors
210 CHAPTER 7. BOOLEAN ALGEBRA
does not agree with the other two.
If all three sensors are operating properly, they should detect flame with equal accuracy.
Thus, they should either all register ”low” (000: no flame) or all register ”high” (111: good
flame). Any other output combination (001, 010, 011, 100, 101, or 110) constitutes a disagree
ment between sensors, and may therefore serve as an indicator of a potential sensor failure. If
we added circuitry to detect any one of the six ”sensor disagreement” conditions, we could use
the output of that circuitry to activate an alarm. Whoever is monitoring the incinerator would
then exercise judgment in either continuing to operate with a possible failed sensor (inputs:
011, 101, or 110), or shut the incinerator down to be absolutely safe. Also, if the incinerator
is shut down (no flame), and one or more of the sensors still indicates flame (001, 010, 011,
100, 101, or 110) while the other(s) indicate(s) no flame, it will be known that a definite sensor
problem exists.
The first step in designing this ”sensor disagreement” detection circuit is to write a truth
table describing its behavior. Since we already have a truth table describing the output of the
”good flame” logic circuit, we can simply add another output column to the table to represent
the second circuit, and make a table representing the entire logic system:
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Good
flame
0
0
1
1
1
1
1
1
Output
Sensor
disagreement
Output = 0
(sensors agree)
Output = 1
(sensors disagree)
While it is possible to generate a SumOfProducts expression for this new truth table col
umn, it would require six terms, of three variables each! Such a Boolean expression would
require many steps to simplify, with a large potential for making algebraic errors:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 211
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Good
flame
0
0
1
1
1
1
1
1
Output
Sensor
disagreement
Output = 0
(sensors agree)
Output = 1
(sensors disagree)
ABC
ABC
ABC
ABC
ABC
ABC
Output = ABC + ABC + ABC + ABC + ABC + ABC
An alternative to generating a SumOfProducts expression to account for all the ”high”
(1) output conditions in the truth table is to generate a ProductOfSums, or POS, expression,
to account for all the ”low” (0) output conditions instead. Being that there are much fewer
instances of a ”low” output in the last truth table column, the resulting ProductOfSums ex
pression should contain fewer terms. As its name suggests, a ProductOfSums expression is
a set of added terms (sums), which are multiplied (product) together. An example of a POS
expression would be (A + B)(C + D), the product of the sums ”A + B” and ”C + D”.
To begin, we identify which rows in the last truth table column have ”low” (0) outputs, and
write a Boolean sum term that would equal 0 for that row’s input conditions. For instance, in
the first row of the truth table, where A=0, B=0, and C=0, the sum term would be (A + B + C),
since that term would have a value of 0 if and only if A=0, B=0, and C=0:
212 CHAPTER 7. BOOLEAN ALGEBRA
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Good
flame
0
0
1
1
1
1
1
1
Output
Sensor
disagreement
Output = 0
(sensors agree)
Output = 1
(sensors disagree)
(A + B + C)
Only one other row in the last truth table column has a ”low” (0) output, so all we need is one
more sum term to complete our ProductOfSums expression. This last sum term represents a
0 output for an input condition of A=1, B=1 and C=1. Therefore, the term must be written as
(A’ + B’+ C’), because only the sum of the complemented input variables would equal 0 for that
condition only:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 213
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Good
flame
0
0
1
1
1
1
1
1
Output
Sensor
disagreement
Output = 0
(sensors agree)
Output = 1
(sensors disagree)
(A + B + C)
(A + B + C)
The completed ProductOfSums expression, of course, is the multiplicative combination of
these two sum terms:
BA C Output
000
100
0 1 0
10 1
01 0
101
1 1 0
11 1 1
0
0
0
0
sensor
inputs
Output = 1
(open valve)
Output = 0
(close valve)
1
1
1
Good
flame
0
0
1
1
1
1
1
1
Output
Sensor
disagreement
Output = 0
(sensors agree)
Output = 1
(sensors disagree)
(A + B + C)
(A + B + C)
Output = (A + B + C)(A + B + C)
214 CHAPTER 7. BOOLEAN ALGEBRA
Whereas a SumOfProducts expression could be implemented in the form of a set of AND
gates with their outputs connecting to a single OR gate, a ProductOfSums expression can be
implemented as a set of OR gates feeding into a single AND gate:
A B C
Output = (A + B + C)(A + B + C)
(A + B + C)
(A + B + C)
Correspondingly, whereas a SumOfProducts expression could be implemented as a paral
lel collection of seriesconnected relay contacts, a ProductOfSums expression can be imple
mented as a series collection of parallelconnected relay contacts:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 215
A
B
C
L1 L2
CR1
CR2
CR3
Output
CR2CR2
CR3
Output = (A + B + C)(A + B + C)
CR3
CR1 CR1
(A + B + C) (A + B + C)
The previous two circuits represent different versions of the ”sensor disagreement” logic
circuit only, not the ”good flame” detection circuit(s). The entire logic system would be the com
bination of both ”good flame” and ”sensor disagreement” circuits, shown on the same diagram.
Implemented in a Programmable Logic Controller (PLC), the entire logic system might
resemble something like this:
216 CHAPTER 7. BOOLEAN ALGEBRA
PLC
X1
X2
X3
X4
X5
X6
L1 L2 Y1
Y2
Y3
Y4
Y5
Y6
Programming
port
Common Source
L1 L2
Personal
computer
Programming
cableX1 Y1
display
Sensor
A
Sensor
Sensor
B
C
Waste valve
solenoid
Sensor
disagreement
alarm lamp
X2
X1
X2 X3
X3
X1
X2
X3
X1
X2
X3
Y2
As you can see, both the SumOfProducts and ProductsOfSums standard Boolean forms
are powerful tools when applied to truth tables. They allow us to derive a Boolean expression
– and ultimately, an actual logic circuit – from nothing but a truth table, which is a written
specification for what we want a logic circuit to do. To be able to go from a written specification
to an actual circuit using simple, deterministic procedures means that it is possible to automate
the design process for a digital circuit. In other words, a computer could be programmed to
design a custom logic circuit from a truth table specification! The steps to take from a truth
table to the final circuit are so unambiguous and direct that it requires little, if any, creativity
or other original thought to execute them.
• REVIEW:
7.9. CONVERTING TRUTH TABLES INTO BOOLEAN EXPRESSIONS 217
• SumOfProducts, or SOP, Boolean expressions may be generated from truth tables quite
easily, by determining which rows of the table have an output of 1, writing one product
term for each row, and finally summing all the product terms. This creates a Boolean
expression representing the truth table as a whole.
• SumOfProducts expressions lend themselves well to implementation as a set of AND
gates (products) feeding into a single OR gate (sum).
• ProductOfSums, or POS, Boolean expressions may also be generated from truth tables
quite easily, by determining which rows of the table have an output of 0, writing one
sum term for each row, and finally multiplying all the sum terms. This creates a Boolean
expression representing the truth table as a whole.
• ProductOfSums expressions lend themselves well to implementation as a set of OR
gates (sums) feeding into a single AND gate (product).
218 CHAPTER 7. BOOLEAN ALGEBRA
Chapter 8
KARNAUGH MAPPING
Contents
8.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219
8.2 Venn diagrams and sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220
8.3 Boolean Relationships on Venn Diagrams . . . . . . . . . . . . . . . . . . . 223
8.4 Making a Venn diagram look like a Karnaugh map . . . . . . . . . . . . . 228
8.5 Karnaugh maps, truth tables, and Boolean expressions . . . . . . . . . . 231
8.6 Logic simplification with Karnaugh maps . . . . . . . . . . . . . . . . . . . 238
8.7 Larger 4variable Karnaugh maps . . . . . . . . . . . . . . . . . . . . . . . . 245
8.8 Minterm vs maxterm solution . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
8.9 Σ (sum) and Π (product) notation . . . . . . . . . . . . . . . . . . . . . . . . . 261
8.10 Don’t care cells in the Karnaugh map . . . . . . . . . . . . . . . . . . . . . . 262
8.11 Larger 5 & 6variable Karnaugh maps . . . . . . . . . . . . . . . . . . . . . . 265
Original author: Dennis Crunkilton
A
C BC BC BC
A
0
1
00 01 1110B A BC
8.1 Introduction
Why learn about Karnaughmaps? The Karnaugh map, like Boolean algebra, is a simplification
tool applicable to digital logic. See the ”Toxic waste incinerator” in the Boolean algebra chapter
for an example of Boolean simplification of digital logic. The Karnaugh Map will simplify logic
faster and more easily in most cases.
219
220 CHAPTER 8. KARNAUGH MAPPING
Boolean simplification is actually faster than the Karnaugh map for a task involving two
or fewer Boolean variables. It is still quite usable at three variables, but a bit slower. At four
input variables, Boolean algebra becomes tedious. Karnaugh maps are both faster and easier.
Karnaugh maps work well for up to six input variables, are usable for up to eight variables.
For more than six to eight variables, simplification should be by CAD (computer automated
design).
Variables Boolean algebra Karnaugh map computer automated
Recommended logic simplification vs number of inputs
12
3
4
56
78
>8
X
X X
X
X
X
X
?
?
X
?
?
?
In theory any of the three methods will work. However, as a practical matter, the above
guidelines work well. We would not normally resort to computer automation to simplify a
three input logic block. We could sooner solve the problem with pencil and paper. However,
if we had seven of these problems to solve, say for a BCD (Binary Coded Decimal) to seven
segment decoder, we might want to automate the process. A BCD to seven segment decoder
generates the logic signals to drive a seven segment LED (light emitting diode) display.
Examples of computer automated design languages for simplification of logic are PALASM,
ABEL, CUPL, Verilog, and VHDL. These programs accept a hardware descriptor language
input file which is based on Boolean equations and produce an output file describing a reduced
(or simplified) Boolean solution. We will not require such tools in this chapter. Let’s move on
to Venn diagrams as an introduction to Karnaugh maps.
8.2 Venn diagrams and sets
Mathematicians use Venn diagrams to show the logical relationships of sets (collections of
objects) to one another. Perhaps you have already seen Venn diagrams in your algebra or other
mathematics studies. If you have, you may remember overlapping circles and the union and
intersection of sets. We will review the overlapping circles of the Venn diagram. We will adopt
the terms OR and AND instead of union and intersection since that is the terminology used in
digital electronics.
The Venn diagram bridges the Boolean algebra from a previous chapter to the Karnaugh
Map. We will relate what you already know about Boolean algebra to Venn diagrams, then
transition to Karnaugh maps.
A set is a collection of objects out of a universe as shown below. The members of the set
are the objects contained within the set. The members of the set usually have something in
common; though, this is not a requirement. Out of the universe of real numbers, for example,
the set of all positive integers {1,2,3...} is a set. The set {3,4,5} is an example of a smaller set,
or subset of the set of all positive integers. Another example is the set of all males out of the
universe of college students. Can you think of some more examples of sets?
8.2. VENN DIAGRAMS AND SETS 221
A
U B
B
A
A
Above left, we have a Venn diagram showing the set A in the circle within the universe U,
the rectangular area. If everything inside the circle is A, then anything outside of the circle is
not A. Thus, above center, we label the rectangular area outside of the circle A as Anot instead
of U. We show B and Bnot in a similar manner.
What happens if both A and B are contained within the same universe? We show four
possibilities.
BA
A
B
A B A
B
Let’s take a closer look at each of the the four possibilities as shown above.
A B
The first example shows that set A and set B have nothing in common according to the
Venn diagram. There is no overlap between the A and B circular hatched regions. For example,
suppose that sets A and B contain the following members:
set A = {1,2,3,4} set B = {5,6,7,8}
None of the members of set A are contained within set B, nor are any of the members of B
contained within A. Thus, there is no overlap of the circles.
222 CHAPTER 8. KARNAUGH MAPPING
A
B
In the second example in the above Venn diagram, Set A is totally contained within set B
How can we explain this situation? Suppose that sets A and B contain the following members:
set A = {1,2}
set B = {1,2,3,4,5,6,7,8}
All members of set A are also members of set B. Therefore, set A is a subset of Set B. Since
all members of set A are members of set B, set A is drawn fully within the boundary of set B.
There is a fifth case, not shown, with the four examples. Hint: it is similar to the last
(fourth) example. Draw a Venn diagram for this fifth case.
A B
The third example above shows perfect overlap between set A and set B. It looks like both
sets contain the same identical members. Suppose that sets A and B contain the following:
set A = {1,2,3,4} set B = {1,2,3,4}
Therefore,
Set A = Set B
Sets And B are identically equal because they both have the same identical members. The
A and B regions within the corresponding Venn diagram above overlap completely. If there is
any doubt about what the above patterns represent, refer to any figure above or below to be
sure of what the circular regions looked like before they were overlapped.
8.3. BOOLEAN RELATIONSHIPS ON VENN DIAGRAMS 223
B
A
The fourth example above shows that there is something in common between set A and set
B in the overlapping region. For example, we arbitrarily select the following sets to illustrate
our point:
set A = {1,2,3,4} set B = {3,4,5,6}
Set A and Set B both have the elements 3 and 4 in common These elements are the reason
for the overlap in the center common to A and B. We need to take a closer look at this situation
8.3 Boolean Relationships on Venn Diagrams
The fourth example has A partially overlapping B. Though, we will first look at the whole
of all hatched area below, then later only the overlapping region. Let’s assign some Boolean
expressions to the regions above as shown below. Below left there is a red horizontal hatched
area for A. There is a blue vertical hatched area for B.
A+B
A+B
B
A
If we look at the whole area of both, regardless of the hatch style, the sum total of all hatched
areas, we get the illustration above right which corresponds to the inclusive OR function of A,
B. The Boolean expression is A+B. This is shown by the 45o hatched area. Anything outside of
the hatched area corresponds to (A+B)not as shown above. Let’s move on to next part of the
fourth example
The other way of looking at a Venn diagram with overlapping circles is to look at just the
part common to both A and B, the double hatched area below left. The Boolean expression for
this common area corresponding to the AND function is AB as shown below right. Note that
everything outside of double hatched AB is ABnot.
224 CHAPTER 8. KARNAUGH MAPPING
AB
A
B
AB
AB
Note that some of the members of A, above, are members of (AB)’. Some of the members
of B are members of (AB)’. But, none of the members of (AB)’ are within the doubly hatched
area AB.
A
B A
B
We have repeated the second example above left. Your fifth example, which you previously
sketched, is provided above right for comparison. Later we will find the occasional element, or
group of elements, totally contained within another group in a Karnaugh map.
Next, we show the development of a Boolean expression involving a complemented variable
below.
A
A
B
AB
A
A
B
B
AB
AB
8.3. BOOLEAN RELATIONSHIPS ON VENN DIAGRAMS 225
Example: (above)
Show a Venn diagram for A’B (Anot AND B).
Solution:
Starting above top left we have red horizontal shaded A’ (Anot), then, top right, B. Next,
lower left, we form the AND function A’B by overlapping the two previous regions. Most people
would use this as the answer to the example posed. However, only the double hatched A’B is
shown far right for clarity. The expression A’B is the region where both A’ and B overlap. The
clear region outside of A’B is (A’B)’, which was not part of the posed example.
Let’s try something similar with the Boolean OR function.
Example:
Find B’+A
B
B
A
B
Solution:
Above right we start out with B which is complemented to B’. Finally we overlay A on top
of B’. Since we are interested in forming the OR function, we will be looking for all hatched
area regardless of hatch style. Thus, A+B’ is all hatched area above right. It is shown as a
single hatch region below left for clarity.
A+B A+B
A+B
A B Double negation
DeMorgans theoremA B AB
Example:
Find (A+B’)’
Solution:
The green 45o A+B’ hatched area was the result of the previous example. Moving on to a
to,(A+B’)’ ,the present example, above left, let us find the complement of A+B’, which is the
226 CHAPTER 8. KARNAUGH MAPPING
white clear area above left corresponding to (A+B’)’. Note that we have repeated, at right, the
AB’ double hatched result from a previous example for comparison to our result. The regions
corresponding to (A+B’)’ and AB’ above left and right respectively are identical. This can be
proven with DeMorgan’s theorem and double negation.
This brings up a point. Venn diagrams don’t actually prove anything. Boolean algebra is
needed for formal proofs. However, Venn diagrams can be used for verification and visualiza
tion. We have verified and visualized DeMorgan’s theorem with a Venn diagram.
Example:
What does the Boolean expression A’+B’ look like on a Venn Diagram?
A
B
B
B
A
A
A B double hatch
A+B any hatch
A+B clear area
Solution: above figure
Start out with red horizontal hatched A’ and blue vertical hatched B’ above. Superimpose
the diagrams as shown. We can still see the A’ red horizontal hatch superimposed on the other
hatch. It also fills in what used to be part of the B (Btrue) circle, but only that part of the B
open circle not common to the A open circle. If we only look at the B’ blue vertical hatch, it fills
that part of the open A circle not common to B. Any region with any hatch at all, regardless of
type, corresponds to A’+B’. That is, everything but the open white space in the center.
Example:
What does the Boolean expression (A’+B’)’ look like on a Venn Diagram?
Solution: above figure, lower left
Looking at the white open space in the center, it is everything NOT in the previous solution
of A’+B’, which is (A’+B’)’.
Example:
8.3. BOOLEAN RELATIONSHIPS ON VENN DIAGRAMS 227
Show that (A’+B’) = AB
Solution: below figure, lower left
BA
A+B = A B = A B
A+B no hatch
AB
A
B
AB
AB
We previously showed on the above right diagram that the white open region is (A’+B’)’.
On an earlier example we showed a doubly hatched region at the intersection (overlay) of AB.
This is the left and middle figures repeated here. Comparing the two Venn diagrams, we see
that this open region , (A’+B’)’, is the same as the doubly hatched region AB (A AND B). We
can also prove that (A’+B’)’=AB by DeMorgan’s theorem and double negation as shown above.
A
B
C
AB
AC
BC
ABC
Three variable Venn diagram
We show a three variable Venn diagram above with regions A (red horizontal), B (blue
vertical), and, C (green 45o). In the very center note that all three regions overlap representing
Boolean expression ABC. There is also a larger petal shaped region where A and B overlap
corresponding to Boolean expression AB. In a similar manner A and C overlap producing
Boolean expression AC. And B and C overlap producing Boolean expression BC.
Looking at the size of regions described by AND expressions above, we see that region size
varies with the number of variables in the associated AND expression.
228 CHAPTER 8. KARNAUGH MAPPING
• A, 1variable is a large circular region.
• AB, 2variable is a smaller petal shaped region.
• ABC, 3variable is the smallest region.
• The more variables in the AND term, the smaller the region.
8.4 Making a Venn diagram look like a Karnaugh map
Starting with circle A in a rectangular A’ universe in figure (a) below, we morph a Venn
diagram into almost a Karnaugh map.
A A
A A
A A
A A
A
A
AA
a b c
d e f
We expand circle A at (b) and (c), conform to the rectangular A’ universe at (d), and change
A to a rectangle at (e). Anything left outside of A is A’ . We assign a rectangle to A’ at (f).
Also, we do not use shading in Karnaugh maps. What we have so far resembles a 1variable
Karnaugh map, but is of little utility. We need multiple variables.
A A A A
B
B
B
B
a b c
Figure (a) above is the same as the previous Venn diagram showing A and A’ above except
that the labelsA andA’ are above the diagram instead of inside the respective regions. Imagine
8.4. MAKING A VENN DIAGRAM LOOK LIKE A KARNAUGH MAP 229
that we have go through a process similar to figures (af) to get a ”square Venn diagram” for B
and B’ as we show in middle figure (b). We will now superimpose the diagrams in Figures (a)
and (b) to get the result at (c), just like we have been doing for Venn diagrams. The reason we
do this is so that we may observe that which may be common to two overlapping regions−−
say where A overlaps B. The lower right cell in figure (c) corresponds to AB where A overlaps
B.
A A
B
B
0 1
0
1
A
B
We don’t waste time drawing a Karnaugh map like (c) above, sketching a simplified version
as above left instead. The column of two cells under A’ is understood to be associated with A’,
and the heading A is associated with the column of cells under it. The row headed by B’ is
associated with the cells to the right of it. In a similar manner B is associated with the cells to
the right of it. For the sake of simplicity, we do not delineate the various regions as clearly as
with Venn diagrams.
The Karnaugh map above right is an alternate form used in most texts. The names of
the variables are listed next to the diagonal line. The A above the diagonal indicates that the
variable A (andA’) is assigned to the columns. The 0 is a substitute forA’, and the 1 substitutes
for A. The B below the diagonal is associated with the rows: 0 for B’, and 1 for B
Example:
Mark the cell corresponding to the Boolean expression AB in the Karnaugh map above with
a 1
A A
B
B
A A
B
B 1
A A
B
B 1
Solution:
Shade or circle the region corresponding to A. Then, shade or enclose the region correspond
ing to B. The overlap of the two regions is AB. Place a 1 in this cell. We do not necessarily
enclose the A and B regions as at above left.
230 CHAPTER 8. KARNAUGH MAPPING
A
A
BB
CC C
A
A
CC C
BB
We develop a 3variable Karnaugh map above, starting with Venn diagram like regions.
The universe (inside the black rectangle) is split into two narrow narrow rectangular regions
for A’ and A. The variables B’ and B divide the universe into two square regions. C occupies a
square region in the middle of the rectangle, with C’ split into two vertical rectangles on each
side of the C square.
In the final figure, we superimpose all three variables, attempting to clearly label the vari
ous regions. The regions are less obvious without color printing, more obvious when compared
to the other three figures. This 3variable KMap (Karnaugh map) has 23 = 8 cells, the small
squares within the map. Each individual cell is uniquely identified by the three Boolean Vari
ables (A, B, C). For example, ABC’ uniquely selects the lower right most cell(*), A’B’C’ selects
the upper left most cell (x).
A
A
CC C
BB
A
A
B B B BC CCC
*
x
We don’t normally label the Karnaugh map as shown above left. Though this figure clearly
shows map coverage by single boolean variables of a 4cell region. Karnaugh maps are labeled
like the illustration at right. Each cell is still uniquely identified by a 3variable product term,
8.5. KARNAUGH MAPS, TRUTH TABLES, AND BOOLEAN EXPRESSIONS 231
a Boolean AND expression. Take, for example, ABC’ following the A row across to the right
and the BC’ column down, both intersecting at the lower right cell ABC’. See (*) above figure.
A
C BC BC BC
A
0
1
00 01 1110B A BC
The above two different forms of a 3variable Karnaugh map are equivalent, and is the
final form that it takes. The version at right is a bit easier to use, since we do not have to write
down so many boolean alphabetic headers and complement bars, just 1s and 0s Use the form
of map on the right and look for the the one at left in some texts. The column headers on the
left B’C’, B’C, BC, BC’ are equivalent to 00, 01, 11, 10 on the right. The row headers A, A’ are
equivalent to 0, 1 on the right map.
8.5 Karnaugh maps, truth tables, and Boolean expres
sions
Maurice Karnaugh, a telecommunications engineer, developed the Karnaugh map at Bell Labs
in 1953 while designing digital logic based telephone switching circuits.
Now that we have developed the Karnaugh map with the aid of Venn diagrams, let’s put
it to use. Karnaugh maps reduce logic functions more quickly and easily compared to Boolean
algebra. By reduce we mean simplify, reducing the number of gates and inputs. We like to
simplify logic to a lowest cost form to save costs by elimination of components. We define
lowest cost as being the lowest number of gates with the lowest number of inputs per gate.
Given a choice, most students do logic simplification with Karnaugh maps rather than
Boolean algebra once they learn this tool.
232 CHAPTER 8. KARNAUGH MAPPING
A B Output
00
0 1
01
1 1
0 1
0
1
A
B
α
β
χ
δ
α β
χ δ
B
A
Output
·
·
·
unspecified logic
B
A 1 CR1
CR1 CR2 . . .
.
.
.
L2L1
CR2
Output = ABC + ABC + . . . ABC
.
.
.
We show five individual items above, which are just different ways of representing the same
thing: an arbitrary 2input digital logic function. First is relay ladder logic, then logic gates,
a truth table, a Karnaugh map, and a Boolean equation. The point is that any of these are
equivalent. Two inputs A and B can take on values of either 0 or 1, high or low, open or closed,
True or False, as the case may be. There are 22= 4 combinations of inputs producing an output.
This is applicable to all five examples.
These four outputs may be observed on a lamp in the relay ladder logic, on a logic probe
on the gate diagram. These outputs may be recorded in the truth table, or in the Karnaugh
map. Look at the Karnaugh map as being a rearranged truth table. The Output of the Boolean
equation may be computed by the laws of Boolean algebra and transfered to the truth table or
Karnaugh map. Which of the five equivalent logic descriptions should we use? The one which
is most useful for the task to be accomplished.
8.5. KARNAUGH MAPS, TRUTH TABLES, AND BOOLEAN EXPRESSIONS 233
A B Output
00
0 1
01
1 1
0 1
0
1
A
B
α
β
χ
δ
α β
χ δ
The outputs of a truth table correspond on a onetoone basis to Karnaugh map entries.
Starting at the top of the truth table, the A=0, B=0 inputs produce an output α. Note that this
same output α is found in the Karnaugh map at the A=0, B=0 cell address, upper left corner
of Kmap where the A=0 row and B=0 column intersect. The other truth table outputs β, χ, δ
from inputs AB=01, 10, 11 are found at corresponding Kmap locations.
Below, we show the adjacent 2cell regions in the 2variable Kmap with the aid of previous
rectangular Venn diagram like Boolean regions.
0 1
0
1
A
B
α β
χ δ
A
A
B B
0 1
0
1
A
B
α β
χ δ
0 1
0
1
A
B
α β
χ δ
0 1
0
1
A
B
α β
χ δ
A
A
B B
Cells α and χ are adjacent in the Kmap as ellipses in the left most Kmap below. Referring
to the previous truth table, this is not the case. There is another truth table entry (β) between
them. Which brings us to the whole point of the organizing the Kmap into a square array,
cells with any Boolean variables in common need to be close to one another so as to present a
pattern that jumps out at us. For cells α and χ they have the Boolean variable B’ in common.
We know this because B=0 (same as B’) for the column above cells α and χ. Compare this to
the square Venn diagram above the Kmap.
A similar line of reasoning shows that β and δ have Boolean B (B=1) in common. Then, α
and β have Boolean A’ (A=0) in common. Finally, χ and δ have Boolean A (A=1) in common.
Compare the last two maps to the middle square Venn diagram.
To summarize, we are looking for commonality of Boolean variables among cells. The Kar
naugh map is organized so that we may see that commonality. Let’s try some examples.
234 CHAPTER 8. KARNAUGH MAPPING
A B Output
00
0 1
01
1 1
0 1
0
1
AB
0
0
1
1
Example:
Transfer the contents of the truth table to the Karnaugh map above.
A B Output
00
0 1
01
1 1
0 1
0
1
AB
0
0
1
1
1
1
A=
B=
A=
B=
0
1
1
1
Solution:
The truth table contains two 1s. the K map must have both of them. locate the first 1 in
the 2nd row of the truth table above.
• note the truth table AB address
• locate the cell in the Kmap having the same address
• place a 1 in that cell
Repeat the process for the 1 in the last line of the truth table.
Example:
For the Karnaugh map in the above problem, write the Boolean expression. Solution is
below.
8.5. KARNAUGH MAPS, TRUTH TABLES, AND BOOLEAN EXPRESSIONS 235
0 1
0
1
AB
1
1
A
A
B B
Out = B
Solution:
Look for adjacent cells, that is, above or to the side of a cell. Diagonal cells are not adjacent.
Adjacent cells will have one or more Boolean variables in common.
• Group (circle) the two 1s in the column
• Find the variable(s) top and/or side which are the same for the group, Write this as the
Boolean result. It is B in our case.
• Ignore variable(s) which are not the same for a cell group. In our case A varies, is both 1
and 0, ignore Boolean A.
• Ignore any variable not associated with cells containing 1s. B’ has no ones under it.
Ignore B’
• Result Out = B
This might be easier to see by comparing to the Venn diagrams to the right, specifically the
B column.
Example:
Write the Boolean expression for the Karnaugh map below.
0 1
0
1
AB
1 1
A
A
B B
Out = A
Solution: (above)
• Group (circle) the two 1’s in the row
236 CHAPTER 8. KARNAUGH MAPPING
• Find the variable(s) which are the same for the group, Out = A’
Example:
For the Truth table below, transfer the outputs to the Karnaugh, then write the Boolean
expression for the result.
A B Output
00
0 1
01
1 1
0 1
0
1
AB
1
1
1
1
1 1
0 1
0
1
AB
1
1 1
Output= A + B
0
0 1
0
1
AB
1
1 1
Wrong Output= AB + B
Solution:
Transfer the 1s from the locations in the Truth table to the corresponding locations in the
Kmap.
• Group (circle) the two 1’s in the column under B=1
• Group (circle) the two 1’s in the row right of A=1
• Write product term for first group = B
• Write product term for second group = A
• Write SumOfProducts of above two terms Output = A+B
The solution of the Kmap in the middle is the simplest or lowest cost solution. A less desirable
solution is at far right. After grouping the two 1s, we make the mistake of forming a group of
1cell. The reason that this is not desirable is that:
• The single cell has a product term of AB’
• The corresponding solution is Output = AB’ + B
• This is not the simplest solution
The way to pick up this single 1 is to form a group of two with the 1 to the right of it as shown
in the lower line of the middle Kmap, even though this 1 has already been included in the
column group (B). We are allowed to reuse cells in order to form larger groups. In fact, it is
desirable because it leads to a simpler result.
We need to point out that either of the above solutions, Output or Wrong Output, are logi
cally correct. Both circuits yield the same output. It is a matter of the former circuit being the
lowest cost solution.
Example:
8.5. KARNAUGH MAPS, TRUTH TABLES, AND BOOLEAN EXPRESSIONS 237
Fill in the Karnaugh map for the Boolean expression below, then write the Boolean expres
sion for the result.
0 1
0
1
AB
1
1 1
0 1
0
1
AB
1
1 1
Output= A + B
AOut= AB + AB + AB
01 10 11
Solution: (above)
The Boolean expression has three product terms. There will be a 1 entered for each product
term. Though, in general, the number of 1s per product term varies with the number of vari
ables in the product term compared to the size of the Kmap. The product term is the address
of the cell where the 1 is entered. The first product term, A’B, corresponds to the 01 cell in the
map. A 1 is entered in this cell. The other two Pterms are entered for a total of three 1s
Next, proceed with grouping and extracting the simplified result as in the previous truth
table problem.
Example:
Simplify the logic diagram below.
B
A
A B
A
B
Out
Solution: (Figure below)
• Write the Boolean expression for the original logic diagram as shown below
• Transfer the product terms to the Karnaugh map
• Form groups of cells as in previous examples
• Write Boolean expression for groups as in previous examples
• Draw simplified logic diagram
238 CHAPTER 8. KARNAUGH MAPPING
0 1
0
1
AB
1
1 1
0 1
0
1
AB
1
1 1
Out= A + B
AOut= AB + AB + AB
01 10 11
B
AB
AB
A
AB
A
B
A
B
Out
Out
A
B
Example:
Simplify the logic diagram below.
0 1
0
1
AB
1
1
Out= AB + AB
01 10
B
AB
A
AB
A
B
Out Out
ExclusiveOR
A
B
Solution:
• Write the Boolean expression for the original logic diagram shown above
• Transfer the product terms to the Karnaugh map.
• It is not possible to form groups.
• No simplification is possible; leave it as it is.
No logic simplification is possible for the above diagram. This sometimes happens. Neither
the methods of Karnaugh maps nor Boolean algebra can simplify this logic further. We show
an ExclusiveOR schematic symbol above; however, this is not a logical simplification. It just
makes a schematic diagram look nicer. Since it is not possible to simplify the ExclusiveOR
logic and it is widely used, it is provided by manufacturers as a basic integrated circuit (7486).
8.6 Logic simplification with Karnaugh maps
The logic simplification examples that we have done so could have been performed with Boolean
algebra about as quickly. Real world logic simplification problems call for larger Karnaugh
maps so that we may do serious work. We will work some contrived examples in this section,
leaving most of the real world applications for the Combinatorial Logic chapter. By contrived,
8.6. LOGIC SIMPLIFICATION WITH KARNAUGH MAPS 239
we mean examples which illustrate techniques. This approach will develop the tools we need
to transition to the more complex applications in the Combinatorial Logic chapter.
We show our previously developed Karnaugh map. We will use the form on the right.
A
C BC BC BC
A
0
1
00 01 1110B A BC
Note the sequence of numbers across the top of the map. It is not in binary sequence which
would be 00, 01, 10, 11. It is 00, 01, 11 10, which is Gray code sequence. Gray code sequence
only changes one binary bit as we go from one number to the next in the sequence, unlike
binary. That means that adjacent cells will only vary by one bit, or Boolean variable. This is
what we need to organize the outputs of a logic function so that we may view commonality.
Moreover, the column and row headings must be in Gray code order, or the map will not work
as a Karnaugh map. Cells sharing common Boolean variables would no longer be adjacent, nor
show visual patterns. Adjacent cells vary by only one bit because a Gray code sequence varies
by only one bit.
If we sketch our own Karnaugh maps, we need to generate Gray code for any size map that
we may use. This is how we generate Gray code of any size.
240 CHAPTER 8. KARNAUGH MAPPING
00
01
1
0
0
1
1
0
0
1
0
1
00
01
11
10
00
01
11
10
00
01
11
10
10
11
01
00
000
001
011
010
10
11
01
00
000
001
011
010
110
111
101
100
How to generate Gray code.
1. Write 0,1 in a column.
2. Draw a mirror under the column.
3. Reflect the numbers about the mirror.
4. Distinguish the numbers above the mirror with leading zeros.
5. Distinguish those below the mirror with
leading ones.
6. Finished 2bit Gray code.
7. Need 3bit Gray code? Draw
mirror under column of four
2bit codes,reflect about
mirror.
8. Distinguish upper 4numbers with leading zeros.
9. Distinguish lower 4numbers with leading ones.
Note that the Gray code sequence, above right, only varies by one bit as we go down the list,
or bottom to top up the list. This property of Gray code is often useful in digital electronics in
general. In particular, it is applicable to Karnaugh maps.
Let us move on to some examples of simplification with 3variable Karnaugh maps. We
show how to map the product terms of the unsimplified logic to the Kmap. We illustrate
how to identify groups of adjacent cells which leads to a SumofProducts simplification of the
digital logic.
00 0111 10
1
0
A
BC
1 1
Out= ABC + ABC
Out= AB
⋅⋅
Above we, place the 1’s in the Kmap for each of the product terms, identify a group of two,
then write a pterm (product term) for the sole group as our simplified result.
8.6. LOGIC SIMPLIFICATION WITH KARNAUGH MAPS 241
00 0111 10
1
0
A
BC
1 1 1 1
Out= ABC + ABC + ABC + A BC
Out= A
⋅
Mapping the four product terms above yields a group of four covered by Boolean A’
00 0111 10
1
0
A
BC
1 1
11
Out= C
Out= ABC + A BC + ABC + ABC
⋅
Mapping the four pterms yields a group of four, which is covered by one variable C.
00 0111 10
1
0
A
BC
1 1 1 1
1 1
Out= A + B
Out= ABC+ABC+ABC+ABC+ABC+ABC
After mapping the six pterms above, identify the upper group of four, pick up the lower two
cells as a group of four by sharing the two with two more from the other group. Covering these
two with a group of four gives a simpler result. Since there are two groups, there will be two
pterms in the SumofProducts result A’+B
00 0111 10
1
0
A
BC
1
1
Out= ABC + ABC
Out= BC
⋅
The two product terms above form one group of two and simplifies to BC
242 CHAPTER 8. KARNAUGH MAPPING
00 0111 10
1
0
A
BC
1 1
1 1
Out= ABC + ABC + ABC + ABC
Out= B
⋅
Mapping the four pterms yields a single group of four, which is B
00 0111 10
1
0
A
BC
1
1 1
1
Out= ABC + ABC + ABC + ABC
Out= C
00
01 11
10
1
11
1
1
11
1
⋅
Mapping the four pterms above yields a group of four. Visualize the group of four by rolling
up the ends of the map to form a cylinder, then the cells are adjacent. We normally mark the
group of four as above left. Out of the variables A, B, C, there is a common variable: C’. C’ is a
0 over all four cells. Final result is C’.
00 0111 10
1
0
A
BC
1 1
1 1
Out= A + C
1 1
Out= ABC+ABC+ABC+ABC+ABC+ABC
The six cells above from the unsimplified equation can be organized into two groups of four.
These two groups should give us two pterms in our simplified result of A’ + C’.
Below, we revisit the Toxic Waste Incinerator from the Boolean algebra chapter. See Boolean
algebra chapter for details on this example. We will simplify the logic using a Karnaugh map.
8.6. LOGIC SIMPLIFICATION WITH KARNAUGH MAPS 243
Output = ABC + ABC + ABC + ABC
A B C
ABC
ABC
ABC
ABC
00 0111 10
1
0
A
BC
1
1 1 1
Output = AB + BC + AC
The Boolean equation for the output has four product terms. Map four 1’s corresponding to
the pterms. Forming groups of cells, we have three groups of two. There will be three pterms
in the simplified result, one for each group. See ”Toxic Waste Incinerator”, Boolean algebra
chapter for a gate diagram of the result, which is reproduced below.
A B C
Output = AB + BC + AC
AB
BC
AC
Below we repeat the Boolean algebra simplification of Toxic waste incinerator for compari
son.
244 CHAPTER 8. KARNAUGH MAPPING
ABC + ABC + ABC + ABC
Factoring BC out of 1st and 4th terms
BC(A + A) + ABC + ABC
Applying identity A + A = 1
BC(1) + ABC + ABC
Applying identity 1A = A
BC + ABC + ABC
Factoring B out of 1st and 3rd terms
B(C + AC) + ABC
Applying rule A + AB = A + B to
the C + AC term
B(C + A) + ABC
Distributing terms
BC + AB + ABC
Factoring A out of 2nd and 3rd terms
BC + A(B + BC)
Applying rule A + AB = A + B to
the B + BC term
BC + A(B + C)
Distributing terms
BC + AB + AC
or
AB + BC + AC
Simplified result
Below we repeat the Toxic waste incinerator Karnaugh map solution for comparison to the
above Boolean algebra simplification. This case illustrates why the Karnaugh map is widely
used for logic simplification.
00 0111 10
1
0
A
BC
1
1 1 1
Output = AB + BC + AC
The Karnaugh map method looks easier than the previous page of boolean algebra.
8.7. LARGER 4VARIABLE KARNAUGH MAPS 245
8.7 Larger 4variable Karnaugh maps
Knowing how to generate Gray code should allow us to build larger maps. Actually, all we need
to do is look at the left to right sequence across the top of the 3variable map, and copy it down
the left side of the 4variable map. See below.
00 01 11 10
00
01
11
10
AB
CD
⋅
The following four variable Karnaugh maps illustrate reduction of Boolean expressions too
tedious for Boolean algebra. Reductions could be done with Boolean algebra. However, the
Karnaugh map is faster and easier, especially if there are many logic reductions to do.
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1 1 1
⋅
Out= AB + CD
Out= ABCD+ABCD+ABCD+ABCD+ABCD+ABCD+ABCD
The above Boolean expression has seven product terms. They are mapped top to bottom
and left to right on the Kmap above. For example, the first Pterm A’B’CD is first row 3rd cell,
corresponding to map location A=0, B=0, C=1, D=1. The other product terms are placed in a
similar manner. Encircling the largest groups possible, two groups of four are shown above.
The dashed horizontal group corresponds the the simplified product term AB. The vertical
group corresponds to Boolean CD. Since there are two groups, there will be two product terms
in the SumOfProducts result of Out=AB+CD.
Fold up the corners of the map below like it is a napkin to make the four cells physically
adjacent.
246 CHAPTER 8. KARNAUGH MAPPING
⋅
00 01 11 10
00
01
11
10
AB
CD
1 1
1 1
Out= BD
Out= ABCD + ABCD + ABCD + ABCD
1 1
1 1
The four cells above are a group of four because they all have the Boolean variables B’ and
D’ in common. In other words, B=0 for the four cells, and D=0 for the four cells. The other
variables (A, B) are 0 in some cases, 1 in other cases with respect to the four corner cells. Thus,
these variables (A, B) are not involved with this group of four. This single group comes out of
the map as one product term for the simplified result: Out=B’C’
For the Kmap below, roll the top and bottom edges into a cylinder forming eight adjacent
cells.
00 01 11 10
00
01
11
10
AB
CD
11 1 1
11 1 1
⋅
Out= B
Out= ABCD + ABCD + ABCD + ABCD
+ ABCD +ABCD + ABCD + ABCD
The above group of eight has one Boolean variable in common: B=0. Therefore, the one
group of eight is covered by one pterm: B’. The original eight term Boolean expression simpli
fies to Out=B’
The Boolean expression below has nine pterms, three of which have three Booleans instead
of four. The difference is that while four Boolean variable product terms cover one cell, the
three Boolean pterms cover a pair of cells each.
8.7. LARGER 4VARIABLE KARNAUGH MAPS 247
00 01 11 10
00
01
11
10
AB
CD
11 1 1
11 1 1
1
1
1
1
⋅
Out= ABCD + ABCD + ABCD + ABCD
+ BCD + BCD + ABCD + ABD + ABCD
Out= B + D
00 01 11 10
00
01
11
10
AB
CD
11 1 1
11 1 1
1
1
1
1
B
D
The six product terms of four Boolean variables map in the usual manner above as single
cells. The three Boolean variable terms (three each) map as cell pairs, which is shown above.
Note that we are mapping pterms into the Kmap, not pulling them out at this point.
For the simplification, we form two groups of eight. Cells in the corners are shared with
both groups. This is fine. In fact, this leads to a better solution than forming a group of eight
and a group of four without sharing any cells. Final Solution is Out=B’+D’
Below we map the unsimplified Boolean expression to the Karnaugh map.
00 01 11 10
00
01
11
10
AB
CD
1
1 1
Out= BCD + ABD + ABCD
Out= ABCD + ABCD + ABCD + ABCD
1
⋅
Above, three of the cells form into a groups of two cells. A fourth cell cannot be combined
with anything, which often happens in ”real world” problems. In this case, the Boolean pterm
ABCD is unchanged in the simplification process. Result: Out= B’C’D’+A’B’D’+ABCD
Often times there is more than one minimum cost solution to a simplification problem. Such
is the case illustrated below.
248 CHAPTER 8. KARNAUGH MAPPING
Out= BCD + ACD + BCD + ACD
Out= ABCD + ABCD + ABCD + ABCD
+ ABCD + ABCD + ABCD + ABCD
⋅
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1 1
1
1
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1 1
1
1
Out= ABC + ABD + ABC + AB D
Both results above have four product terms of three Boolean variable each. Both are equally
valid minimal cost solutions. The difference in the final solution is due to how the cells are
grouped as shown above. A minimal cost solution is a valid logic design with the minimum
number of gates with the minimum number of inputs.
Below we map the unsimplified Boolean equation as usual and form a group of four as a
first simplification step. It may not be obvious how to pick up the remaining cells.
AB
CD
1
Out= AC + AD + BC + BD
Out= ABCD + ABCD + ABCD
+ ABCD + ABCD + ABCD
+ ABCD + ABCD + ABCD
1
1
1
1
⋅
1
1 1 1
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10
Pick up three more cells in a group of four, center above. There are still two cells remaining.
the minimal cost method to pick up those is to group them with neighboring cells as groups of
four as at above right.
On a cautionary note, do not attempt to form groups of three. Groupings must be powers of
2, that is, 1, 2, 4, 8 ...
Below we have another example of two possible minimal cost solutions. Start by forming a
couple of groups of four after mapping the cells.
8.8. MINTERM VS MAXTERM SOLUTION 249
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
⋅
Out= CD + CD+ ABC
Out= ABCD+ABCD+ABCD+ABCD+ABCD
+ABCD + ABCD + ABCD + ABCD
1
1
1
1
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1
1
1
Out= CD + CD+ ABD
The two solutions depend on whether the single remaining cell is grouped with the first or
the second group of four as a group of two cells. That cell either comes out as either ABC’ or
ABD, your choice. Either way, this cell is covered by either Boolean product term. Final results
are shown above.
Below we have an example of a simplification using the Karnaugh map at left or Boolean
algebra at right. Plot C’ on the map as the area of all cells covered by address C=0, the 8cells
on the left of the map. Then, plot the single ABCD cell. That single cell forms a group of 2cell
as shown, which simplifies to Pterm ABD, for an end result of Out = C’ + ABD.
00 01 11 10
00
01
11
10
AB
CD
11 1
⋅
Out= C + ABD
Out= C +ABCD
1 1
1 1
1 1
Applying rule A + AB = A + B to
the C + ABCD term
Out= C +ABCD
Out= C + ABD
Simplification by Boolean
Algebra
This (above) is a rare example of a four variable problem that can be reduced with Boolean
algebra without a lot of work, assuming that you remember the theorems.
8.8 Minterm vs maxterm solution
So far we have been finding SumOfProduct (SOP) solutions to logic reduction problems. For
each of these SOP solutions, there is also a ProductOfSums solution (POS), which could be
250 CHAPTER 8. KARNAUGH MAPPING
more useful, depending on the application. Before working a ProductOfSums solution, we
need to introduce some new terminology. The procedure below for mapping product terms
is not new to this chapter. We just want to establish a formal procedure for minterms for
comparison to the new procedure for maxterms.
00 0111 10
1
0
A
BC
Out= AB C
0
00 0111 10
1
0
A
BC
0
1
1
Out= AB C
0 0 0
000 000
0 0 0
Numeric= 1 1 1
Minterm= A B C
Out= AB C Out= AB C
Minterm= A B C
Numeric= 0 1 0
A minterm is a Boolean expression resulting in 1 for the output of a single cell, and 0s for all
other cells in a Karnaugh map, or truth table. If a minterm has a single 1 and the remaining
cells as 0s, it would appear to cover a minimum area of 1s. The illustration above left shows
the minterm ABC, a single product term, as a single 1 in a map that is otherwise 0s. We
have not shown the 0s in our Karnaugh maps up to this point, as it is customary to omit them
unless specifically needed. Another minterm A’BC’ is shown above right. The point to review
is that the address of the cell corresponds directly to the minterm being mapped. That is, the
cell 111 corresponds to the minterm ABC above left. Above right we see that the minterm
A’BC’ corresponds directly to the cell 010. A Boolean expression or map may have multiple
minterms.
Referring to the above figure, Let’s summarize the procedure for placing a minterm in a
Kmap:
• Identify the minterm (product term) term to be mapped.
• Write the corresponding binary numeric value.
• Use binary value as an address to place a 1 in the Kmap
• Repeat steps for other minterms (Pterms within a SumOfProducts).
Out= AB C +ABC
00 0111 10
1
0
A
BC
1
000
0 0 0
1
.
Numeric= 0 1 0 111
Minterm= A B C ABC
Out= AB C + ABC
8.8. MINTERM VS MAXTERM SOLUTION 251
A Boolean expression will more often than not consist of multiple minterms corresponding
to multiple cells in a Karnaugh map as shown above. The multiple minterms in this map are
the individual minterms which we examined in the previous figure above. The point we review
for reference is that the 1s come out of the Kmap as a binary cell address which converts
directly to one or more product terms. By directly we mean that a 0 corresponds to a comple
mented variable, and a 1 corresponds to a true variable. Example: 010 converts directly to
A’BC’. There was no reduction in this example. Though, we do have a SumOfProducts result
from the minterms.
Referring to the above figure, Let’s summarize the procedure for writing the SumOfProducts
reduced Boolean equation from a Kmap:
• Form largest groups of 1s possible covering all minterms. Groups must be a power of 2.
• Write binary numeric value for groups.
• Convert binary value to a product term.
• Repeat steps for other groups. Each group yields a pterms within a SumOfProducts.
Nothing new so far, a formal procedure has been written down for dealing with minterms.
This serves as a pattern for dealing with maxterms.
Next we attack the Boolean function which is 0 for a single cell and 1s for all others.
00 0111 10
1
0
A
BC
10 1 1
1111
Numeric = 1 1 1
Maxterm = A+B+C
Out = ( A +B+C)
Complement = 0 0 0
. .
A maxterm is a Boolean expression resulting in a 0 for the output of a single cell expression,
and 1s for all other cells in the Karnaugh map, or truth table. The illustration above left shows
the maxterm (A+B+C), a single sum term, as a single 0 in a map that is otherwise 1s. If a
maxterm has a single 0 and the remaining cells as 1s, it would appear to cover a maximum
area of 1s.
There are some differences now that we are dealing with something new, maxterms. The
maxterm is a 0, not a 1 in the Karnaugh map. A maxterm is a sum term, (A+B+C) in our
example, not a product term.
It also looks strange that (A+B+C) is mapped into the cell 000. For the equationOut=(A+B+C)=0,
all three variables (A, B, C)must individually be equal to 0. Only (0+0+0)=0will equal 0. Thus
we place our sole 0 for minterm (A+B+C) in cell A,B,C=000 in the Kmap, where the inputs
are all0 . This is the only case which will give us a 0 for our maxterm. All other cells contain
1s because any input values other than ((0,0,0) for (A+B+C) yields 1s upon evaluation.
Referring to the above figure, the procedure for placing a maxterm in the Kmap is:
252 CHAPTER 8. KARNAUGH MAPPING
• Identify the Sum term to be mapped.
• Write corresponding binary numeric value.
• Form the complement
• Use the complement as an address to place a 0 in the Kmap
• Repeat for other maxterms (Sum terms within ProductofSums expression).
00 0111 10
1
0
A
BC
0
1 1
1 1
1
1
1
Numeric = 0 0 0
Maxterm = A+B+ C
Out = (A+B+ C)
Complement = 1 1 1
. .
Another maxterm A’+B’+C’ is shown above. Numeric 000 corresponds to A’+B’+C’. The
complement is 111. Place a 0 for maxterm (A’+B’+C’) in this cell (1,1,1) of the Kmap as
shown above.
Why should (A’+B’+C’) cause a 0 to be in cell 111? When A’+B’+C’ is (1’+1’+1’), all 1s in,
which is (0+0+0) after taking complements, we have the only condition that will give us a 0.
All the 1s are complemented to all 0s, which is 0 when ORed.
00 0111 10
1
0
A
BC
1
0 0 11
111
Out = ( A +B+C)(A+B+ C)
Maxterm= (A+B+C)
Numeric= 1 1 1
Complement= 0 0 0
Maxterm= (A+B+C)
Numeric= 1 1 0
Complement= 0 0 1
A Boolean ProductOfSums expression or map may have multiple maxterms as shown
above. Maxterm (A+B+C) yields numeric 111 which complements to 000, placing a 0 in cell
(0,0,0). Maxterm (A+B+C’) yields numeric 110 which complements to 001, placing a 0 in cell
(0,0,1).
Now that we have the kmap setup, what we are really interested in is showing how to
write a ProductOfSums reduction. Form the 0s into groups. That would be a group of two
below. Write the binary value corresponding to the sumterm which is (0,0,X). Both A and B
8.8. MINTERM VS MAXTERM SOLUTION 253
are 0 for the group. But, C is both 0 and 1 so we write an X as a place holder for C. Form
the complement (1,1,X). Write the Sumterm (A+B) discarding the C and the X which held its’
place. In general, expect to have more sumterms multiplied together in the ProductOfSums
result. Though, we have a simple example here.
00 0111 10
1
0
A
BC
1
Out =(A +B)
0 0 1
⋅
1
111
A B C = 0 0 X
Complement = 1 1 X
Sumterm =(A +B)
Out = ( A +B+C)(A+B+ C)
⋅
Let’s summarize the procedure for writing the ProductOfSums Boolean reduction for a
Kmap:
• Form largest groups of 0s possible, covering all maxterms. Groups must be a power of 2.
• Write binary numeric value for group.
• Complement binary numeric value for group.
• Convert complement value to a sumterm.
• Repeat steps for other groups. Each group yields a sumterm within a ProductOfSums
result.
Example:
Simplify the ProductOfSums Boolean expression below, providing a result in POS form.
Out= (A+B+C+D)(A+B+C+D)(A+B+C+D)(A+B+C+D)
(A+B+C+D)(A+B+C+D)(A +B+C+D)
Solution:
Transfer the seven maxterms to the map below as 0s. Be sure to complement the input
variables in finding the proper cell location.
254 CHAPTER 8. KARNAUGH MAPPING
00 01 11 10
00
01
11
10
⋅
Out= (A+B+C+D)(A+B+C+D)(A+B+C+D)(A+B+C+D)
(A+B+C+D)(A+B+C+D)(A +B+C+D)
0
0
0
0
0
0
0
AB
CD
We map the 0s as they appear left to right top to bottom on the map above. We locate the
last three maxterms with leader lines..
Once the cells are in place above, form groups of cells as shown below. Larger groups will
give a sumterm with fewer inputs. Fewer groups will yield fewer sumterms in the result.
00 01 11 10
00
01
11
10
0
0
0
0
0
0
0
Out= (B+C+D)(A+C+D)(C+D)
AB
CD
ABCD = XX10 > XX01 > (C+D )
ABCD = 0X01 > 1X10 > ( A+ C+D )
ABCD = X001 > X110 > ( B+ C+D )
input complement Sumterm
We have three groups, so we expect to have three sumterms in our POS result above. The
group of 4cells yields a 2variable sumterm. The two groups of 2cells give us two 3variable
sumterms. Details are shown for how we arrived at the Sumterms above. For a group, write
the binary group input address, then complement it, converting that to the Boolean sumterm.
The final result is product of the three sums.
Example:
Simplify the ProductOfSums Boolean expression below, providing a result in SOP form.
Out= (A+B+C+D)(A+B+C+D)(A+B+C+D)(A+B+C+D)
(A+B+C+D)(A+B+C+D)(A +B+C+D)
Solution:
This looks like a repeat of the last problem. It is except that we ask for a SumOfProducts
Solution instead of the ProductOfSums which we just finished. Map the maxterm 0s from the
ProductOfSums given as in the previous problem, below left.
8.8. MINTERM VS MAXTERM SOLUTION 255
00 01 11 10
00
01
11
10
⋅
Out= (A+B+C+D)(A+B+C+D)(A+B+C+D)(A+B+C+D)
(A+B+C+D)(A+B+C+D)(A +B+C+D)
0
0
0
0
0
0
0
AB
CD
00 01 11 10
00
01
11
10
0
0
0
0
0
0
0
AB
CD
1 1
11
1 11
11
Then fill in the implied 1s in the remaining cells of the map above right.
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1
1
1
Out= CD + CD+ ABD
0
0
0
0
0
0
0
.
Form groups of 1s to cover all 1s. Then write the SumOfProducts simplified result as in
the previous section of this chapter. This is identical to a previous problem.
00 01 11 10
00
01
11
10
Out= (A+B+C+D)(A+B+C+D)(A+B+C+D)(A+B+C+D)
(A+B+C+D)(A+B+C+D)(A +B+C+D)
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1
1
1
Out= CD + CD+ ABD
0
0
0
0
0
0
0
Out= (B+C+D)(A+C+D)(C+D)
AB
CD
Above we show both the ProductOfSums solution, from the previous example, and the
SumOfProducts solution from the current problem for comparison. Which is the simpler
256 CHAPTER 8. KARNAUGH MAPPING
solution? The POS uses 3OR gates and 1AND gate, while the SOP uses 3AND gates and
1OR gate. Both use four gates each. Taking a closer look, we count the number of gate inputs.
The POS uses 8inputs; the SOP uses 7inputs. By the definition of minimal cost solution, the
SOP solution is simpler. This is an example of a technically correct answer that is of little use
in the real world.
The better solution depends on complexity and the logic family being used. The SOP solu
tion is usually better if using the TTL logic family, as NAND gates are the basic building block,
which works well with SOP implementations. On the other hand, A POS solution would be
acceptable when using the CMOS logic family since all sizes of NOR gates are available.
Out= CD + CD+ ABDOut= (B+C+D)(A+C+D)(C+D)
C
D
A
B
Out
D
C
B
A
Out
The gate diagrams for both cases are shown above, ProductOfSums left, and SumOf
Products right.
Below, we take a closer look at the SumOfProducts version of our example logic, which is
repeated at left.
Out= CD + CD+ ABD
C
D
A
B
Out
Out= CD + CD+ ABD
C
D
A
B
Out
Above all AND gates at left have been replaced by NAND gates at right.. The OR gate at
the output is replaced by a NAND gate. To prove that ANDOR logic is equivalent to NAND
NAND logic, move the inverter invert bubbles at the output of the 3NAND gates to the input
of the final NAND as shown in going from above right to below left.
8.8. MINTERM VS MAXTERM SOLUTION 257
C
D
A
B
Out
Out
OutX
Y
Z
X
Y
Z
Out= X+Y+Z
Out= X Y Z
Out= X+Y+Z
Out= X+Y+Z
DeMorgans
Double negation
Above right we see that the output NAND gate with inverted inputs is logically equivalent
to an OR gate by DeMorgan’s theorem and double negation. This information is useful in
building digital logic in a laboratory setting where TTL logic family NAND gates are more
readily available in a wide variety of configurations than other types.
The Procedure for constructing NANDNAND logic, in place of ANDOR logic is as follows:
• Produce a reduced SumOfProducts logic design.
• When drawing the wiring diagram of the SOP, replace all gates (both AND and OR) with
NAND gates.
• Unused inputs should be tied to logic High.
• In case of troubleshooting, internal nodes at the first level of NAND gate outputs do
NOT match ANDOR diagram logic levels, but are inverted. Use the NANDNAND logic
diagram. Inputs and final output are identical, though.
• Label any multiple packages U1, U2,.. etc.
• Use data sheet to assign pin numbers to inputs and outputs of all gates.
Example:
Let us revisit a previous problem involving an SOP minimization. Produce a ProductOf
Sums solution. Compare the POS solution to the previous SOP.
258 CHAPTER 8. KARNAUGH MAPPING
AB
CD
1
Out= AC + AD + BC + BD
Out= A B C D + A B C D + A BCD
+ A B C D + A B CD + A BCD
+ A B C D + A B C D + ABCD
1
1
1
1
⋅
1
1 1 1
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
Out= (A+B)( C +D)
Solution:
Above left we have the original problem starting with a 9minterm Boolean unsimplified
expression. Reviewing, we formed four groups of 4cells to yield a 4productterm SOP result,
lower left.
In the middle figure, above, we fill in the empty spaces with the implied 0s. The 0s form two
groups of 4cells. The solid blue group is (A’+B), the dashed red group is (C’+D). This yields
two sumterms in the ProductOfSums result, above right Out = (A’+B)(C’+D)
Comparing the previous SOP simplification, left, to the POS simplification, right, shows
that the POS is the least cost solution. The SOP uses 5gates total, the POS uses only 3gates.
This POS solution even looks attractive when using TTL logic due to simplicity of the result.
We can find AND gates and an OR gate with 2inputs.
C
D
A
B
Out
Out= AC + AD + BC + BD Out= (A+B)( C +D)
Out
D
A
B
C
The SOP and POS gate diagrams are shown above for our comparison problem.
Given the pinouts for the TTL logic family integrated circuit gates below, label the maxterm
diagram above right with Circuit designators (U1a, U1b, U2a, etc), and pin numbers.
8.8. MINTERM VS MAXTERM SOLUTION 259
1 2 3 4 5 6 7
891011121314
7408
GND
VCC
1 2 3 4 5 6 7
891011121314
7432
GND
VCC
1 2 3 4 5 6 7
891011121314
7404
GND
VCC
Out= (A+B)( C +D)
Out
D
A
B
C
Each integrated circuit package that we use will receive a circuit designator: U1, U2, U3.
To distinguish between the individual gates within the package, they are identified as a, b, c,
d, etc. The 7404 hexinverter package is U1. The individual inverters in it are are U1a, U1b,
U1c, etc. U2 is assigned to the 7432 quad OR gate. U3 is assigned to the 7408 quad AND gate.
With reference to the pin numbers on the package diagram above, we assign pin numbers to
all gate inputs and outputs on the schematic diagram below.
We can now build this circuit in a laboratory setting. Or, we could design a printed circuit
board for it. A printed circuit board contains copper foil ”wiring” backed by a non conductive
substrate of phenolic, or epoxyfiberglass. Printed circuit boards are used to mass produce
electronic circuits. Ground the inputs of unused gates.
Out= (A+B)( C +D)
Out
D
A
B
C
1
2
3
4
5
6
1
2
3
1 2
3 4
U1a
U1b
U2a
U2b
U3a
U1 = 7404
U2 = 7432
U3 = 7408
Label the previous POS solution diagram above left (third figure back) with Circuit desig
nators and pin numbers. This will be similar to what we just did.
260 CHAPTER 8. KARNAUGH MAPPING
1 2 3 4 5 6 7
891011121314
7420
GND
VCC
1 2 3 4 5 6 7
891011121314
7400
GND
VCC
We can find 2input AND gates, 7408 in the previous example. However, we have trouble
finding a 4input OR gate in our TTL catalog. The only kind of gate with 4inputs is the 7420
NAND gate shown above right.
We can make the 4input NAND gate into a 4input OR gate by inverting the inputs to the
NAND gate as shown below. So we will use the 7420 4input NAND gate as an OR gate by
inverting the inputs.
Y = A B = A+B DeMorgan’s
Y =A+B Double negation =
We will not use discrete inverters to invert the inputs to the 7420 4input NAND gate, but
will drive it with 2input NAND gates in place of the AND gates called for in the SOP, minterm,
solution. The inversion at the output of the 2input NAND gates supply the inversion for the
4input OR gate.
C
D
A
B
Out
Out= AC + AD + BC + BD
U1b
U1a
U2d
U3a
U2c
U2b
U2a
12
3 4
5
6
2
1
3
4
89
10
1112
13
1
2
4
5
6
U1 = 7404
U2 = 7400
U3 = 7420
Out= ( AC )( AD )( BC )( BD )
Out= AC + AD + BC + BD
Boolean from diagram
DeMorgan’s
Double negation
8.9. Σ (SUM) AND Π (PRODUCT) NOTATION 261
The result is shown above. It is the only practical way to actually build it with TTL gates
by using NANDNAND logic replacing ANDOR logic.
8.9 Σ (sum) and Π (product) notation
For reference, this section introduces the terminology used in some texts to describe the minterms
and maxterms assigned to a Karnaugh map. Otherwise, there is no new material here.
Σ (sigma) indicates sum and lower case ”m” indicates minterms. Σm indicates sum of
minterms. The following example is revisited to illustrate our point. Instead of a Boolean
equation description of unsimplified logic, we list the minterms.
f(A,B,C,D) = Σ m(1, 2, 3, 4, 5, 7, 8, 9, 11, 12, 13, 15)
or
f(A,B,C,D) = Σ(m1,m2,m3,m4,m5,m7,m8,m9,m11,m12,m13,m15)
The numbers indicate cell location, or address, within a Karnaugh map as shown below
right. This is certainly a compact means of describing a list of minterms or cells in a Kmap.
AB
CD
1
f(A,B,C,D) = AC + AD + BC + BD
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10
AB
CD
0
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
1 3 2
6754
12 13 1415
101198
Out= A B C D + A B C D + A BCD
+ A B C D + A B CD + A BCD
+ A B C D + A B C D + ABCD
f(A,B,C,D)=∑ m(0,1,3,4,5,7,12,13,15)
The SumOfProducts solution is not affected by the new terminology. The minterms, 1s, in
the map have been grouped as usual and a SumOFProducts solution written.
Below, we show the terminology for describing a list of maxterms. Product is indicated by
the Greek Π (pi), and upper case ”M” indicates maxterms. ΠM indicates product of maxterms.
The same example illustrates our point. The Boolean equation description of unsimplified
logic, is replaced by a list of maxterms.
262 CHAPTER 8. KARNAUGH MAPPING
f(A,B,C,D) = Π M(2, 6, 8, 9, 10, 11, 14)
or
f(A,B,C,D) = Π(M2, M6, M8, M9, M10, M11, M14)
Once again, the numbers indicate Kmap cell address locations. For maxterms this is the
location of 0s, as shown below. A ProductOFSums solution is completed in the usual manner.
AB
CD
0
00 01 11 10
00
01
11
10
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
f(A,B,C,D)= (A+B)( C +D)
1 3 2
6754
12 13 1415
101198
f(A,B,C,D)= Π M(2,6,8,9,10,11,14)
Out= (A+B+C +D)(A+B+C+D)(A+B+C+D)(A +B+C+D)
(A+B+C+D)( A+B+C+ D )(A +B+C+D)
AB
CD
1
1
1
1
1
1
1 1 1
00 01 11 10
00
01
11
10 0 0 0 0
0
0
0
8.10 Don’t care cells in the Karnaugh map
Up to this point we have considered logic reduction problems where the input conditions were
completely specified. That is, a 3variable truth table or Karnaugh map had 2n = 23 or 8
entries, a full table or map. It is not always necessary to fill in the complete truth table for
some realworld problems. We may have a choice to not fill in the complete table.
For example, when dealing with BCD (Binary Coded Decimal) numbers encoded as four
bits, we may not care about any codes above the BCD range of (0, 1, 2...9). The 4bit binary
codes for the hexadecimal numbers (Ah, Bh, Ch, Eh, Fh) are not valid BCD codes. Thus, we
do not have to fill in those codes at the end of a truth table, or Kmap, if we do not care to.
We would not normally care to fill in those codes because those codes (1010, 1011, 1100, 1101,
1110, 1111) will never exist as long as we are dealing only with BCD encoded numbers. These
six invalid codes are don’t cares as far as we are concerned. That is, we do not care what output
our logic circuit produces for these don’t cares.
8.10. DON’T CARE CELLS IN THE KARNAUGH MAP 263
Don’t cares in a Karnaugh map, or truth table, may be either 1s or 0s, as long as we don’t
care what the output is for an input condition we never expect to see. We plot these cells with
an asterisk, *, among the normal 1s and 0s. When forming groups of cells, treat the don’t care
cell as either a 1 or a 0, or ignore the don’t cares. This is helpful if it allows us to form a larger
group than would otherwise be possible without the don’t cares. There is no requirement to
group all or any of the don’t cares. Only use them in a group if it simplifies the logic.
A
BC
0
1
00 01 11 10
* *
0 0 0 0
0 1
Out = A C
A
BC
0
1
00 01 11 10
0 0 0 0
0 1
Out = A B C
A
BC
0
1
00 01 11 10
* *
0 0 0 0
0 1
ABC = 0XX > 1XX > A
ABC = XX0 > XX1 > C
input comp Sum
Out = A C (POS)
lement term
**
Above is an example of a logic function where the desired output is 1 for input ABC = 101
over the range from 000 to 101. We do not care what the output is for the other possible
inputs (110, 111). Map those two as don’t cares. We show two solutions. The solution on the
right Out = AB’C is the more complex solution since we did not use the don’t care cells. The
solution in the middle, Out=AC, is less complex because we grouped a don’t care cell with the
single 1 to form a group of two. The third solution, a ProductOfSums on the right, results
from grouping a don’t care with three zeros forming a group of four 0s. This is the same, less
complex, Out=AC. We have illustrated that the don’t care cells may be used as either 1s or 0s,
whichever is useful.
A to D L1
L2
L3
L4
L5
generator
lamp logictach ometer
The electronics class of Lightning State College has been asked to build the lamp logic for a
stationary bicycle exhibit at the local science museum. As a rider increases his pedaling speed,
lamps will light on a bar graph display. No lamps will light for no motion. As speed increases,
the lower lamp, L1 lights, then L1 and L2, then, L1, L2, and L3, until all lamps light at the
highest speed. Once all the lamps illuminate, no further increase in speed will have any effect
on the display.
A small DC generator coupled to the bicycle tire outputs a voltage proportional to speed. It
drives a tachometer board which limits the voltage at the high end of speed where all lamps
light. No further increase in speed can increase the voltage beyond this level. This is crucial
264 CHAPTER 8. KARNAUGH MAPPING
because the downstream A to D (Analog to Digital) converter puts out a 3bit code, ABC, 23 or
8codes, but we only have five lamps. A is the most significant bit, C the least significant bit.
The lamp logic needs to respond to the six codes out of the A to D. For ABC=000, no motion,
no lamps light. For the five codes (001 to 101) lamps L1, L1&L2, L1&L2&L3, up to all lamps
will light, as speed, voltage, and the A to D code (ABC) increases. We do not care about the
response to input codes (110, 111) because these codes will never come out of the A to D due
to the limiting in the tachometer block. We need to design five logic circuits to drive the five
lamps.
A
BC
0
1
00 01 11 10
L1
A
BC
0
1
00 01 11 10
L2
A
BC
0
1
00 01 11 10
L4
A
BC
0
1
00 01 11 10
L3
A
BC
0
1
00 01 11 10
L5
* * * * * *
* * * *
0 0
0
0
0
1 1
1 1 1 1 1
1 1
0
1 1
10 0
1 1
0 0 0 0 0 0
0 1
L1 = A + B + C L2 = A + B L3 = A + B C
L5 = A C L4 = A
Since, none of the lamps light for ABC=000 out of the A to D, enter a 0 in all Kmaps for
cell ABC=000. Since we don’t care about the never to be encountered codes (110, 111), enter
asterisks into those two cells in all five Kmaps.
Lamp L5 will only light for code ABC=101. Enter a 1 in that cell and five 0s into the
remaining empty cells of L5 Kmap.
L4 will light initially for code ABC=100, and will remain illuminated for any code greater,
ABC=101, because all lamps below L5 will light when L5 lights. Enter 1s into cells 100 and
101 of the L4 map so that it will light for those codes. Four 0’s fill the remaining L4 cells
L3 will initially light for code ABC=011. It will also light whenever L5 and L4 illuminate.
Enter three 1s into cells 011, 100, 101 for L3 map. Fill three 0s into the remaining L3 cells.
L2 lights for ABC=010 and codes greater. Fill 1s into cells 010, 011, 100, 101, and two 0s
in the remaining cells.
The only time L1 is not lighted is for no motion. There is already a 0 in cell ABC=000. All
the other five cells receive 1s.
Group the 1’s as shown above, using don’t cares whenever a larger group results. The L1
map shows three product terms, corresponding to three groups of 4cells. We used both don’t
cares in two of the groups and one don’t care on the third group. The don’t cares allowed us to
form groups of four.
In a similar manner, the L2 and L4 maps both produce groups of 4cells with the aid of
the don’t care cells. The L4 reduction is striking in that the L4 lamp is controlled by the most
significant bit from the A to D converter, L5=A. No logic gates are required for lamp L4. In
8.11. LARGER 5 & 6VARIABLE KARNAUGH MAPS 265
the L3 and L5 maps, single cells form groups of two with don’t care cells. In all five maps, the
reduced Boolean equation is less complex than without the don’t cares.
A
C
A
B
+5VDC
AC
A+BC
A+B
A+B+C
L5
L4
L3
L2
L1
5x470 Ω
U1a
U2aU1b
U3a
U2b
U2c
U3b
U3c
U3d
U3e
7408
7432
7406
The gate diagram for the circuit is above. The outputs of the five Kmap equations drive
inverters. Note that the L1 OR gate is not a 3input gate but a 2input gate having inputs
(A+B), C, outputting A+B+C The open collector inverters, 7406, are desirable for driving
LEDs, though, not part of the Kmap logic design. The output of an open collecter gate or
inverter is open circuited at the collector internal to the integrated circuit package so that all
collector current may flow through an external load. An active high into any of the inverters
pulls the output low, drawing current through the LED and the current limiting resistor. The
LEDs would likely be part of a solid state relay driving 120VAC lamps for a museum exhibit,
not shown here.
8.11 Larger 5 & 6variable Karnaugh maps
Larger Karnaugh maps reduce larger logic designs. How large is large enough? That depends
on the number of inputs, fanins, to the logic circuit under consideration. One of the large
programmable logic companies has an answer.
Altera’s own data, extracted from its library of customer designs, supports the
value of heterogeneity. By examining logic cones, mapping them onto LUTbased
nodes and sorting them by the number of inputs that would be best at each node,
Altera found that the distribution of fanins was nearly flat between two and six
inputs, with a nice peak at five.
266 CHAPTER 8. KARNAUGH MAPPING
The answer is no more than six inputs for most all designs, and five inputs for the average
logic design. The five variable Karnaugh map follows.
00
01
11
10
AB
CDE
000 001 011 010 110 111 101 100
5 variable Karnaugh map (Gray code)
The older version of the five variable Kmap, a Gray Code map or reflection map, is shown
above. The top (and side for a 6variable map) of the map is numbered in full Gray code. The
Gray code reflects about the middle of the code. This style map is found in older texts. The
newer preferred style is below.
00
01
11
10
AB
CDE
000 001 011 010 110111101100
5 variable Karnaugh map (overlay)
The overlay version of the Karnaugh map, shown above, is simply two (four for a 6variable
map) identical maps except for the most significant bit of the 3bit address across the top. If
we look at the top of the map, we will see that the numbering is different from the previous
Gray code map. If we ignore the most significant digit of the 3digit numbers, the sequence 00,
01, 11, 10 is at the heading of both sub maps of the overlay map. The sequence of eight 3digit
numbers is not Gray code. Though the sequence of four of the least significant two bits is.
Let’s put our 5variable Karnaugh Map to use. Design a circuit which has a 5bit binary
input (A, B, C, D, E), with A being the MSB (Most Significant Bit). It must produce an output
logic High for any prime number detected in the input data.
8.11. LARGER 5 & 6VARIABLE KARNAUGH MAPS 267
00
01
11
10
AB
CDE
000 001 011 010 110 111 101 100
5 variable Karnaugh map (Gray code)
1 1
1 1
1 1
1 1 1
11
A BE
BCE ACD E
A B CEABD E
mirror line
1
A B C D
We show the solution above on the older Gray code (reflection) map for reference. The
prime numbers are (1,2,3,5,7,11,13,17,19,23,29,31). Plot a 1 in each corresponding cell. Then,
proceed with grouping of the cells. Finish by writing the simplified result. Note that 4cell
group A’B’E consists of two pairs of cell on both sides of the mirror line. The same is true of
the 2cell group AB’DE. It is a group of 2cells by being reflected about the mirror line. When
using this version of the Kmap look for mirror images in the other half of the map.
Out = A’B’E + B’C’E + A’C’DE + A’CD’E + ABCE + AB’DE + A’B’C’D
Below we show the more common version of the 5variable map, the overlay map.
AB
CDE
000 001 011 010 110111101100
5 variable Karnaugh map (overlay)
00
01
11
10
1 1 1 1
1 1
1 1 1
1 1
A BE
B CE
ACDE
ABDE
A CDE
ABCE
A B CD
1
If we compare the patterns in the two maps, some of the cells in the right half of the map
are moved around since the addressing across the top of the map is different. We also need to
268 CHAPTER 8. KARNAUGH MAPPING
take a different approach at spotting commonality between the two halves of the map. Overlay
one half of the map atop the other half. Any overlap from the top map to the lower map is a
potential group. The figure below shows that group AB’DE is composed of two stacked cells.
Group A’B’E consists of two stacked pairs of cells.
For the A’B’E group of 4cells ABCDE = 00xx1 for the group. That is A,B,E are the same
001 respectively for the group. And, CD=xx that is it varies, no commonality in CD=xx for the
group of 4cells. Since ABCDE = 00xx1, the group of 4cells is covered by A’B’XXE = A’B’E.
000 001 011 010
110111101
00
01
11
10
11
10
1 1
1
1 1
1 1
1
1 1
1
A BE
A BDE
1
The above 5variable overlay map is shown stacked.
An example of a six variable Karnaugh map follows. We have mentally stacked the four sub
maps to see the group of 4cells corresponding to Out = C’F’
8.11. LARGER 5 & 6VARIABLE KARNAUGH MAPS 269
000 001 011
110
000
001
011
010
110
110
010
010
110
1
1
1
1
010
ABC
DEF
Out = CF
A magnitude comparator (used to illustrate a 6variable Kmap) compares two binary num
bers, indicating if they are equal, greater than, or less than each other on three respective
outputs. A three bit magnitude comparator has two inputs A2A1A0 and B2B1B0 An integrated
circuit magnitude comparator (7485) would actually have four inputs, But, the Karnaugh map
below needs to be kept to a reasonable size. We will only solve for the A>B output.
Below, a 6variable Karnaugh map aids simplification of the logic for a 3bit magnitude
comparator. This is an overlay type of map. The binary address code across the top and down
the left side of the map is not a full 3bit Gray code. Though the 2bit address codes of the
four sub maps is Gray code. Find redundant expressions by stacking the four sub maps atop
one another (shown above). There could be cells common to all four maps, though not in the
example below. It does have cells common to pairs of sub maps.
A<B
A=B
A>B
A
B
Magnitude
Comparator
The A>B output above is ABC>XYZ on the map below.
270 CHAPTER 8. KARNAUGH MAPPING
6 variable Karnaugh map (overlay)
00
01
11
10
000 001 011 010 110111101100
00
01
11
10
ABC
XYZ
1
1
1
1
0
0
0
0
1
1
1 1
1 1
1 1 1 1
1 1 1 1
1 1 1 1
1 1 11
1
1 1
11
1
AX
BXY
ABY ABCZ
BCXZ
CXYZ
ACYZ
Out = AX+ABY+BXY+ABCZ+ACYZ+BCXZ+CXYZ
Where ever ABC is greater than XYZ, a 1 is plotted. In the first line ABC=000 cannot be
greater than any of the values of XYZ. No 1s in this line. In the second line, ABC=001, only
the first cell ABCXYZ= 001000 is ABC greater than XYZ. A single 1 is entered in the first cell
of the second line. The fourth line, ABC=010, has a pair of 1s. The third line, ABC=011 has
three 1s. Thus, the map is filled with 1s in any cells where ABC is greater than XXZ.
In grouping cells, form groups with adjacent sub maps if possible. All but one group of
16cells involves cells from pairs of the sub maps. Look for the following groups:
• 1 group of 16cells
• 2 groups of 8cells
• 4 groups of 4cells
The group of 16cells, AX’ occupies all of the lower right sub map; though, we don’t circle it on
the figure above.
One group of 8cells is composed of a group of 4cells in the upper sub map overlaying a
similar group in the lower left map. The second group of 8cells is composed of a similar group
of 4cells in the right sub map overlaying the same group of 4cells in the lower left map.
The four groups of 4cells are shown on the Karnaugh map above with the associated prod
uct terms. Along with the product terms for the two groups of 8cells and the group of 16cells,
the final SumOfProducts reduction is shown, all seven terms. Counting the 1s in the map,
8.11. LARGER 5 & 6VARIABLE KARNAUGH MAPS 271
there is a total of 16+6+6=28 ones. Before the Kmap logic reduction there would have been
28 product terms in our SOP output, each with 6inputs. The Karnaugh map yielded seven
product terms of four or less inputs. This is really what Karnaugh maps are all about!
The wiring diagram is not shown. However, here is the parts list for the 3bit magnitude
comparator for ABC>XYZ using 4 TTL logic family parts:
• 1 ea 7410 triple 3input NAND gate AX’, ABY’, BX’Y’
• 2 ea 7420 dual 4input NAND gate ABCZ’, ACY’Z’, BCX’Z’, CX’Y’Z’
• 1 ea 7430 8input NAND gate for output of 7Pterms
• REVIEW:
• Boolean algebra, Karnaugh maps, and CAD (Computer Aided Design) are methods of
logic simplification. The goal of logic simplification is a minimal cost solution.
• A minimal cost solution is a valid logic reduction with the minimum number of gates with
the minimum number of inputs.
• Venn diagrams allow us to visualize Boolean expressions, easing the transition to Kar
naugh maps.
• Karnaugh map cells are organized in Gray code order so that we may visualize redun
dancy in Boolean expressions which results in simplification.
• The more common SumOfProducts (Sum of Minters) expressions are implemented as
AND gates (products) feeding a single OR gate (sum).
• SumOfProducts expressions (ANDOR logic) are equivalent to a NANDNAND imple
mentation. All AND gates and OR gates are replaced by NAND gates.
• Less often used, ProductOfSums expressions are implemented as OR gates (sums) feed
ing into a single AND gate (product). ProductOfSums expressions are based on the 0s,
maxterms, in a Karnaugh map.
272 CHAPTER 8. KARNAUGH MAPPING
Chapter 9
COMBINATIONAL LOGIC
FUNCTIONS
Contents
9.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
9.2 A HalfAdder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
9.3 A FullAdder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
9.4 Decoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
9.5 Encoder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
9.6 Demultiplexers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
9.7 Multiplexers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
9.8 Using multiple combinational circuits . . . . . . . . . . . . . . . . . . . . . 294
Original author: David Zitzelsberger
9.1 Introduction
The term ”combinational” comes to us from mathematics. In mathematics a combination is an
unordered set, which is a formal way to say that nobody cares which order the items came in.
Most games work this way, if you rolled dice one at a time and get a 2 followed by a 3 it is the
same as if you had rolled a 3 followed by a 2. With combinational logic, the circuit produces
the same output regardless of the order the inputs are changed.
There are circuits which depend on the when the inputs change, these circuits are called
sequential logic. Even though you will not find the term ”sequential logic” in the chapter titles,
the next several chapters will discuss sequential logic.
Practical circuits will have a mix of combinational and sequential logic, with sequential
logic making sure everything happens in order and combinational logic performing functions
like arithmetic, logic, or conversion.
273
274 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
You have already used combinational circuits. Each logic gate discussed previously is a
combinational logic function. Let’s follow how two NAND gate works if we provide them inputs
in different orders.
We begin with both inputs being 0.
0 0
1
1
We then set one input high.
0
1
1
1
We then set the other input high.
11
0
0
So NAND gates do not care about the order of the inputs, and you will find the same true
of all the other gates covered up to this point (AND, XOR, OR, NOR, XNOR, and NOT).
9.2 A HalfAdder
As a first example of useful combinational logic, let’s build a device that can add two binary
digits together. We can quickly calculate what the answers should be:
9.3. A FULLADDER 275
0 + 0 = 0 0 + 1 = 1 1 + 0 = 1 1 + 1 = 102
So we well need two inputs (a and b) and two outputs. The low order output will be called
Σ because it represents the sum, and the high order output will be called Cout because it
represents the carry out.
The truth table is
ΣA B Cout
0 0 0 0
0 1 1 0
1 0 1 0
1 1 0 1
Simplifying boolean equations or making some Karnaugh map will produce the same circuit
shown below, but start by looking at the results. The Σ column is our familiar XOR gate, while
the Cout column is the AND gate. This device is called a halfadder for reasons that will make
sense in the next section.
A
B
Cout
Σ
or in ladder logic
L1 L2
A B
A B
A B Σ
Cout
9.3 A FullAdder
The halfadder is extremely useful until you want to add more that one binary digit quantities.
The slow way to develop a two binary digit adders would be to make a truth table and reduce it.
Then when you decide to make a three binary digit adder, do it again. Then when you decide to
276 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
make a four digit adder, do it again. Then when ... The circuits would be fast, but development
time would be slow.
Looking at a two binary digit sum shows what we need to extend addition to multiple binary
digits.
11
11
11

110
Look at how many inputs the middle column uses. Our adder needs three inputs; a, b, and
the carry from the previous sum, and we can use our twoinput adder to build a three input
adder.
Σ is the easy part. Normal arithmetic tells us that if Σ = a + b + Cin and Σ1 = a + b, then Σ
= Σ1 + Cin.
HA
A
B
Σ
Σ1
C1 HA
Cin
C2
L1 L2
A B
A B
A B
Σ
Σ1
Σ1
Σ1
C1
Σ1
Cin
Cin
Cin C2
What do we do with C1 and C2? Let’s look at three input sums and quickly calculate:
Cin + a + b = ?
0 + 0 + 0 = 0 0 + 0 + 1 = 1 0 + 1 + 0 = 1 0 +
1 + 1 = 10
9.3. A FULLADDER 277
1 + 0 + 0 = 1 1 + 0 + 1 = 10 1 + 1 + 0 = 10 1 +
1 + 1 = 11
If you have any concern about the low order bit, please confirm that the circuit and ladder
calculate it correctly.
In order to calculate the high order bit, notice that it is 1 in both cases when a + b produces
a C1. Also, the high order bit is 1 when a + b produces a Σ1 and Cin is a 1. So We will have a
carry when C1 OR (Σ1 AND Cin). Our complete three input adder is:
HA
A
B
Σ
Σ1
HA
Cin
Cout
L1 L2
A B
A B
A B
Σ
Σ1
Σ1
Σ1
C1
Σ1
Cin
Cin
Cin C2
C1 C2 Cout
For some designs, being able to eliminate one or more types of gates can be important, and
you can replace the final OR gate with an XOR gate without changing the results.
We can now connect two adders to add 2 bit quantities.
278 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
HA
FA
A0
A1
B0
B1
Σ0
Σ1
Cout
L1 L2
Σ1
C1
C1
A0
A0
A0
Σ0B0
B0
B0
A1
A1
A1
B1
B1
B1
TΣ1
TC1
TΣ1 C1
TΣ1 C1
TΣ1
TC1
TP1
TP1 Cout
A0 is the low order bit of A, A1 is the high order bit of A, B0 is the low order bit of B, B1 is
the high order bit of B, Σ0is the low order bit of the sum, Σ1 is the high order bit of the sum,
and Cout is the Carry.
A two binary digit adder would never be made this way. Instead the lowest order bits would
also go through a full adder.
9.3. A FULLADDER 279
FA
A0
A1
B0
B1
Σ0
Σ1
Cout
FA
Cin
280 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
L1 L2
Σ1
C1
C1
A0
A0
A0
Σ0
B0
B0
B0
A1
A1
A1
B1
B1
B1
TΣ1
TC1
TΣ1 C1
TΣ1 C1
TΣ1
TC1
TP1
TP1 Cout
TΣ0
TC0
TΣ0 Cin
TΣ0 Cin
TC0 Cin TP0
TC0 TP0
There are several reasons for this, one being that we can then allow a circuit to determine
whether the lowest order carry should be included in the sum. This allows for the chaining of
even larger sums. Consider two different ways to look at a four bit sum.
111 1<+ 11<+
0110  01  10
1011  10  11
9.3. A FULLADDER 281
     
10001 1 +100 +101
If we allow the program to add a two bit number and remember the carry for later, then
use that carry in the next sum the program can add any number of bits the user wants even
though we have only provided a twobit adder. Small PLCs can also be chained together for
larger numbers.
These full adders can also can be expanded to any number of bits space allows. As an
example, here’s how to do an 8 bit adder.
A0 B0
Σ0
FA
Cin
FA FA FA FA FA FA FA
A1 A2 A3 A4 A5 A6 A7B1 B2 B3 B4 B5 B6 B7
Σ1 Σ2 Σ3 Σ4 Σ5 Σ6 Σ7 Cout
This is the same result as using the two 2bit adders to make a 4bit adder and then us
ing two 4bit adders to make an 8bit adder or reduplicating ladder logic and updating the
numbers.
A0 B0
Σ0
FACin FA FA FA FA FA FA FA
A1 A2 A3 A4 A5 A6 A7B1 B2 B3 B4 B5 B6 B7
Σ1 Σ2 Σ3 Σ4 Σ5 Σ6 Σ7 Cout
2+ 2+ 2+ 2+
4+ 4+
Each ”2+” is a 2bit adder and made of two full adders. Each ”4+” is a 4bit adder and made
of two 2bit adders. And the result of two 4bit adders is the same 8bit adder we used full
adders to build.
For any large combinational circuit there are generally two approaches to design: you can
take simpler circuits and replicate them; or you can design the complex circuit as a complete
device.
Using simpler circuits to build complex circuits allows a you to spend less time designing
but then requires more time for signals to propagate through the transistors. The 8bit adder
design above has to wait for all the Cxout signals to move from A0 + B0 up to the inputs of Σ7.
If a designer builds an 8bit adder as a complete device simplified to a sum of products, then
each signal just travels through one NOT gate, one AND gate and one OR gate. A seventeen
input device has a truth table with 131,072 entries, and reducing 131,072 entries to a sum of
products will take some time.
When designing for systems that have a maximum allowed response time to provide the
final result, you can begin by using simpler circuits and then attempt to replace portions of the
circuit that are too slow. That way you spend most of your time on the portions of a circuit that
matter.
282 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
9.4 Decoder
A decoder is a circuit that changes a code into a set of signals. It is called a decoder because
it does the reverse of encoding, but we will begin our study of encoders and decoders with
decoders because they are simpler to design.
A common type of decoder is the line decoder which takes an ndigit binary number and
decodes it into 2n data lines. The simplest is the 1to2 line decoder. The truth table is
A D0D1
0
1
0
1 0
1
A is the address and D is the dataline. D0 is NOT A and D1 is A. The circuit looks like
D0
D1
A
L1 L2
A
A
D0
D1
Only slightly more complex is the 2to4 line decoder. The truth table is
A0A1 D3 D2 D1 D0
0 0 0 0
0
1
0
10
0
1 0
1
0
0
1
0
01
1
1
0
0 0
Developed into a circuit it looks like
9.4. DECODER 283
A0
A1
D0
D1
D2
D3
L1 L2
A1 A0 D0
A1
A1
A1
A0
A0
A0
D1
D2
D3
Larger line decoders can be designed in a similar fashion, but just like with the binary
adder there is a way to make larger decoders by combining smaller decoders. An alternate
circuit for the 2to4 line decoder is
284 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
1to2
Decoder
1to2
Decoder
A0
A1
D0
D1
D2
D3
Replacing the 1to2 Decoders with their circuits will show that both circuits are equivalent.
In a similar fashion a 3to8 line decoder can be made from a 1to2 line decoder and a 2to4
line decoder, and a 4to16 line decoder can be made from two 2to4 line decoders.
You might also consider making a 2to4 decoder ladder from 1to2 decoder ladders. If you
do it might look something like this:
9.4. DECODER 285
L1 L2
A0 T0,0
A1
A0 T0,1
T1,0
A1 T1,1
T1,0 T0,0 D0
D1
D2
D3
T1,0
T0,0
T0,1
T0,1
T1,1
T1,1
For some logic it may be required to build up logic like this. For an eightbit adder we only
know how to sum eight bits by summing one bit at a time. Usually it is easier to design ladder
logic from boolean equations or truth tables rather than design logic gates and then ”translate”
that into ladder logic.
A typical application of a line decoder circuit is to select among multiple devices. A circuit
needing to select among sixteen devices could have sixteen control lines to select which device
should ”listen”. With a decoder only four control lines are needed.
286 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
9.5 Encoder
An encoder is a circuit that changes a set of signals into a code. Let’s begin making a 2to1
line encoder truth table by reversing the 1to2 decoder truth table.
AD0D1
0
1
0
1 0
1
table.
This truth table is a little short. A complete truth table would be
D1 D0
0 0
0
1
0
0
1
11
1
A
One question we need to answer is what to do with those other inputs? Do we ignore them?
Do we have them generate an additional error output? In many circuits this problem is solved
by adding sequential logic in order to know not just what input is active but also which order
the inputs became active.
A more useful application of combinational encoder design is a binary to 7segment encoder.
The seven segments are given according
D0
D1 D2
D3
D4 D5
D6
Our truth table is:
9.5. ENCODER 287
I3 I2 I1 I0 D6 D5 D4 D3 D2 D1 D0
1 0 0 1 1 1 1 0 1 1
1 0 0 0 1 1 1 1 1 1
0
0
0
0
0
0
0
0
1
1
1
1
0
0
0
0 0 0 1 1 1 0 1 1 1
1
1
0
0
1
1
0 1 0 0 1 0 0 1 0
1
0
1
0
1
0 1 0 1 1 1 0 1
1
0
1
1
1
0 1 1 0 1 1
0
1
1
1 1 1 0 1 0
1
0
0 1 0 1 1
1
1
1
0
1 1 1
0 1 0
Deciding what to do with the remaining six entries of the truth table is easier with this
circuit. This circuit should not be expected to encode an undefined combination of inputs, so
we can leave them as ”don’t care” when we design the circuit. The boolean equations are
D0 = I3 + I1 + I3I2I1I0 + I3I2I1I0
D1 = I3 + I2I1 + I2I1 + I2I0
D2 = I2 + I3I2I1I0 + I3I2I1I0
D3 = I3 + I1I0 + I2I1
D4 = I1I0 + I2I1I0
D5 = I3 + I2 + I0
D6 = I3 + I1I0 + I3I2I1 + I3I2I1I0 + I3I2I1I0
and the circuit is
(See Figure 9.1.)
288 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
I 0 I 1 I 2 I 3
D
0
D
1 D
2
D
3
D
4 D
5
D
6
Figure 9.1: Sevensegment decoder gate level diagram.
9.6. DEMULTIPLEXERS 289
L1 L2
D0
D1
D2
D3
D4
D5
D6
I0
I0
I0
I0
I0
I0
I0
I0
I0
I0
I0
I0
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I3
I3
I3
I3
I3
I3
I3
I3
I3
I3
I3
I3
9.6 Demultiplexers
A demultiplexer, sometimes abbreviated dmux, is a circuit that has one input and more than
one output. It is used when a circuit wishes to send a signal to one of many devices. This
290 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
description sounds similar to the description given for a decoder, but a decoder is used to select
among many devices while a demultiplexer is used to send a signal among many devices.
A demultiplexer is used often enough that it has its own schematic symbol
D0
D1
I
A
The truth table for a 1to2 demultiplexer is
0 0 0 0
0 01 0
01 01
1 01 1
AI D0 D1
Using our 1to2 decoder as part of the circuit, we can express this circuit easily
1to2 line
decoder
I
A
D0
D1
9.6. DEMULTIPLEXERS 291
L1 L2
A
A
D0
D1
I
This circuit can be expanded two different ways. You can increase the number of signals
that get transmitted, or you can increase the number of inputs that get passed through. To
increase the number of inputs that get passed through just requires a larger line decoder.
Increasing the number of signals that get transmitted is even easier.
As an example, a device that passes one set of two signals among four signals is a ”twobit
1to2 demultiplexer”. Its circuit is
1to2 1to2
decoder decoder
I1
I0
A
D0
D2
D1
D3
or by expressing the circuit as
292 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
I0
I1
A
D0
D1
D2
D3
A
A
D0
D1
I0
A
A
D2
D3
I1
L1 L2
shows that it could be two onebit 1to2 demultiplexers without changing its expected be
havior.
A 1to4 demultiplexer can easily be built from 1to2 demultiplexers as follows.
9.7. MULTIPLEXERS 293
9.7 Multiplexers
A multiplexer, abbreviated mux, is a device that has multiple inputs and one output.
The schematic symbol for multiplexers is
I0
I1
A
D
The truth table for a 2to1 multiplexer is
0 0 0 0
0 0 1 0
0 1 0 0
0 1 1 1
1 0 0 1
1 0 1 0
1 1 0 1
1 1 1 1
I1 I0 A D
Using a 1to2 decoder as part of the circuit, we can express this circuit easily.
1to2 line
decoder
A
I0
I1
D
294 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
L1 L2
I1
I0 DA
A
Multiplexers can also be expanded with the same naming conventions as demultiplexers. A
4to1 multiplexer circuit is
That is the formal definition of a multiplexer. Informally, there is a lot of confusion. Both
demultiplexers and multiplexers have similar names, abbreviations, schematic symbols and
circuits, so confusion is easy. The term multiplexer, and the abbreviation mux, are often used
to also mean a demultiplexer, or a multiplexer and a demultiplexer working together. So when
you hear about a multiplexer, it may mean something quite different.
9.8 Using multiple combinational circuits
As an example of using several circuits together, we are going to make a device that will have
16 inputs, representing a four digit number, to a four digit 7segment display but using just
one binaryto7segment encoder.
First, the overall architecture of our circuit provides what looks like our the description
provided.
9.8. USING MULTIPLE COMBINATIONAL CIRCUITS 295
7segment
encoder
Follow this circuit through and you can confirm that it matches the description given above.
There are 16 primary inputs. There are two more inputs used to select which digit will be
displayed. There are 28 outputs to control the four digit 7segment display. Only four of the
primary inputs are encoded at a time. You may have noticed a potential question though.
When one of the digits are selected, what do the other three digits display? Review the
circuit for the demultiplexers and notice that any line not selected by the A input is zero. So
the other three digits are blank. We don’t have a problem, only one digit displays at a time.
Let’s get a perspective on just how complex this circuit is by looking at the equivalent ladder
logic.
296 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
D6
I0
I0
I0
I1
I1
I1
I1
I2
I2
I2
I3
I3
I3
I3
A1 A0 D0
A0A1
Q0,0
Q0,1
A0
A0
A1
A1 Q0,2
Q0,3
A1 A0 Q1,0
A1 A0 D1 Q1,1
A1 A0 Q1,2
A1 A0 Q1,3
A1 A0 Q2,0
A1 A0 Q2,1
A1 A0 D2 Q2,2
A1 A0 Q2,3
A1 A0 Q3,0
A1 A0 Q3,1
A1 A0 Q3,2
A1 A0 D3 Q3,3
A1 A0 Q4,0
A1 A0 Q4,1
A1 A0 Q4,2
A1 A0 Q4,3
A1 A0 Q5,0
A1 A0 Q5,1
A1 A0 Q5,2
A1 A0 Q5,3
A1 A0 Q6,0
A1 A0 Q6,1
A1 A0 Q6,2
A1 A0 Q6,3
D0
D0
D0
D1
D1
D1
D2
D2
D2
D3
D3
D3
D4
D5
D6
D4
D4
D4
D5
D5
D5
D6
D6
D6
L1 L2
A1 A0 D0,0 I0
A1 A0 D1,0 I0
A1 A0 D2,0 I0
A1 A0 D3,0 I0
A1 A0 D0,1 I1
A1 A0 D1,1 I1
A1 A0 D2,1 I1
A1 A0 D3,1 I1
A1 A0 D0,2 I2
A1 A0 D1,2 I2
A1 A0 D2,2 I2
A1 A0 D3,2 I2
A1 A0 D0,3 I3
A1 A0 D1,3 I3
A1 A0 D2,3 I3
A1 A0 D3,3 I3
D0
D1
D2
D3
D4
D5
I0
I0
I0
I0
I0
I0
I0
I0
I0
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I1
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I2
I3
I3
I3
I3
I3
I3
I3
I3
x
xx
xxx
Notice how quickly this large circuit was developed from smaller parts. This is true of most
complex circuits: they are composed of smaller parts allowing a designer to abstract away
some complexity and understand the circuit as a whole. Sometimes a designer can even take
components that others have designed and remove the detail design work.
In addition to the added quantity of gates, this design suffers from one additional weakness.
You can only see one display one digit at a time. If there was some way to rotate through the
9.8. USING MULTIPLE COMBINATIONAL CIRCUITS 297
four digits quickly, you could have the appearance of all four digits being displayed at the same
time. That is a job for a sequential circuit, which is the subject of the next several chapters.
298 CHAPTER 9. COMBINATIONAL LOGIC FUNCTIONS
Chapter 10
MULTIVIBRATORS
Contents
10.1 Digital logic with feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
10.2 The SR latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
10.3 The gated SR latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
10.4 The D latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
10.5 Edgetriggered latches: FlipFlops . . . . . . . . . . . . . . . . . . . . . . . . 310
10.6 The JK flipflop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
10.7 Asynchronous flipflop inputs . . . . . . . . . . . . . . . . . . . . . . . . . . . 317
10.8 Monostable multivibrators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
10.1 Digital logic with feedback
With simple gate and combinational logic circuits, there is a definite output state for any given
input state. Take the truth table of an OR gate, for instance:
299
300 CHAPTER 10. MULTIVIBRATORS
L1 L2
A
B
A B Output
00
0 1
01
1 1
0
1
1
1
A
B
Output
Output
For each of the four possible combinations of input states (00, 01, 10, and 11), there is
one, definite, unambiguous output state. Whether we’re dealing with a multitude of cascaded
gates or a single gate, that output state is determined by the truth table(s) for the gate(s) in
the circuit, and nothing else.
However, if we alter this gate circuit so as to give signal feedback from the output to one of
the inputs, strange things begin to happen:
10.1. DIGITAL LOGIC WITH FEEDBACK 301
L1 L2
A
A
Output
Output
A
0
1
Output
?
1
CR1
CR1
We know that if A is 1, the output must be 1, as well. Such is the nature of an OR gate: any
”high” (1) input forces the output ”high” (1). If A is ”low” (0), however, we cannot guarantee the
logic level or state of the output in our truth table. Since the output feeds back to one of the OR
gate’s inputs, and we know that any 1 input to an OR gates makes the output 1, this circuit
will ”latch” in the 1 output state after any time that A is 1. When A is 0, the output could be
either 0 or 1, depending on the circuit’s prior state! The proper way to complete the above truth
table would be to insert the word latch in place of the question mark, showing that the output
maintains its last state when A is 0.
Any digital circuit employing feedback is called a multivibrator. The example we just ex
plored with the OR gate was a very simple example of what is called a bistable multivibrator. It
is called ”bistable” because it can hold stable in one of two possible output states, either 0 or 1.
There are also monostable multivibrators, which have only one stable output state (that other
state being momentary), which we’ll explore later; and astable multivibrators, which have no
stable state (oscillating back and forth between an output of 0 and 1).
A very simple astable multivibrator is an inverter with the output fed directly back to the
input:
302 CHAPTER 10. MULTIVIBRATORS
L1 L2
Output
CR1 CR1
Inverter with feedback
When the input is 0, the output switches to 1. That 1 output gets fed back to the input as a
1. When the input is 1, the output switches to 0. That 0 output gets fed back to the input as a
0, and the cycle repeats itself. The result is a high frequency (several megahertz) oscillator, if
implemented with a solidstate (semiconductor) inverter gate:
If implemented with relay logic, the resulting oscillator will be considerably slower, cycling
at a frequency well within the audio range. The buzzer or vibrator circuit thus formed was
used extensively in early radio circuitry, as a way to convert steady, lowvoltage DC power
into pulsating DC power which could then be stepped up in voltage through a transformer to
produce the high voltage necessary for operating the vacuum tube amplifiers. Henry Ford’s
engineers also employed the buzzer/transformer circuit to create continuous high voltage for
operating the spark plugs on Model T automobile engines:
"Model T" highvoltage
ignition coil
Borrowing terminology from the old mechanical buzzer (vibrator) circuits, solidstate circuit
10.2. THE SR LATCH 303
engineers referred to any circuit with two or more vibrators linked together as a multivibrator.
The astable multivibrator mentioned previously, with only one ”vibrator,” is more commonly
implemented with multiple gates, as we’ll see later.
The most interesting and widely used multivibrators are of the bistable variety, so we’ll
explore them in detail now.
10.2 The SR latch
A bistable multivibrator has two stable states, as indicated by the prefix bi in its name. Typi
cally, one state is referred to as set and the other as reset. The simplest bistable device, there
fore, is known as a setreset, or SR, latch.
To create an SR latch, we can wire two NOR gates in such a way that the output of one
feeds back to the input of another, and vice versa, like this:
R
S
Q
Q
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
The Q and notQ outputs are supposed to be in opposite states. I say ”supposed to” because
making both the S and R inputs equal to 1 results in both Q and notQ being 0. For this reason,
having both S and R equal to 1 is called an invalid or illegal state for the SR multivibrator.
Otherwise, making S=1 and R=0 ”sets” the multivibrator so that Q=1 and notQ=0. Conversely,
making R=1 and S=0 ”resets” the multivibrator in the opposite state. When S and R are
both equal to 0, the multivibrator’s outputs ”latch” in their prior states. Note how the same
multivibrator function can be implemented in ladder logic, with the same results:
304 CHAPTER 10. MULTIVIBRATORS
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
L1 L2
CR1R
CR2
S CR2
CR1
CR1
CR2
Q
Q
By definition, a condition of Q=1 and notQ=0 is set. A condition of Q=0 and notQ=1 is
reset. These terms are universal in describing the output states of any multivibrator circuit.
The astute observer will note that the initial powerup condition of either the gate or ladder
variety of SR latch is such that both gates (coils) start in the deenergized mode. As such,
one would expect that the circuit will start up in an invalid condition, with both Q and not
Q outputs being in the same state. Actually, this is true! However, the invalid condition is
unstable with both S and R inputs inactive, and the circuit will quickly stabilize in either the
set or reset condition because one gate (or relay) is bound to react a little faster than the other.
If both gates (or coils) were precisely identical, they would oscillate between high and low like
an astable multivibrator upon powerup without ever reaching a point of stability! Fortunately
for cases like this, such a precise match of components is a rare possibility.
It must be noted that although an astable (continually oscillating) condition would be ex
tremely rare, there will most likely be a cycle or two of oscillation in the above circuit, and the
final state of the circuit (set or reset) after powerup would be unpredictable. The root of the
problem is a race condition between the two relays CR1 and CR2.
A race condition occurs when two mutuallyexclusive events are simultaneously initiated
through different circuit elements by a single cause. In this case, the circuit elements are
relays CR1 and CR2, and their deenergized states are mutually exclusive due to the normally
closed interlocking contacts. If one relay coil is deenergized, its normallyclosed contact will
keep the other coil energized, thus maintaining the circuit in one of two states (set or reset).
Interlocking prevents both relays from latching. However, if both relay coils start in their de
energized states (such as after the whole circuit has been deenergized and is then powered
up) both relays will ”race” to become latched on as they receive power (the ”single cause”)
through the normallyclosed contact of the other relay. One of those relays will inevitably
reach that condition before the other, thus opening its normallyclosed interlocking contact and
deenergizing the other relay coil. Which relay ”wins” this race is dependent on the physical
10.2. THE SR LATCH 305
characteristics of the relays and not the circuit design, so the designer cannot ensure which
state the circuit will fall into after powerup.
Race conditions should be avoided in circuit design primarily for the unpredictability that
will be created. One way to avoid such a condition is to insert a timedelay relay into the
circuit to disable one of the competing relays for a short time, giving the other one a clear
advantage. In other words, by purposely slowing down the deenergization of one relay, we
ensure that the other relay will always ”win” and the race results will always be predictable.
Here is an example of how a timedelay relay might be applied to the above circuit to avoid the
race condition:
L1 L2
CR1R
CR2
S CR2
CR1
CR1
CR2
Q
Q
TD1
TD1
1 second
When the circuit powers up, timedelay relay contact TD1 in the fifth rung down will delay
closing for 1 second. Having that contact open for 1 second prevents relay CR2 from energizing
through contact CR1 in its normallyclosed state after powerup. Therefore, relay CR1 will be
allowed to energize first (with a 1second head start), thus opening the normallyclosed CR1
contact in the fifth rung, preventing CR2 from being energized without the S input going active.
The end result is that the circuit powers up cleanly and predictably in the reset state with S=0
and R=0.
It should be mentioned that race conditions are not restricted to relay circuits. Solidstate
logic gate circuits may also suffer from the ill effects of race conditions if improperly designed.
Complex computer programs, for that matter, may also incur race problems if improperly de
signed. Race problems are a possibility for any sequential system, and may not be discovered
until some time after initial testing of the system. They can be very difficult problems to detect
and eliminate.
A practical application of an SR latch circuit might be for starting and stopping a motor,
306 CHAPTER 10. MULTIVIBRATORS
using normallyopen, momentary pushbutton switch contacts for both start (S) and stop (R)
switches, then energizing a motor contactor with either a CR1 or CR2 contact (or using a con
tactor in place of CR1 or CR2). Normally, a much simpler ladder logic circuit is employed, such
as this:
L1 L2
CR1
CR1
Start Stop
CR1
Motor "on"
In the above motor start/stop circuit, the CR1 contact in parallel with the start switch con
tact is referred to as a ”sealin” contact, because it ”seals” or latches control relay CR1 in the
energized state after the start switch has been released. To break the ”seal,” or to ”unlatch”
or ”reset” the circuit, the stop pushbutton is pressed, which deenergizes CR1 and restores the
sealin contact to its normally open status. Notice, however, that this circuit performs much the
same function as the SR latch. Also note that this circuit has no inherent instability problem
(if even a remote possibility) as does the doublerelay SR latch design.
In semiconductor form, SR latches come in prepackaged units so that you don’t have to
build them from individual gates. They are symbolized as such:
S Q
QR
• REVIEW:
• A bistable multivibrator is one with two stable output states.
• In a bistable multivibrator, the condition of Q=1 and notQ=0 is defined as set. A condition
of Q=0 and notQ=1 is conversely defined as reset. If Q and notQ happen to be forced to
the same state (both 0 or both 1), that state is referred to as invalid.
• In an SR latch, activation of the S input sets the circuit, while activation of the R input
resets the circuit. If both S and R inputs are activated simultaneously, the circuit will be
in an invalid condition.
10.3. THE GATED SR LATCH 307
• A race condition is a state in a sequential system where two mutuallyexclusive events
are simultaneously initiated by a single cause.
10.3 The gated SR latch
It is sometimes useful in logic circuits to have a multivibrator which changes state only when
certain conditions are met, regardless of its S and R input states. The conditional input is
called the enable, and is symbolized by the letter E. Study the following example to see how
this works:
R
S
Q
Q
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
E
0
0
0
0
1
1
1
1
E
When the E=0, the outputs of the two AND gates are forced to 0, regardless of the states of
either S or R. Consequently, the circuit behaves as though S and R were both 0, latching the
Q and notQ outputs in their last states. Only when the enable input is activated (1) will the
latch respond to the S and R inputs. Note the identical function in ladder logic:
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
E
0
0
0
0
1
1
1
1
L1 L2
CR1R
CR2
S CR2
CR1
CR1
CR2
Q
Q
E
E
308 CHAPTER 10. MULTIVIBRATORS
A practical application of this might be the same motor control circuit (with two normally
open pushbutton switches for start and stop), except with the addition of a master lockout input
(E) that disables both pushbuttons from having control over the motor when its low (0).
Once again, these multivibrator circuits are available as prepackaged semiconductor de
vices, and are symbolized as such:
S Q
QR
E
It is also common to see the enable input designated by the letters ”EN” instead of just ”E.”
• REVIEW:
• The enable input on a multivibrator must be activated for either S or R inputs to have
any effect on the output state.
• This enable input is sometimes labeled ”E”, and other times as ”EN”.
10.4 The D latch
Since the enable input on a gated SR latch provides a way to latch the Q and notQ outputs
without regard to the status of S or R, we can eliminate one of those inputs to create a multi
vibrator latch circuit with no ”illegal” input states. Such a circuit is called a D latch, and its
internal logic looks like this:
Q
Q
Q Q
latch latch
latch latch0 0
0 1
1 0
1 1
E
D
E D
0
01
1
Note that the R input has been replaced with the complement (inversion) of the old S input,
and the S input has been renamed to D. As with the gated SR latch, the D latch will not
respond to a signal input if the enable input is 0 – it simply stays latched in its last state.
When the enable input is 1, however, the Q output follows the D input.
Since the R input of the SR circuitry has been done away with, this latch has no ”invalid”
or ”illegal” state. Q and notQ are always opposite of one another. If the above diagram is
confusing at all, the next diagram should make the concept simpler:
10.4. THE D LATCH 309
S Q
QR
E
D
E
Like both the SR and gated SR latches, the D latch circuit may be found as its own
prepackaged circuit, complete with a standard symbol:
Q
QD
E
The D latch is nothing more than a gated SR latch with an inverter added to make R the
complement (inverse) of S. Let’s explore the ladder logic equivalent of a D latch, modified from
the basic ladder diagram of an SR latch:
0 0
0
0
1
1
1 1
latch latch
Q Q
10
L1 L2
CR1
CR2
CR2
CR1
CR1
CR2
Q
Q
E D
latch latch
1 0
D
D E
E
An application for the D latch is a 1bit memory circuit. You can ”write” (store) a 0 or 1 bit
in this latch circuit by making the enable input high (1) and setting D to whatever you want
the stored bit to be. When the enable input is made low (0), the latch ignores the status of
the D input and merrily holds the stored bit value, outputting at the stored value at Q, and its
inverse on output notQ.
310 CHAPTER 10. MULTIVIBRATORS
• REVIEW:
• A D latch is like an SR latch with only one input: the ”D” input. Activating the D input
sets the circuit, and deactivating the D input resets the circuit. Of course, this is only
if the enable input (E) is activated as well. Otherwise, the output(s) will be latched,
unresponsive to the state of the D input.
• D latches can be used as 1bit memory circuits, storing either a ”high” or a ”low” state
when disabled, and ”reading” new data from the D input when enabled.
10.5 Edgetriggered latches: FlipFlops
So far, we’ve studied both SR and D latch circuits with an enable inputs. The latch responds
to the data inputs (SR or D) only when the enable input is activated. In many digital appli
cations, however, it is desirable to limit the responsiveness of a latch circuit to a very short
period of time instead of the entire duration that the enabling input is activated. One method
of enabling a multivibrator circuit is called edge triggering, where the circuit’s data inputs have
control only during the time that the enable input is transitioning from one state to another.
Let’s compare timing diagrams for a normal D latch versus one that is edgetriggered:
D
E
Q
Q
Outputs respond to input (D)
Regular Dlatch response
during these time periods
10.5. EDGETRIGGERED LATCHES: FLIPFLOPS 311
D
E
Q
Q
Outputs respond to input (D)
Positive edgetriggered Dlatch response
only when enable signal transitions
from low to high
In the first timing diagram, the outputs respond to input D whenever the enable (E) input
is high, for however long it remains high. When the enable signal falls back to a low state, the
circuit remains latched. In the second timing diagram, we note a distinctly different response
in the circuit output(s): it only responds to the D input during that brief moment of time when
the enable signal changes, or transitions, from low to high. This is known as positive edge
triggering.
There is such a thing as negative edge triggering as well, and it produces the following
response to the same input signals:
D
E
Q
Q
Outputs respond to input (D)
only when enable signal transitions
Negative edgetriggered Dlatch response
from high to low
Whenever we enable a multivibrator circuit on the transitional edge of a squarewave en
able signal, we call it a flipflop instead of a latch. Consequently, and edgetriggered SR circuit
is more properly known as an SR flipflop, and an edgetriggered D circuit as a D flipflop. The
enable signal is renamed to be the clock signal. Also, we refer to the data inputs (S, R, and D,
312 CHAPTER 10. MULTIVIBRATORS
respectively) of these flipflops as synchronous inputs, because they have effect only at the time
of the clock pulse edge (transition), thereby synchronizing any output changes with that clock
pulse, rather than at the whim of the data inputs.
But, how do we actually accomplish this edgetriggering? To create a ”gated” SR latch from
a regular SR latch is easy enough with a couple of AND gates, but how do we implement logic
that only pays attention to the rising or falling edge of a changing digital signal? What we need
is a digital circuit that outputs a brief pulse whenever the input is activated for an arbitrary
period of time, and we can use the output of this circuit to briefly enable the latch. We’re
getting a little ahead of ourselves here, but this is actually a kind of monostable multivibrator,
which for now we’ll call a pulse detector.
Pulse detector
circuit
Output
Input
Input
Output
The duration of each output pulse is set by components in the pulse circuit itself. In ladder
logic, this can be accomplished quite easily through the use of a timedelay relay with a very
short delay time:
L1 L2
Input TD1
TD1 Input
Output
Implementing this timing function with semiconductor components is actually quite easy,
as it exploits the inherent time delay within every logic gate (known as propagation delay).
What we do is take an input signal and split it up two ways, then place a gate or a series
of gates in one of those signal paths just to delay it a bit, then have both the original signal
and its delayed counterpart enter into a twoinput gate that outputs a high signal for the brief
moment of time that the delayed signal has not yet caught up to the lowtohigh change in
the nondelayed signal. An example circuit for producing a clock pulse on a lowtohigh input
signal transition is shown here:
10.5. EDGETRIGGERED LATCHES: FLIPFLOPS 313
Input
Output
Input
Delayed input
Delayed input
Output
Brief period of time when
are high
both inputs of the AND gate
This circuit may be converted into a negativeedge pulse detector circuit with only a change
of the final gate from AND to NOR:
Input
Output
Input
Delayed input
Delayed input
Output
Brief period of time when
are low
both inputs of the NOR gate
Now that we know how a pulse detector can be made, we can show it attached to the enable
input of a latch to turn it into a flipflop. In this case, the circuit is a SR flipflop:
R
S
Q
Q
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
Pulse
detectorC
C
x
x
x
x
Only when the clock signal (C) is transitioning from low to high is the circuit responsive to
314 CHAPTER 10. MULTIVIBRATORS
the S and R inputs. For any other condition of the clock signal (”x”) the circuit will be latched.
A ladder logic version of the SR flipflop is shown here:
TD1C
C CR3
S R
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
0 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
L1 L2
CR1R
CR2
S CR2
CR1
CR1
CR2
Q
Q
CR3
CR3
x
x
x
x
TD1
C
Relay contact CR3 in the ladder diagram takes the place of the old E contact in the SR latch
circuit, and is closed only during the short time that both C is closed and timedelay contact
TR1 is closed. In either case (gate or ladder circuit), we see that the inputs S and R have no
effect unless C is transitioning from a low (0) to a high (1) state. Otherwise, the flipflop’s
outputs latch in their previous states.
It is important to note that the invalid state for the SR flipflop is maintained only for the
short period of time that the pulse detector circuit allows the latch to be enabled. After that
brief time period has elapsed, the outputs will latch into either the set or the reset state. Once
again, the problem of a race condition manifests itself. With no enable signal, an invalid output
state cannot be maintained. However, the valid ”latched” states of the multivibrator – set and
reset – are mutually exclusive to one another. Therefore, the two gates of the multivibrator
circuit will ”race” each other for supremacy, and whichever one attains a high output state first
will ”win.”
The block symbols for flipflops are slightly different from that of their respective latch
counterparts:
10.6. THE JK FLIPFLOP 315
S Q
QR
C
D
C
Q
Q
The triangle symbol next to the clock inputs tells us that these are edgetriggered devices,
and consequently that these are flipflops rather than latches. The symbols above are positive
edgetriggered: that is, they ”clock” on the rising edge (lowtohigh transition) of the clock
signal. Negative edgetriggered devices are symbolized with a bubble on the clock input line:
S
C
R
Q
Q
D
C
Q
Q
Both of the above flipflops will ”clock” on the falling edge (hightolow transition) of the
clock signal.
• REVIEW:
• A flipflop is a latch circuit with a ”pulse detector” circuit connected to the enable (E)
input, so that it is enabled only for a brief moment on either the rising or falling edge of
a clock pulse.
• Pulse detector circuits may be made from timedelay relays for ladder logic applications,
or from semiconductor gates (exploiting the phenomenon of propagation delay).
10.6 The JK flipflop
Another variation on a theme of bistable multivibrators is the JK flipflop. Essentially, this is
a modified version of an SR flipflop with no ”invalid” or ”illegal” output state. Look closely at
the following diagram to see how this is accomplished:
Q
Q
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
Pulse
detectorC
C
x
x
x
x
K
J
J K
toggle toggle
316 CHAPTER 10. MULTIVIBRATORS
What used to be the S and R inputs are now called the J and K inputs, respectively. The
old twoinput AND gates have been replaced with 3input AND gates, and the third input of
each gate receives feedback from the Q and notQ outputs. What this does for us is permit
the J input to have effect only when the circuit is reset, and permit the K input to have effect
only when the circuit is set. In other words, the two inputs are interlocked, to use a relay logic
term, so that they cannot both be activated simultaneously. If the circuit is ”set,” the J input
is inhibited by the 0 status of notQ through the lower AND gate; if the circuit is ”reset,” the K
input is inhibited by the 0 status of Q through the upper AND gate.
When both J and K inputs are 1, however, something unique happens. Because of the
selective inhibiting action of those 3input AND gates, a ”set” state inhibits input J so that the
flipflop acts as if J=0 while K=1 when in fact both are 1. On the next clock pulse, the outputs
will switch (”toggle”) from set (Q=1 and notQ=0) to reset (Q=0 and notQ=1). Conversely, a
”reset” state inhibits input K so that the flipflop acts as if J=1 and K=0 when in fact both are
1. The next clock pulse toggles the circuit again from reset to set.
See if you can follow this logical sequence with the ladder logic equivalent of the JK flip
flop:
TD1C
C CR3
0 0
0
0
1
1
1 1
latch latch
Q Q
0 1
1 0
latch latch
latch latch
latch latch
latch latch0 0
0 1
1 0
1 1
L1 L2
CR1
CR2
CR2
CR1
CR1
CR2
Q
Q
CR3
CR3
x
x
x
x
TD1
J
K
CR2
CR1
KJ
toggle toggle
C
The end result is that the SR flipflop’s ”invalid” state is eliminated (along with the race
condition it engendered) and we get a useful feature as a bonus: the ability to toggle between
the two (bistable) output states with every transition of the clock input signal.
There is no such thing as a JK latch, only JK flipflops. Without the edgetriggering of
10.7. ASYNCHRONOUS FLIPFLOP INPUTS 317
the clock input, the circuit would continuously toggle between its two output states when both
J and K were held high (1), making it an astable device instead of a bistable device in that
circumstance. If we want to preserve bistable operation for all combinations of input states,
we must use edgetriggering so that it toggles only when we tell it to, one step (clock pulse) at
a time.
The block symbol for a JK flipflop is a whole lot less frightening than its internal circuitry,
and just like the SR and D flipflops, JK flipflops come in two clock varieties (negative and
positive edgetriggered):
J Q
Q
C
K
J
C
K Q
Q
• REVIEW:
• A JK flipflop is nothing more than an SR flipflop with an added layer of feedback. This
feedback selectively enables one of the two set/reset inputs so that they cannot both carry
an active signal to the multivibrator circuit, thus eliminating the invalid condition.
• When both J and K inputs are activated, and the clock input is pulsed, the outputs (Q and
notQ) will swap states. That is, the circuit will toggle from a set state to a reset state, or
vice versa.
10.7 Asynchronous flipflop inputs
The normal data inputs to a flip flop (D, S and R, or J and K) are referred to as synchronous
inputs because they have effect on the outputs (Q and notQ) only in step, or in sync, with
the clock signal transitions. These extra inputs that I now bring to your attention are called
asynchronous because they can set or reset the flipflop regardless of the status of the clock
signal. Typically, they’re called preset and clear:
S Q
QR
C
D
C
Q
Q J Q
Q
C
K
PRE
CLR
PRE
CLR
PRE
CLR
318 CHAPTER 10. MULTIVIBRATORS
When the preset input is activated, the flipflop will be set (Q=1, notQ=0) regardless of
any of the synchronous inputs or the clock. When the clear input is activated, the flipflop will
be reset (Q=0, notQ=1), regardless of any of the synchronous inputs or the clock. So, what
happens if both preset and clear inputs are activated? Surprise, surprise: we get an invalid
state on the output, where Q and notQ go to the same state, the same as our old friend, the
SR latch! Preset and clear inputs find use when multiple flipflops are ganged together to
perform a function on a multibit binary word, and a single line is needed to set or reset them
all at once.
Asynchronous inputs, just like synchronous inputs, can be engineered to be activehigh or
activelow. If they’re activelow, there will be an inverting bubble at that input lead on the
block symbol, just like the negative edgetrigger clock inputs.
S Q
QR
C
D
C
Q
Q J Q
Q
C
K
PRE
CLR
PRE
CLR
PRE
CLR
Sometimes the designations ”PRE” and ”CLR” will be shown with inversion bars above
them, to further denote the negative logic of these inputs:
S Q
QR
C
D
C
Q
Q J Q
Q
C
K
PRE
CLR
PRE
CLR
PRE
CLR
• REVIEW:
• Asynchronous inputs on a flipflop have control over the outputs (Q and notQ) regardless
of clock input status.
• These inputs are called the preset (PRE) and clear (CLR). The preset input drives the
flipflop to a set state while the clear input drives it to a reset state.
10.8. MONOSTABLE MULTIVIBRATORS 319
• It is possible to drive the outputs of a JK flipflop to an invalid condition using the asyn
chronous inputs, because all feedback within the multivibrator circuit is overridden.
10.8 Monostable multivibrators
We’ve already seen one example of a monostable multivibrator in use: the pulse detector used
within the circuitry of flipflops, to enable the latch portion for a brief time when the clock
input signal transitions from either low to high or high to low. The pulse detector is classified
as a monostable multivibrator because it has only one stable state. By stable, I mean a state of
output where the device is able to latch or hold to forever, without external prodding. A latch
or flipflop, being a bistable device, can hold in either the ”set” or ”reset” state for an indefinite
period of time. Once its set or reset, it will continue to latch in that state unless prompted to
change by an external input. A monostable device, on the other hand, is only able to hold in
one particular state indefinitely. Its other state can only be held momentarily when triggered
by an external input.
A mechanical analogy of a monostable device would be a momentary contact pushbutton
switch, which springreturns to its normal (stable) position when pressure is removed from its
button actuator. Likewise, a standard wall (toggle) switch, such as the type used to turn lights
on and off in a house, is a bistable device. It can latch in one of two modes: on or off.
All monostable multivibrators are timed devices. That is, their unstable output state will
hold only for a certain minimum amount of time before returning to its stable state. With
semiconductor monostable circuits, this timing function is typically accomplished through the
use of resistors and capacitors, making use of the exponential charging rates of RC circuits. A
comparator is often used to compare the voltage across the charging (or discharging) capacitor
with a steady reference voltage, and the on/off output of the comparator used for a logic signal.
With ladder logic, time delays are accomplished with timedelay relays, which can be con
structed with semiconductor/RC circuits like that just mentioned, or mechanical delay devices
which impede the immediate motion of the relay’s armature. Note the design and operation of
the pulse detector circuit in ladder logic:
L1 L2
Input TD1
TD1 Input
Output
Input
Output
1 second
1 second
320 CHAPTER 10. MULTIVIBRATORS
No matter how long the input signal stays high (1), the output remains high for just 1
second of time, then returns to its normal (stable) low state.
For some applications, it is necessary to have a monostable device that outputs a longer
pulse than the input pulse which triggers it. Consider the following ladder logic circuit:
L1 L2
Input TD1
TD1
Output
Input
Output
10 seconds
10 seconds
10 seconds 10 seconds
When the input contact closes, TD1 contact immediately closes, and stays closed for 10
seconds after the input contact opens. No matter how short the input pulse is, the output
stays high (1) for exactly 10 seconds after the input drops low again. This kind of monostable
multivibrator is called a oneshot. More specifically, it is a retriggerable oneshot, because the
timing begins after the input drops to a low state, meaning that multiple input pulses within
10 seconds of each other will maintain a continuous high output:
Input
Output
10 seconds
"Retriggering" action
One application for a retriggerable oneshot is that of a single mechanical contact de
bouncer. As you can see from the above timing diagram, the output will remain high despite
”bouncing” of the input signal from a mechanical switch. Of course, in a reallife switch de
bouncer circuit, you’d probably want to use a time delay of much shorter duration than 10
seconds, as you only need to ”debounce” pulses that are in the millisecond range.
10.8. MONOSTABLE MULTIVIBRATORS 321
+V
Rpulldown
Oneshot
"Dirty" signal
"Clean" signal
Switch
actuated
momentarily
What if we only wanted a 10 second timed pulse output from a relay logic circuit, regardless
of how many input pulses we received or how longlived they may be? In that case, we’d have
to couple a pulsedetector circuit to the retriggerable oneshot time delay circuit, like this:
L1 L2
Input
TD1
Output
Input
Output
10 seconds
Input
TD2
TD2
0.5 second
TD1 TD2
10 sec. 10 sec. 10 sec.
Time delay relay TD1 provides an ”on” pulse to time delay relay coil TD2 for an arbitrarily
short moment (in this circuit, for at least 0.5 second each time the input contact is actuated).
As soon as TD2 is energized, the normallyclosed, timedclosed TD2 contact in series with it
prevents coil TD2 from being reenergized as long as its timing out (10 seconds). This effectively
makes it unresponsive to any more actuations of the input switch during that 10 second period.
Only after TD2 times out does the normallyclosed, timedclosed TD2 contact in series with
it allow coil TD2 to be energized again. This type of oneshot is called a nonretriggerable one
shot.
322 CHAPTER 10. MULTIVIBRATORS
Oneshot multivibrators of both the retriggerable and nonretriggerable variety find wide
application in industry for siren actuation and machine sequencing, where an intermittent
input signal produces an output signal of a set time.
• REVIEW:
• A monostable multivibrator has only one stable output state. The other output state can
only be maintained temporarily.
• Monostable multivibrators, sometimes called oneshots, come in two basic varieties: re
triggerable and nonretriggerable.
• Oneshot circuits with very short time settings may be used to debounce the ”dirty” sig
nals created by mechanical switch contacts.
Chapter 11
SEQUENTIAL
CIRCUITSCOUNTERS
Contents
11.1 Binary count sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323
11.2 Asynchronous counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 325
11.3 Synchronous counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 332
11.4 Counter modulus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
11.5 Finite State Machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
*** INCOMPLETE ***
11.1 Binary count sequence
If we examine a fourbit binary count sequence from 0000 to 1111, a definite pattern will be
evident in the ”oscillations” of the bits between 0 and 1:
323
324 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
0 0 0 0
0 0 0 1
0 0
0 0
1 0
1 1
1
1
1
1
0
0
0
0
0 0
0 1
1 0
1 1
1
1
1
1
1
1
1
1
0 0 0
0 0 1
0
0
1 0
1 1
1
1
1
1
0 0
0 1
1 0
1 1
Note how the least significant bit (LSB) toggles between 0 and 1 for every step in the count
sequence, while each succeeding bit toggles at onehalf the frequency of the one before it. The
most significant bit (MSB) only toggles once during the entire sixteenstep count sequence: at
the transition between 7 (0111) and 8 (1000).
If we wanted to design a digital circuit to ”count” in fourbit binary, all we would have
to do is design a series of frequency divider circuits, each circuit dividing the frequency of a
squarewave pulse by a factor of 2:
(LSB)
(MSB)
0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
JK flipflops are ideally suited for this task, because they have the ability to ”toggle” their
output state at the command of a clock pulse when both J and K inputs are made ”high” (1):
11.2. ASYNCHRONOUS COUNTERS 325
J Q
Q
C
K
Vdd
signal A
signal B
A
B
0 1 0 1 0 1 0 1 0 1 0 1
0 0 0 0 01 1 1 1 1 1 0
If we consider the two signals (A and B) in this circuit to represent two bits of a binary
number, signal A being the LSB and signal B being the MSB, we see that the count sequence
is backward: from 11 to 10 to 01 to 00 and back again to 11. Although it might not be counting
in the direction we might have assumed, at least it counts!
The following sections explore different types of counter circuits, all made with JK flip
flops, and all based on the exploitation of that flipflop’s toggle mode of operation.
• REVIEW:
• Binary count sequences follow a pattern of octave frequency division: the frequency of
oscillation for each bit, from LSB to MSB, follows a dividebytwo pattern. In other words,
the LSB will oscillate at the highest frequency, followed by the next bit at onehalf the
LSB’s frequency, and the next bit at onehalf the frequency of the bit before it, etc.
• Circuits may be built that ”count” in a binary sequence, using JK flipflops set up in the
”toggle” mode.
11.2 Asynchronous counters
In the previous section, we saw a circuit using one JK flipflop that counted backward in a two
bit binary sequence, from 11 to 10 to 01 to 00. Since it would be desirable to have a circuit that
could count forward and not just backward, it would be worthwhile to examine a forward count
sequence again and look for more patterns that might indicate how to build such a circuit.
Since we know that binary count sequences follow a pattern of octave (factor of 2) fre
quency division, and that JK flipflop multivibrators set up for the ”toggle” mode are capable
of performing this type of frequency division, we can envision a circuit made up of several JK
flipflops, cascaded to produce four bits of output. The main problem facing us is to determine
how to connect these flipflops together so that they toggle at the right times to produce the
326 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
proper binary sequence. Examine the following binary count sequence, paying attention to
patterns preceding the ”toggling” of a bit between 0 and 1:
0 0 0 0
0 0 0 1
0 0
0 0
1 0
1 1
1
1
1
1
0
0
0
0
0 0
0 1
1 0
1 1
1
1
1
1
1
1
1
1
0 0 0
0 0 1
0
0
1 0
1 1
1
1
1
1
0 0
0 1
1 0
1 1
Note that each bit in this fourbit sequence toggles when the bit before it (the bit having a
lesser significance, or placeweight), toggles in a particular direction: from 1 to 0. Small arrows
indicate those points in the sequence where a bit toggles, the head of the arrow pointing to the
previous bit transitioning from a ”high” (1) state to a ”low” (0) state:
0 0 0 0
0 0 0 1
0 0
0 0
1 0
1 1
1
1
1
1
0
0
0
0
0 0
0 1
1 0
1 1
1
1
1
1
1
1
1
1
0 0 0
0 0 1
0
0
1 0
1 1
1
1
1
1
0 0
0 1
1 0
1 1
Starting with four JK flipflops connected in such a way to always be in the ”toggle” mode,
we need to determine how to connect the clock inputs in such a way so that each succeeding bit
toggles when the bit before it transitions from 1 to 0. The Q outputs of each flipflop will serve
11.2. ASYNCHRONOUS COUNTERS 327
as the respective binary bits of the final, fourbit count:
J Q
Q
C
K
Vdd
J Q
Q
C
K
Vdd
?
J Q
Q
C
K
Vdd
?
J Q
Q
C
K
Vdd
?
Q0 Q1 Q2 Q3
If we used flipflops with negativeedge triggering (bubble symbols on the clock inputs), we
could simply connect the clock input of each flipflop to the Q output of the flipflop before it,
so that when the bit before it changes from a 1 to a 0, the ”falling edge” of that signal would
”clock” the next flipflop to toggle the next bit:
J Q
Q
C
K
Vdd Vdd Vdd Vdd
Q0 Q1 Q2 Q3
J
C
K Q
Q J
C
K Q
Q J
C
K Q
Q
A fourbit "up" counter
This circuit would yield the following output waveforms, when ”clocked” by a repetitive
source of pulses from an oscillator:
0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
Q0
Q1
Q2
Q3
Clock
The first flipflop (the one with the Q0 output), has a positiveedge triggered clock input,
so it toggles with each rising edge of the clock signal. Notice how the clock signal in this
example has a duty cycle less than 50%. I’ve shown the signal in this manner for the purpose of
328 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
demonstrating how the clock signal need not be symmetrical to obtain reliable, ”clean” output
bits in our fourbit binary sequence. In the very first flipflop circuit shown in this chapter, I
used the clock signal itself as one of the output bits. This is a bad practice in counter design,
though, because it necessitates the use of a square wave signal with a 50% duty cycle (”high”
time = ”low” time) in order to obtain a count sequence where each and every step pauses for
the same amount of time. Using one JK flipflop for each output bit, however, relieves us of
the necessity of having a symmetrical clock signal, allowing the use of practically any variety
of high/low waveform to increment the count sequence.
As indicated by all the other arrows in the pulse diagram, each succeeding output bit is
toggled by the action of the preceding bit transitioning from ”high” (1) to ”low” (0). This is the
pattern necessary to generate an ”up” count sequence.
A less obvious solution for generating an ”up” sequence using positiveedge triggered flip
flops is to ”clock” each flipflop using the Q’ output of the preceding flipflop rather than the Q
output. Since the Q’ output will always be the exact opposite state of the Q output on a JK
flipflop (no invalid states with this type of flipflop), a hightolow transition on the Q output
will be accompanied by a lowtohigh transition on the Q’ output. In other words, each time the
Q output of a flipflop transitions from 1 to 0, the Q’ output of the same flipflop will transition
from 0 to 1, providing the positivegoing clock pulse we would need to toggle a positiveedge
triggered flipflop at the right moment:
J Q
Q
C
K
Vdd Vdd Vdd Vdd
Q0 Q1 Q2 Q3
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
A different way of making a fourbit "up" counter
One way we could expand the capabilities of either of these two counter circuits is to regard
the Q’ outputs as another set of four binary bits. If we examine the pulse diagram for such
a circuit, we see that the Q’ outputs generate a downcounting sequence, while the Q outputs
generate an upcounting sequence:
11.2. ASYNCHRONOUS COUNTERS 329
J Q
Q
C
K
Vdd Vdd Vdd Vdd
Q0 Q1 Q2 Q3
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3
A simultaneous "up" and "down" counter
0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
Q0
Q1
Q2
Q3
1 0 1 0 1 0 1 0 1 0 1 0 1 0 1 0
1 1 0 0 1 1 0 0 1 1 0 0 1 1 0 0
1 1 1 1 0 0 0 0 0 0 0 01 1 1 1
1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0
Q0
Q1
Q2
Q3
"Up" count sequence
"Down" count sequence
Unfortunately, all of the counter circuits shown thusfar share a common problem: the ripple
effect. This effect is seen in certain types of binary adder and data conversion circuits, and is
due to accumulative propagation delays between cascaded gates. When the Q output of a flip
flop transitions from 1 to 0, it commands the next flipflop to toggle. If the next flipflop toggle
is a transition from 1 to 0, it will command the flipflop after it to toggle as well, and so on.
However, since there is always some small amount of propagation delay between the command
to toggle (the clock pulse) and the actual toggle response (Q and Q’ outputs changing states),
any subsequent flipflops to be toggled will toggle some time after the first flipflop has toggled.
Thus, when multiple bits toggle in a binary count sequence, they will not all toggle at exactly
330 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
the same time:
0 1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
Q0
Q1
Q2
Q3
Accumulated
delay
Pulse diagram showing (exaggerated) propagation delays
As you can see, the more bits that toggle with a given clock pulse, the more severe the
accumulated delay time from LSB to MSB. When a clock pulse occurs at such a transition
point (say, on the transition from 0111 to 1000), the output bits will ”ripple” in sequence from
LSB to MSB, as each succeeding bit toggles and commands the next bit to toggle as well, with
a small amount of propagation delay between each bit toggle. If we take a closeup look at this
effect during the transition from 0111 to 1000, we can see that there will be false output counts
generated in the brief time period that the ”ripple” effect takes place:
Q0
Q1
Q2
Q3
1
0
1
1
1
0
0
00 0 0
001
1
0
1
0 0
0
7 8
Count CountFalse
counts
delay
delay
delay
Instead of cleanly transitioning from a ”0111” output to a ”1000” output, the counter circuit
will very quickly ripple from 0111 to 0110 to 0100 to 0000 to 1000, or from 7 to 6 to 4 to 0 and
11.2. ASYNCHRONOUS COUNTERS 331
then to 8. This behavior earns the counter circuit the name of ripple counter, or asynchronous
counter.
In many applications, this effect is tolerable, since the ripple happens very, very quickly
(the width of the delays has been exaggerated here as an aid to understanding the effects). If
all we wanted to do was drive a set of lightemitting diodes (LEDs) with the counter’s outputs,
for example, this brief ripple would be of no consequence at all. However, if we wished to
use this counter to drive the ”select” inputs of a multiplexer, index a memory pointer in a
microprocessor (computer) circuit, or perform some other task where false outputs could cause
spurious errors, it would not be acceptable. There is a way to use this type of counter circuit in
applications sensitive to false, ripplegenerated outputs, and it involves a principle known as
strobing.
Most decoder and multiplexer circuits are equipped with at least one input called the ”en
able.” The output(s) of such a circuit will be active only when the enable input is made active.
We can use this enable input to strobe the circuit receiving the ripple counter’s output so that
it is disabled (and thus not responding to the counter output) during the brief period of time in
which the counter outputs might be rippling, and enabled only when sufficient time has passed
since the last clock pulse that all rippling will have ceased. In most cases, the strobing signal
can be the same clock pulse that drives the counter circuit:
J Q
Q
C
K
Vdd Vdd Vdd Vdd
Q0 Q1 Q2 Q3
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
EN
Binary
count
input
Outputs
Receiving circuit
Counter circuit
Clock signal
With an activelow Enable input, the receiving circuit will respond to the binary count
of the fourbit counter circuit only when the clock signal is ”low.” As soon as the clock pulse
goes ”high,” the receiving circuit stops responding to the counter circuit’s output. Since the
counter circuit is positiveedge triggered (as determined by the first flipflop clock input), all
the counting action takes place on the lowtohigh transition of the clock signal, meaning that
the receiving circuit will become disabled just before any toggling occurs on the counter circuit’s
four output bits. The receiving circuit will not become enabled until the clock signal returns
to a low state, which should be a long enough time after all rippling has ceased to be ”safe” to
allow the new count to have effect on the receiving circuit. The crucial parameter here is the
clock signal’s ”high” time: it must be at least as long as the maximum expected ripple period of
the counter circuit. If not, the clock signal will prematurely enable the receiving circuit, while
some rippling is still taking place.
332 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
Another disadvantage of the asynchronous, or ripple, counter circuit is limited speed. While
all gate circuits are limited in terms of maximum signal frequency, the design of asynchronous
counter circuits compounds this problem by making propagation delays additive. Thus, even if
strobing is used in the receiving circuit, an asynchronous counter circuit cannot be clocked at
any frequency higher than that which allows the greatest possible accumulated propagation
delay to elapse well before the next pulse.
The solution to this problem is a counter circuit that avoids ripple altogether. Such a
counter circuit would eliminate the need to design a ”strobing” feature into whatever digi
tal circuits use the counter output as an input, and would also enjoy a much greater operating
speed than its asynchronous equivalent. This design of counter circuit is the subject of the next
section.
• REVIEW:
• An ”up” counter may be made by connecting the clock inputs of positiveedge triggered
JK flipflops to the Q’ outputs of the preceding flipflops. Another way is to use negative
edge triggered flipflops, connecting the clock inputs to the Q outputs of the preceding
flipflops. In either case, the J and K inputs of all flipflops are connected to Vcc or Vdd so
as to always be ”high.”
• Counter circuits made from cascaded JK flipflops where each clock input receives its
pulses from the output of the previous flipflop invariably exhibit a ripple effect, where
false output counts are generated between some steps of the count sequence. These types
of counter circuits are called asynchronous counters, or ripple counters.
• Strobing is a technique applied to circuits receiving the output of an asynchronous (ripple)
counter, so that the false counts generated during the ripple time will have no ill effect.
Essentially, the enable input of such a circuit is connected to the counter’s clock pulse in
such a way that it is enabled only when the counter outputs are not changing, and will
be disabled during those periods of changing counter outputs where ripple occurs.
11.3 Synchronous counters
A synchronous counter, in contrast to an asynchronous counter, is one whose output bits change
state simultaneously, with no ripple. The only way we can build such a counter circuit from
JK flipflops is to connect all the clock inputs together, so that each and every flipflop receives
the exact same clock pulse at the exact same time:
11.3. SYNCHRONOUS COUNTERS 333
J Q
Q
C
K
J Q
Q
C
K
?
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3
?
?
?
?
?
?
?
Now, the question is, what do we do with the J and K inputs? We know that we still have to
maintain the same dividebytwo frequency pattern in order to count in a binary sequence, and
that this pattern is best achieved utilizing the ”toggle” mode of the flipflop, so the fact that the
J and K inputs must both be (at times) ”high” is clear. However, if we simply connect all the
J and K inputs to the positive rail of the power supply as we did in the asynchronous circuit,
this would clearly not work because all the flipflops would toggle at the same time: with each
and every clock pulse!
J Q
Q
C
K
Vdd
J Q
Q
C
K
Vdd
J Q
Q
C
K
Vdd
J Q
Q
C
K
Vdd
Q0 Q1 Q2 Q3
This circuit will not function as a counter!
Let’s examine the fourbit binary counting sequence again, and see if there are any other
patterns that predict the toggling of a bit. Asynchronous counter circuit design is based on the
fact that each bit toggle happens at the same time that the preceding bit toggles from a ”high”
to a ”low” (from 1 to 0). Since we cannot clock the toggling of a bit based on the toggling of
a previous bit in a synchronous counter circuit (to do so would create a ripple effect) we must
find some other pattern in the counting sequence that can be used to trigger a bit toggle:
Examining the fourbit binary count sequence, another predictive pattern can be seen. No
tice that just before a bit toggles, all preceding bits are ”high:”
334 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
0 0 0 0
0 0 0 1
0 0
0 0
1 0
1 1
1
1
1
1
0
0
0
0
0 0
0 1
1 0
1 1
1
1
1
1
1
1
1
1
0 0 0
0 0 1
0
0
1 0
1 1
1
1
1
1
0 0
0 1
1 0
1 1
This pattern is also something we can exploit in designing a counter circuit. If we enable
each JK flipflop to toggle based on whether or not all preceding flipflop outputs (Q) are ”high,”
we can obtain the same counting sequence as the asynchronous circuit without the ripple effect,
since each flipflop in this circuit will be clocked at exactly the same time:
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3
Vdd
This flipflop This flipflop This flipflop This flipflop
toggles on every
clock pulse
toggles only if
Q0 is "high"
toggles only if toggles only if
Q0 AND Q1
are "high"
Q0 AND Q1 AND Q2
are "high"
A fourbit synchronous "up" counter
The result is a fourbit synchronous ”up” counter. Each of the higherorder flipflops are
made ready to toggle (both J and K inputs ”high”) if the Q outputs of all previous flipflops are
”high.” Otherwise, the J and K inputs for that flipflop will both be ”low,” placing it into the
”latch” mode where it will maintain its present output state at the next clock pulse. Since the
first (LSB) flipflop needs to toggle at every clock pulse, its J and K inputs are connected to Vcc
or Vdd, where they will be ”high” all the time. The next flipflop need only ”recognize” that the
first flipflop’s Q output is high to be made ready to toggle, so no AND gate is needed. However,
11.3. SYNCHRONOUS COUNTERS 335
the remaining flipflops should be made ready to toggle only when all lowerorder output bits
are ”high,” thus the need for AND gates.
To make a synchronous ”down” counter, we need to build the circuit to recognize the appro
priate bit patterns predicting each toggle state while counting down. Not surprisingly, when
we examine the fourbit binary count sequence, we see that all preceding bits are ”low” prior
to a toggle (following the sequence from bottom to top):
0 0 0 0
0 0 0 1
0 0
0 0
1 0
1 1
1
1
1
1
0
0
0
0
0 0
0 1
1 0
1 1
1
1
1
1
1
1
1
1
0 0 0
0 0 1
0
0
1 0
1 1
1
1
1
1
0 0
0 1
1 0
1 1
Since each JK flipflop comes equipped with a Q’ output as well as a Q output, we can use
the Q’ outputs to enable the toggle mode on each succeeding flipflop, being that each Q’ will be
”high” every time that the respective Q is ”low:”
336 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3
Vdd
This flipflop This flipflop This flipflop This flipflop
toggles on every
clock pulse
toggles only if
Q0 is "high"
toggles only if toggles only if
Q0 AND Q1
are "high"
Q0 AND Q1 AND Q2
are "high"
A fourbit synchronous "down" counter
Taking this idea one step further, we can build a counter circuit with selectable between
”up” and ”down” count modes by having dual lines of AND gates detecting the appropriate
bit conditions for an ”up” and a ”down” counting sequence, respectively, then use OR gates to
combine the AND gate outputs to the J and K inputs of each succeeding flipflop:
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3Vdd
A fourbit synchronous "up/down" counter
J Q
Q
C
K
J Q
Q
C
K
Up/Down
This circuit isn’t as complex as it might first appear. The Up/Down control input line simply
enables either the upper string or lower string of AND gates to pass the Q/Q’ outputs to the
succeeding stages of flipflops. If the Up/Down control line is ”high,” the top AND gates become
enabled, and the circuit functions exactly the same as the first (”up”) synchronous counter
circuit shown in this section. If the Up/Down control line is made ”low,” the bottom AND gates
become enabled, and the circuit functions identically to the second (”down” counter) circuit
shown in this section.
To illustrate, here is a diagram showing the circuit in the ”up” counting mode (all disabled
circuitry shown in grey rather than black):
11.3. SYNCHRONOUS COUNTERS 337
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3Vdd
J Q
Q
C
K
J Q
Q
C
K
Vdd
Up Down/
Counter in "up" counting mode
Here, shown in the ”down” counting mode, with the same grey coloring representing dis
abled circuitry:
J Q
Q
C
K
J Q
Q
C
K
Q0 Q1 Q2 Q3Vdd
J Q
Q
C
K
J Q
Q
C
K
Vdd
Up Down/
Counter in "down" counting mode
Up/down counter circuits are very useful devices. A common application is in machine
motion control, where devices called rotary shaft encoders convert mechanical rotation into a
series of electrical pulses, these pulses ”clocking” a counter circuit to track total motion:
Rotary shaft encoder
LED
Light sensor
(phototransistor)
Counter
Q0 Q1 Q2 Q3
As the machine moves, it turns the encoder shaft, making and breaking the light beam
between LED and phototransistor, thereby generating clock pulses to increment the counter
circuit. Thus, the counter integrates, or accumulates, total motion of the shaft, serving as
an electronic indication of how far the machine has moved. If all we care about is tracking
total motion, and do not care to account for changes in the direction of motion, this arrange
ment will suffice. However, if we wish the counter to increment with one direction of motion
and decrement with the reverse direction of motion, we must use an up/down counter, and an
338 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
encoder/decoding circuit having the ability to discriminate between different directions.
If we redesign the encoder to have two sets of LED/phototransistor pairs, those pairs
aligned such that their squarewave output signals are 90o out of phase with each other, we
have what is known as a quadrature output encoder (the word ”quadrature” simply refers to
a 90o angular separation). A phase detection circuit may be made from a Dtype flipflop, to
distinguish a clockwise pulse sequence from a counterclockwise pulse sequence:
Rotary shaft encoder
LED
Light sensor
(phototransistor)
Counter
Q0 Q1 Q2 Q3
D
C
Q
Q
Up/Down
(quadrature output)
When the encoder rotates clockwise, the ”D” input signal squarewave will lead the ”C”
input squarewave, meaning that the ”D” input will already be ”high” when the ”C” transitions
from ”low” to ”high,” thus setting the Dtype flipflop (making the Q output ”high”) with every
clock pulse. A ”high” Q output places the counter into the ”Up” count mode, and any clock
pulses received by the clock from the encoder (from either LED) will increment it. Conversely,
when the encoder reverses rotation, the ”D” input will lag behind the ”C” input waveform,
meaning that it will be ”low” when the ”C” waveform transitions from ”low” to ”high,” forcing
the Dtype flipflop into the reset state (making the Q output ”low”) with every clock pulse.
This ”low” signal commands the counter circuit to decrement with every clock pulse from the
encoder.
This circuit, or something very much like it, is at the heart of every positionmeasuring
circuit based on a pulse encoder sensor. Such applications are very common in robotics, CNC
machine tool control, and other applications involving the measurement of reversible, mechan
ical motion.
11.4 Counter modulus
INCOMPLETE
11.5 Finite State Machines
Up to now, every circuit that was presented was a combinatorial circuit. That means that its
output was dependent only by its current inputs. Previous inputs for that type of circuits have
no effect on the output.
However, there are many applications where there is a need for our circuits to have ”mem
ory”; to remember previous inputs and calculate their outputs according to them. A circuit
11.5. FINITE STATE MACHINES 339
whose output depends not only on the present input but also on the history of the input is
called a sequential circuit.
In this section we will learn how to design and build such sequential circuits. In order to
see how this procedure works, we will use an example, on which we will study our topic.
So let’s suppose we have a digital quiz game that works on a clock and reads an input from
a manual button. However, we want the switch to transmit only one HIGH pulse to the circuit.
If we hook the button directly on the game circuit it will transmit HIGH for as few clock cycles
as our finger can achieve. On a common clock frequency our finger can never be fast enough.
The desing procedure has specific steps that must be followed in order to get the work done:
Step 1
The first step of the design procedure is to define with simple but clear words what we want
our circuit to do:
”Our mission is to design a secondary circuit that will transmit a HIGH pulse with duration
of only one cycle when the manual button is pressed, and won’t transmit another pulse until the
button is depressed and pressed again.”
Step 2
The next step is to design a State Diagram. This is a diagram that is made from circles and
arrows and describes visually the operation of our circuit. In mathematic terms, this diagram
that describes the operation of our sequential circuit is a Finite State Machine.
The State Diagram of our circuit is the following: (Figure 11.1)
Every circle represents a ”state”, a welldefined condition that our machine can be found at.
In the upper half of the circle we describe that condition. The description helps us remember
what our circuit is supposed to do at that condition.
• The first circle is the ”standby” condition. This is where our circuit starts from and where
it waits for another button press
• The second circle is the condition where the button is just pressed and our circuit needs
to transmit a HIGH pulse.
• The third circle is the condition where our circuit waits for the button to be depressed
before it returns to the ”standby” condition.
In the lower part of the circle is the output of our circuit. If we want our circuit to transmit
a HIGH on a specific state, we put a 1 on that state. Ohterwise we put a 0.
Every arrow represents a ”transition” from one state to another. A transition happens once
every clock cycle. Depending on the current Input, we may go to a different state each time.
Notice the number in the middle of every arrow. This is the current Input.
For example, when we are in the ”InitialStan by” state and we ”read” a 1, the diagram
tells us that we have to go to the ”Activate Pulse” state. If we read a 0 we must stay on the
”InitialStand by” state.
So, what does our ”Machine” do exactly? It starts from the ”Initial  Stand by” state and
waits until a 1 is read at the Input. Then it goes to the ”Activate Pulse” state and transmits a
HIGH pulse on its output. If the button keeps being pressed, the circuit goes to the third state,
340 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
Figure 11.1: A State Diagram
11.5. FINITE STATE MACHINES 341
the ”Wait Loop”. There it waits until the button is depressed (Input goes 0) while transmitting
a LOW on the output. Then it’s all over again!
This is possibly the most difficult part of the design procedure, because it cannot be de
scribed by simple steps. It takes exprerience and a bit of sharp thinking in order to set up a
State Diagram, but the rest is just a set of predetermined steps.
Step 3
Next, we replace the words that describe the different states of the diagram with binary
numbers. We start the enumeration from 0 which is assigned always on the initial state. We
then continue the enumeration with any state we like, until all states have their number.
The result looks something like this: (Figure 11.2)
Figure 11.2: A State Diagram with Coded States
Step 4
Afterwards, we fill the State Table. This table has a very specific form. I will give the table
of our example and use it to explain how will fill it in. (Figure 11.3)
The first columns are as many as the bits of the highest number we assigned the State
Diagram. If we had 5 states, we would have used up to the number 100, which means we
would use 3 columns. For our example, we used up to the number 10, so only 2 columns will be
needed. These columns describe the Current State of our circuit.
To the right of the Current State columns we write the Input Columns. These will be as
many as our Input variables. Our example has only one Input.
342 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
Figure 11.3: A State Table
Next, we write the Next State Columns. These are as many as the Current State columns.
Finally, we write the Outputs Columns. These are as many as our outputs. Our example
has only one output. In this example, the output is dependant only on the current states. This
is the reason the outputs column has two 1: to result in an output Boolean function that is
independant of input I. Keep on reading for further details.
The Current State and Input columns are the Inputs of our table. We fill them in with all
the binary numbers from 0 to.
It is simpler than it sounds fortunately. Usually there will be more rows than the actual
States we have created in the State Diagram, but that’s ok.
Each row of the Next State columns is filled as follows: We fill it in with the state that we
reach when, in the State Diagram, from the Current State of the same row we follow the Input
of the same row. If we get to a row whose Current State number doesn’t correspond to any
actual State in the State Diagram we fill it with Don’t Care terms (X). After all, we don’t care
where we can go from a State that doesn’t exist. We wouldn’t be there in the first place! Again
it is simpler than it sounds.
The outputs column is filled by the output of the corresponding Current State in the State
Diagram.
The State Table is complete! It describes the behaviour of our circuit as fully as the State
Diagram does.
Step 5a
The next step is to take that theoretical ”Machine” and implement it in a circuit. Most
often than not, this implementation involves Flip Flops. This guide is dedicated to this kind
of implementation and will describe the procedure for both D  Flip Flops as well as JK  Flip
Flops. T  Flip Flops will not be included as they are too similar to the two previous cases.
11.5. FINITE STATE MACHINES 343
The selection of the Flip Flop to use is arbitrary and usually is determined by cost fac
tors. The best choice is to perform both analysis and decide which type of Flip Flop results in
minimum number of logic gates and lesser cost.
First we will examine how we implement our ”Machine” with DFlip Flops.
We will need as many D  Flip Flops as the State columns, 2 in our example. For every Flip
Flop we will add one one more column in our State table (Figure 11.4) with the name of the
Flip Flop’s input, ”D” for this case. The column that corresponds to each Flip Flop describes
what input we must give the Flip Flop in order to go from the Current State to the
Next State. For the D  Flip Flop this is easy: The necessary input is equal to the Next State.
In the rows that contain X’s we fill X’s in this column as well.
Figure 11.4: A State Table with D  Flip Flop Excitations
Step 5b
We can do the same steps with JK  Flip Flops. There are some differences however. A JK 
Flip Flop has two inputs, therefore we need to add two columns for each Flip Flop. The content
of each cell is dictated by the JK’s excitation table: (Figure 11.5)
JK  Flip Flop Excitation Table
This table says that if we want to go from State Q to State Qnext, we need to use the specific
input for each terminal. For example, to go from 0 to 1, we need to feed J with 1 and we don’t
care wich input we feed to terminal K.
Step 6
We are in the final stage of our procedure. What remains, is to determine the Boolean
functions that determine the inputs of our Flip Flops and the Output. We will extract one
Boolean funtion for each Flip Flop input we have. This can be done with a Karnaugh Map. The
input variables of this map are the Current State variables as well as the Inputs.
344 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
Figure 11.5: A State Table with JK  Flip Flop Excitations
Figure 11.6: Karnaugh Maps for the D  Flip Flop Inputs
11.5. FINITE STATE MACHINES 345
That said, the input functions for our D  Flip Flops are the following: (Figure 11.6)
If we chose to use JK  Flip Flops our functions would be the following: (Figure 11.7)
Figure 11.7: Karnaugh Map for the JK  Flip Flop Input
A Karnaugh Map will be used to determine the function of the Output as well: (Figure 11.8)
Figure 11.8: Karnaugh Map for the Output variable Y
Step 7
346 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
We design our circuit. We place the Flip Flops and use logic gates to form the Boolean
functions that we calculated. The gates take input from the output of the Flip Flops and the
Input of the circuit. Don’t forget to connect the clock to the Flip Flops!
The D  Flip Flop version: (Figure 11.9)
Figure 11.9: The completed D  Flip Flop Sequential Circuit
The JK  Flip Flop version: (Figure 11.10)
This is it! We have successfully designed and constructed a Sequential Circuit. At first it
might seem a daunting task, but after practice and repetition the procedure will become trivial.
Sequential Circuits can come in handy as control parts of bigger circuits and can perform any
sequential logic task that we can think of. The sky is the limit! (or the circuit board, at least)
• REVIEW:
• A Sequential Logic function has a ”memory” feature and takes into account past inputs
in order to decide on the output.
• The Finite State Machine is an abstract mathematical model of a sequential logic func
tion. It has finite inputs, outputs and number of states.
BIBLIOGRAPHY 347
Figure 11.10: The completed JK  Flip Flop Sequential Circuit
• FSMs are implemented in reallife circuits through the use of Flip Flops
• The implementation procedure needs a specific order of steps (algorithm), in order to be
carried out.
Contributors
Contributors to this chapter are listed in chronological order of their contributions, from most
recent to first. See Appendix 2 (Contributor List) for dates and contact information.
George Zogopoulos Papaliakos (November 2010): Author of Finite State Machines sec
tion.
Bibliography
[1] C. L. Liu, Elements of Discrete Mathematics, 2nd Edition
[2] M. Morris Mano, Digital Design, 3rd Edition
[3] “Sequential logic” at http://en.wikipedia.org/wiki/Sequential%5Fcircuit
[4] “JK flipflop”, Flipflop (electronics) at http://en.wikipedia.org/wiki/
JK%5Fflip%5Fflop%23JK%5Fflipflop
348 CHAPTER 11. SEQUENTIAL CIRCUITSCOUNTERS
Chapter 12
SHIFT REGISTERS
Contents
12.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
12.2 Serialin/serialout shift register . . . . . . . . . . . . . . . . . . . . . . . . . 352
12.2.1 Serialin/serialout devices . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
12.3 Parallelin, serialout shift register . . . . . . . . . . . . . . . . . . . . . . . . 361
12.3.1 Parallelin/serialout devices . . . . . . . . . . . . . . . . . . . . . . . . . . 364
12.3.2 Practical applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
12.4 Serialin, parallelout shift register . . . . . . . . . . . . . . . . . . . . . . . 372
12.4.1 Serialin/ parallelout devices . . . . . . . . . . . . . . . . . . . . . . . . . 373
12.4.2 Practical applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 379
12.5 Parallelin, parallelout, universal shift register . . . . . . . . . . . . . . . 381
12.5.1 Parallelin/ parallelout and universal devices . . . . . . . . . . . . . . . . 386
12.5.2 Practical applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
12.6 Ring counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
12.6.1 Johnson counters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
12.7 references . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Original author: Dennis Crunkilton
12.1 Introduction
Shift registers, like counters, are a form of sequential logic. Sequential logic, unlike combina
tional logic is not only affected by the present inputs, but also, by the prior history. In other
words, sequential logic remembers past events.
Shift registers produce a discrete delay of a digital signal or waveform. A waveform syn
chronized to a clock, a repeating square wave, is delayed by ”n” discrete clock times, where
”n” is the number of shift register stages. Thus, a four stage shift register delays ”data in”
349
350 CHAPTER 12. SHIFT REGISTERS
by four clocks to ”data out”. The stages in a shift register are delay stages, typically type ”D”
FlipFlops or type ”JK” Flipflops.
Formerly, very long (several hundred stages) shift registers served as digital memory. This
obsolete application is reminiscent of the acoustic mercury delay lines used as early computer
memory.
Serial data transmission, over a distance of meters to kilometers, uses shift registers to
convert parallel data to serial form. Serial data communications replaces many slow parallel
data wires with a single serial high speed circuit.
Serial data over shorter distances of tens of centimeters, uses shift registers to get data
into and out of microprocessors. Numerous peripherals, including analog to digital converters,
digital to analog converters, display drivers, and memory, use shift registers to reduce the
amount of wiring in circuit boards.
Some specialized counter circuits actually use shift registers to generate repeating wave
forms. Longer shift registers, with the help of feedback generate patterns so long that they
look like random noise, pseudonoise.
Basic shift registers are classified by structure according to the following types:
• Serialin/serialout
• Parallelin/serialout
• Serialin/parallelout
• Universal parallelin/parallelout
• Ring counter
clock
data in data out
stage A stage B stage C stage D
Serialin, serialout shift register with 4stages
Above we show a block diagram of a serialin/serialout shift register, which is 4stages long.
Data at the input will be delayed by four clock periods from the input to the output of the shift
register.
Data at ”data in”, above, will be present at the Stage A output after the first clock pulse.
After the second pulse stage A data is transfered to stage B output, and ”data in” is transfered
to stage A output. After the third clock, stage C is replaced by stage B; stage B is replaced by
stage A; and stage A is replaced by ”data in”. After the fourth clock, the data originally present
at ”data in” is at stage D, ”output”. The ”first in” data is ”first out” as it is shifted from ”data
in” to ”data out”.
12.1. INTRODUCTION 351
clock
data out
stage A stage B stage C stage D
Parallelin, serialout shift register with 4stages
DA DB DC DD
data in
Data is loaded into all stages at once of a parallelin/serialout shift register. The data is
then shifted out via ”data out” by clock pulses. Since a 4 stage shift register is shown above,
four clock pulses are required to shift out all of the data. In the diagram above, stage D data
will be present at the ”data out” up until the first clock pulse; stage C data will be present
at ”data out” between the first clock and the second clock pulse; stage B data will be present
between the second clock and the third clock; and stage A data will be present between the
third and the fourth clock. After the fourth clock pulse and thereafter, successive bits of ”data
in” should appear at ”data out” of the shift register after a delay of four clock pulses.
If four switches were connected to DA through DD, the status could be read into a micro
processor using only one data pin and a clock pin. Since adding more switches would require
no additional pins, this approach looks attractive for many inputs.
clock
data out
stage A stage B stage C stage D
Serialin, parallelout shift register with 4stages
QA QB QC QD
data in
Above, four data bits will be shifted in from ”data in” by four clock pulses and be available
at QA through QD for driving external circuitry such as LEDs, lamps, relay drivers, and horns.
After the first clock, the data at ”data in” appears at QA. After the second clock, The old
QA data appears at QB ; QA receives next data from ”data in”. After the third clock, QB data is
at QC . After the fourth clock, QC data is at QD. This stage contains the data first present at
”data in”. The shift register should now contain four data bits.
352 CHAPTER 12. SHIFT REGISTERS
clock
data out
stage A stage B stage C stage D
Parallelin, parallelout shift register with 4stages
QA QB QC QD
data in
DA DB DC DD
mode
A parallelin/parallelout shift register combines the function of the parallelin, serialout
shift register with the function of the serialin, parallelout shift register to yield the univer
sal shift register. The ”do anything” shifter comes at a price– the increased number of I/O
(Input/Output) pins may reduce the number of stages which can be packaged.
Data presented at DA through DD is parallel loaded into the registers. This data at QA
through QD may be shifted by the number of pulses presented at the clock input. The shifted
data is available at QA through QD. The ”mode” input, which may be more than one input,
controls parallel loading of data from DA through DD, shifting of data, and the direction of
shifting. There are shift registers which will shift data either left or right.
clock
data in
data out
stage A stage B stage C stage D
Ring Counter, shift register output fed back to input
QD
If the serial output of a shift register is connected to the serial input, data can be perpetually
shifted around the ring as long as clock pulses are present. If the output is inverted before
being fed back as shown above, we do not have to worry about loading the initial data into the
”ring counter”.
12.2 Serialin/serialout shift register
Serialin, serialout shift registers delay data by one clock time for each stage. They will store
a bit of data for each register. A serialin, serialout shift register may be one to 64 bits in
length, longer if registers or packages are cascaded.
Below is a single stage shift register receiving data which is not synchronized to the register
clock. The ”data in” at the D pin of the type D FF (FlipFlop) does not change levels when the
12.2. SERIALIN/SERIALOUT SHIFT REGISTER 353
clock changes for low to high. We may want to synchronize the data to a system wide clock in
a circuit board to improve the reliability of a digital logic circuit.
D
C
Q
Qdata in
clock
Data present at clock time is transfered from D to Q.
QA
data in
clock
t1 t2 t3 t4
The obvious point (as compared to the figure below) illustrated above is that whatever ”data
in” is present at the D pin of a type D FF is transfered from D to output Q at clock time. Since
our example shift register uses positive edge sensitive storage elements, the output Q follows
the D input when the clock transitions from low to high as shown by the up arrows on the
diagram above. There is no doubt what logic level is present at clock time because the data is
stable well before and after the clock edge. This is seldom the case in multistage shift registers.
But, this was an easy example to start with. We are only concerned with the positive, low to
high, clock edge. The falling edge can be ignored. It is very easy to see Q follow D at clock
time above. Compare this to the diagram below where the ”data in” appears to change with
the positive clock edge.
D
C
Q
Qdata in
clock
Does the clock t1 see a 0 or a 1 at data in at D? Which output is correct, QC or QW ?
QC
data in
clock
t1 t2 t3
QW
???
?
Since ”data in” appears to changes at clock time t1 above, what does the type D FF see at
clock time? The short over simplified answer is that it sees the data that was present at D
prior to the clock. That is what is transfered to Q at clock time t1. The correct waveform is QC .
At t1 Q goes to a zero if it is not already zero. The D register does not see a one until time t2,
at which time Q goes high.
354 CHAPTER 12. SHIFT REGISTERS
D
C
Q
Qdata in
clock
Data present tH before clock time at D is transfered to Q.
QA
data in
clock
t1 t2 t3 t4
delay of 1 clock
period
Since data, above, present at D is clocked to Q at clock time, and Q cannot change until the
next clock time, the D FF delays data by one clock period, provided that the data is already
synchronized to the clock. The QA waveform is the same as ”data in” with a one clock period
delay.
A more detailed look at what the input of the type D FlipFlop sees at clock time follows.
Refer to the figure below. Since ”data in” appears to changes at clock time (above), we need
further information to determine what the D FF sees. If the ”data in” is from another shift
register stage, another same type D FF, we can draw some conclusions based on data sheet
information. Manufacturers of digital logic make available information about their parts in
data sheets, formerly only available in a collection called a data book. Data books are still
available; though, the manufacturer’s web site is the modern source.
clock
tHtS
tPQ
data in D
Data must be present (tS) before the clock and after(tH) the clock. Data is
delayed from D to Q by propagation delay (tP)
The following data was extracted from the CD4006b data sheet for operation at 5VDC ,
which serves as an example to illustrate timing.
(http://focus.ti.com/docs/prod/folders/print/cd4006b.html)
• tS=100ns
• tH=60ns
• tP=200400ns typ/max
12.2. SERIALIN/SERIALOUT SHIFT REGISTER 355
tS is the setup time, the time data must be present before clock time. In this case data must
be present at D 100ns prior to the clock. Furthermore, the data must be held for hold time
tH=60ns after clock time. These two conditions must be met to reliably clock data from D to Q
of the FlipFlop.
There is no problem meeting the setup time of 60ns as the data at D has been there for
the whole previous clock period if it comes from another shift register stage. For example, at
a clock frequency of 1 Mhz, the clock period is 1000 µs, plenty of time. Data will actually be
present for 1000µs prior to the clock, which is much greater than the minimum required tS of
60ns.
The hold time tH=60ns is met because D connected to Q of another stage cannot change any
faster than the propagation delay of the previous stage tP=200ns. Hold time is met as long as
the propagation delay of the previous D FF is greater than the hold time. Data at D driven by
another stage Q will not change any faster than 200ns for the CD4006b.
To summarize, output Q follows input D at nearly clock time if FlipFlops are cascaded into
a multistage shift register.
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QA QB QC
data in data out
clock
Serialin, serialout shift register using type "D" storage elements
Three type D FlipFlops are cascaded Q to D and the clocks paralleled to form a three stage
shift register above.
J
C
K Q
Q J
C
K Q
Q J
C
K Q
Qdata in
clock
Serialin, serialout shift register using type "JK" storage elements
QA QB QC
data out
Type JK FFs cascaded Q to J, Q’ to K with clocks in parallel to yield an alternate form of
the shift register above.
A serialin/serialout shift register has a clock input, a data input, and a data output from
the last stage. In general, the other stage outputs are not available Otherwise, it would be a
serialin, parallelout shift register..
356 CHAPTER 12. SHIFT REGISTERS
The waveforms below are applicable to either one of the preceding two versions of the serial
in, serialout shift register. The three pairs of arrows show that a three stage shift register
temporarily stores 3bits of data and delays it by three clock periods from input to output.
QA
QB
QC
data in
clock
t1 t2 t3 t4
At clock time t1 a ”data in” of 0 is clocked from D to Q of all three stages. In particular, D
of stage A sees a logic 0, which is clocked to QA where it remains until time t2.
At clock time t2 a ”data in” of 1 is clocked from D to QA. At stages B and C, a 0, fed from
preceding stages is clocked to QB and QC .
At clock time t3 a ”data in” of 0 is clocked from D to QA. QA goes low and stays low for
the remaining clocks due to ”data in” being 0. QB goes high at t3 due to a 1 from the previous
stage. QC is still low after t3 due to a low from the previous stage.
QC finally goes high at clock t4 due to the high fed to D from the previous stage QB . All
earlier stages have 0s shifted into them. And, after the next clock pulse at t5, all logic 1s will
have been shifted out, replaced by 0s
12.2.1 Serialin/serialout devices
We will take a closer look at the following parts available as integrated circuits, courtesy of
Texas Instruments. For complete device data sheets follow the links.
• CD4006b 18bit serialin/ serialout shift register
(http://focus.ti.com/docs/prod/folders/print/cd4006b.html)
• CD4031b 64bit serialin/ serialout shift register
(http://focus.ti.com/docs/prod/folders/print/cd4031b.html)
• CD4517b dual 64bit serialin/ serialout shift register
(http://focus.ti.com/docs/prod/folders/print/cd4517b.html)
The following serialin/ serialout shift registers are 4000 series CMOS (Complementary
Metal Oxide Semiconductor) family parts. As such, They will accept a VDD, positive power
supply of 3Volts to 15Volts. The VSS pin is grounded. The maximum frequency of the shift
clock, which varies with VDD, is a few megahertz. See the full data sheet for details.
12.2. SERIALIN/SERIALOUT SHIFT REGISTER 357
clock
4 12
CL
1 13D1 +4D1
D2 D2 +5
CL
6 9
5 10D3 +4D3
D4 D4 +5
3
11
D2 +4 D4 +4
8
2
latch
CL
CL and CL to all 18stages & latch.
CL
VSS ( pin 7) = Gnd, VDD (pin 14) = +3 to +18 VDC
CD4006b Serialin/ serialout shift register
CL
CL
CL
CL
CL
CL
The 18bit CD4006b consists of two stages of 4bits and two more stages of 5bits with a an
output tap at 4bits. Thus, the 5bit stages could be used as 4bit shift registers. To get a full
18bit shift register the output of one shift register must be cascaded to the input of another
and so on until all stages create a single shift register as shown below.
14 13 12 11 10 9 8
7654321
VSS
VDD
D1 D2 D3 D4
Clock
D1+4
D1+4 delayed
D4+4
D4+5
D3+4
D2+4
D2+5
cl
CD4006b 18bit serialin/ serialout shift register
D
Q
+5V
Clock
Data
Out
358 CHAPTER 12. SHIFT REGISTERS
A CD4031 64bit serialin/ serialout shift register is shown below. A number of pins are not
connected (nc). Both Q and Q’ are available from the 64th stage, actually Q64 and Q’64. There
is also a Q64 ”delayed” from a half stage which is delayed by half a clock cycle. A major feature
is a data selector which is at the data input to the shift register.
1 2 3 4 5 6 7 8
910111213141516
VDD
VSSData 2
clock Q64 Q64
Q64 delayed
mode controlData 1
D Q
Q
DQ
64stages
nc nc nc nc
nc nc
CLD
delayed clock out
C
CD4031 64bit serialin/ serialout shift register
The ”mode control” selects between two inputs: data 1 and data 2. If ”mode control” is high,
data will be selected from ”data 2” for input to the shift register. In the case of ”mode control”
being logic low, the ”data 1” is selected. Examples of this are shown in the two figures below.
12.2. SERIALIN/SERIALOUT SHIFT REGISTER 359
1 2 3 4 5 6 7 8
910111213141516
VSSData 2
clock Q64 Q64
mode control
= logic highData 1
D Q
Q
DQ
64stages
nc nc nc nc
nc nc
CLD
delayed clock out
C
CD4031 64bit serialin/ serialout shift register recirculating data.
VDD
VDD
mode control
clock
t1 t2 t3 t4
clock
Q64
64 clocks 64 clocks
The ”data 2” above is wired to the Q64 output of the shift register. With ”mode control” high,
the Q64 output is routed back to the shifter data input D. Data will recirculate from output to
input. The data will repeat every 64 clock pulses as shown above. The question that arises is
how did this data pattern get into the shift register in the first place?
1 2 3 4 5 6 7 8
910111213141516
VSS
Data
2 clock Q64 Q64
mode control
= logic low
data 1
D Q
Q
DQ
64stages
nc nc nc nc
nc nc
CLD
delayed clock out
C
CD4031 64bit serialin/ serialout shift register load new data at Data 1.
VDD
VDD
mode control
Q64
data 1
clock
t1 t2 t64 t65
360 CHAPTER 12. SHIFT REGISTERS
With ”mode control” low, the CD4031 ”data 1” is selected for input to the shifter. The output,
Q64, is not recirculated because the lower data selector gate is disabled. By disabled we mean
that the logic low ”mode select” inverted twice to a low at the lower NAND gate prevents it for
passing any signal on the lower pin (data 2) to the gate output. Thus, it is disabled.
1 2 3 4 5 6 7 8
910111213141516
VDD
VSSQ32A DA
D
DB
C
CD4517b dual 64bit serialin/ serialout shift register
Q48A WEA CLA Q64A
Q16B Q48B WEB CL8 Q64B Q32B
Q16A
D
C
16 32 48 64
16 32 48 64
WE
WE
3 to 18 VDC
GND
A CD4517b dual 64bit shift register is shown above. Note the taps at the 16th, 32nd, and
48th stages. That means that shift registers of those lengths can be configured from one of the
64bit shifters. Of course, the 64bit shifters may be cascaded to yield an 80bit, 96bit, 112bit,
or 128bit shift register. The clock CLA and CLB need to be paralleled when cascading the two
shifters. WEB and WEB are grounded for normal shifting operations. The data inputs to the
shift registers A and B are DA and DB respectively.
Suppose that we require a 16bit shift register. Can this be configured with the CD4517b?
How about a 64shift register from the same part?
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 361
1 2 3 4 5 6 7 8
910111213141516
VDD
VSSQ32A DA
D
DB
C
CD4517b dual 64bit serialin/ serialout shift register, wired for
16shift register, 64bit shift register
Q48A WEA CLA
Q16B Q48B
WEB CL8 Q64B Q32B
Q16A
D
C
16 32 48 64
16 32 48 64
WE
WE
5V
Q16B out
Q64A
Q64A out
clock
data in
clock
data in
Above we show A CD4517b wired as a 16bit shift register for section B. The clock for section
B is CLB . The data is clocked in at CLB . And the data delayed by 16clocks is picked of off
Q16B . WEB , the write enable, is grounded.
Above we also show the same CD4517b wired as a 64bit shift register for the independent
section A. The clock for section A is CLA. The data enters at CLA. The data delayed by 64clock
pulses is picked up from Q64A. WEA, the write enable for section A, is grounded.
12.3 Parallelin, serialout shift register
Parallelin/ serialout shift registers do everything that the previous serialin/ serialout shift
registers do plus input data to all stages simultaneously. The parallelin/ serialout shift reg
ister stores data, shifts it on a clock by clock basis, and delays it by the number of stages
times the clock period. In addition, parallelin/ serialout really means that we can load data
in parallel into all stages before any shifting ever begins. This is a way to convert data from a
parallel format to a serial format. By parallel format we mean that the data bits are present
simultaneously on individual wires, one for each data bit as shown below. By serial format we
mean that the data bits are presented sequentially in time on a single wire or circuit as in the
case of the ”data out” on the block diagram below.
362 CHAPTER 12. SHIFT REGISTERS
clock
data out
stage A stage B stage C stage D
Parallelin, serialout shift register with 4stages
DA DB DC DD
data in
Below we take a close look at the internal details of a 3stage parallelin/ serialout shift
register. A stage consists of a type D FlipFlop for storage, and an ANDOR selector to de
termine whether data will load in parallel, or shift stored data to the right. In general, these
elements will be replicated for the number of stages required. We show three stages due to
space limitations. Four, eight or sixteen bits is normal for real parts.
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QC
SI
CLK
SHIFT/LD = 0
SO
Parallelin/ serialout shift register showing parallel load path
QA QB
DB DCDA
Above we show the parallel load path when SHIFT/LD’ is logic low. The upper NAND gates
serving DA DB DC are enabled, passing data to the D inputs of type D FlipFlops QA QB DC
respectively. At the next positive going clock edge, the data will be clocked from D to Q of the
three FFs. Three bits of data will load into QA QB DC at the same time.
The type of parallel load just described, where the data loads on a clock pulse is known
as synchronous load because the loading of data is synchronized to the clock. This needs to
be differentiated from asynchronous load where loading is controlled by the preset and clear
pins of the FlipFlops which does not require the clock. Only one of these load methods is used
within an individual device, the synchronous load being more common in newer devices.
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QA QB QC
DB DCDA
SI
CLK
SHIFT/LD = 1
SO
Parallelin/ serialout shift register showing shift path
DAA
DBB DCC
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 363
The shift path is shown above when SHIFT/LD’ is logic high. The lower AND gates of the
pairs feeding the OR gate are enabled giving us a shift register connection of SI to DA , QA
to DB , QB to DC , QC to SO. Clock pulses will cause data to be right shifted out to SO on
successive pulses.
The waveforms below show both parallel loading of three bits of data and serial shifting of
this data. Parallel data at DA DB DC is converted to serial data at SO.
DA
DB
DC
data in
clock
t1 t2 t3 t4
SHIFT/LD
Parallelin/ serialout shift register load/shift waveforms
QA
QB
QC (SO)
1
0
1
1
0
1
What we previously described with words for parallel loading and shifting is now set down
as waveforms above. As an example we present 101 to the parallel inputs DAA DBB DCC .
Next, the SHIFT/LD’ goes low enabling loading of data as opposed to shifting of data. It needs
to be low a short time before and after the clock pulse due to setup and hold requirements. It is
considerably wider than it has to be. Though, with synchronous logic it is convenient to make
it wide. We could have made the active low SHIFT/LD’ almost two clocks wide, low almost a
clock before t1 and back high just before t3. The important factor is that it needs to be low
around clock time t1 to enable parallel loading of the data by the clock.
Note that at t1 the data 101 at DA DB DC is clocked from D to Q of the FlipFlops as shown
at QA QB QC at time t1. This is the parallel loading of the data synchronous with the clock.
364 CHAPTER 12. SHIFT REGISTERS
DA
DB
DC
data in
clock
t1 t2 t3 t4
SHIFT/LD
Parallelin/ serialout shift register load/shift waveforms
QA
QB
1
0
1
1
0
1 0 1QC (SO)
Now that the data is loaded, we may shift it provided that SHIFT/LD’ is high to enable
shifting, which it is prior to t2. At t2 the data 0 at QC is shifted out of SO which is the same as
the QC waveform. It is either shifted into another integrated circuit, or lost if there is nothing
connected to SO. The data at QB , a 0 is shifted to QC . The 1 at QA is shifted into QB . With
”data in” a 0, QA becomes 0. After t2, QA QB QC = 010.
After t3, QA QB QC = 001. This 1, which was originally present at QA after t1, is now
present at SO and QC . The last data bit is shifted out to an external integrated circuit if it
exists. After t4 all data from the parallel load is gone. At clock t5 we show the shifting in of a
data 1 present on the SI, serial input.
Why provide SI and SO pins on a shift register? These connections allow us to cascade
shift register stages to provide large shifters than available in a single IC (Integrated Circuit)
package. They also allow serial connections to and from other ICs like microprocessors.
12.3.1 Parallelin/serialout devices
Let’s take a closer look at parallelin/ serialout shift registers available as integrated circuits,
courtesy of Texas Instruments. For complete device data sheets follow these the links.
• SN74ALS166 parallelin/ serialout 8bit shift register, synchronous load
(http://wwws.ti.com/sc/ds/sn74als166.pdf)
• SN74ALS165 parallelin/ serialout 8bit shift register, asynchronous load
(http://wwws.ti.com/sc/ds/sn74als165.pdf)
• CD4014B parallelin/ serialout 8bit shift register, synchronous load
(http://wwws.ti.com/sc/ds/cd4014b.pdf)
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 365
• SN74LS647 parallelin/ serialout 16bit shift register, synchronous load
(http://wwws.ti.com/sc/ds/sn74ls674.pdf)
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QH
SER
CLK
SH/LD
SN74ALS166 Parallelin/ serialout 8bit shift register
QA QB
B HA
CLR
CLKINH
The SN74ALS166 shown above is the closest match of an actual part to the previous parallel
in/ serial out shifter figures. Let us note the minor changes to our figure above. First of all,
there are 8stages. We only show three. All 8stages are shown on the data sheet available
at the link above. The manufacturer labels the data inputs A, B, C, and so on to H. The
SHIFT/LOAD control is called SH/LD’. It is abbreviated from our previous terminology, but
works the same: parallel load if low, shift if high. The shift input (serial data in) is SER on the
ALS166 instead of SI. The clock CLK is controlled by an inhibit signal, CLKINH. If CLKINH
is high, the clock is inhibited, or disabled. Otherwise, this ”real part” is the same as what we
have looked at in detail.
SRG8R
M1 [Shift]
M2 [Load]
≥1 C3/1
1, 3D
2, 3D
2, 3D
13 QH
9
15
6
7
2
3
4
5
10
11
12
14
1SER
A
B
C
D
E
F
G
H
CLR
SH/LD
CLKINH
CLK
SN74ALS166 ANSI Symbol
Above is the ANSI (American National Standards Institute) symbol for the SN74ALS166
as provided on the data sheet. Once we know how the part operates, it is convenient to hide
the details within a symbol. There are many general forms of symbols. The advantage of the
ANSI symbol is that the labels provide hints about how the part operates.
366 CHAPTER 12. SHIFT REGISTERS
The large notched block at the top of the ’74ASL166 is the control section of the ANSI
symbol. There is a reset indicted by R. There are three control signals: M1 (Shift), M2 (Load),
and C3/1 (arrow) (inhibited clock). The clock has two functions. First, C3 for shifting parallel
data wherever a prefix of 3 appears. Second, whenever M1 is asserted, as indicated by the 1
of C3/1 (arrow), the data is shifted as indicated by the right pointing arrow. The slash (/) is
a separator between these two functions. The 8shift stages, as indicated by title SRG8, are
identified by the external inputs A, B, C, to H. The internal 2, 3D indicates that data, D, is
controlled by M2 [Load] and C3 clock. In this case, we can conclude that the parallel data is
loaded synchronously with the clock C3. The upper stage at A is a wider block than the others
to accommodate the input SER. The legend 1, 3D implies that SER is controlled by M1 [Shift]
and C3 clock. Thus, we expect to clock in data at SER when shifting as opposed to parallel
loading.
≥1 ≥1 &
ANSI gate symbols
&
The ANSI/IEEE basic gate rectangular symbols are provided above for comparison to the
more familiar shape symbols so that we may decipher the meaning of the symbology associated
with the CLKINH and CLK pins on the previous ANSI SN74ALS166 symbol. The CLK and
CLKINH feed an OR gate on the SN74ALS166 ANSI symbol. OR is indicated by => on the
rectangular inset symbol. The long triangle at the output indicates a clock. If there was a
bubble with the arrow this would have indicated shift on negative clock edge (high to low).
Since there is no bubble with the clock arrow, the register shifts on the positive (low to high
transition) clock edge. The long arrow, after the legend C3/1 pointing right indicates shift
right, which is down the symbol.
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 367
D
C
Q
Q D
C
Q
QSER
CLK
SH/LD
SN74ALS165 Parallelin/ serialout 8bit shift register,
asynchronous load
QA QB
CLKINH
R
S
R
S D
C
Q
Q
QH
R
S
HBA
Part of the internal logic of the SN74ALS165 parallelin/ serialout, asynchronous load shift
register is reproduced from the data sheet above. See the link at the beginning of this section
the for the full diagram. We have not looked at asynchronous loading of data up to this point.
First of all, the loading is accomplished by application of appropriate signals to the Set (preset)
and Reset (clear) inputs of the FlipFlops. The upper NAND gates feed the Set pins of the
FFs and also cascades into the lower NAND gate feeding the Reset pins of the FFs. The lower
NAND gate inverts the signal in going from the Set pin to the Reset pin.
First, SH/LD’ must be pulled Low to enable the upper and lower NAND gates. If SH/LD’
were at a logic high instead, the inverter feeding a logic low to all NAND gates would force
a High out, releasing the ”active low” Set and Reset pins of all FFs. There would be no
possibility of loading the FFs.
With SH/LD’ held Low, we can feed, for example, a data 1 to parallel inputA, which inverts
to a zero at the upper NAND gate output, setting FF QA to a 1. The 0 at the Set pin is fed
to the lower NAND gate where it is inverted to a 1 , releasing the Reset pin of QA. Thus, a
data A=1 sets QA=1. Since none of this required the clock, the loading is asynchronous with
respect to the clock. We use an asynchronous loading shift register if we cannot wait for a clock
to parallel load data, or if it is inconvenient to generate a single clock pulse.
The only difference in feeding a data 0 to parallel input A is that it inverts to a 1 out of the
upper gate releasing Set. This 1 at Set is inverted to a 0 at the lower gate, pulling Reset to a
Low, which resets QA=0.
368 CHAPTER 12. SHIFT REGISTERS
SRG8
C1 [LOAD]
≥1 C2/
2D
1D
1D
6 QH
1
15
2
11
12
13
14
3
4
5
6
10SER
A
B
C
D
E
F
G
H
SH/LD
CLKINH
CLK
SN74ALS165 ANSI Symbol
1D
7 QH
The ANSI symbol for the SN74ALS166 above has two internal controls C1 [LOAD] and C2
clock from theOR function of (CLKINH, CLK). SRG8 says 8stage shifter. The arrow afterC2
indicates shifting right or down. SER input is a function of the clock as indicated by internal
label 2D. The parallel data inputs A, B, C to H are a function of C1 [LOAD], indicated by
internal label 1D. C1 is asserted when sh/LD’ =0 due to the halfarrow inverter at the input.
Compare this to the control of the parallel data inputs by the clock of the previous synchronous
ANSI SN75ALS166. Note the differences in the ANSI Data labels.
SRG8
M1 [Shift]
M2 [Load]
C3/1
1, 3D
2, 3D
2, 3D
3 Q8
9
10
7
6
5
4
13
14
15
1
11SER
P1
P2
P3
P4
P5
P6
P7
P8
CLK
12
2
Q7
Q6
LD/SH
SRG8
C1 [Load]
≥1 C2/
2D
1D
3 Q8
9
10
7
6
5
4
13
14
15
1
11SER
P1
P2
P3
P4
P5
P6
P7
P8
CLK
12
2
Q7
Q6
LD/SH
CD4021B, asynchronous load
CMOS Parallelin/ serialout shift registers, 8bit ANSI symbols
1D
1D2, 3D
CD4014B, synchronous load
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 369
On the CD4014B above, M1 is asserted when LD/SH’=0. M2 is asserted when LD/SH’=1.
Clock C3/1 is used for parallel loading data at 2, 3D when M2 is active as indicated by the 2,3
prefix labels. Pins P3 to P7 are understood to have the smae internal 2,3 prefix labels as P2
and P8. At SER, the 1,3D prefix implies that M1 and clock C3 are necessary to input serial
data. Right shifting takes place when M1 active is as indicated by the 1 in C3/1 arrow.
The CD4021B is a similar part except for asynchronous parallel loading of data as implied
by the lack of any 2 prefix in the data label 1D for pins P1, P2, to P8. Of course, prefix 2 in
label 2D at input SER says that data is clocked into this pin. The OR gate inset shows that
the clock is controlled by LD/SH’.
SRG 16
G1/2 EN
& C4/3
3,4D
6 SER/Q15
1
8
9
14
15
P1
P2
P3
P4
P5
P6
P7
P8
CLK
CS
SN74LS674, parallelin/serialout, synchronous load
MODE
R/W
74LS674
3
5
2
1,2 M3
G2
3,4D
3,4D
7
13
11
10
16
3,4D
P9
P10 18
20
22
17
19
21
23
P11
P12
P13
P14
P15
P0
The above SN74LS674 internal label SRG 16 indicates 16bit shift register. The MODE
input to the control section at the top of the symbol is labeled 1,2 M3. Internal M3 is a function
of input MODE and G1 and G2 as indicated by the 1,2 preceding M3. The base label G
indicates an AND function of any such G inputs. Input R/W’ is internally labeled G1/2 EN.
This is an enable EN (controlled by G1 AND G2) for tristate devices used elsewhere in the
symbol. We note that CS’ on (pin 1) is internal G2. Chip select CS’ also is ANDed with the
input CLK to give internal clock C4. The bubble within the clock arrow indicates that activity
is on the negative (high to low transition) clock edge. The slash (/) is a separator implying two
370 CHAPTER 12. SHIFT REGISTERS
functions for the clock. Before the slash, C4 indicates control of anything with a prefix of 4.
After the slash, the 3’ (arrow) indicates shifting. The 3’ of C4/3’ implies shifting when M3 is
deasserted (MODE=0). The long arrow indicates shift right (down).
Moving down below the control section to the data section, we have external inputs P0
P15, pins (711, 1323). The prefix 3,4 of internal label 3,4D indicates that M3 and the clock
C4 control loading of parallel data. The D stands for Data. This label is assumed to apply to all
the parallel inputs, though not explicitly written out. Locate the label 3’,4D on the right of the
P0 (pin7) stage. The complemented3 indicates that M3=MODE=0 inputs (shifts) SER/Q15
(pin5) at clock time, (4 of 3’,4D) corresponding to clock C4. In other words, with MODE=0, we
shift data into Q0 from the serial input (pin 6). All other stages shift right (down) at clock time.
Moving to the bottom of the symbol, the triangle pointing right indicates a buffer between
Q and the output pin. The Triangle pointing down indicates a tristate device. We previously
stated that the tristate is controlled by enable EN, which is actually G1 AND G2 from the
control section. If R/W=0, the tristate is disabled, and we can shift data into Q0 via SER (pin
6), a detail we omitted above. We actually need MODE=0, R/W’=0, CS’=0
The internal logic of the SN74LS674 and a table summarizing the operation of the control
signals is available in the link in the bullet list, top of section.
If R/W’=1, the tristate is enabled, Q15 shifts out SER/Q15 (pin 6) and recirculates to the Q0
stage via the right hand wire to 3’,4D. We have assumed that CS’ was low giving us clock C4/3’
and G2 to ENable the tristate.
12.3.2 Practical applications
An application of a parallelin/ serialout shift register is to read data into a microprocessor.
1 2 3
4 5 6
87 9
Keypad Alarm
Gnd
Clock
Serial data
+5V
Alarm with remote keypad
The Alarm above is controlled by a remote keypad. The alarm box supplies +5V and ground
to the remote keypad to power it. The alarm reads the remote keypad every few tens of mil
liseconds by sending shift clocks to the keypad which returns serial data showing the status of
the keys via a parallelin/ serialout shift register. Thus, we read nine key switches with four
wires. How many wires would be required if we had to run a circuit for each of the nine keys?
12.3. PARALLELIN, SERIALOUT SHIFT REGISTER 371
SRG8
C1 [Load]
≥1 C2/
2D
1D
3 Q8
9
10
7
6
5
4
13
14
15
1
11
SER
P1
P2
P3
P4
P5
P6
P7
P8
CLK
12
2
Q7
Q6
LD/SH
CD4021B
1D
1D
+5V
+5V
microprocessor
+5V
Shift clock
Load/shift
Serial data in
Reading switches into microprocessor
Other serial or gnd
Serial data in
Load/shift
Shift clock
P8
P7P6
P4
P3
P2
P1
P5
A practical application of a parallelin/ serialout shift register is to read many switch clo
sures into a microprocessor on just a few pins. Some low end microprocessors only have 6I/O
(Input/Output) pins available on an 8pin package. Or, we may have used most of the pins
on an 84pin package. We may want to reduce the number of wires running around a circuit
board, machine, vehicle, or building. This will increase the reliability of our system. It has
been reported that manufacturers who have reduced the number of wires in an automobile
produce a more reliable product. In any event, only three microprocessor pins are required to
read in 8bits of data from the switches in the figure above.
We have chosen an asynchronous loading device, the CD4021B because it is easier to control
the loading of data without having to generate a single parallel load clock. The parallel data
inputs of the shift register are pulled up to +5V with a resistor on each input. If all switches are
open, all 1s will be loaded into the shift register when the microprocessor moves the LD/SH’
line from low to high, then back low in anticipation of shifting. Any switch closures will apply
logic 0s to the corresponding parallel inputs. The data pattern at P1P7 will be parallel loaded
by the LD/SH’=1 generated by the microprocessor software.
The microprocessor generates shift pulses and reads a data bit for each of the 8bits. This
process may be performed totally with software, or larger microprocessors may have one or
more serial interfaces to do the task more quickly with hardware. With LD/SH’=0, the micro
processor generates a 0 to 1 transition on the Shift clock line, then reads a data bit on the
Serial data in line. This is repeated for all 8bits.
The SER line of the shift register may be driven by another identical CD4021B circuit if
more switch contacts need to be read. In which case, the microprocessor generates 16shift
pulses. More likely, it will be driven by something else compatible with this serial data format,
for example, an analog to digital converter, a temperature sensor, a keyboard scanner, a serial
readonly memory. As for the switch closures, they may be limit switches on the carriage of a
372 CHAPTER 12. SHIFT REGISTERS
machine, an overtemperature sensor, a magnetic reed switch, a door or window switch, an air
or water pressure switch, or a solid state optical interrupter.
12.4 Serialin, parallelout shift register
A serialin/parallelout shift register is similar to the serialin/ serialout shift register in that
it shifts data into internal storage elements and shifts data out at the serialout, dataout, pin.
It is different in that it makes all the internal stages available as outputs. Therefore, a serial
in/parallelout shift register converts data from serial format to parallel format. If four data
bits are shifted in by four clock pulses via a single wire at datain, below, the data becomes
available simultaneously on the four Outputs QA to QD after the fourth clock pulse.
clock
data out
stage A stage B stage C stage D
Serialin, parallelout shift register with 4stages
QA QB QC QD
data in
The practical application of the serialin/parallelout shift register is to convert data from
serial format on a single wire to parallel format on multiple wires. Perhaps, we will illuminate
four LEDs (Light Emitting Diodes) with the four outputs (QA QB QC QD ).
R
1D
R
QA
R
QB QD
SI
CLK
CLR
R
QC
SO
Serialin/ Parallel out shift register details
1D 1D 1D
The above details of the serialin/parallelout shift register are fairly simple. It looks like
a serialin/ serialout shift register with taps added to each stage output. Serial data shifts in
at SI (Serial Input). After a number of clocks equal to the number of stages, the first data bit
12.4. SERIALIN, PARALLELOUT SHIFT REGISTER 373
in appears at SO (QD) in the above figure. In general, there is no SO pin. The last stage (QD
above) serves as SO and is cascaded to the next package if it exists.
If a serialin/parallelout shift register is so similar to a serialin/ serialout shift register,
why do manufacturers bother to offer both types? Why not just offer the serialin/parallelout
shift register? They actually only offer the serialin/parallelout shift register, as long as it has
no more than 8bits. Note that serialin/ serialout shift registers come in bigger than 8bit
lengths of 18 to to 64bits. It is not practical to offer a 64bit serialin/parallelout shift register
requiring that many output pins. See waveforms below for above shift register.
SI
CLK
t1 t2 t3 t4
Serialin/ parallelout shift register waveforms
QA
QB
QC
1 0 1
1 0 1
1 0
1
CLR
1
QD 1
0
1
1
t5 t6 t7 t8
The shift register has been cleared prior to any data by CLR’, an active low signal, which
clears all type D FlipFlops within the shift register. Note the serial data 1011 pattern pre
sented at the SI input. This data is synchronized with the clock CLK. This would be the case
if it is being shifted in from something like another shift register, for example, a parallelin/
serialout shift register (not shown here). On the first clock at t1, the data 1 at SI is shifted
from D to Q of the first shift register stage. After t2 this first data bit is at QB . After t3 it is at
QC . After t4 it is at QD. Four clock pulses have shifted the first data bit all the way to the last
stage QD. The second data bit a 0 is at QC after the 4th clock. The third data bit a 1 is at QB .
The fourth data bit another 1 is at QA. Thus, the serial data input pattern 1011 is contained
in (QD QC QB QA). It is now available on the four outputs.
It will available on the four outputs from just after clock t4 to just before t5. This parallel
data must be used or stored between these two times, or it will be lost due to shifting out the
QD stage on following clocks t5 to t8 as shown above.
12.4.1 Serialin/ parallelout devices
Let’s take a closer look at Serialin/ parallelout shift registers available as integrated circuits,
courtesy of Texas Instruments. For complete device data sheets follow the links.
374 CHAPTER 12. SHIFT REGISTERS
• SN74ALS164A serialin/ parallelout 8bit shift register
(http://wwws.ti.com/sc/ds/sn74als164a.pdf)
• SN74AHC594 serialin/ parallelout 8bit shift register with output register
(http://wwws.ti.com/sc/ds/sn74ahct594.pdf)
• SN74AHC595 serialin/ parallelout 8bit shift register with output register
(http://wwws.ti.com/sc/ds/sn74ahct595.pdf)
• CD4094 serialin/ parallelout 8bit shift register with output register
(http://wwws.ti.com/sc/ds/cd4094b.pdf)
(http://www.st.com/stonline/books/pdf/docs/2069.pdf)
R
1R
1S
R
1R
1S
QA
R
1R
1S
QB QH
A
B
CLK
CLR
Serialin/ Parallel out shift register details
The 74ALS164A is almost identical to our prior diagram with the exception of the two serial
inputs A and B. The unused input should be pulled high to enable the other input. We do not
show all the stages above. However, all the outputs are shown on the ANSI symbol below,
along with the pin numbers.
12.4. SERIALIN, PARALLELOUT SHIFT REGISTER 375
SRG8
R
C1/
9
8
11
12
13
3
4
5
6
10
A
QA
QB
QC
QD
QE
QF
QG
QH
CLR
CLK
SN74ALS164A ANSI Symbol
2
1
B & 1D
The CLK input to the control section of the above ANSI symbol has two internal functions
C1, control of anything with a prefix of 1. This would be clocking in of data at 1D. The second
function, the arrow after after the slash (/) is right (down) shifting of data within the shift
register. The eight outputs are available to the right of the eight registers below the control
section. The first stage is wider than the others to accommodate the A&B input.
R
QA QB
SER
SRCLK
SRCLR
R
74AHC594 Serialin/ Parallel out 8bit shift register with output registers
1D 1D
R
1D
R
1D
R
1D
R
1D
QH’
RCLR
RCLK
QH
14
11
10
13
12
15 1 7
QC QD QE QF QG
2 3 4 5 6
9
QB’QA’
376 CHAPTER 12. SHIFT REGISTERS
The above internal logic diagram is adapted from the TI (Texas Instruments) data sheet for
the 74AHC594. The type ”D” FFs in the top row comprise a serialin/ parallelout shift register.
This section works like the previously described devices. The outputs (QA’ QB ’ to QH ’ ) of the
shift register half of the device feed the type ”D” FFs in the lower half in parallel. QH ’ (pin 9)
is shifted out to any optional cascaded device package.
A single positive clock edge at RCLK will transfer the data from D to Q of the lower
FFs. All 8bits transfer in parallel to the output register (a collection of storage elements).
The purpose of the output register is to maintain a constant data output while new data is
being shifted into the upper shift register section. This is necessary if the outputs drive relays,
valves, motors, solenoids, horns, or buzzers. This feature may not be necessary when driving
LEDs as long as flicker during shifting is not a problem.
Note that the 74AHC594 has separate clocks for the shift register (SRCLK) and the output
register ( RCLK). Also, the shifter may be cleared by SRCLR and, the output register by
RCLR. It desirable to put the outputs in a known state at poweron, in particular, if driving
relays, motors, etc. The waveforms below illustrate shifting and latching of data.
SI
SRCLK
t1 t2 t3 t4
Waveforms for 74AHC594 serialin/ parallelout shift registe rwith latch
QA’
QB’
QC’
1 0 1
1 0 1
1 0
1
RCLR
1
QD’ 1
0
1
1
t5 t6 t7 t8
QA
QB
QC
QD
RCLK
1
0
1
1
CLR
12.4. SERIALIN, PARALLELOUT SHIFT REGISTER 377
The above waveforms show shifting of 4bits of data into the first four stages of 74AHC594,
then the parallel transfer to the output register. In actual fact, the 74AHC594 is an 8bit shift
register, and it would take 8clocks to shift in 8bits of data, which would be the normal mode of
operation. However, the 4bits we show saves space and adequately illustrates the operation.
We clear the shift register half a clock prior to t0 with SRCLR’=0. SRCLR’ must be
released back high prior to shifting. Just prior to t0 the output register is cleared by RCLR’=0.
It, too, is released ( RCLR’=1).
Serial data 1011 is presented at the SI pin between clocks t0 and t4. It is shifted in by clocks
t1 t2 t3 t4 appearing at internal shift stages QA’ QB ’ QC ’ QD’ . This data is present at these
stages between t4 and t5. After t5 the desired data (1011) will be unavailable on these internal
shifter stages.
Between t4 and t5 we apply a positive going RCLK transferring data 1011 to register out
puts QA QB QC QD . This data will be frozen here as more data (0s) shifts in during the
succeeding SRCLKs (t5 to t8). There will not be a change in data here until another RCLK is
applied.
R
QA QB
SER
SRCLK
SRCLR
R
74AHC595 Serialin/ Parallel out 8bit shift register with output registers
1D 1D1D
1D 1D 1D
QH’
RCLK
QH
14
11
10
12
15 1 7
QC QD QE QF QG
2 3 4 5 6
9
OE13
QQ Q
Q Q Q
The 74AHC595 is identical to the ’594 except that the RCLR’ is replaced by an OE’ en
abling a tristate buffer at the output of each of the eight output register bits. Though the
output register cannot be cleared, the outputs may be disconnected by OE’=1. This would al
low external pullup or pulldown resistors to force any relay, solenoid, or valve drivers to a
known state during a system powerup. Once the system is poweredup and, say, a micropro
cessor has shifted and latched data into the ’595, the output enable could be asserted (OE’=0)
to drive the relays, solenoids, and valves with valid data, but, not before that time.
378 CHAPTER 12. SHIFT REGISTERS
SRG8
C3/
13
11
5
6
7
15
1
2
3
4
SER QA
QB
QC
QD
QE
QF
QG
QH
SRCLK
SN74AHC594 ANSI Symbol
14 2,3D
QH’9
RCLK 12 C4
1,4D
2D
2D
R1RCLR
R2SRCLR 10
1,4D
SRG8
C3/
13
11
5
6
7
15
1
2
3
4
QA
QB
QC
QD
QE
QF
QG
QH
SN74AHC595 ANSI Symbol
14 2,3D
QH’9
12 C4
4D
2D
2D
EN
R210
4D
SER
SRCLK
RCLK
OE
SRCLR
Above are the proposed ANSI symbols for these devices. C3 clocks data into the serial input
(external SER) as indicate by the 3 prefix of 2,3D. The arrow after C3/ indicates shifting right
(down) of the shift register, the 8stages to the left of the ’595symbol below the control section.
The 2 prefix of 2,3D and 2D indicates that these stages can be reset by R2 (external SRCLR’).
The 1 prefix of 1,4D on the ’594 indicates that R1 (external RCLR’) may reset the output
register, which is to the right of the shift register section. The ’595, which has an EN at ex
ternal OE’ cannot reset the output register. But, the EN enables tristate (inverted triangle)
output buffers. The right pointing triangle of both the ’594 and’595 indicates internal buffer
ing. Both the ’594 and’595 output registers are clocked by C4 as indicated by 4 of 1,4D and 4D
respectively.
SRG8
C1/
13
11
5
6
7
15
2
3
4
SERIAL OUT
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
CLOCK
CD4094B/ 74HCT4094 ANSI Symbol
14
1D
QS’
9
STROBE
12
C2
2D
OUTPUT
10
3
1
EN3ENABLE
QS
SERIAL OUT
SERIAL
IN
The CD4094B is a 3 to 15VDC capable latching shift register alternative to the previous
74AHC594 devices. CLOCK, C1, shifts data in at SERIAL IN as implied by the 1 prefix
12.4. SERIALIN, PARALLELOUT SHIFT REGISTER 379
of 1D. It is also the clock of the right shifting shift register (left half of the symbol body) as
indicated by the /(rightarrow) of C1/(arrow) at the CLOCK input.
STROBE, C2 is the clock for the 8bit output register to the right of the symbol body. The
2 of 2D indicates that C2 is the clock for the output register. The inverted triangle in the
output latch indicates that the output is tristated, being enabled by EN3. The 3 preceding
the inverted triangle and the 3 of EN3 are often omitted, as any enable (EN) is understood to
control the tristate outputs.
QS and QS ’ are nonlatched outputs of the shift register stage. QS could be cascaded to
SERIAL IN of a succeeding device.
12.4.2 Practical applications
A realworld application of the serialin/ parallelout shift register is to output data from a
microprocessor to a remote panel indicator. Or, another remote output device which accepts
serial format data.
1 2 3
4 5 6
87 9
Keypad Alarm
Gnd
Clock
Serial data
+5V
Alarm with remote keypad and display
Remote display
The figure ”Alarm with remote key pad” is repeated here from the parallelin/ serialout
section with the addition of the remote display. Thus, we can display, for example, the status
of the alarm loops connected to the main alarm box. If the Alarm detects an open window, it
can send serial data to the remote display to let us know. Both the keypad and the display
would likely be contained within the same remote enclosure, separate from the main alarm
box. However, we will only look at the display panel in this section.
If the display were on the same board as the Alarm, we could just run eight wires to the
eight LEDs along with two wires for power and ground. These eight wires are much less
desirable on a long run to a remote panel. Using shift registers, we only need to run five
wires clock, serial data, a strobe, power, and ground. If the panel were just a few inches
away from the main board, it might still be desirable to cut down on the number of wires in
a connecting cable to improve reliability. Also, we sometimes use up most of the available
pins on a microprocessor and need to use serial techniques to expand the number of outputs.
Some integrated circuit output devices, such as Digital to Analog converters contain serialin/
parallelout shift registers to receive data from microprocessors. The techniques illustrated
here are applicable to those parts.
380 CHAPTER 12. SHIFT REGISTERS
microprocessor
+5V
Shift clock
Output to LEDs from microprocessor
SRG8
C3/
13
11
5
6
7
15
1
2
3
4
SER QA
QB
QC
QD
QE
QF
QG
QH
SRCLK
14 2,3D
QH’9
RCLK 12 C4
1,4D
2D
2D
R1RCLR
R2SRCLR 10
1,4D
74AHC594
Serial data out
+5V
Latch LED data
+5V 470Ω x8
We have chosen the 74AHC594 serialin/ parallelout shift register with output register;
though, it requires an extra pin, RCLK, to parallel load the shiftedin data to the output pins.
This extra pin prevents the outputs from changing while data is shifting in. This is not much
of a problem for LEDs. But, it would be a problem if driving relays, valves, motors, etc.
Code executed within the microprocessor would start with 8bits of data to be output. One
bit would be output on the ”Serial data out” pin, driving SER of the remote 74AHC594. Next,
the microprocessor generates a low to high transition on ”Shift clock”, driving SRCLK of the
’595 shift register. This positive clock shifts the data bit at SER from ”D” to ”Q” of the first
shift register stage. This has no effect on the QA LED at this time because of the internal 8bit
output register between the shift register and the output pins (QA to QH ). Finally, ”Shift clock”
is pulled back low by the microprocessor. This completes the shifting of one bit into the ’595.
The above procedure is repeated seven more times to complete the shifting of 8bits of data
from the microprocessor into the 74AHC594 serialin/ parallelout shift register. To transfer
the 8bits of data within the internal ’595 shift register to the output requires that the micro
processor generate a low to high transition on RCLK, the output register clock. This applies
new data to the LEDs. The RCLK needs to be pulled back low in anticipation of the next 8bit
transfer of data.
The data present at the output of the ’595 will remain until the process in the above two
paragraphs is repeated for a new 8bits of data. In particular, new data can be shifted into the
’595 internal shift register without affecting the LEDs. The LEDs will only be updated with
new data with the application of the RCLK rising edge.
What if we need to drive more than eight LEDs? Simply cascade another 74AHC594 SER
pin to theQH ’ of the existing shifter. Parallel the SRCLK andRCLK pins. The microprocessor
would need to transfer 16bits of data with 16clocks before generating an RCLK feeding both
devices.
The discrete LED indicators, which we show, could be 7segment LEDs. Though, there are
LSI (Large Scale Integration) devices capable of driving several 7segment digits. This device
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 381
accepts data from a microprocessor in a serial format, driving more LED segments than it has
pins by by multiplexing the LEDs. For example, see link below for MAX6955
(http://www.maximic.com/appnotes.cfm/appnote number/3211)
12.5 Parallelin, parallelout, universal shift register
The purpose of the parallelin/ parallelout shift register is to take in parallel data, shift it,
then output it as shown below. A universal shift register is a doeverything device in addition
to the parallelin/ parallelout function.
clock
data out
stage A stage B stage C stage D
Parallelin, parallelout shift register with 4stages
QA QB QC QD
data in
DA DB DC DD
mode
Above we apply four bit of data to a parallelin/ parallelout shift register at DA DB DC
DD. The mode control, which may be multiple inputs, controls parallel loading vs shifting.
The mode control may also control the direction of shifting in some real devices. The data will
be shifted one bit position for each clock pulse. The shifted data is available at the outputs
QA QB QC QD . The ”data in” and ”data out” are provided for cascading of multiple stages.
Though, above, we can only cascade data for right shifting. We could accommodate cascading
of leftshift data by adding a pair of left pointing signals, ”data in” and ”data out”, above.
The internal details of a right shifting parallelin/ parallelout shift register are shown be
low. The tristate buffers are not strictly necessary to the parallelin/ parallelout shift register,
but are part of the realworld device shown below.
382 CHAPTER 12. SHIFT REGISTERS
QA QB
DBDA
SER
CLK
LD/SH
DC
QD
CLR
OC
QD’
Cascade
74LS395 parallelin/ parallelout shift register with tristate output
QC
DD
The 74LS395 so closely matches our concept of a hypothetical right shifting parallelin/
parallelout shift register that we use an overly simplified version of the data sheet details
above. See the link to the full data sheet more more details, later in this chapter.
LD/SH’ controls the ANDOR multiplexer at the data input to the FF’s. If LD/SH’=1, the
upper four AND gates are enabled allowing application of parallel inputs DA DB DC DD to the
four FF data inputs. Note the inverter bubble at the clock input of the four FFs. This indicates
that the 74LS395 clocks data on the negative going clock, which is the high to low transition.
The four bits of data will be clocked in parallel from DA DB DC DD to QA QB QC QD at the
next negative going clock. In this ”real part”, OC’ must be low if the data needs to be available
at the actual output pins as opposed to only on the internal FFs.
The previously loaded data may be shifted right by one bit position if LD/SH’=0 for the
succeeding negative going clock edges. Four clocks would shift the data entirely out of our
4bit shift register. The data would be lost unless our device was cascaded from QD’ to SER of
another device.
QA QB QC QD
1 1 0 1
X 1 1 0
X X 1 1
load
shift
shift
Load and shift Load and 2shifts
load
shift
Parallelin/ parallelout shift register
QA QB QC QD
1 1 0 1
X 1 1 0
DA DB DC DD
1 1 0 1data
DA DB DC DD
1 1 0 1data
Above, a data pattern is presented to inputsDA DB DC DD. The pattern is loaded toQA QB
QC QD . Then it is shifted one bit to the right. The incoming data is indicated by X, meaning
the we do no know what it is. If the input (SER) were grounded, for example, we would know
what data (0) was shifted in. Also shown, is right shifting by two positions, requiring two
clocks.
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 383
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QA QB QC
SI SO
Shift right
CLK
The above figure serves as a reference for the hardware involved in right shifting of data.
It is too simple to even bother with this figure, except for comparison to more complex figures
to follow.
Load and right shift
load
shift
QA QB QC
1 1 0
X 1 1
Right shifting of data is provided above for reference to the previous right shifter.
D
C
Q
Q D
C
Q
Q D
C
Q
Q
QA QB QC
SI
SO
Shift left
CLK
If we need to shift left, the FFs need to be rewired. Compare to the previous right shifter.
Also, SI and SO have been reversed. SI shifts to QC . QC shifts to QB . QB shifts to QA.
QA leaves on the SO connection, where it could cascade to another shifter SI. This left shift
sequence is backwards from the right shift sequence.
Load and left shift
load
shift
QA QB QC
1 1 0
1 0 X
Above we shift the same data pattern left by one bit.
384 CHAPTER 12. SHIFT REGISTERS
There is one problem with the ”shift left” figure above. There is no market for it. Nobody
manufactures a shiftleft part. A ”real device” which shifts one direction can be wired exter
nally to shift the other direction. Or, should we say there is no left or right in the context of
a device which shifts in only one direction. However, there is a market for a device which will
shift left or right on command by a control line. Of course, left and right are valid in that
context.
QB
CLK
L/R
Shift left/ right, right action
SR
D Q
Ck
D Q
Ck
LR
QC’
SR
cascade
D Q
Ck
LR LR
SL
cascade
SL
QA QC
(L/R=1)
What we have above is a hypothetical shift register capable of shifting either direction
under the control of L’/R. It is setup with L’/R=1 to shift the normal direction, right. L’/R=1
enables the multiplexer AND gates labeled R. This allows data to follow the path illustrated
by the arrows, when a clock is applied. The connection path is the same as the”too simple”
”shift right” figure above.
Data shifts in at SR, to QA, to QB , to QC , where it leaves at SR cascade. This pin could
drive SR of another device to the right.
What if we change L’/R to L’/R=0?
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 385
QB
CLK
L/R
Shift left/ right register, left action
SR
D Q
Ck
D Q
Ck
LR
QC’
SR
cascade
D Q
Ck
LR LR
SL
cascade
SL
QA QC
(L/R=0)
With L’/R=0, the multiplexer AND gates labeled L are enabled, yielding a path, shown by
the arrows, the same as the above ”shift left” figure. Data shifts in at SL, to QC , to QB , to QA,
where it leaves at SL cascade. This pin could drive SL of another device to the left.
The prime virtue of the above two figures illustrating the ”shift left/ right register” is sim
plicity. The operation of the left right control L’/R=0 is easy to follow. A commercial part needs
the parallel data loading implied by the section title. This appears in the figure below.
DA
CLK
L/R
Shift left/ right/ load
SH/LD
SR
D Q
Ck
D Q
Ck
loadLR
DB
loadLR
SL
D Q
Ck
DC
loadLR
SR
cascade
SL
cascade
QA QB QC
Now that we can shift both left and right via L’/R, let us add SH/LD’, shift/ load, and the
AND gates labeled ”load” to provide for parallel loading of data from inputs DA DB DC . When
386 CHAPTER 12. SHIFT REGISTERS
SH/LD’=0, AND gates R and L are disabled, AND gates ”load” are enabled to pass data DA
DB DC to the FF data inputs. the next clock CLK will clock the data to QA QB QC . As long
as the same data is present it will be reloaded on succeeding clocks. However, data present
for only one clock will be lost from the outputs when it is no longer present on the data inputs.
One solution is to load the data on one clock, then proceed to shift on the next four clocks. This
problem is remedied in the 74ALS299 by the addition of another AND gate to the multiplexer.
If SH/LD’ is changed to SH/LD’=1, the AND gates labeled ”load” are disabled, allowing the
left/ right control L’/R to set the direction of shift on the L or R AND gates. Shifting is as in
the previous figures.
The only thing needed to produce a viable integrated device is to add the fourth AND gate
to the multiplexer as alluded for the 74ALS299. This is shown in the next section for that part.
12.5.1 Parallelin/ parallelout and universal devices
Let’s take a closer look at Serialin/ parallelout shift registers available as integrated circuits,
courtesy of Texas Instruments. For complete device data sheets, follow the links.
• SN74LS395A parallelin/ parallelout 4bit shift register
(http://wwws.ti.com/sc/ds/sn74ls395a.pdf)
• SN74ALS299 parallelin/ parallelout 8bit universal shift register
(http://wwws.ti.com/sc/ds/sn74als299.pdf)
SRG4R
C3/2
1
7
15
14
13
12
SER QA
QB
QC
QD
CLR
CLK
SN74LS395A ANSI Symbol
3
2
A
EN4OC
LD/SH
10
9
B
C
D
4
5
6
QD’11
M2 (SHIFT)
M1 (LOAD)
2,3D
4
1,3D
41,3D
We have already looked at the internal details of the SN74LS395A, see above previous
figure, 74LS395 parallelin/ parallelout shift register with tristate output. Directly above is
the ANSI symbol for the 74LS395.
Why only 4bits, as indicated by SRG4 above? Having both parallel inputs, and parallel
outputs, in addition to control and power pins, does not allow for any more I/O (Input/Output)
bits in a 16pin DIP (Dual Inline Package).
R indicates that the shift register stages are reset by input CLR’ (active low inverting half
arrow at input) of the control section at the top of the symbol. OC’, when low, (invert arrow
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 387
again) will enable (EN4) the four tristate output buffers (QA QB QC QD ) in the data section.
Load/shift’ (LD/SH’) at pin (7) corresponds to internals M1 (load) and M2 (shift). Look for
prefixes of 1 and 2 in the rest of the symbol to ascertain what is controlled by these.
The negative edge sensitive clock (indicated by the invert arrow at pin10) C3/2has two
functions. First, the 3 of C3/2 affects any input having a prefix of 3, say 2,3D or 1,3D in the
data section. This would be parallel load at A, B, C, D attributed to M1 and C3 for 1,3D.
Second, 2 of C3/2rightarrow indicates data clocking wherever 2 appears in a prefix (2,3D at
pin2). Thus we have clocking of data at SER into QA with mode 2 . The right arrow after
C3/2 accounts for shifting at internal shift register stages QA QB QC QD.
The right pointing triangles indicate buffering; the inverted triangle indicates tristate,
controlled by the EN4. Note, all the 4s in the symbol associated with the EN are frequently
omitted. Stages QB QC are understood to have the same attributes as QD. QD’ cascades to
the next package’s SER to the right.
mode mux
activity S1 S0 clock gate
hold 0 0 ↑ hold
shift left 0 1 ↑ L
shift right 1 0 ↑ R
load 1 1 ↑ load
S1 S0 OE2 OE1 tristate
X X X 1 disable
X X 1 X disable
0 0 0 0 enable
0 1 0 0 enable
1 0 0 0 enable
1 1 X X disable
The table above, condensed from the data ’299 data sheet, summarizes the operation of the
74ALS299 universal shift/ storage register. Follow the ’299 link above for full details. The
Multiplexer gates R, L, load operate as in the previous ”shift left/ right register” figures. The
difference is that the mode inputs S1 and S0 select shift left, shift right, and load with mode
set to S1 S0 = to 01, 10, and 11respectively as shown in the table, enabling multiplexer gates
L, R, and load respectively. See table. A minor difference is the parallel load path from the
tristate outputs. Actually the tristate buffers are (must be) disabled by S1 S0 = 11 to float
the I/O bus for use as inputs. A bus is a collection of similar signals. The inputs are applied
to A, B through H (same pins as QA, QB , through QH ) and routed to the load gate in the
multiplexers, and on the the D inputs of the FFs. Data is parallel load on a clock pulse.
The one new multiplexer gate is the AND gate labeled hold, enabled by S1 S0 = 00. The
hold gate enables a path from the Q output of the FF back to the hold gate, to the D input
of the same FF. The result is that with mode S1 S0 = 00, the output is continuously reloaded
with each new clock pulse. Thus, data is held. This is summarized in the table.
To read data from outputs QA, QB , through QH , the tristate buffers must be enabled by
OE2’, OE1’ =00 and mode =S1 S0 = 00, 01, or 10. That is, mode is anything except load. See
second table.
388 CHAPTER 12. SHIFT REGISTERS
QH’
DA
OE1
CLK
S0
74ALS299 universal shift/ storage register with tristate outputs
H/QH
QA’
OE2
S1
SR
SL
CLR
A/QA
D Q
R
Ck
D Q
R
Ck
R L load R L load
6stages
omitted
hold hold
Right shift data from a package to the left, shifts in on the SR input. Any data shifted
out to the right from stage QH cascades to the right via QH ’. This output is unaffected by the
tristate buffers. The shift right sequence for S1 S0 = 10 is:
SR > QA > QB > QC > QD > QE > QF > QG > QH (QH ’)
Left shift data from a package to the right shifts in on the SL input. Any data shifted out
to the left from stage QA cascades to the left via QA’, also unaffected by the tristate buffers.
The shift left sequence for S1 S0 = 01 is:
(QA’) QA < QB < QC < QD < QE < QF < QG < QH (QSL’)
Shifting may take place with the tristate buffers disabled by one of OE2’ or OE1’ = 1.
Though, the register contents outputs will not be accessible. See table.
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 389
C4/1
SN74ALS299 ANSI Symbol
3,4D 1,4D
6,13
QH’17
SRG8
& 3EN13
0 0
1 3
M
R
3,4D
1,4D
3,4D
12,13
2,4D
QA’8
Z5
Z6
Z12
CLR 9
OE1 2
OE2 3
SO 1
S1 19
CLK 12
SR 11
A/QA 7
6
14
5
13
4
16
15
SL 18
5,13
B/QB
C/QC
D/QD
E/QE
F/QF
G/QG
H/QH
/2
The ”clean” ANSI symbol for the SN74ALS299 parallelin/ parallelout 8bit universal shift
register with tristate output is shown for reference above.
C4/1
SN74ALS299 ANSI Symbol, annotated
3,4D 1,4D
6,13
QH’17
SRG8
& 3EN13
0 0
1 3
M
R
3,4D
1,4D
3,4D
12,13
2,4D
QA’8
Z5
Z6
Z12
CLR 9
OE1 2
OE2 3
SO 1
S1 19
CLK 12
SR 11
A/QA 7
6
14
5
13
4
16
15
SL 18
5,13
B/QB
C/QC
D/QD
E/QE
F/QF
G/QG
H/QH
/2
mode3 SO S1 M
O O O
O 1 1
21 O
1 1 3
function
hold
shift right
load
shift left
EN13=mode3 & OE1 & OE2
enable tristate buffers
prefix 1 implies right shift
prefix 2 implies left shift
prefix 3,4D implies mode3 parallel
load by C4
Z5, Z6 to Z12 are tristate outputs of the
shift register stages associated with the
I/O pins A/QA, B/QB, to Q/QH as implied
by prefixes 5,13; 6,13; to 12,13 respectively.
4 as a prefix (4D) implies clocking of
data by C4, as opposed to shifting
A double arrow implies bidirectional
data, equivalent to the input (no arrow)
and output (single arrow).
is buffer
is tristate
390 CHAPTER 12. SHIFT REGISTERS
The annotated version of the ANSI symbol is shown to clarify the terminology contained
therein. Note that the ANSI mode (S0 S1) is reversed from the order (S1 S0) used in the
previous table. That reverses the decimal mode numbers (1 & 2). In any event, we are in
complete agreement with the official data sheet, copying this inconsistency.
12.5.2 Practical applications
The Alarm with remote keypad block diagram is repeated below. Previously, we built the
keypad reader and the remote display as separate units. Now we will combine both the keypad
and display into a single unit using a universal shift register. Though separate in the diagram,
the Keypad and Display are both contained within the same remote enclosure.
1 2 3
4 5 6
87 9
Keypad Alarm
Gnd
Clock
Serial data
+5V
Alarm with remote keypad and display
Remote display
We will parallel load the keyboard data into the shift register on a single clock pulse, then
shift it out to the main alarm box. At the same time , we will shift LED data from the main
alarm to the remote shift register to illuminate the LEDs. We will be simultaneously shifting
keyboard data out and LED data into the shift register.
12.5. PARALLELIN, PARALLELOUT, UNIVERSAL SHIFT REGISTER 391
C4/1
74ALS299 universal shift register reads switches, drives LEDs
3,4D 1,4D
6,13
QH’
17
SRG8
& 3EN13
0 0
1 3
M
R
3,4D
1,4D
3,4D
12,13
2,4D
QA’8
Z5
Z6
Z12
CLR 9
OE1 2
OE2
3
SO 1
S1 19
CLK 12
SR 11
A/QA 7
6
14
5
13
4
16
15
SL 18
5,13
B/QB
C/QC
D/QD
E/QE
F/QF
G/QG
H/QH
/2
470 Ω x8
+5V
S1
CLK
SO
SR
2.2K Ω
+5V
S1 S0 mode
0 0 hold
0 1 L
1 0 R
1 1 load
1
19
2
18 17 16 15 14 13 12 11
3 4 5 6 7 8 9
74ALS541
74ALS299
Eight LEDs and current limiting resistors are connected to the eight I/O pins of the 74ALS299
universal shift register. The LEDS can only be driven during Mode 3 with S1=0 S0=0. The
OE1’ and OE2’ tristate enables are grounded to permenantly enable the tristate outputs dur
ing modes 0, 1, 2. That will cause the LEDS to light (flicker) during shifting. If this were a
problem the EN1’ and EN2’ could be ungrounded and paralleled with S1 and S0 respectively
to only enable the tristate buffers and light the LEDS during hold, mode 3. Let’s keep it simple
for this example.
During parallel loading, S0=1 inverted to a 0, enables the octal tristate buffers to ground
the switch wipers. The upper, open, switch contacts are pulled up to logic high by the resister
LED combination at the eight inputs. Any switch closure will short the input low. We parallel
load the switch data into the ’299 at clock t0 when both S0 and S1 are high. See waveforms
below.
392 CHAPTER 12. SHIFT REGISTERS
t0
S1
S0
shift right
load
holdhold
S1 S0 mode
0 0 hold
0 1 L
1 0 R
1 1 load
Load (t0) & shift (t1t8) switches out of QH’, shift LED data into SR
t1 t2 t3 t4 t5 t6 t7 t8 t9 t10 t11
Once S0 goes low, eight clocks (t0 tot8) shift switch closure data out of the ’299 via the Qh’
pin. At the same time, new LED data is shifted in at SR of the 299 by the same eight clocks.
The LED data replaces the switch closure data as shifting proceeds.
After the 8th shift clock, t8, S1 goes low to yield hold mode (S1 S0 = 00). The data in the
shift register remains the same even if there are more clocks, for example, T9, t10, etc. Where
do the waveforms come from? They could be generated by a microprocessor if the clock rate
were not over 100 kHz, in which case, it would be inconvenient to generate any clocks after t8.
If the clock was in the megahertz range, the clock would run continuously. The clock, S1 and
S0 would be generated by digital logic, not shown here.
12.6 Ring counters
If the output of a shift register is fed back to the input. a ring counter results. The data
pattern contained within the shift register will recirculate as long as clock pulses are applied.
For example, the data pattern will repeat every four clock pulses in the figure below. However,
we must load a data pattern. All 0’s or all 1’s doesn’t count. Is a continuous logic level from
such a condition useful?
clock
data in
data out
stage A stage B stage C stage D
Ring Counter, shift register output fed back to input
QD
We make provisions for loading data into the parallelin/ serialout shift register configured
as a ring counter below. Any random pattern may be loaded. The most generally useful pattern
is a single 1.
12.6. RING COUNTERS 393
clock
data out
stage A stage B stage C stage D
Parallelin, serialout shift register configured as
a ring counter
DA DB DC DD
data in
1 0 0 0
Loading binary 1000 into the ring counter, above, prior to shifting yields a viewable pattern.
The data pattern for a single stage repeats every four clock pulses in our 4stage example. The
waveforms for all four stages look the same, except for the one clock time delay from one stage
to the next. See figure below.
clock
t1 t2 t3 t4
SHIFT/LD
Load 1000 into 4stage ring counter and shift
QA
QB
QC
QD
t8 t12
1
0
0
0
The circuit above is a divide by 4 counter. Comparing the clock input to any one of the
outputs, shows a frequency ratio of 4:1. How may stages would we need for a divide by 10 ring
counter? Ten stages would recirculate the 1 every 10 clock pulses.
394 CHAPTER 12. SHIFT REGISTERS
D
C
Q
Q D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
SET
Set one stage, clear three stages
An alternate method of initializing the ring counter to 1000 is shown above. The shift
waveforms are identical to those above, repeating every fourth clock pulse. The requirement
for initialization is a disadvantage of the ring counter over a conventional counter. At a min
imum, it must be initialized at powerup since there is no way to predict what state flipflops
will power up in. In theory, initialization should never be required again. In actual prac
tice, the flipflops could eventually be corrupted by noise, destroying the data pattern. A ”self
correcting” counter, like a conventional synchronous binary counter would be more reliable.
J Q
Q
C
K
J Q
Q
C
K
5V
CLK
QB QA GA GB GC GD
Compare to binary synchronous counter with decoder
J K Q
0 0 hold
0 1 0
1 0 1
1 1 toggle
The above binary synchronous counter needs only two stages, but requires decoder gates.
The ring counter had more stages, but was self decoding, saving the decode gates above. An
other disadvantage of the ring counter is that it is not ”self starting”. If we need the decoded
outputs, the ring counter looks attractive, in particular, if most of the logic is in a single shift
register package. If not, the conventional binary counter is less complex without the decoder.
12.6. RING COUNTERS 395
clock
t1 t2 t3 t4
Compare to binary synchronous counter with decode, waveforms
GA
GB
GC
GD
t8 t12
1
0
0
0
QA
QB 1
0
The waveforms decoded from the synchronous binary counter are identical to the previous
ring counter waveforms. The counter sequence is (QA QB) = (00 01 10 11).
12.6.1 Johnson counters
The switchtail ring counter, also know as the Johnson counter, overcomes some of the limita
tions of the ring counter. Like a ring counter a Johnson counter is a shift register fed back on
its’ self. It requires half the stages of a comparable ring counter for a given division ratio. If
the complement output of a ring counter is fed back to the input instead of the true output,
a Johnson counter results. The difference between a ring counter and a Johnson counter is
which output of the last stage is fed back (Q or Q’). Carefully compare the feedback connection
below to the previous ring counter.
D
C
Q
Q D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
RESET
Johnson counter (note the QD to DA feedback connection)
QA QB QC QDQAQBQCQD
0 0 0 0
1 0 0 0
1 1 0 0
1 1 1 0
1 1 1 1
0 1 1 1
0 0 1 1
0 0 0 1
repeat
This ”reversed” feedback connection has a profound effect upon the behavior of the other
wise similar circuits. Recirculating a single 1 around a ring counter divides the input clock by
a factor equal to the number of stages. Whereas, a Johnson counter divides by a factor equal
to twice the number of stages. For example, a 4stage ring counter divides by 4. A 4stage
Johnson counter divides by 8.
396 CHAPTER 12. SHIFT REGISTERS
Start a Johnson counter by clearing all stages to 0s before the first clock. This is often done
at powerup time. Referring to the figure below, the first clock shifts three 0s from ( QA QB
QC) to the right into ( QB QC QD). The 1 at QD’ (the complement of Q) is shifted back into
QA. Thus, we start shifting 1s to the right, replacing the 0s. Where a ring counter recirculated
a single 1, the 4stage Johnson counter recirculates four 0s then four 1s for an 8bit pattern,
then repeats.
clock
t1 t2 t3 t4
Four stage Johnson counter waveforms
QA
QB
QC
QD
t8 t12 t16
RESET
t5 t6 t7t0
The above waveforms illustrates that multiphase square waves are generated by a Johnson
counter. The 4stage unit above generates four overlapping phases of 50% duty cycle. How
many stages would be required to generate a set of three phase waveforms? For example, a
three stage Johnson counter, driven by a 360 Hertz clock would generate three 120o phased
square waves at 60 Hertz.
The outputs of the flopflops in a Johnson counter are easy to decode to a single state. Below
for example, the eight states of a 4stage Johnson counter are decoded by no more than a two
input gate for each of the states. In our example, eight of the two input gates decode the states
for our example Johnson counter.
12.6. RING COUNTERS 397
D
C
Q
Q D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
CLR
Johnson counter with decoder (CD4022B)
QA QB QC QD
QAQBQCQD G
0 0 0 0 G0
1 0 0 0 G1
1 1 0 0 G2
1 1 1 0 G3
1 1 1 1 G4
0 1 1 1 G5
0 0 1 1 G6
0 0 0 1 G7
G0 G5 G2 G7
G4 G1 G6 G3
No matter how long the Johnson counter, only 2input decoder gates are needed. Note, we
could have used uninverted inputs to the AND gates by changing the gate inputs from true to
inverted at the FFs, Q to Q’, (and vice versa). However, we are trying to make the diagram
above match the data sheet for the CD4022B, as closely as practical.
clock
t1 t2 t3 t4
Four stage (8state) Johnson counter decoder waveforms
QA
QB
QC
QD
t8 t12 t16
G1=QAQB
G2=QBQC
G3=QCQD
G4=QAQD
G5=QAQB
G6=QBQC
G7=QCQD
G0=QAQD
t0 t5 t6 t7
Above, our four phased square waves QA to QD are decoded to eight signals (G0 to G7)
398 CHAPTER 12. SHIFT REGISTERS
active during one clock period out of a complete 8clock cycle. For example, G0 is active high
when both QA and QD are low. Thus, pairs of the various register outputs define each of the
eight states of our Johnson counter example.
D
C
Q
Q D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
RESET
CD4022B modulo8 Johnson counter with unused state detector
QA QB QC QD
QAQBQCQD G
0 0 0 0 G0
1 0 0 0 G1
1 1 0 0 G2
1 1 1 0 G3
1 1 1 1 G4
0 1 1 1 G5
0 0 1 1 G6
0 0 0 1 G7
G0 G5 G2 G7
G4 G1 G6 G3
Cout
11 1 5 7 12
2 4 3 10
14
13
10
CLOCK
ENABLE
NOR gate unused state detector: QA QB QC = 010 forces the 1 to a 0
Above is the more complete internal diagram of the CD4022B Johnson counter. See the
manufacturers’ data sheet for minor details omitted. The major new addition to the diagram
as compared to previous figures is the disallowed state detector composed of the twoNOR gates.
Take a look at the inset state table. There are 8permissible states as listed in the table. Since
our shifter has four flipflops, there are a total of 16states, of which there are 8disallowed
states. That would be the ones not listed in the table.
In theory, we will not get into any of the disallowed states as long as the shift register is
RESET before first use. However, in the ”real world” after many days of continuous opera
tion due to unforeseen noise, power line disturbances, near lightning strikes, etc, the Johnson
counter could get into one of the disallowed states. For high reliability applications, we need
to plan for this slim possibility. More serious is the case where the circuit is not cleared at
powerup. In this case there is no way to know which of the 16states the circuit will power up
in. Once in a disallowed state, the Johnson counter will not return to any of the permissible
states without intervention. That is the purpose of the NOR gates.
12.6. RING COUNTERS 399
Examine the table for the sequence (QA QB QC) = (010). Nowhere does this sequence
appear in the table of allowed states. Therefore (010) is disallowed. It should never occur. If it
does, the Johnson counter is in a disallowed state, which it needs to exit to any allowed state.
Suppose that (QA QB QC) = (010). The second NOR gate will replace QB = 1 with a 0 at the
D input to FF QC . In other words, the offending 010 is replaced by 000. And 000, which does
appear in the table, will be shifted right. There are may triple0 sequences in the table. This
is how the NOR gates get the Johnson counter out of a disallowed state to an allowed state.
Not all disallowed states contain a 010 sequence. However, after a few clocks, this sequence
will appear so that any disallowed states will eventually be escaped. If the circuit is powered
up without a RESET, the outputs will be unpredictable for a few clocks until an allowed state
is reached. If this is a problem for a particular application, be sure to RESET on powerup.
Johnson counter devices
A pair of integrated circuit Johnson counter devices with the output states decoded is available.
We have already looked at the CD4017 internal logic in the discussion of Johnson counters. The
4000 series devices can operate from 3V to 15V power supplies. The the 74HC’ part, designed
for a TTL compatiblity, can operate from a 2V to 6V supply, count faster, and has greater output
drive capability. For complete device data sheets, follow the links.
• CD4017 Johnson counter with 10 decoded outputs
CD4022 Johnson counter with 8 decoded outputs
(http://wwws.ti.com/sc/ds/cd4017b.pdf)
• 74HC4017 Johnson counter, 10 decoded outputs
(http://wwws.ti.com/sc/ds/cd74hc4017.pdf)
&
CTR DIV 10/
DEC 0
1
2
3
4
5
6
7
8
9
CD<5
CT=0
CKEN
CK
CLR
13
14
15
3
4
7
2
10
5
6
1
9
12
11
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
CO
CD4017B, 74HC4017
&
CTR DIV 8/
OCTAL 0
1
2
3
4
5
6
7
CD<4
CT=0
CKEN
CK
CLR
13
14
15
3
4
7
2
10
5
1
12
11
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
CO
CD4022B
The ANSI symbols for the modulo10 (divide by 10) and modulo8 Johnson counters are
shown above. The symbol takes on the characteristics of a counter rather than a shift register
derivative, which it is. Waveforms for the CD4022 modulo8 and operation were shown previ
ously. The CD4017B/ 74HC4017 decade counter is a 5stage Johnson counter with ten decoded
400 CHAPTER 12. SHIFT REGISTERS
outputs. The operation and waveforms are similar to the CD4017. In fact, the CD4017 and
CD4022 are both detailed on the same data sheet. See above links. The 74HC4017 is a more
modern version of the decade counter.
These devices are used where decoded outputs are needed instead of the binary or BCD
(Binary Coded Decimal) outputs found on normal counters. By decoded, we mean one line
out of the ten lines is active at a time for the ’4017 in place of the four bit BCD code out of
conventional counters. See previous waveforms for 1of8 decoding for the ’4022 Octal Johnson
counter.
Practical applications
&
CTR DIV 10/
DEC 0
1
2
3
4
5
6
7
8
9
CD<5
CT=0
CKEN
CK
CLR
13
14
15
3
4
7
2
10
5
6
1
9
12
11
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
CO
74HC4017
555
10K
100K
1µF 0.1µF1
2
6
7
8 4
3
5
R1
5V
R2
C1
C2
1.46f = (R1 +2R2)C
U1
U2
5V
16
8
R=820 Ω x10
Decoded ring counter drives walking LED
The above Johnson counter shifts a lighted LED each fifth of a second around the ring of ten.
Note that the 74HC4017 is used instead of the ’40017 because the former part has more current
drive capability. From the data sheet, (at the link above) operating at VCC= 5V, the VOH= 4.6V
at 4ma. In other words, the outputs can supply 4 ma at 4.6 V to drive the LEDs. Keep in mind
that LEDs are normally driven with 10 to 20 ma of current. Though, they are visible down to 1
ma. This simple circuit illustrates an application of the ’HC4017. Need a bright display for an
exhibit? Then, use inverting buffers to drive the cathodes of the LEDs pulled up to the power
supply by lower value anode resistors.
The 555 timer, serving as an astable multivibrator, generates a clock frequency determined
by R1 R2 C1. This drives the 74HC4017 a step per clock as indicated by a single LED illumi
nated on the ring. Note, if the 555 does not reliably drive the clock pin of the ’4015, run it
through a single buffer stage between the 555 and the ’4017. A variable R2 could change the
step rate. The value of decoupling capacitor C2 is not critical. A similar capacitor should be
applied across the power and ground pins of the ’4017.
12.6. RING COUNTERS 401
D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
QA QB QC
Three phase square/ sine wave generator.
low pass filter
low pass filter
low pass filter
φ2
φ3
φ1P1
CLOCK
P2
P3
P1
P2
P3
QA
QC Disallowed state
The Johnson counter above generates 3phase square waves, phased 60o apart with respect
to (QA QB QC). However, we need 120o phased waveforms of power applications (see Volume
II, AC). Choosing P1=QA P2=QC P3=QB ’ yields the 120o phasing desired. See figure below. If
these (P1 P2 P3) are lowpass filtered to sine waves and amplified, this could be the beginnings
of a 3phase power supply. For example, do you need to drive a small 3phase 400 Hz aircraft
motor? Then, feed 6x 400Hz to the above circuit CLOCK. Note that all these waveforms are
50% duty cycle.
clock
t1 t2 t3 t4
3stage Johnson counter generates 3Ø waveform.
QA
QC
t8t0 t5 t6 t7
P1=QA
QB
P3=QB’
P2=QC
402 CHAPTER 12. SHIFT REGISTERS
The circuit below produces 3phase nonoverlapping, less than 50% duty cycle, waveforms
for driving 3phase stepper motors.
D
C
Q
Q D
C
Q
Q D
C
Q
Q
CLOCK
QA QB QC
P0
P1
P2
P1=QBQC
P2=QAQB
P0=QAQC
3stage (6state) Johnson counter decoded for 3φ stepper
motor.
9
1
2
3 14
16
15
VMOTOR
8
U3A
U3B
U3C
part of
ULN2003
QA
QC
Disallowed state
Above we decode the overlapping outputs QA QB QC to nonoverlapping outputs P0 P1 P2
as shown below. These waveforms drive a 3phase stepper motor after suitable amplification
from the milliamp level to the fractional amp level using the ULN2003 drivers shown above, or
the discrete component Darlington pair driver shown in the circuit which follow. Not counting
the motor driver, this circuit requires three IC (Integrated Circuit) packages: two dual type
”D” FF packages and a quad NAND gate.
clock
t1 t2 t3 t4
3stage Johnson counter generates 3Ø stepper
waveform.
QA
QC
t8t0 t5 t6 t7
QB
P1=QBQC
P2=QAQB
P0=QAQC
12.6. RING COUNTERS 403
&
CTR DIV 10/
DEC 0
1
2
3
4
5
6
7
8
9
CD<5
CT=0
CKEN
CK
CLR
13
14
15
3
4
7
2
10
5
6
1
9
12
11
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
CO
CD4017B, 74HC4017
G0
G1
G2
VCC
8
16
VM
Johnson sequence terminated early by reset at Q3, which is high.
for nano seconds
A single CD4017, above, generates the required 3phase stepper waveforms in the circuit
above by clearing the Johnson counter at count 3. Count 3 persists for less than a microsecond
before it clears its’ self. The other counts (Q0=G0 Q1=G1 Q2=G2) remain for a full clock period
each.
The Darlington bipolar transistor drivers shown above are a substitute for the internal
circuitry of the ULN2003. The design of drivers is beyond the scope of this digital electronics
chapter. Either driver may be used with either waveform generator circuit.
clock
t1 t2 t3 t4
CD4017B 5stage (10state) Johnson counter resetting
at QC QB QA= 100 generates 3−φ stepper waveform.
QA
QB
QC
t8
Q1=G1=QAQB
Q1=G2=QBQC
Q0=G0=QAQD
t0 t5 t6 t7
G3=QCQD
The above waceforms make the most sense in the context of the internal logic of the CD4017
shown earlier in this section. Though, the AND gating equations for the internal decoder are
shown. The signals QA QB QC are Johnson counter direct shift register outputs not available
404 CHAPTER 12. SHIFT REGISTERS
on pinouts. The QD waveform shows resetting of the ’4017 every three clocks. Q0 Q1 Q2, etc.
are decoded outputs which actually are available at output pins.
&
CTR DIV 10/
DEC 0
1
2
3
4
5
6
7
8
9
CD<5
CT=0
CKEN
CK
CLR
13
14
15
3
4
7
2
10
5
6
1
9
12
11
Q0
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
Q9
CO
CD4017B, 74HC4017
VCC
8
16
VM
Johnson counter drives unipolar stepper motor.
9
1
2
3 14
16
15
8 part of
ULN2003
4 13
VM
Above we generate waveforms for driving a unipolar stepper motor, which only requires one
polarity of driving signal. That is, we do not have to reverse the polarity of the drive to the
windings. This simplifies the power driver between the ’4017 and the motor. Darlington pairs
from a prior diagram may be substituted for the ULN3003.
clock
t1 t2 t3 t4 t8
Q1
Q2
Q0
t0 t5 t6 t7
Q4
Q3
Johnson counter unipolar stepper motor waveforms.
Once again, the CD4017B generates the required waveforms with a reset after the teminal
count. The decoded outputs Q0 Q1 Q2 Q3 sucessively drive the stepper motor windings, with
Q4 reseting the counter at the end of each group of four pulses.
12.7. REFERENCES 405
12.7 references
DataSheetCatalog.com http://www.datasheetcatalog.com/
http://www.st.com/stonline/psearch/index.htm select standard logics
http://www.st.com/stonline/books/pdf/docs/2069.pdf
http://www.ti.com/ (Products, Logic, Product Tree)
406 CHAPTER 12. SHIFT REGISTERS
Chapter 13
DIGITALANALOG
CONVERSION
Contents
13.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
13.2 The R/2nR DAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 409
13.3 The R/2R DAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
13.4 Flash ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414
13.5 Digital ramp ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
13.6 Successive approximation ADC . . . . . . . . . . . . . . . . . . . . . . . . . . 419
13.7 Tracking ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
13.8 Slope (integrating) ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 422
13.9 DeltaSigma (∆Σ) ADC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
13.10Practical considerations of ADC circuits . . . . . . . . . . . . . . . . . . . . 427
13.1 Introduction
Connecting digital circuitry to sensor devices is simple if the sensor devices are inherently
digital themselves. Switches, relays, and encoders are easily interfaced with gate circuits due
to the on/off nature of their signals. However, when analog devices are involved, interfacing
becomes much more complex. What is needed is a way to electronically translate analog sig
nals into digital (binary) quantities, and vice versa. An analogtodigital converter, or ADC,
performs the former task while a digitaltoanalog converter, or DAC, performs the latter.
An ADC inputs an analog electrical signal such as voltage or current and outputs a binary
number. In block diagram form, it can be represented as such:
407
408 CHAPTER 13. DIGITALANALOG CONVERSION
ADC
Vdd
input
Binary
outputsignal
Analog
ADAC, on the other hand, inputs a binary number and outputs an analog voltage or current
signal. In block diagram form, it looks like this:
Vdd
input
Binary
output
DAC signal
Analog
Together, they are often used in digital systems to provide complete interface with ana
log sensors and output devices for control systems such as those used in automotive engine
controls:
ADC
Vdd
input
signal
Analog
VddVdd
Control
computer DAC Analog
signal
output
Digital control system with
analog I/O
It is much easier to convert a digital signal into an analog signal than it is to do the reverse.
Therefore, we will begin with DAC circuitry and then move to ADC circuitry.
13.2. THE R/2NR DAC 409
13.2 The R/2nR DAC
This DAC circuit, otherwise known as the binaryweightedinput DAC, is a variation on the
inverting summer opamp circuit. If you recall, the classic inverting summer circuit is an
operational amplifier using negative feedback for controlled gain, with several voltage inputs
and one voltage output. The output voltage is the inverted (opposite polarity) sum of all input
voltages:
−
+
R
R
R
V1
V2
V3
Vout
R
I1
I3
I2
I1 + I2 + I30 V
0 V
Inverting summer circuit
Vout =  (V1 + V2 + V3)
For a simple inverting summer circuit, all resistors must be of equal value. If any of the
input resistors were different, the input voltages would have different degrees of effect on the
output, and the output voltage would not be a true sum. Let’s consider, however, intentionally
setting the input resistors at different values. Suppose we were to set the input resistor values
at multiple powers of two: R, 2R, and 4R, instead of all the same value R:
−
+
V1
V2
V3
Vout
R
I1
I3
I2
I1 + I2 + I30 V
0 V
R
2R
4R
Vout =  (V1 + V22 +
V3
4
)
Starting from V1 and going through V3, this would give each input voltage exactly half the
effect on the output as the voltage before it. In other words, input voltage V1 has a 1:1 effect on
410 CHAPTER 13. DIGITALANALOG CONVERSION
the output voltage (gain of 1), while input voltage V2 has half that much effect on the output
(a gain of 1/2), and V3 half of that (a gain of 1/4). These ratios are were not arbitrarily chosen:
they are the same ratios corresponding to place weights in the binary numeration system. If
we drive the inputs of this circuit with digital gates so that each input is either 0 volts or full
supply voltage, the output voltage will be an analog representation of the binary value of these
three bits.
−
+
V1
V2
V3
Vout
R
I1
I3
I2
I1 + I2 + I30 V
0 V
R
2R
4R
Binary
input
MSB
LSB
If we chart the output voltages for all eight combinations of binary bits (000 through 111)
input to this circuit, we will get the following progression of voltages:

 Binary  Output voltage 

 000  0.00 V 

 001  1.25 V 

 010  2.50 V 

 011  3.75 V 

 100  5.00 V 

 101  6.25 V 

 110  7.50 V 

 111  8.75 V 

Note that with each step in the binary count sequence, there results a 1.25 volt change in
the output. This circuit is very easy to simulate using SPICE. In the following simulation, I
set up the DAC circuit with a binary input of 110 (note the first node numbers for resistors R1,
R2, and R3: a node number of ”1” connects it to the positive side of a 5 volt battery, and a node
number of ”0” connects it to ground). The output voltage appears on node 6 in the simulation:
13.2. THE R/2NR DAC 411
−
+
6
0
5
0
1
5 V
R1
R2
R3
Rfeedbk
1 kΩ
2 kΩ
4 kΩ
5 6Rbogus
99 kΩ
binaryweighted dac
v1 1 0 dc 5
rbogus 1 0 99k
r1 1 5 1k
r2 1 5 2k
r3 0 5 4k
rfeedbk 5 6 1k
e1 6 0 5 0 999k
.end
node voltage node voltage node voltage
(1) 5.0000 (5) 0.0000 (6) 7.5000
We can adjust resistors values in this circuit to obtain output voltages directly correspond
ing to the binary input. For example, by making the feedback resistor 800 Ω instead of 1 kΩ,
the DAC will output 1 volt for the binary input 001, 4 volts for the binary input 100, 7 volts
for the binary input 111, and so on.
(with feedback resistor set at 800 ohms)

 Binary  Output voltage 

 000  0.00 V 

 001  1.00 V 

 010  2.00 V 

 011  3.00 V 

 100  4.00 V 

 101  5.00 V 

412 CHAPTER 13. DIGITALANALOG CONVERSION
 110  6.00 V 

 111  7.00 V 

If we wish to expand the resolution of this DAC (add more bits to the input), all we need to
do is add more input resistors, holding to the same poweroftwo sequence of values:
−
+
Vout
R
2R
4R
Binary
input
MSB
LSB
8R
16R
32R
Rfeedback
6bit binaryweighted DAC
It should be noted that all logic gates must output exactly the same voltages when in the
”high” state. If one gate is outputting +5.02 volts for a ”high” while another is outputting
only +4.86 volts, the analog output of the DAC will be adversely affected. Likewise, all ”low”
voltage levels should be identical between gates, ideally 0.00 volts exactly. It is recommended
that CMOS output gates are used, and that input/feedback resistor values are chosen so as to
minimize the amount of current each gate has to source or sink.
13.3 The R/2R DAC
An alternative to the binaryweightedinput DAC is the socalled R/2R DAC, which uses fewer
unique resistor values. A disadvantage of the former DAC design was its requirement of sev
eral different precise input resistor values: one unique value per binary input bit. Manufacture
may be simplified if there are fewer different resistor values to purchase, stock, and sort prior
to assembly.
Of course, we could take our last DAC circuit and modify it to use a single input resistance
value, by connecting multiple resistors together in series:
13.3. THE R/2R DAC 413
−
+
Vout
R
Binary
input
MSB
LSB
RR
RRRR
Unfortunately, this approach merely substitutes one type of complexity for another: volume
of components over diversity of component values. There is, however, a more efficient design
methodology.
By constructing a different kind of resistor network on the input of our summing circuit,
we can achieve the same kind of binary weighting with only two kinds of resistor values, and
with only a modest increase in resistor count. This ”ladder” network looks like this:
−
+
Vout
Binary
input
MSB
LSB
RR
2R 2R 2R
2R
R/2R "ladder" DAC
2R
Mathematically analyzing this ladder network is a bit more complex than for the previous
circuit, where each input resistor provided an easilycalculated gain for that bit. For those who
are interested in pursuing the intricacies of this circuit further, you may opt to use Thevenin’s
theorem for each binary input (remember to consider the effects of the virtual ground), and/or
use a simulation program like SPICE to determine circuit response. Either way, you should
obtain the following table of figures:

 Binary  Output voltage 

 000  0.00 V 

414 CHAPTER 13. DIGITALANALOG CONVERSION
 001  1.25 V 

 010  2.50 V 

 011  3.75 V 

 100  5.00 V 

 101  6.25 V 

 110  7.50 V 

 111  8.75 V 

As was the case with the binaryweighted DAC design, we can modify the value of the
feedback resistor to obtain any ”span” desired. For example, if we’re using +5 volts for a ”high”
voltage level and 0 volts for a ”low” voltage level, we can obtain an analog output directly
corresponding to the binary input (011 = 3 volts, 101 = 5 volts, 111 = 7 volts, etc.) by using a
feedback resistance with a value of 1.6R instead of 2R.
13.4 Flash ADC
Also called the parallel A/D converter, this circuit is the simplest to understand. It is formed of
a series of comparators, each one comparing the input signal to a unique reference voltage. The
comparator outputs connect to the inputs of a priority encoder circuit, which then produces a
binary output. The following illustration shows a 3bit flash ADC circuit:
13.4. FLASH ADC 415
Vref
R
R
R
R
R
R
R

+
Vin
8line to
3line
priority
encoder
Binary output
Vdd
+

+

+

+

+

+

Vref is a stable reference voltage provided by a precision voltage regulator as part of the
converter circuit, not shown in the schematic. As the analog input voltage exceeds the refer
ence voltage at each comparator, the comparator outputs will sequentially saturate to a high
state. The priority encoder generates a binary number based on the highestorder active input,
ignoring all other active inputs.
When operated, the flash ADC produces an output that looks something like this:
Time
Analog
input
Time
Digital
output
For this particular application, a regular priority encoder with all its inherent complexity
isn’t necessary. Due to the nature of the sequential comparator output states (each comparator
saturating ”high” in sequence from lowest to highest), the same ”highestorderinput selection”
effect may be realized through a set of ExclusiveOR gates, allowing the use of a simpler, non
priority encoder:
416 CHAPTER 13. DIGITALANALOG CONVERSION
Vref
R
R
R
R
R
R
R

+
Vin
8line to
3line
encoder
Binary output
Vdd
+

+

+

+

+

+

And, of course, the encoder circuit itself can be made from a matrix of diodes, demonstrating
just how simply this converter design may be constructed:
13.5. DIGITAL RAMP ADC 417
Vref
R
R
R
R
R
R
R
R

+
Vin
Binary output
7
6
5
4
3
2
1
Pulldown
resistors
+

+

+

+

+

+

Not only is the flash converter the simplest in terms of operational theory, but it is the most
efficient of the ADC technologies in terms of speed, being limited only in comparator and gate
propagation delays. Unfortunately, it is the most componentintensive for any given number
of output bits. This threebit flash ADC requires seven comparators. A fourbit version would
require 15 comparators. With each additional output bit, the number of required comparators
doubles. Considering that eight bits is generally considered the minimum necessary for any
practical ADC (255 comparators needed!), the flash methodology quickly shows its weakness.
An additional advantage of the flash converter, often overlooked, is the ability for it to pro
duce a nonlinear output. With equalvalue resistors in the reference voltage divider network,
each successive binary count represents the same amount of analog signal increase, providing
a proportional response. For special applications, however, the resistor values in the divider
network may be made nonequal. This gives the ADC a custom, nonlinear response to the ana
log input signal. No other ADC design is able to grant this signalconditioning behavior with
just a few component value changes.
13.5 Digital ramp ADC
Also known as the stairstepramp, or simply counter A/D converter, this is also fairly easy to
understand but unfortunately suffers from several limitations.
418 CHAPTER 13. DIGITALANALOG CONVERSION
The basic idea is to connect the output of a freerunning binary counter to the input of a
DAC, then compare the analog output of the DAC with the analog input signal to be digitized
and use the comparator’s output to tell the counter when to stop counting and reset. The
following schematic shows the basic idea:
CTR
DAC
Vdd Vdd
Vdd
SRG−
+Vin
Load
Binary
output
As the counter counts up with each clock pulse, the DAC outputs a slightly higher (more
positive) voltage. This voltage is compared against the input voltage by the comparator. If the
input voltage is greater than the DAC output, the comparator’s output will be high and the
counter will continue counting normally. Eventually, though, the DAC output will exceed the
input voltage, causing the comparator’s output to go low. This will cause two things to happen:
first, the hightolow transition of the comparator’s output will cause the shift register to ”load”
whatever binary count is being output by the counter, thus updating the ADC circuit’s output;
secondly, the counter will receive a low signal on the activelow LOAD input, causing it to reset
to 00000000 on the next clock pulse.
The effect of this circuit is to produce a DAC output that ramps up to whatever level the
analog input signal is at, output the binary number corresponding to that level, and start over
again. Plotted over time, it looks like this:
13.6. SUCCESSIVE APPROXIMATION ADC 419
Time
Analog
input
Time
Digital
output
Note how the time between updates (new digital output values) changes depending on how
high the input voltage is. For low signal levels, the updates are rather closespaced. For higher
signal levels, they are spaced further apart in time:
Digital
output
longer
time
shorter
time
For many ADC applications, this variation in update frequency (sample time) would not
be acceptable. This, and the fact that the circuit’s need to count all the way from 0 at the
beginning of each count cycle makes for relatively slow sampling of the analog signal, places
the digitalramp ADC at a disadvantage to other counter strategies.
13.6 Successive approximation ADC
One method of addressing the digital ramp ADC’s shortcomings is the socalled successive
approximation ADC. The only change in this design is a very special counter circuit known as
a successiveapproximation register. Instead of counting up in binary sequence, this register
counts by trying all values of bits starting with the mostsignificant bit and finishing at the
leastsignificant bit. Throughout the count process, the register monitors the comparator’s
output to see if the binary count is less than or greater than the analog signal input, adjusting
the bit values accordingly. The way the register counts is identical to the ”trialandfit” method
of decimaltobinary conversion, whereby different values of bits are tried from MSB to LSB to
get a binary number that equals the original decimal number. The advantage to this counting
strategy is much faster results: the DAC output converges on the analog signal input in much
larger steps than with the 0tofull count sequence of a regular counter.
Without showing the inner workings of the successiveapproximation register (SAR), the
circuit looks like this:
420 CHAPTER 13. DIGITALANALOG CONVERSION
DAC
Vdd Vdd
Vdd
SRG−
+Vin Binary
output
SAR
>/<
Done
It should be noted that the SAR is generally capable of outputting the binary number in
serial (one bit at a time) format, thus eliminating the need for a shift register. Plotted over
time, the operation of a successiveapproximation ADC looks like this:
Time
Analog
input
Time
Digital
output
Note how the updates for this ADC occur at regular intervals, unlike the digital ramp ADC
circuit.
13.7. TRACKING ADC 421
13.7 Tracking ADC
A third variation on the counterDACbased converter theme is, in my estimation, the most el
egant. Instead of a regular ”up” counter driving the DAC, this circuit uses an up/down counter.
The counter is continuously clocked, and the up/down control line is driven by the output of the
comparator. So, when the analog input signal exceeds the DAC output, the counter goes into
the ”count up” mode. When the DAC output exceeds the analog input, the counter switches
into the ”count down” mode. Either way, the DAC output always counts in the proper direction
to track the input signal.
CTR
DAC
Vdd Vdd
−
+Vin Binary
output
U/D
Notice how no shift register is needed to buffer the binary count at the end of a cycle. Since
the counter’s output continuously tracks the input (rather than counting to meet the input and
then resetting back to zero), the binary output is legitimately updated with every clock pulse.
An advantage of this converter circuit is speed, since the counter never has to reset. Note
the behavior of this circuit:
422 CHAPTER 13. DIGITALANALOG CONVERSION
Time
Analog
input
Time
Digital
output
Note the much faster update time than any of the other ”counting” ADC circuits. Also note
how at the very beginning of the plot where the counter had to ”catch up” with the analog
signal, the rate of change for the output was identical to that of the first counting ADC. Also,
with no shift register in this circuit, the binary output would actually ramp up rather than
jump from zero to an accurate count as it did with the counter and successive approximation
ADC circuits.
Perhaps the greatest drawback to this ADC design is the fact that the binary output is
never stable: it always switches between counts with every clock pulse, even with a perfectly
stable analog input signal. This phenomenon is informally known as bit bobble, and it can be
problematic in some digital systems.
This tendency can be overcome, though, through the creative use of a shift register. For
example, the counter’s output may be latched through a parallelin/parallelout shift register
only when the output changes by two or more steps. Building a circuit to detect two or more
successive counts in the same direction takes a little ingenuity, but is worth the effort.
13.8 Slope (integrating) ADC
So far, we’ve only been able to escape the sheer volume of components in the flash converter by
using a DAC as part of our ADC circuitry. However, this is not our only option. It is possible
to avoid using a DAC if we substitute an analog ramping circuit and a digital counter with
precise timing.
The is the basic idea behind the socalled singleslope, or integrating ADC. Instead of using
a DAC with a ramped output, we use an opamp circuit called an integrator to generate a
sawtooth waveform which is then compared against the analog input by a comparator. The
time it takes for the sawtooth waveform to exceed the input signal voltage level is measured
by means of a digital counter clocked with a precisefrequency square wave (usually from a
crystal oscillator). The basic schematic diagram is shown here:
13.8. SLOPE (INTEGRATING) ADC 423
CTR
Vdd
−
+
Vin
Binary
output
−
+
Vref CLR
Vdd
SRG
R C
The IGFET capacitordischarging transistor scheme shown here is a bit oversimplified. In
reality, a latching circuit timed with the clock signal would most likely have to be connected
to the IGFET gate to ensure full discharge of the capacitor when the comparator’s output goes
high. The basic idea, however, is evident in this diagram. When the comparator output is low
(input voltage greater than integrator output), the integrator is allowed to charge the capacitor
in a linear fashion. Meanwhile, the counter is counting up at a rate fixed by the precision clock
frequency. The time it takes for the capacitor to charge up to the same voltage level as the input
depends on the input signal level and the combination of Vref , R, and C. When the capacitor
reaches that voltage level, the comparator output goes high, loading the counter’s output into
the shift register for a final output. The IGFET is triggered ”on” by the comparator’s high
output, discharging the capacitor back to zero volts. When the integrator output voltage falls
to zero, the comparator output switches back to a low state, clearing the counter and enabling
the integrator to ramp up voltage again.
This ADC circuit behaves very much like the digital ramp ADC, except that the comparator
reference voltage is a smooth sawtooth waveform rather than a ”stairstep:”
Time
Analog
input
Time
Digital
output
The singleslope ADC suffers all the disadvantages of the digital ramp ADC, with the added
drawback of calibration drift. The accurate correspondence of this ADC’s output with its input
is dependent on the voltage slope of the integrator being matched to the counting rate of the
counter (the clock frequency). With the digital ramp ADC, the clock frequency had no effect
on conversion accuracy, only on update time. In this circuit, since the rate of integration and
424 CHAPTER 13. DIGITALANALOG CONVERSION
the rate of count are independent of each other, variation between the two is inevitable as it
ages, and will result in a loss of accuracy. The only good thing to say about this circuit is that
it avoids the use of a DAC, which reduces circuit complexity.
An answer to this calibration drift dilemma is found in a design variation called the dual
slope converter. In the dualslope converter, an integrator circuit is driven positive and nega
tive in alternating cycles to ramp down and then up, rather than being reset to 0 volts at the
end of every cycle. In one direction of ramping, the integrator is driven by the positive analog
input signal (producing a negative, variable rate of output voltage change, or output slope) for
a fixed amount of time, as measured by a counter with a precision frequency clock. Then, in the
other direction, with a fixed reference voltage (producing a fixed rate of output voltage change)
with time measured by the same counter. The counter stops counting when the integrator’s
output reaches the same voltage as it was when it started the fixedtime portion of the cycle.
The amount of time it takes for the integrator’s capacitor to discharge back to its original out
put voltage, as measured by the magnitude accrued by the counter, becomes the digital output
of the ADC circuit.
The dualslope method can be thought of analogously in terms of a rotary spring such as
that used in a mechanical clock mechanism. Imagine we were building a mechanism to mea
sure the rotary speed of a shaft. Thus, shaft speed is our ”input signal” to be measured by
this device. The measurement cycle begins with the spring in a relaxed state. The spring is
then turned, or ”wound up,” by the rotating shaft (input signal) for a fixed amount of time.
This places the spring in a certain amount of tension proportional to the shaft speed: a greater
shaft speed corresponds to a faster rate of winding. and a greater amount of spring tension
accumulated over that period of time. After that, the spring is uncoupled from the shaft and
allowed to unwind at a fixed rate, the time for it to unwind back to a relaxed state measured
by a timer device. The amount of time it takes for the spring to unwind at that fixed rate will
be directly proportional to the speed at which it was wound (input signal magnitude) during
the fixedtime portion of the cycle.
This technique of analogtodigital conversion escapes the calibration drift problem of the
singleslope ADC because both the integrator’s integration coefficient (or ”gain”) and the counter’s
rate of speed are in effect during the entire ”winding” and ”unwinding” cycle portions. If the
counter’s clock speed were to suddenly increase, this would shorten the fixed time period where
the integrator ”winds up” (resulting in a lesser voltage accumulated by the integrator), but it
would also mean that it would count faster during the period of time when the integrator was
allowed to ”unwind” at a fixed rate. The proportion that the counter is counting faster will
be the same proportion as the integrator’s accumulated voltage is diminished from before the
clock speed change. Thus, the clock speed error would cancel itself out and the digital output
would be exactly what it should be.
Another important advantage of this method is that the input signal becomes averaged as
it drives the integrator during the fixedtime portion of the cycle. Any changes in the analog
signal during that period of time have a cumulative effect on the digital output at the end of
that cycle. Other ADC strategies merely ”capture” the analog signal level at a single point in
time every cycle. If the analog signal is ”noisy” (contains significant levels of spurious voltage
spikes/dips), one of the other ADC converter technologies may occasionally convert a spike or
dip because it captures the signal repeatedly at a single point in time. A dualslope ADC, on
the other hand, averages together all the spikes and dips within the integration period, thus
providing an output with greater noise immunity. Dualslope ADCs are used in applications
13.9. DELTASIGMA (∆Σ) ADC 425
demanding high accuracy.
13.9 DeltaSigma (∆Σ) ADC
One of the more advanced ADC technologies is the socalled deltasigma, or ∆Σ (using the
proper Greek letter notation). In mathematics and physics, the capital Greek letter delta (∆)
represents difference or change, while the capital letter sigma (Σ) represents summation: the
adding of multiple terms together. Sometimes this converter is referred to by the same Greek
letters in reverse order: sigmadelta, or Σ∆.
In a ∆Σ converter, the analog input voltage signal is connected to the input of an integrator,
producing a voltage rateofchange, or slope, at the output corresponding to input magnitude.
This ramping voltage is then compared against ground potential (0 volts) by a comparator. The
comparator acts as a sort of 1bit ADC, producing 1 bit of output (”high” or ”low”) depending on
whether the integrator output is positive or negative. The comparator’s output is then latched
through a Dtype flipflop clocked at a high frequency, and fed back to another input channel
on the integrator, to drive the integrator in the direction of a 0 volt output. The basic circuit
looks like this:
−
+
−
+
R
R
C
Vin
+V
V
+V
V
D
C
Q
Q
−
+
+V
V
+V
Output
+V
The leftmost opamp is the (summing) integrator. The next opamp the integrator feeds into
is the comparator, or 1bit ADC. Next comes the Dtype flipflop, which latches the comparator’s
output at every clock pulse, sending either a ”high” or ”low” signal to the next comparator at
the top of the circuit. This final comparator is necessary to convert the singlepolarity 0V /
5V logic level output voltage of the flipflop into a +V / V voltage signal to be fed back to the
integrator.
If the integrator output is positive, the first comparator will output a ”high” signal to the
D input of the flipflop. At the next clock pulse, this ”high” signal will be output from the Q
line into the noninverting input of the last comparator. This last comparator, seeing an input
voltage greater than the threshold voltage of 1/2 +V, saturates in a positive direction, sending
426 CHAPTER 13. DIGITALANALOG CONVERSION
a full +V signal to the other input of the integrator. This +V feedback signal tends to drive
the integrator output in a negative direction. If that output voltage ever becomes negative, the
feedback loop will send a corrective signal (V) back around to the top input of the integrator to
drive it in a positive direction. This is the deltasigma concept in action: the first comparator
senses a difference (∆) between the integrator output and zero volts. The integrator sums (Σ)
the comparator’s output with the analog input signal.
Functionally, this results in a serial stream of bits output by the flipflop. If the analog input
is zero volts, the integrator will have no tendency to ramp either positive or negative, except in
response to the feedback voltage. In this scenario, the flipflop output will continually oscillate
between ”high” and ”low,” as the feedback system ”hunts” back and forth, trying to maintain
the integrator output at zero volts:
Flipflop output
Integrator output
0 1 0 1 0 1 0 1 0 1 0 1 0 1
∆Σ converter operation with
0 volt analog input
If, however, we apply a negative analog input voltage, the integrator will have a tendency
to ramp its output in a positive direction. Feedback can only add to the integrator’s ramping
by a fixed voltage over a fixed time, and so the bit stream output by the flipflop will not be
quite the same:
Flipflop output
Integrator output
0 1 0 1 0 1 0 1 0 1 0 1
∆Σ converter operation with
1
small negative analog input
0
By applying a larger (negative) analog input signal to the integrator, we force its output to
ramp more steeply in the positive direction. Thus, the feedback system has to output more 1’s
than before to bring the integrator output back to zero volts:
13.10. PRACTICAL CONSIDERATIONS OF ADC CIRCUITS 427
Flipflop output
Integrator output
0 1 0 1 0 1
∆Σ converter operation with
1 0 1 1 0 1 1 0
medium negative analog input
As the analog input signal increases in magnitude, so does the occurrence of 1’s in the
digital output of the flipflop:
Flipflop output
Integrator output
0 1 0 1 0 1
∆Σ converter operation with
1 1 1 0 1 101
large negative analog input
A parallel binary number output is obtained from this circuit by averaging the serial stream
of bits together. For example, a counter circuit could be designed to collect the total number of
1’s output by the flipflop in a given number of clock pulses. This count would then be indicative
of the analog input voltage.
Variations on this theme exist, employing multiple integrator stages and/or comparator
circuits outputting more than 1 bit, but one concept common to all ∆Σ converters is that of
oversampling. Oversampling is when multiple samples of an analog signal are taken by an
ADC (in this case, a 1bit ADC), and those digitized samples are averaged. The end result
is an effective increase in the number of bits resolved from the signal. In other words, an
oversampled 1bit ADC can do the same job as an 8bit ADC with onetime sampling, albeit at
a slower rate.
13.10 Practical considerations of ADC circuits
Perhaps the most important consideration of an ADC is its resolution. Resolution is the number
of binary bits output by the converter. Because ADC circuits take in an analog signal, which
is continuously variable, and resolve it into one of many discrete steps, it is important to know
how many of these steps there are in total.
428 CHAPTER 13. DIGITALANALOG CONVERSION
For example, an ADC with a 10bit output can represent up to 1024 (210) unique conditions
of signal measurement. Over the range of measurement from 0% to 100%, there will be exactly
1024 unique binary numbers output by the converter (from 0000000000 to 1111111111, inclu
sive). An 11bit ADC will have twice as many states to its output (2048, or 211), representing
twice as many unique conditions of signal measurement between 0% and 100%.
Resolution is very important in data acquisition systems (circuits designed to interpret and
record physical measurements in electronic form). Suppose we were measuring the height of
water in a 40foot tall storage tank using an instrument with a 10bit ADC. 0 feet of water in
the tank corresponds to 0% of measurement, while 40 feet of water in the tank corresponds
to 100% of measurement. Because the ADC is fixed at 10 bits of binary data output, it will
interpret any given tank level as one out of 1024 possible states. To determine how much
physical water level will be represented in each step of the ADC, we need to divide the 40 feet
of measurement span by the number of steps in the 0to1024 range of possibilities, which is
1023 (one less than 1024). Doing this, we obtain a figure of 0.039101 feet per step. This equates
to 0.46921 inches per step, a little less than half an inch of water level represented for every
binary count of the ADC.
Water
tank
LT ADC 10bitoutput0 ft
40 ft
20 ft
10 ft
30 ft
00000000002 = 0 feet of water level
11111111112 = 40 feet of water level
00000000012 = 0.039101 feet of water level
00000000102 = 0.07820 feet of water level
.
.
.
1 step
Binary output: Equivalent measurement:
1024 states
Level "transmitter"
AtoD converter
This step value of 0.039101 feet (0.46921 inches) represents the smallest amount of tank
level change detectable by the instrument. Admittedly, this is a small amount, less than 0.1%
of the overall measurement span of 40 feet. However, for some applications it may not be
fine enough. Suppose we needed this instrument to be able to indicate tank level changes
down to onetenth of an inch. In order to achieve this degree of resolution and still maintain a
measurement span of 40 feet, we would need an instrument with more than ten ADC bits.
To determine how many ADC bits are necessary, we need to first determine how many 1/10
inch steps there are in 40 feet. The answer to this is 40/(0.1/12), or 4800 1/10 inch steps in 40
feet. Thus, we need enough bits to provide at least 4800 discrete steps in a binary counting
sequence. 10 bits gave us 1023 steps, and we knew this by calculating 2 to the power of 10 (210
= 1024) and then subtracting one. Following the same mathematical procedure, 2111 = 2047,
13.10. PRACTICAL CONSIDERATIONS OF ADC CIRCUITS 429
2121 = 4095, and 2131 = 8191. 12 bits falls shy of the amount needed for 4800 steps, while 13
bits is more than enough. Therefore, we need an instrument with at least 13 bits of resolution.
Another important consideration of ADC circuitry is its sample frequency, or conversion
rate. This is simply the speed at which the converter outputs a new binary number. Like
resolution, this consideration is linked to the specific application of the ADC. If the converter
is being used to measure slowchanging signals such as level in a water storage tank, it could
probably have a very slow sample frequency and still perform adequately. Conversely, if it is
being used to digitize an audio frequency signal cycling at several thousand times per second,
the converter needs to be considerably faster.
Consider the following illustration of ADC conversion rate versus signal type, typical of a
successiveapproximation ADC with regular sample intervals:
Time
Sample timesAnalog
input
Time
Digital
output
Here, for this slowchanging signal, the sample rate is more than adequate to capture its
general trend. But consider this example with the same sample time:
Time
Sample timesAnalog
input
Time
Digital
output
When the sample period is too long (too slow), substantial details of the analog signal will
be missed. Notice how, especially in the latter portions of the analog signal, the digital output
430 CHAPTER 13. DIGITALANALOG CONVERSION
utterly fails to reproduce the true shape. Even in the first section of the analog waveform, the
digital reproduction deviates substantially from the true shape of the wave.
It is imperative that an ADC’s sample time is fast enough to capture essential changes in
the analog waveform. In data acquisition terminology, the highestfrequency waveform that
an ADC can theoretically capture is the socalled Nyquist frequency, equal to onehalf of the
ADC’s sample frequency. Therefore, if an ADC circuit has a sample frequency of 5000 Hz, the
highestfrequency waveform it can successfully resolve will be the Nyquist frequency of 2500
Hz.
If an ADC is subjected to an analog input signal whose frequency exceeds the Nyquist
frequency for that ADC, the converter will output a digitized signal of falsely low frequency.
This phenomenon is known as aliasing. Observe the following illustration to see how aliasing
occurs:
Time
Sample timesAnalog
input
Time
Digital
output
Aliasing
Note how the period of the output waveform is much longer (slower) than that of the input
waveform, and how the two waveform shapes aren’t even similar:
Analog
input
Digital
output
Period
Period
13.10. PRACTICAL CONSIDERATIONS OF ADC CIRCUITS 431
It should be understood that the Nyquist frequency is an absolutemaximum frequency limit
for an ADC, and does not represent the highest practical frequency measurable. To be safe,
one shouldn’t expect an ADC to successfully resolve any frequency greater than onefifth to
onetenth of its sample frequency.
A practical means of preventing aliasing is to place a lowpass filter before the input of the
ADC, to block any signal frequencies greater than the practical limit. This way, the ADC cir
cuitry will be prevented from seeing any excessive frequencies and thus will not try to digitize
them. It is generally considered better that such frequencies go unconverted than to have them
be ”aliased” and appear in the output as false signals.
Yet another measure of ADC performance is something called step recovery. This is a mea
sure of how quickly an ADC changes its output to match a large, sudden change in the analog
input. In some converter technologies especially, step recovery is a serious limitation. One
example is the tracking converter, which has a typically fast update period but a dispropor
tionately slow step recovery.
An ideal ADC has a great many bits for very fine resolution, samples at lightningfast
speeds, and recovers from steps instantly. It also, unfortunately, doesn’t exist in the real world.
Of course, any of these traits may be improved through additional circuit complexity, either in
terms of increased component count and/or special circuit designs made to run at higher clock
speeds. Different ADC technologies, though, have different strengths. Here is a summary of
them ranked from best to worst:
Resolution/complexity ratio:
Singleslope integrating, dualslope integrating, counter, tracking, successive approxima
tion, flash.
Speed:
Flash, tracking, successive approximation, singleslope integrating & counter, dualslope
integrating.
Step recovery:
Flash, successiveapproximation, singleslope integrating & counter, dualslope integrating,
tracking.
Please bear in mind that the rankings of these different ADC technologies depend on other
factors. For instance, how an ADC rates on step recovery depends on the nature of the step
change. A tracking ADC is equally slow to respond to all step changes, whereas a singleslope
or counter ADC will register a hightolow step change quicker than a lowtohigh step change.
Successiveapproximation ADCs are almost equally fast at resolving any analog signal, but a
tracking ADC will consistently beat a successiveapproximation ADC if the signal is changing
slower than one resolution step per clock pulse. I ranked integrating converters as having a
greater resolution/complexity ratio than counter converters, but this assumes that precision
analog integrator circuits are less complex to design and manufacture than precision DACs
required within counterbased converters. Others may not agree with this assumption.
432 CHAPTER 13. DIGITALANALOG CONVERSION
Chapter 14
DIGITAL COMMUNICATION
Contents
14.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
14.2 Networks and busses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437
14.2.1 Shortdistance busses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 439
14.2.2 Extendeddistance networks . . . . . . . . . . . . . . . . . . . . . . . . . . 440
14.3 Data flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 441
14.4 Electrical signal types . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 442
14.5 Optical data communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . 446
14.6 Network topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
14.6.1 Pointtopoint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.6.2 Bus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.6.3 Star . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.6.4 Ring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.7 Network protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 450
14.8 Practical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 453
14.1 Introduction
In the design of large and complex digital systems, it is often necessary to have one device com
municate digital information to and from other devices. One advantage of digital information
is that it tends to be far more resistant to transmitted and interpreted errors than information
symbolized in an analog medium. This accounts for the clarity of digitallyencoded telephone
connections, compact audio disks, and for much of the enthusiasm in the engineering com
munity for digital communications technology. However, digital communication has its own
unique pitfalls, and there are multitudes of different and incompatible ways in which it can be
sent. Hopefully, this chapter will enlighten you as to the basics of digital communication, its
advantages, disadvantages, and practical considerations.
433
434 CHAPTER 14. DIGITAL COMMUNICATION
Suppose we are given the task of remotely monitoring the level of a water storage tank. Our
job is to design a system to measure the level of water in the tank and send this information
to a distant location so that other people may monitor it. Measuring the tank’s level is quite
easy, and can be accomplished with a number of different types of instruments, such as float
switches, pressure transmitters, ultrasonic level detectors, capacitance probes, strain gauges,
or radar level detectors.
For the sake of this illustration, we will use an analog levelmeasuring device with an
output signal of 420 mA. 4 mA represents a tank level of 0%, 20 mA represents a tank level of
100%, and anything in between 4 and 20 mA represents a tank level proportionately between
0% and 100%. If we wanted to, we could simply send this 420 milliamp analog current signal
to the remote monitoring location by means of a pair of copper wires, where it would drive a
panel meter of some sort, the scale of which was calibrated to reflect the depth of water in the
tank, in whatever units of measurement preferred.
Panel
meter
24 VDC
"Transmitter"
Water
tank
Analog tanklevel measurement "loop"
This analog communication system would be simple and robust. For many applications,
it would suffice for our needs perfectly. But, it is not the only way to get the job done. For
the purposes of exploring digital techniques, we’ll explore other methods of monitoring this
hypothetical tank, even though the analog method just described might be the most practical.
The analog system, as simple as it may be, does have its limitations. One of them is the
problem of analog signal interference. Since the tank’s water level is symbolized by the mag
nitude of DC current in the circuit, any ”noise” in this signal will be interpreted as a change in
the water level. With no noise, a plot of the current signal over time for a steady tank level of
50% would look like this:
12 mA
0 mA
Time
Plot of signal at 50% tank level
14.1. INTRODUCTION 435
If the wires of this circuit are arranged too close to wires carrying 60 Hz AC power, for
example, inductive and capacitive coupling may create a false ”noise” signal to be introduced
into this otherwise DC circuit. Although the low impedance of a 420 mA loop (250 Ω, typi
cally) means that small noise voltages are significantly loaded (and thereby attenuated by the
inefficiency of the capacitive/inductive coupling formed by the power wires), such noise can be
significant enough to cause measurement problems:
12 mA
0 mA
Time
Plot of signal at 50% tank level
(with 60 Hz interference)
The above example is a bit exaggerated, but the concept should be clear: any electrical noise
introduced into an analog measurement system will be interpreted as changes in the measured
quantity. One way to combat this problem is to symbolize the tank’s water level by means of a
digital signal instead of an analog signal. We can do this really crudely by replacing the analog
transmitter device with a set of water level switches mounted at different heights on the tank:
Water
tank
Tank level measurement with switches
L1 L2
Each of these switches is wired to close a circuit, sending current to individual lamps
mounted on a panel at the monitoring location. As each switch closed, its respective lamp
would light, and whoever looked at the panel would see a 5lamp representation of the tank’s
level.
Being that each lamp circuit is digital in nature – either 100% on or 100% off – electrical
interference from other wires along the run have much less effect on the accuracy of mea
surement at the monitoring end than in the case of the analog signal. A huge amount of
interference would be required to cause an ”off” signal to be interpreted as an ”on” signal, or
vice versa. Relative resistance to electrical interference is an advantage enjoyed by all forms
of digital communication over analog.
436 CHAPTER 14. DIGITAL COMMUNICATION
Now that we know digital signals are far more resistant to error induced by ”noise,” let’s
improve on this tank level measurement system. For instance, we could increase the resolution
of this tank gauging system by adding more switches, for more precise determination of water
level. Suppose we install 16 switches along the tank’s height instead of five. This would
significantly improve our measurement resolution, but at the expense of greatly increasing the
quantity of wires needing to be strung between the tank and the monitoring location. One way
to reduce this wiring expense would be to use a priority encoder to take the 16 switches and
generate a binary number which represented the same information:
16line
to
4line
priority
encoder
.
.
.
.
.
.
.
.
.
.
.
.
Q0Q1Q2Q3
Q0Q1Q2Q3
0 0 0 0
0 0 0 1
0 0 1 0
0 0 1 1
0 1 0 0
0 1 0 1
0 1 1 0
0 1 1 1
1 0 0 0
1 0 0 1
1 0 1 0
1 0 1 1
1 1 0 0
1 1 0 1
1 1 1 0
1 1 1 1
Switch 0
Switch 15
Now, only 4 wires (plus any ground and power wires necessary) are needed to communicate
the information, as opposed to 16 wires (plus any ground and power wires). At the monitoring
location, we would need some kind of display device that could accept the 4bit binary data and
generate an easytoread display for a person to view. A decoder, wired to accept the 4bit data
as its input and light 1of16 output lamps, could be used for this task, or we could use a 4bit
decoder/driver circuit to drive some kind of numerical digit display.
Still, a resolution of 1/16 tank height may not be good enough for our application. To better
resolve the water level, we need more bits in our binary output. We could add still more
switches, but this gets impractical rather quickly. A better option would be to reattach our
original analog transmitter to the tank and electronically convert its 420 milliamp analog
output into a binary number with far more bits than would be practical using a set of discrete
level switches. Since the electrical noise we’re trying to avoid is encountered along the long run
of wire from the tank to the monitoring location, this A/D conversion can take place at the tank
(where we have a ”clean” 420 mA signal). There are a variety of methods to convert an analog
signal to digital, but we’ll skip an indepth discussion of those techniques and concentrate on
the digital signal communication itself.
The type of digital information being sent from our tank instrumentation to the monitoring
instrumentation is referred to as parallel digital data. That is, each binary bit is being sent
along its own dedicated wire, so that all bits arrive at their destination simultaneously. This
obviously necessitates the use of at least one wire per bit to communicate with the monitoring
location. We could further reduce our wiring needs by sending the binary data along a single
channel (one wire + ground), so that each bit is communicated one at a time. This type of
14.2. NETWORKS AND BUSSES 437
information is referred to as serial digital data.
We could use a multiplexer or a shift register to take the parallel data from the A/D con
verter (at the tank transmitter), and convert it to serial data. At the receiving end (the mon
itoring location) we could use a demultiplexer or another shift register to convert the serial
data to parallel again for use in the display circuitry. The exact details of how the mux/demux
or shift register pairs are maintained in synchronization is, like A/D conversion, a topic for
another lesson. Fortunately, there are digital IC chips called UARTs (Universal Asynchronous
ReceiverTransmitters) that handle all these details on their own and make the designer’s life
much simpler. For now, we must continue to focus our attention on the matter at hand: how to
communicate the digital information from the tank to the monitoring location.
14.2 Networks and busses
This collection of wires that I keep referring to between the tank and the monitoring location
can be called a bus or a network. The distinction between these two terms is more semantic
than technical, and the two may be used interchangeably for all practical purposes. In my
experience, the term ”bus” is usually used in reference to a set of wires connecting digital
components within the enclosure of a computer device, and ”network” is for something that is
physically more widespread. In recent years, however, the word ”bus” has gained popularity in
describing networks that specialize in interconnecting discrete instrumentation sensors over
long distances (”Fieldbus” and ”Profibus” are two examples). In either case, we are making
reference to the means by which two or more digital devices are connected together so that
data can be communicated between them.
Names like ”Fieldbus” or ”Profibus” encompass not only the physical wiring of the bus or
network, but also the specified voltage levels for communication, their timing sequences (espe
cially for serial data transmission), connector pinout specifications, and all other distinguishing
technical features of the network. In other words, when we speak of a certain type of bus or
network by name, we’re actually speaking of a communications standard, roughly analogous
to the rules and vocabulary of a written language. For example, before two or more people can
become penpals, they must be able to write to one another in a common format. To merely
have a mail system that is able to deliver their letters to each other is not enough. If they
agree to write to each other in French, they agree to hold to the conventions of character set,
vocabulary, spelling, and grammar that is specified by the standard of the French language.
Likewise, if we connect two Profibus devices together, they will be able to communicate with
each other only because the Profibus standard has specified such important details as voltage
levels, timing sequences, etc. Simply having a set of wires strung between multiple devices
is not enough to construct a working system (especially if the devices were built by different
manufacturers!).
To illustrate in detail, let’s design our own bus standard. Taking the crude water tank
measurement system with five switches to detect varying levels of water, and using (at least)
five wires to conduct the signals to their destination, we can lay the foundation for the mighty
BogusBus:
438 CHAPTER 14. DIGITAL COMMUNICATION
Water
tank
BogusBusTM
LS
LS
LS
LS
LS
1
2
3
4
5 LS5
LS4
LS3
LS2
LS1
24 V 24 V
5 5
4 4
3 3
2 2
1 1
+V +V
V V
Lamp
5
Lamp
Lamp
Lamp
Lamp
4
3
2
1
Connector
The physical wiring for the BogusBus consists of seven wires between the transmitter de
vice (switches) and the receiver device (lamps). The transmitter consists of all components and
wiring connections to the left of the leftmost connectors (the ”–>>–” symbols). Each connector
symbol represents a complementary male and female element. The bus wiring consists of the
seven wires between the connector pairs. Finally, the receiver and all of its constituent wiring
lies to the right of the rightmost connectors. Five of the network wires (labeled 1 through 5)
carry the data while two of those wires (labeled +V and V) provide connections for DC power
supplies. There is a standard for the 7pin connector plugs, as well. The pin layout is asym
metrical to prevent ”backward” connection.
In order for manufacturers to receive the aweinspiring ”BogusBuscompliant” certification
on their products, they would have to comply with the specifications set by the designers of
BogusBus (most likely another company, which designed the bus for a specific task and ended
up marketing it for a wide variety of purposes). For instance, all devices must be able to use
the 24 Volt DC supply power of BogusBus: the switch contacts in the transmitter must be
rated for switching that DC voltage, the lamps must definitely be rated for being powered by
that voltage, and the connectors must be able to handle it all. Wiring, of course, must be
in compliance with that same standard: lamps 1 through 5, for example, must be wired to
the appropriate pins so that when LS4 of Manufacturer XYZ’s transmitter closes, lamp 4 of
Manufacturer ABC’s receiver lights up, and so on. Since both transmitter and receiver contain
DC power supplies rated at an output of 24 Volts, all transmitter/receiver combinations (from
all certified manufacturers) must have power supplies that can be safely wired in parallel.
Consider what could happen if Manufacturer XYZ made a transmitter with the negative ()
side of their 24VDC power supply attached to earth ground and Manufacturer ABC made a
receiver with the positive (+) side of their 24VDC power supply attached to earth ground. If
both earth grounds are relatively ”solid” (that is, a low resistance between them, such as might
14.2. NETWORKS AND BUSSES 439
be the case if the two grounds were made on the metal structure of an industrial building), the
two power supplies would shortcircuit each other!
BogusBus, of course, is a completely hypothetical and very impractical example of a digital
network. It has incredibly poor data resolution, requires substantial wiring to connect devices,
and communicates in only a single direction (from transmitter to receiver). It does, however,
suffice as a tutorial example of what a network is and some of the considerations associated
with network selection and operation.
There are many types of buses and networks that you might come across in your profes
sion. Each one has its own applications, advantages, and disadvantages. It is worthwhile to
associate yourself with some of the ”alphabet soup” that is used to label the various designs:
14.2.1 Shortdistance busses
PC/AT Bus used in early IBMcompatible computers to connect peripheral devices such as disk
drive and sound cards to the motherboard of the computer.
PCI Another bus used in personal computers, but not limited to IBMcompatibles. Much
faster than PC/AT. Typical data transfer rate of 100Mbytes/second (32 bit) and 200Mbytes/second
(64 bit).
PCMCIA A bus designed to connect peripherals to laptop and notebook sized personal
computers. Has a very small physical ”footprint,” but is considerably slower than other popular
PC buses.
VME A highperformance bus (codesigned by Motorola, and based on Motorola’s earlier
VersaBus standard) for constructing versatile industrial and military computers, where mul
tiple memory, peripheral, and even microprocessor cards could be plugged in to a passive
”rack” or ”card cage” to facilitate custom system designs. Typical data transfer rate of 50
Mbytes/second (64 bits wide).
VXI Actually an expansion of the VME bus, VXI (VME eXtension for Instrumentation)
includes the standard VME bus along with connectors for analog signals between cards in the
rack.
S100 Sometimes called the Altair bus, this bus standard was the product of a conference
in 1976, intended to serve as an interface to the Intel 8080 microprocessor chip. Similar in
philosophy to the VME, where multiple function cards could be plugged in to a passive ”rack,”
facilitating the construction of custom systems.
MC6800 The Motorola equivalent of the Intelcentric S100 bus, designed to interface pe
ripheral devices to the popular Motorola 6800 microprocessor chip.
STD Stands for SimpleToDesign, and is yet another passive ”rack” similar to the PC/AT
bus, and lends itself well toward designs based on IBMcompatible hardware. Designed by Pro
Log, it is 8 bits wide (parallel), accommodating relatively small (4.5 inch by 6.5 inch) circuit
cards.
Multibus I and II Another bus intended for the flexible design of custom computer sys
tems, designed by Intel. 16 bits wide (parallel).
CompactPCI An industrial adaptation of the personal computer PCI standard, designed
as a higherperformance alternative to the older VME bus. At a bus clock speed of 66 MHz,
data transfer rates are 200 Mbytes/ second (32 bit) or 400 Mbytes/sec (64 bit).
Microchannel Yet another bus, this one designed by IBM for their illfated PS/2 series of
computers, intended for the interfacing of PC motherboards to peripheral devices.
440 CHAPTER 14. DIGITAL COMMUNICATION
IDE A bus used primarily for connecting personal computer hard disk drives with the ap
propriate peripheral cards. Widely used in today’s personal computers for hard drive and
CDROM drive interfacing.
SCSI An alternative (technically superior to IDE) bus used for personal computer disk
drives. SCSI stands for Small Computer System Interface. Used in some IBMcompatible PC’s,
as well as Macintosh (Apple), and many mini and mainframe business computers. Used to
interface hard drives, CDROM drives, floppy disk drives, printers, scanners, modems, and a
host of other peripheral devices. Speeds up to 1.5 Mbytes per second for the original standard.
GPIB (IEEE 488) General Purpose Interface Bus, also known as HPIB or IEEE 488, which
was intended for the interfacing of electronic test equipment such as oscilloscopes and mul
timeters to personal computers. 8 bit wide address/data ”path” with 8 additional lines for
communications control.
Centronics parallelWidely used on personal computers for interfacing printer and plotter
devices. Sometimes used to interface with other peripheral devices, such as external ZIP (100
Mbyte floppy) disk drives and tape drives.
USB Universal Serial Bus, which is intended to interconnect many external peripheral de
vices (such as keyboards, modems, mice, etc.) to personal computers. Long used on Macintosh
PC’s, it is now being installed as new equipment on IBMcompatible machines.
FireWire (IEEE 1394) A highspeed serial network capable of operating at 100, 200, or
400 Mbps with versatile features such as ”hot swapping” (adding or removing devices with the
power on) and flexible topology. Designed for highperformance personal computer interfacing.
Bluetooth A radiobased communications network designed for office linking of computer
devices. Provisions for data security designed into this network standard.
14.2.2 Extendeddistance networks
20 mA current loop Not to be confused with the common instrumentation 420 mA analog
standard, this is a digital communications network based on interrupting a 20 mA (or some
times 60 mA) current loop to represent binary data. Although the low impedance gives good
noise immunity, it is susceptible to wiring faults (such as breaks) which would fail the entire
network.
RS232C The most common serial network used in computer systems, often used to link
peripheral devices such as printers and mice to a personal computer. Limited in speed and dis
tance (typically 45 feet and 20 kbps, although higher speeds can be run with shorter distances).
I’ve been able to run RS232 reliably at speeds in excess of 100 kbps, but this was using a cable
only 6 feet long! RS232C is often referred to simply as RS232 (no ”C”).
RS422A/RS485 Two serial networks designed to overcome some of the distance and versa
tility limitations of RS232C. Used widely in industry to link serial devices together in electri
cally ”noisy” plant environments. Much greater distance and speed limitations than RS232C,
typically over half a mile and at speeds approaching 10 Mbps.
Ethernet (IEEE 802.3) A highspeed network which links computers and some types of
peripheral devices together. ”Normal” Ethernet runs at a speed of 10 million bits/second, and
”Fast” Ethernet runs at 100 million bits/second. The slower (10 Mbps) Ethernet has been
implemented in a variety of means on copper wire (thick coax = ”10BASE5”, thin coax =
”10BASE2”, twistedpair = ”10BASET”), radio, and on optical fiber (”10BASEF”). The Fast
14.3. DATA FLOW 441
Ethernet has also been implemented on a few different means (twistedpair, 2 pair = 100BASE
TX; twistedpair, 4 pair = 100BASET4; optical fiber = 100BASEFX).
Token ring Another highspeed network linking computer devices together, using a phi
losophy of communication that is much different from Ethernet, allowing for more precise re
sponse times from individual network devices, and greater immunity to network wiring dam
age.
FDDI A very highspeed network exclusively implemented on fiberoptic cabling.
Modbus/Modbus Plus Originally implemented by the Modicon corporation, a large maker
of Programmable Logic Controllers (PLCs) for linking remote I/O (Input/Output) racks with a
PLC processor. Still quite popular.
Profibus Originally implemented by the Siemens corporation, another large maker of PLC
equipment.
Foundation Fieldbus A highperformance bus expressly designed to allow multiple pro
cess instruments (transmitters, controllers, valve positioners) to communicate with host com
puters and with each other. May ultimately displace the 420 mA analog signal as the standard
means of interconnecting process control instrumentation in the future.
14.3 Data flow
Buses and networks are designed to allow communication to occur between individual devices
that are interconnected. The flow of information, or data, between nodes can take a variety of
forms:
Transmitter Receiver
Simplex communication
With simplex communication, all data flow is unidirectional: from the designated trans
mitter to the designated receiver. BogusBus is an example of simplex communication, where
the transmitter sent information to the remote monitoring location, but no information is ever
sent back to the water tank. If all we want to do is send information oneway, then simplex is
just fine. Most applications, however, demand more:
Duplex communication
Receiver /
Transmitter Transmitter
Receiver /
With duplex communication, the flow of information is bidirectional for each device. Duplex
can be further divided into two subcategories:
442 CHAPTER 14. DIGITAL COMMUNICATION
Receiver /
Transmitter Transmitter
Receiver /
Halfduplex
Transmitter
Receiver
Receiver
Transmitter
Fullduplex
(take turns)
(simultaneous)
Halfduplex communication may be likened to two tin cans on the ends of a single taut
string: Either can may be used to transmit or receive, but not at the same time. Fullduplex
communication is more like a true telephone, where two people can talk at the same time and
hear one another simultaneously, the mouthpiece of one phone transmitting the the earpiece
of the other, and vice versa. Fullduplex is often facilitated through the use of two separate
channels or networks, with an individual set of wires for each direction of communication. It
is sometimes accomplished by means of multiplefrequency carrier waves, especially in radio
links, where one frequency is reserved for each direction of communication.
14.4 Electrical signal types
With BogusBus, our signals were very simple and straightforward: each signal wire (1 through
5) carried a single bit of digital data, 0 Volts representing ”off” and 24 Volts DC representing
”on.” Because all the bits arrived at their destination simultaneously, we would call BogusBus
a parallel network technology. If we were to improve the performance of BogusBus by adding
binary encoding (to the transmitter end) and decoding (to the receiver end), so that more steps
of resolution were available with fewer wires, it would still be a parallel network. If, however,
we were to add a paralleltoserial converter at the transmitter end and a serialtoparallel
converter at the receiver end, we would have something quite different.
It is primarily with the use of serial technology that we are forced to invent clever ways
to transmit data bits. Because serial data requires us to send all data bits through the same
wiring channel from transmitter to receiver, it necessitates a potentially high frequency signal
on the network wiring. Consider the following illustration: a modified BogusBus system is
communicating digital data in parallel, binaryencoded form. Instead of 5 discrete bits like the
original BogusBus, we’re sending 8 bits from transmitter to receiver. The A/D converter on
the transmitter side generates a new output every second. That makes for 8 bits per second
of data being sent to the receiver. For the sake of illustration, let’s say that the transmitter is
bouncing between an output of 10101010 and 10101011 every update (once per second):
14.4. ELECTRICAL SIGNAL TYPES 443
Bit 0
Bit 1
Bit 2
Bit 3
Bit 4
Bit 5
Bit 6
Bit 7
1 second
Since only the least significant bit (Bit 1) is changing, the frequency on that wire (to ground)
is only 1/2 Hertz. In fact, no matter what numbers are being generated by the A/D converter
between updates, the frequency on any wire in this modified BogusBus network cannot exceed
1/2 Hertz, because that’s how fast the A/D updates its digital output. 1/2 Hertz is pretty slow,
and should present no problems for our network wiring.
On the other hand, if we used an 8bit serial network, all data bits must appear on the
single channel in sequence. And these bits must be output by the transmitter within the
1second window of time between A/D converter updates. Therefore, the alternating digital
output of 10101010 and 10101011 (once per second) would look something like this:
1 second
Serial data
10101010
10101011
10101010
10101011
The frequency of our BogusBus signal is now approximately 4 Hertz instead of 1/2 Hertz,
an eightfold increase! While 4 Hertz is still fairly slow, and does not constitute an engineering
problem, you should be able to appreciate what might happen if we were transmitting 32 or
64 bits of data per update, along with the other bits necessary for parity checking and signal
synchronization, at an update rate of thousands of times per second! Serial data network
frequencies start to enter the radio range, and simple wires begin to act as antennas, pairs of
wires as transmission lines, with all their associated quirks due to inductive and capacitive
reactances.
What is worse, the signals that we’re trying to communicate along a serial network are of a
squarewave shape, being binary bits of information. Square waves are peculiar things, being
mathematically equivalent to an infinite series of sine waves of diminishing amplitude and
increasing frequency. A simple square wave at 10 kHz is actually ”seen” by the capacitance
and inductance of the network as a series of multiple sinewave frequencies which extend into
the hundreds of kHz at significant amplitudes. What we receive at the other end of a long
2conductor network won’t look like a clean square wave anymore, even under the best of
conditions!
444 CHAPTER 14. DIGITAL COMMUNICATION
When engineers speak of network bandwidth, they’re referring to the practical frequency
limit of a network medium. In serial communication, bandwidth is a product of data volume
(binary bits per transmitted ”word”) and data speed (”words” per second). The standard mea
sure of network bandwidth is bits per second, or bps. An obsolete unit of bandwidth known
as the baud is sometimes falsely equated with bits per second, but is actually the measure of
signal level changes per second. Many serial network standards use multiple voltage or cur
rent level changes to represent a single bit, and so for these applications bps and baud are not
equivalent.
The general BogusBus design, where all bits are voltages referenced to a common ”ground”
connection, is the worstcase situation for highfrequency square wave data communication.
Everything will work well for short distances, where inductive and capacitive effects can be
held to a minimum, but for long distances this method will surely be problematic:
Input
Signal
Output
Signal
Transmitter Receiver
signal wire
ground wire
Stray capacitance
Groundreferenced voltage signal
A robust alternative to the common ground signal method is the differential voltage method,
where each bit is represented by the difference of voltage between a groundisolated pair of
wires, instead of a voltage between one wire and a common ground. This tends to limit the
capacitive and inductive effects imposed upon each signal and the tendency for the signals to be
corrupted due to outside electrical interference, thereby significantly improving the practical
distance of a serial network:
Input
Signal
Output
Signal
Transmitter Receiver
signal wire
Differential voltage signal
signal wire
Both signal wires isolated
from ground!
diminishing effect.
Capacitance through ground
minimized due to series
The triangular amplifier symbols represent differential amplifiers, which output a voltage
signal between two wires, neither one electrically common with ground. Having eliminated any
14.4. ELECTRICAL SIGNAL TYPES 445
relation between the voltage signal and ground, the only significant capacitance imposed on
the signal voltage is that existing between the two signal wires. Capacitance between a signal
wire and a grounded conductor is of much less effect, because the capacitive path between the
two signal wires via a ground connection is two capacitances in series (from signal wire #1
to ground, then from ground to signal wire #2), and series capacitance values are always less
than any of the individual capacitances. Furthermore, any ”noise” voltage induced between
the signal wires and earth ground by an external source will be ignored, because that noise
voltage will likely be induced on both signal wires in equal measure, and the receiving amplifier
only responds to the differential voltage between the two signal wires, rather than the voltage
between any one of them and earth ground.
RS232C is a prime example of a groundreferenced serial network, while RS422A is a
prime example of a differential voltage serial network. RS232C finds popular application in
office environments where there is little electrical interference and wiring distances are short.
RS422A is more widely used in industrial applications where longer wiring distances and
greater potential for electrical interference from AC power wiring exists.
However, a large part of the problem with digital network signals is the squarewave nature
of such voltages, as was previously mentioned. If only we could avoid square waves all together,
we could avoid many of their inherent difficulties in long, highfrequency networks. One way of
doing this is to modulate a sine wave voltage signal with our digital data. ”Modulation” means
that magnitude of one signal has control over some aspect of another signal. Radio technology
has incorporated modulation for decades now, in allowing an audiofrequency voltage signal
to control either the amplitude (AM) or frequency (FM) of a much higher frequency ”carrier”
voltage, which is then send to the antenna for transmission. The frequencymodulation (FM)
technique has found more use in digital networks than amplitudemodulation (AM), except
that its referred to as Frequency Shift Keying (FSK). With simple FSK, sine waves of two
distinct frequencies are used to represent the two binary states, 1 and 0:
0 (low) 1 (high) 0 (low)
Due to the practical problems of getting the low/high frequency sine waves to begin and end
at the zero crossover points for any given combination of 0’s and 1’s, a variation of FSK called
phasecontinuous FSK is sometimes used, where the consecutive combination of a low/high
frequency represents one binary state and the combination of a high/low frequency represents
the other. This also makes for a situation where each bit, whether it be 0 or 1, takes exactly
the same amount of time to transmit along the network:
0 (low) 1 (high)
With sine wave signal voltages, many of the problems encountered with square wave dig
ital signals are minimized, although the circuitry required to modulate (and demodulate) the
network signals is more complex and expensive.
446 CHAPTER 14. DIGITAL COMMUNICATION
14.5 Optical data communication
A modern alternative to sending (binary) digital information via electric voltage signals is to
use optical (light) signals. Electrical signals from digital circuits (high/low voltages) may be
converted into discrete optical signals (light or no light) with LEDs or solidstate lasers. Like
wise, light signals can be translated back into electrical form through the use of photodiodes
or phototransistors for introduction into the inputs of gate circuits.
Transmitter Receiver
Light pulses
Transmitting digital information in optical form may be done in open air, simply by aiming
a laser at a photodetector at a remote distance, but interference with the beam in the form
of temperature inversion layers, dust, rain, fog, and other obstructions can present significant
engineering problems:
Transmitter Receiver
Interference
One way to avoid the problems of openair optical data transmission is to send the light
pulses down an ultrapure glass fiber. Glass fibers will ”conduct” a beam of light much as a
copper wire will conduct electrons, with the advantage of completely avoiding all the associ
ated problems of inductance, capacitance, and external interference plaguing electrical signals.
Optical fibers keep the light beam contained within the fiber core by a phenomenon known as
total internal reflectance.
An optical fiber is composed of two layers of ultrapure glass, each layer made of glass
with a slightly different refractive index, or capacity to ”bend” light. With one type of glass
concentrically layered around a central glass core, light introduced into the central core cannot
escape outside the fiber, but is confined to travel within the core:
14.5. OPTICAL DATA COMMUNICATION 447
Cladding
Cladding
Core
Light
These layers of glass are very thin, the outer ”cladding” typically 125 microns (1 micron =
1 millionth of a meter, or 10−6 meter) in diameter. This thinness gives the fiber considerable
flexibility. To protect the fiber from physical damage, it is usually given a thin plastic coating,
placed inside of a plastic tube, wrapped with kevlar fibers for tensile strength, and given an
outer sheath of plastic similar to electrical wire insulation. Like electrical wires, optical fibers
are often bundled together within the same sheath to form a single cable.
Optical fibers exceed the datahandling performance of copper wire in almost every regard.
They are totally immune to electromagnetic interference and have very high bandwidths. How
ever, they are not without certain weaknesses.
One weakness of optical fiber is a phenomenon known as microbending. This is where the
fiber is bend around too small of a radius, causing light to escape the inner core, through the
cladding:
Escaping
light
Reflected
light
Sharp
bend
Microbending
Not only does microbending lead to diminished signal strength due to the lost light, but it
also constitutes a security weakness in that a light sensor intentionally placed on the outside
of a sharp bend could intercept digital data transmitted over the fiber.
448 CHAPTER 14. DIGITAL COMMUNICATION
Another problem unique to optical fiber is signal distortion due to multiple light paths,
or modes, having different distances over the length of the fiber. When light is emitted by a
source, the photons (light particles) do not all travel the exact same path. This fact is patently
obvious in any source of light not conforming to a straight beam, but is true even in devices
such as lasers. If the optical fiber core is large enough in diameter, it will support multiple
pathways for photons to travel, each of these pathways having a slightly different length from
one end of the fiber to the other. This type of optical fiber is called multimode fiber:
Light
"Modes" of light traveling in a fiber
A light pulse emitted by the LED taking a shorter path through the fiber will arrive at the
detector sooner than light pulses taking longer paths. The result is distortion of the square
wave’s rising and falling edges, called pulse stretching. This problem becomes worse as the
overall fiber length is increased:
Transmitted
pulse
Received
pulse
"Pulsestretching" in optical fiber
However, if the fiber core is made small enough (around 5 microns in diameter), light modes
are restricted to a single pathway with one length. Fiber so designed to permit only a single
mode of light is known as singlemode fiber. Because singlemode fiber escapes the problem of
pulse stretching experienced in long cables, it is the fiber of choice for longdistance (several
miles or more) networks. The drawback, of course, is that with only one mode of light, single
mode fibers do not conduct as as much light as multimode fibers. Over long distances, this
exacerbates the need for ”repeater” units to boost light power.
14.6 Network topology
If we want to connect two digital devices with a network, we would have a kind of network
known as ”pointtopoint:”
14.6. NETWORK TOPOLOGY 449
Device
1
Device
2
Network
PointtoPoint topology
For the sake of simplicity, the network wiring is symbolized as a single line between the two
devices. In actuality, it may be a twisted pair of wires, a coaxial cable, an optical fiber, or even
a sevenconductor BogusBus. Right now, we’re merely focusing on the ”shape” of the network,
technically known as its topology.
If we want to include more devices (sometimes called nodes) on this network, we have sev
eral options of network configuration to choose from:
Device
1
Device
2
Bus topology
Device Device
3 4
Device
1
Device
2
Device Device
3 4
HubStar topology
Device
1
Device
2
Device Device
3 4
Ring topology
Many network standards dictate the type of topology which is used, while others are more
versatile. Ethernet, for example, is commonly implemented in a ”bus” topology but can also
be implemented in a ”star” or ”ring” topology with the appropriate interconnecting equipment.
Other networks, such as RS232C, are almost exclusively pointtopoint; and token ring (as you
might have guessed) is implemented solely in a ring topology.
Different topologies have different pros and cons associated with them:
450 CHAPTER 14. DIGITAL COMMUNICATION
14.6.1 Pointtopoint
Quite obviously the only choice for two nodes.
14.6.2 Bus
Very simple to install and maintain. Nodes can be easily added or removed with minimal
wiring changes. On the other hand, the one bus network must handle all communication
signals from all nodes. This is known as broadcast networking, and is analogous to a group of
people talking to each other over a single telephone connection, where only one person can talk
at a time (limiting data exchange rates), and everyone can hear everyone else when they talk
(which can be a data security issue). Also, a break in the bus wiring can lead to nodes being
isolated in groups.
14.6.3 Star
With devices known as ”gateways” at branching points in the network, data flow can be re
stricted between nodes, allowing for private communication between specific groups of nodes.
This addresses some of the speed and security issues of the simple bus topology. However,
those branches could easily be cut off from the rest of the ”star” network if one of the gateways
were to fail. Can also be implemented with ”switches” to connect individual nodes to a larger
network on demand. Such a switched network is similar to the standard telephone system.
14.6.4 Ring
This topology provides the best reliability with the least amount of wiring. Since each node
has two connection points to the ring, a single break in any part of the ring doesn’t affect the
integrity of the network. The devices, however, must be designed with this topology in mind.
Also, the network must be interrupted to install or remove nodes. As with bus topology, ring
networks are broadcast by nature.
As you might suspect, two or more ring topologies may be combined to give the ”best of both
worlds” in a particular application. Quite often, industrial networks end up in this fashion
over time, simply from engineers and technicians joining multiple networks together for the
benefit of plantwide information access.
14.7 Network protocols
Aside from the issues of the physical network (signal types and voltage levels, connector
pinouts, cabling, topology, etc.), there needs to be a standardized way in which communica
tion is arbitrated between multiple nodes in a network, even if its as simple as a twonode,
pointtopoint system. When a node ”talks” on the network, it is generating a signal on the
network wiring, be it high and low DC voltage levels, some kind of modulated AC carrier wave
signal, or even pulses of light in a fiber. Nodes that ”listen” are simply measuring that applied
signal on the network (from the transmitting node) and passively monitoring it. If two or more
nodes ”talk” at the same time, however, their output signals may clash (imagine two logic gates
14.7. NETWORK PROTOCOLS 451
trying to apply opposite signal voltages to a single line on a bus!), corrupting the transmitted
data.
The standardized method by which nodes are allowed to transmit to the bus or network
wiring is called a protocol. There are many different protocols for arbitrating the use of a
common network between multiple nodes, and I’ll cover just a few here. However, its good to
be aware of these few, and to understand why some work better for some purposes than others.
Usually, a specific protocol is associated with a standardized type of network. This is merely
another ”layer” to the set of standards which are specified under the titles of various networks.
The International Standards Organization (ISO) has specified a general architecture of
network specifications in their DIS7498 model (applicable to most any digital network). Con
sisting of seven ”layers,” this outline attempts to categorize all levels of abstraction necessary
to communicate digital data.
• Level 1: Physical Specifies electrical and mechanical details of communication: wire
type, connector design, signal types and levels.
• Level 2: Data link Defines formats of messages, how data is to be addressed, and error
detection/correction techniques.
• Level 3: Network Establishes procedures for encapsulation of data into ”packets” for
transmission and reception.
• Level 4: Transport Among other things, the transport layer defines how complete data
files are to be handled over a network.
• Level 5: Session Organizes data transfer in terms of beginning and end of a specific
transmission. Analogous to job control on a multitasking computer operating system.
• Level 6: Presentation Includes definitions for character sets, terminal control, and
graphics commands so that abstract data can be readily encoded and decoded between
communicating devices.
• Level 7: Application The enduser standards for generating and/or interpreting com
municated data in its final form. In other words, the actual computer programs using the
communicated data.
Some established network protocols only cover one or a few of the DIS7498 levels. For ex
ample, the widely used RS232C serial communications protocol really only addresses the first
(”physical”) layer of this sevenlayer model. Other protocols, such as the Xwindows graphical
client/server system developed at MIT for distributed graphicuserinterface computer systems,
cover all seven layers.
Different protocols may use the same physical layer standard. An example of this is the
RS422A and RS485 protocols, both of which use the same differentialvoltage transmitter
and receiver circuitry, using the same voltage levels to denote binary 1’s and 0’s. On a physical
level, these two communication protocols are identical. However, on a more abstract level the
protocols are different: RS422A is pointtopoint only, while RS485 supports a bus topology
”multidrop” with up to 32 addressable nodes.
Perhaps the simplest type of protocol is the one where there is only one transmitter, and all
the other nodes are merely receivers. Such is the case for BogusBus, where a single transmitter
452 CHAPTER 14. DIGITAL COMMUNICATION
generates the voltage signals impressed on the network wiring, and one or more receiver units
(with 5 lamps each) light up in accord with the transmitter’s output. This is always the case
with a simplex network: there’s only one talker, and everyone else listens!
When we have multiple transmitting nodes, we must orchestrate their transmissions in
such a way that they don’t conflict with one another. Nodes shouldn’t be allowed to talk when
another node is talking, so we give each node the ability to ”listen” and to refrain from talking
until the network is silent. This basic approach is called Carrier Sense Multiple Access (CSMA),
and there exists a few variations on this theme. Please note that CSMA is not a standardized
protocol in itself, but rather a methodology that certain protocols follow.
One variation is to simply let any node begin to talk as soon as the network is silent. This is
analogous to a group of people meeting at a round table: anyone has the ability to start talking,
so long as they don’t interrupt anyone else. As soon as the last person stops talking, the next
person waiting to talk will begin. So, what happens when two or more people start talking at
once? In a network, the simultaneous transmission of two or more nodes is called a collision.
With CSMA/CD (CSMA/Collision Detection), the nodes that collide simply reset themselves
with a random delay timer circuit, and the first one to finish its time delay tries to talk again.
This is the basic protocol for the popular Ethernet network.
Another variation of CSMA is CSMA/BA (CSMA/Bitwise Arbitration), where colliding nodes
refer to preset priority numbers which dictate which one has permission to speak first. In
other words, each node has a ”rank” which settles any dispute over who gets to start talking
first after a collision occurs, much like a group of people where dignitaries and common citi
zens are mixed. If a collision occurs, the dignitary is generally allowed to speak first and the
common person waits afterward.
In either of the two examples above (CSMA/CD and CSMA/BA), we assumed that any node
could initiate a conversation so long as the network was silent. This is referred to as the
”unsolicited” mode of communication. There is a variation called ”solicited” mode for either
CSMA/CD or CSMA/BA where the initial transmission is only allowed to occur when a desig
nated master node requests (solicits) a reply. Collision detection (CD) or bitwise arbitration
(BA) applies only to postcollision arbitration as multiple nodes respond to the master device’s
request.
An entirely different strategy for node communication is the Master/Slave protocol, where
a single master device allots time slots for all the other nodes on the network to transmit, and
schedules these time slots so that multiple nodes cannot collide. The master device addresses
each node by name, one at a time, letting that node talk for a certain amount of time. When it
is finished, the master addresses the next node, and so on, and so on.
Yet another strategy is the TokenPassing protocol, where each node gets a turn to talk (one
at a time), and then grants permission for the next node to talk when its done. Permission to
talk is passed around from node to node as each one hands off the ”token” to the next in sequen
tial order. The token itself is not a physical thing: it is a series of binary 1’s and 0’s broadcast
on the network, carrying a specific address of the next node permitted to talk. Although token
passing protocol is often associated with ringtopology networks, it is not restricted to any
topology in particular. And when this protocol is implemented in a ring network, the sequence
of token passing does not have to follow the physical connection sequence of the ring.
Just as with topologies, multiple protocols may be joined together over different segments
of a heterogeneous network, for maximum benefit. For instance, a dedicated Master/Slave net
work connecting instruments together on the manufacturing plant floor may be linked through
14.8. PRACTICAL CONSIDERATIONS 453
a gateway device to an Ethernet network which links multiple desktop computer workstations
together, one of those computer workstations acting as a gateway to link the data to an FDDI
fiber network back to the plant’s mainframe computer. Each network type, topology, and pro
tocol serves different needs and applications best, but through gateway devices, they can all
share the same data.
It is also possible to blend multiple protocol strategies into a new hybrid within a single
network type. Such is the case for Foundation Fieldbus, which combines Master/Slave with a
form of tokenpassing. A Link Active Scheduler (LAS) device sends scheduled ”Compel Data”
(CD) commands to query slave devices on the Fieldbus for timecritical information. In this
regard, Fieldbus is a Master/Slave protocol. However, when there’s time between CD queries,
the LAS sends out ”tokens” to each of the other devices on the Fieldbus, one at a time, giving
them opportunity to transmit any unscheduled data. When those devices are done transmit
ting their information, they return the token back to the LAS. The LAS also probes for new
devices on the Fieldbus with a ”Probe Node” (PN) message, which is expected to produce a
”Probe Response” (PR) back to the LAS. The responses of devices back to the LAS, whether
by PR message or returned token, dictate their standing on a ”Live List” database which the
LAS maintains. Proper operation of the LAS device is absolutely critical to the functioning of
the Fieldbus, so there are provisions for redundant LAS operation by assigning ”Link Master”
status to some of the nodes, empowering them to become alternate Link Active Schedulers if
the operating LAS fails.
Other data communications protocols exist, but these are the most popular. I had the op
portunity to work on an old (circa 1975) industrial control system made by Honeywell where
a master device called the Highway Traffic Director, or HTD, arbitrated all network communi
cations. What made this network interesting is that the signal sent from the HTD to all slave
devices for permitting transmission was not communicated on the network wiring itself, but
rather on sets of individual twistedpair cables connecting the HTD with each slave device.
Devices on the network were then divided into two categories: those nodes connected to the
HTD which were allowed to initiate transmission, and those nodes not connected to the HTD
which could only transmit in response to a query sent by one of the former nodes. Primitive and
slow are the only fitting adjectives for this communication network scheme, but it functioned
adequately for its time.
14.8 Practical considerations
A principal consideration for industrial control networks, where the monitoring and control
of reallife processes must often occur quickly and at set times, is the guaranteed maximum
communication time from one node to another. If you’re controlling the position of a nuclear
reactor coolant valve with a digital network, you need to be able to guarantee that the valve’s
network node will receive the proper positioning signals from the control computer at the right
times. If not, very bad things could happen!
The ability for a network to guarantee data ”throughput” is called determinism. A deter
ministic network has a guaranteed maximum time delay for data transfer from node to node,
whereas a nondeterministic network does not. The preeminent example of a nondeterministic
network is Ethernet, where the nodes rely on random timedelay circuits to reset and re
attempt transmission after a collision. Being that a node’s transmission of data could be
454 CHAPTER 14. DIGITAL COMMUNICATION
delayed indefinitely from a long series of resets and retries after repeated collisions, there
is no guarantee that its data will ever get sent out to the network. Realistically though, the
odds are so astronomically great that such a thing would happen that it is of little practical
concern in a lightlyloaded network.
Another important consideration, especially for industrial control networks, is network
fault tolerance: that is, how susceptible is a particular network’s signaling, topology, and/or
protocol to failures? We’ve already briefly discussed some of the issues surrounding topology,
but protocol impacts reliability just as much. For example, a Master/Slave network, while be
ing extremely deterministic (a good thing for critical controls), is entirely dependent upon the
master node to keep everything going (generally a bad thing for critical controls). If the mas
ter node fails for any reason, none of the other nodes will be able to transmit any data at all,
because they’ll never receive their alloted time slot permissions to do so, and the whole system
will fail.
A similar issue surrounds tokenpassing systems: what happens if the node holding the
token were to fail before passing the token on to the next node? Some tokenpassing systems
address this possibility by having a few designated nodes generate a new token if the network
is silent for too long. This works fine if a node holding the token dies, but it causes problems
if part of a network falls silent because a cable connection comes undone: the portion of the
network that falls silent generates its own token after awhile, and you essentially are left with
two smaller networks with one token that’s getting passed around each of them to sustain
communication. Trouble occurs, however, if that cable connection gets plugged back in: those
two segmented networks are joined in to one again, and now there’s two tokens being passed
around one network, resulting in nodes’ transmissions colliding!
There is no ”perfect network” for all applications. The task of the engineer and technician
is to know the application and know the operations of the network(s) available. Only then can
efficient system design and maintenance become a reality.
Chapter 15
DIGITAL STORAGE (MEMORY)
Contents
15.1 Why digital? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 455
15.2 Digital memory terms and concepts . . . . . . . . . . . . . . . . . . . . . . . 456
15.3 Modern nonmechanical memory . . . . . . . . . . . . . . . . . . . . . . . . . 458
15.4 Historical, nonmechanical memory technologies . . . . . . . . . . . . . . . 460
15.5 Readonly memory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
15.6 Memory with moving parts: ”Drives” . . . . . . . . . . . . . . . . . . . . . . 467
15.1 Why digital?
Although many textbooks provide good introductions to digital memory technology, I intend
to make this chapter unique in presenting both past and present technologies to some degree
of detail. While many of these memory designs are obsolete, their foundational principles
are still quite interesting and educational, and may even find reapplication in the memory
technologies of the future.
The basic goal of digital memory is to provide a means to store and access binary data:
sequences of 1’s and 0’s. The digital storage of information holds advantages over analog tech
niques much the same as digital communication of information holds advantages over analog
communication. This is not to say that digital data storage is unequivocally superior to analog,
but it does address some of the more common problems associated with analog techniques and
thus finds immense popularity in both consumer and industrial applications. Digital data stor
age also complements digital computation technology well, and thus finds natural application
in the world of computers.
The most evident advantage of digital data storage is the resistance to corruption. Suppose
that we were going to store a piece of data regarding the magnitude of a voltage signal by
means of magnetizing a small chunk of magnetic material. Since many magnetic materials
455
456 CHAPTER 15. DIGITAL STORAGE (MEMORY)
retain their strength of magnetization very well over time, this would be a logical media candi
date for longterm storage of this particular data (in fact, this is precisely how audio and video
tape technology works: thin plastic tape is impregnated with particles of ironoxide material,
which can be magnetized or demagnetized via the application of a magnetic field from an elec
tromagnet coil. The data is then retrieved from the tape by moving the magnetized tape past
another coil of wire, the magnetized spots on the tape inducing voltage in that coil, reproducing
the voltage waveform initially used to magnetize the tape).
If we represent an analog signal by the strength of magnetization on spots of the tape, the
storage of data on the tape will be susceptible to the smallest degree of degradation of that
magnetization. As the tape ages and the magnetization fades, the analog signal magnitude
represented on the tape will appear to be less than what it was when we first recorded the
data. Also, if any spurious magnetic fields happen to alter the magnetization on the tape,
even if its only by a small amount, that altering of field strength will be interpreted upon re
play as an altering (or corruption) of the signal that was recorded. Since analog signals have
infinite resolution, the smallest degree of change will have an impact on the integrity of the
data storage.
If we were to use that same tape and store the data in binary digital form, however, the
strength of magnetization on the tape would fall into two discrete levels: ”high” and ”low,”
with no valid inbetween states. As the tape aged or was exposed to spurious magnetic fields,
those same locations on the tape would experience slight alteration of magnetic field strength,
but unless the alterations were extreme, no data corruption would occur upon replay of the
tape. By reducing the resolution of the signal impressed upon the magnetic tape, we’ve gained
significant immunity to the kind of degradation and ”noise” typically plaguing stored analog
data. On the other hand, our data resolution would be limited to the scanning rate and the
number of bits output by the A/D converter which interpreted the original analog signal, so the
reproduction wouldn’t necessarily be ”better” than with analog, merely more rugged. With the
advanced technology of modern A/D’s, though, the tradeoff is acceptable for most applications.
Also, by encoding different types of data into specific binary number schemes, digital stor
age allows us to archive a wide variety of information that is often difficult to encode in analog
form. Text, for example, is represented quite easily with the binary ASCII code, seven bits for
each character, including punctuation marks, spaces, and carriage returns. A wider range of
text is encoded using the Unicode standard, in like manner. Any kind of numerical data can be
represented using binary notation on digital media, and any kind of information that can be
encoded in numerical form (which almost any kind can!) is storable, too. Techniques such as
parity and checksum error detection can be employed to further guard against data corruption,
in ways that analog does not lend itself to.
15.2 Digital memory terms and concepts
When we store information in some kind of circuit or device, we not only need some way to
store and retrieve it, but also to locate precisely where in the device that it is. Most, if not
all, memory devices can be thought of as a series of mail boxes, folders in a file cabinet, or
some other metaphor where information can be located in a variety of places. When we refer
to the actual information being stored in the memory device, we usually refer to it as the data.
The location of this data within the storage device is typically called the address, in a manner
15.2. DIGITAL MEMORY TERMS AND CONCEPTS 457
reminiscent of the postal service.
With some types of memory devices, the address in which certain data is stored can be called
up by means of parallel data lines in a digital circuit (we’ll discuss this in more detail later in
this lesson). With other types of devices, data is addressed in terms of an actual physical
location on the surface of some type of media (the tracks and sectors of circular computer disks,
for instance). However, some memory devices such as magnetic tapes have a onedimensional
type of data addressing: if you want to play your favorite song in the middle of a cassette tape
album, you have to fastforward to that spot in the tape, arriving at the proper spot by means
of trialanderror, judging the approximate area by means of a counter that keeps track of tape
position, and/or by the amount of time it takes to get there from the beginning of the tape.
The access of data from a storage device falls roughly into two categories: random access and
sequential access. Random access means that you can quickly and precisely address a specific
data location within the device, and nonrandom simply means that you cannot. A vinyl record
platter is an example of a randomaccess device: to skip to any song, you just position the
stylus arm at whatever location on the record that you want (compact audio disks so the same
thing, only they do it automatically for you). Cassette tape, on the other hand, is sequential.
You have to wait to go past the other songs in sequence before you can access or address the
song that you want to skip to.
The process of storing a piece of data to a memory device is called writing, and the process
of retrieving data is called reading. Memory devices allowing both reading and writing are
equipped with a way to distinguish between the two tasks, so that no mistake is made by the
user (writing new information to a device when all you wanted to do is see what was stored
there). Some devices do not allow for the writing of new data, and are purchased ”prewritten”
from the manufacturer. Such is the case for vinyl records and compact audio disks, and this
is typically referred to in the digital world as readonly memory, or ROM. Cassette audio and
video tape, on the other hand, can be rerecorded (rewritten) or purchased blank and recorded
fresh by the user. This is often called readwrite memory.
Another distinction to be made for any particular memory technology is its volatility, or
data storage permanence without power. Many electronic memory devices store binary data
by means of circuits that are either latched in a ”high” or ”low” state, and this latching effect
holds only as long as electric power is maintained to those circuits. Such memory would be
properly referred to as volatile. Storage media such as magnetized disk or tape is nonvolatile,
because no source of power is needed to maintain data storage. This is often confusing for
new students of computer technology, because the volatile electronic memory typically used for
the construction of computer devices is commonly and distinctly referred to as RAM (Random
Access Memory). While RAM memory is typically randomlyaccessed, so is virtually every
other kind of memory device in the computer! What ”RAM” really refers to is the volatility of
the memory, and not its mode of access. Nonvolatile memory integrated circuits in personal
computers are commonly (and properly) referred to as ROM (ReadOnly Memory), but their
data contents are accessed randomly, just like the volatile memory circuits!
Finally, there needs to be a way to denote how much data can be stored by any particular
memory device. This, fortunately for us, is very simple and straightforward: just count up the
number of bits (or bytes, 1 byte = 8 bits) of total data storage space. Due to the high capacity of
modern data storage devices, metric prefixes are generally affixed to the unit of bytes in order
to represent storage space: 1.6 Gigabytes is equal to 1.6 billion bytes, or 12.8 billion bits, of
data storage capacity. The only caveat here is to be aware of rounded numbers. Because the
458 CHAPTER 15. DIGITAL STORAGE (MEMORY)
storage mechanisms of many randomaccess memory devices are typically arranged so that the
number of ”cells” in which bits of data can be stored appears in binary progression (powers of
2), a ”one kilobyte” memory device most likely contains 1024 (2 to the power of 10) locations for
data bytes rather than exactly 1000. A ”64 kbyte” memory device actually holds 65,536 bytes
of data (2 to the 16th power), and should probably be called a ”66 Kbyte” device to be more
precise. When we round numbers in our base10 system, we fall out of step with the round
equivalents in the base2 system.
15.3 Modern nonmechanical memory
Now we can proceed to studying specific types of digital storage devices. To start, I want to
explore some of the technologies which do not require any moving parts. These are not nec
essarily the newest technologies, as one might suspect, although they will most likely replace
movingpart technologies in the future.
A very simple type of electronic memory is the bistable multivibrator. Capable of storing
a single bit of data, it is volatile (requiring power to maintain its memory) and very fast. The
Dlatch is probably the simplest implementation of a bistable multivibrator for memory usage,
the D input serving as the data ”write” input, the Q output serving as the ”read” output, and
the enable input serving as the read/write control line:
Q
QD
E
Data write Data read
Write/Read
If we desire more than one bit’s worth of storage (and we probably do), we’ll have to have
many latches arranged in some kind of an array where we can selectively address which one
(or which set) we’re reading from or writing to. Using a pair of tristate buffers, we can connect
both the data write input and the data read output to a common data bus line, and enable
those buffers to either connect the Q output to the data line (READ), connect the D input to
the data line (WRITE), or keep both buffers in the HighZ state to disconnect D and Q from the
data line (unaddressed mode). One memory ”cell” would look like this, internally:
15.3. MODERN NONMECHANICAL MEMORY 459
Q
QD
E
Write/Read
Memory cell circuit
Enable
Address
Data
in/out
When the address enable input is 0, both tristate buffers will be placed in highZ mode, and
the latch will be disconnected from the data input/output (bus) line. Only when the address
enable input is active (1) will the latch be connected to the data bus. Every latch circuit, of
course, will be enabled with a different ”address enable” (AE) input line, which will come from
a 1ofn output decoder:
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
. . .
AE Data
W/R
4line to
16line
decoder
A0A1
A3
A2
AE
W/R
Data
Memory cell 15
Memory cell 0
.
.
.
.
. . .
. . .
. . .
. . .
. . .
1bit
data
bus
cell 14
cell 13
cell 12
cell 3
cell 2
cell 1
.
.
.
.
.
.
16 x 1 bit memory
Write/Read
In the above circuit, 16 memory cells are individually addressed with a 4bit binary code
input into the decoder. If a cell is not addressed, it will be disconnected from the 1bit data bus
460 CHAPTER 15. DIGITAL STORAGE (MEMORY)
by its internal tristate buffers: consequently, data cannot be either written or read through the
bus to or from that cell. Only the cell circuit that is addressed by the 4bit decoder input will
be accessible through the data bus.
This simple memory circuit is randomaccess and volatile. Technically, it is known as a
static RAM. Its total memory capacity is 16 bits. Since it contains 16 addresses and has a data
bus that is 1 bit wide, it would be designated as a 16 x 1 bit static RAM circuit. As you can
see, it takes an incredible number of gates (and multiple transistors per gate!) to construct a
practical static RAM circuit. This makes the static RAM a relatively lowdensity device, with
less capacity than most other types of RAM technology per unit IC chip space. Because each
cell circuit consumes a certain amount of power, the overall power consumption for a large
array of cells can be quite high. Early static RAM banks in personal computers consumed a
fair amount of power and generated a lot of heat, too. CMOS IC technology has made it possible
to lower the specific power consumption of static RAM circuits, but low storage density is still
an issue.
To address this, engineers turned to the capacitor instead of the bistable multivibrator as a
means of storing binary data. A tiny capacitor could serve as a memory cell, complete with a
single MOSFET transistor for connecting it to the data bus for charging (writing a 1), discharg
ing (writing a 0), or reading. Unfortunately, such tiny capacitors have very small capacitances,
and their charge tends to ”leak” away through any circuit impedances quite rapidly. To com
bat this tendency, engineers designed circuits internal to the RAM memory chip which would
periodically read all cells and recharge (or ”refresh”) the capacitors as needed. Although this
added to the complexity of the circuit, it still required far less componentry than a RAM built
of multivibrators. They called this type of memory circuit a dynamic RAM, because of its need
of periodic refreshing.
Recent advances in IC chip manufacturing has led to the introduction of flash memory,
which works on a capacitive storage principle like the dynamic RAM, but uses the insulated
gate of a MOSFET as the capacitor itself.
Before the advent of transistors (especially the MOSFET), engineers had to implement dig
ital circuitry with gates constructed from vacuum tubes. As you can imagine, the enormous
comparative size and power consumption of a vacuum tube as compared to a transistor made
memory circuits like static and dynamic RAM a practical impossibility. Other, rather inge
nious, techniques to store digital data without the use of moving parts were developed.
15.4 Historical, nonmechanical memory technologies
Perhaps the most ingenious technique was that of the delay line. A delay line is any kind of
device which delays the propagation of a pulse or wave signal. If you’ve ever heard a sound
echo back and forth through a canyon or cave, you’ve experienced an audio delay line: the noise
wave travels at the speed of sound, bouncing off of walls and reversing direction of travel. The
delay line ”stores” data on a very temporary basis if the signal is not strengthened periodically,
but the very fact that it stores data at all is a phenomenon exploitable for memory technology.
Early computer delay lines used long tubes filled with liquid mercury, which was used as
the physical medium through which sound waves traveled along the length of the tube. An
electrical/sound transducer was mounted at each end, one to create sound waves from electrical
impulses, and the other to generate electrical impulses from sound waves. A stream of serial
15.4. HISTORICAL, NONMECHANICAL MEMORY TECHNOLOGIES 461
binary data was sent to the transmitting transducer as a voltage signal. The sequence of sound
waves would travel from left to right through the mercury in the tube and be received by the
transducer at the other end. The receiving transducer would receive the pulses in the same
order as they were transmitted:
Mercury
−
+
−
+
Amplifier AmplifierData pulses moving at speed of sound
Data pulses moving at speed of light
Mercury tube delayline memory
A feedback circuit connected to the receiving transducer would drive the transmitting trans
ducer again, sending the same sequence of pulses through the tube as sound waves, storing the
data as long as the feedback circuit continued to function. The delay line functioned like a first
infirstout (FIFO) shift register, and external feedback turned that shift register behavior into
a ring counter, cycling the bits around indefinitely.
The delay line concept suffered numerous limitations from the materials and technology
that were then available. The EDVAC computer of the early 1950’s used 128 mercuryfilled
tubes, each one about 5 feet long and storing a maximum of 384 bits. Temperature changes
would affect the speed of sound in the mercury, thus skewing the time delay in each tube and
causing timing problems. Later designs replaced the liquid mercury medium with solid rods of
glass, quartz, or special metal that delayed torsional (twisting) waves rather than longitudinal
(lengthwise) waves, and operated at much higher frequencies.
One such delay line used a special nickelirontitanium wire (chosen for its good tempera
ture stability) about 95 feet in length, coiled to reduce the overall package size. The total delay
time from one end of the wire to the other was about 9.8 milliseconds, and the highest practical
clock frequency was 1 MHz. This meant that approximately 9800 bits of data could be stored in
the delay line wire at any given time. Given different means of delaying signals which wouldn’t
be so susceptible to environmental variables (such as serial pulses of light within a long optical
fiber), this approach might someday find reapplication.
Another approach experimented with by early computer engineers was the use of a cathode
ray tube (CRT), the type commonly used for oscilloscope, radar, and television viewscreens, to
store binary data. Normally, the focused and directed electron beam in a CRT would be used
to make bits of phosphor chemical on the inside of the tube glow, thus producing a viewable
image on the screen. In this application, however, the desired result was the creation of an
electric charge on the glass of the screen by the impact of the electron beam, which would then
be detected by a metal grid placed directly in front of the CRT. Like the delay line, the socalled
Williams Tube memory needed to be periodically refreshed with external circuitry to retain
its data. Unlike the delay line mechanisms, it was virtually immune to the environmental
factors of temperature and vibration. The IBM model 701 computer sported a Williams Tube
memory with 4 Kilobyte capacity and a bad habit of ”overcharging” bits on the tube screen with
successive rewrites so that false ”1” states might overflow to adjacent spots on the screen.
The next major advance in computer memory came when engineers turned to magnetic
462 CHAPTER 15. DIGITAL STORAGE (MEMORY)
materials as a means of storing binary data. It was discovered that certain compounds of iron,
namely ”ferrite,” possessed hysteresis curves that were almost square:
Field intensity (H)
Flux density
(B)
Hysteresis curve for ferrite
Shown on a graph with the strength of the applied magnetic field on the horizontal axis
(field intensity), and the actual magnetization (orientation of electron spins in the ferrite mate
rial) on the vertical axis (flux density), ferrite won’t become magnetized one direction until the
applied field exceeds a critical threshold value. Once that critical value is exceeded, the elec
trons in the ferrite ”snap” into magnetic alignment and the ferrite becomes magnetized. If the
applied field is then turned off, the ferrite maintains full magnetism. To magnetize the ferrite
in the other direction (polarity), the applied magnetic field must exceed the critical value in the
opposite direction. Once that critical value is exceeded, the electrons in the ferrite ”snap” into
magnetic alignment in the opposite direction. Once again, if the applied field is then turned off,
the ferrite maintains full magnetism. To put it simply, the magnetization of a piece of ferrite
is ”bistable.”
Exploiting this strange property of ferrite, we can use this natural magnetic ”latch” to store
a binary bit of data. To set or reset this ”latch,” we can use electric current through a wire
or coil to generate the necessary magnetic field, which will then be applied to the ferrite. Jay
Forrester of MIT applied this principle in inventing the magnetic ”core” memory, which became
the dominant computer memory technology during the 1970’s.
15.4. HISTORICAL, NONMECHANICAL MEMORY TECHNOLOGIES 463
. . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
Column wire drivers
Row
wire
drivers
8 x 8
magnetic
core memory
array
A grid of wires, electrically insulated from one another, crossed through the center of many
ferrite rings, each of which being called a ”core.” As DC current moved through any wire from
the power supply to ground, a circular magnetic field was generated around that energized
wire. The resistor values were set so that the amount of current at the regulated power supply
voltage would produce slightly more than 1/2 the critical magnetic field strength needed to
magnetize any one of the ferrite rings. Therefore, if column #4 wire was energized, all the
cores on that column would be subjected to the magnetic field from that one wire, but it would
not be strong enough to change the magnetization of any of those cores. However, if column #4
wire and row #5 wire were both energized, the core at that intersection of column #4 and row
#5 would be subjected to a sum of those two magnetic fields: a magnitude strong enough to
”set” or ”reset” the magnetization of that core. In other words, each core was addressed by the
intersection of row and column. The distinction between ”set” and ”reset” was the direction of
the core’s magnetic polarity, and that bit value of data would be determined by the polarity of
the voltages (with respect to ground) that the row and column wires would be energized with.
The following photograph shows a core memory board from a Data General brand, ”Nova”
model computer, circa late 1960’s or early 1970’s. It had a total storage capacity of 4 kbytes
(that’s kilobytes, not megabytes!). A ballpoint pen is shown for size comparison:
464 CHAPTER 15. DIGITAL STORAGE (MEMORY)
The electronic components seen around the periphery of this board are used for ”driving”
the column and row wires with current, and also to read the status of a core. A closeup
photograph reveals the ringshaped cores, through which the matrix wires thread. Again, a
ballpoint pen is shown for size comparison:
A core memory board of later design (circa 1971) is shown in the next photograph. Its cores
are much smaller and more densely packed, giving more memory storage capacity than the
former board (8 kbytes instead of 4 kbytes):
15.4. HISTORICAL, NONMECHANICAL MEMORY TECHNOLOGIES 465
And, another closeup of the cores:
Writing data to core memory was easy enough, but reading that data was a bit of a trick. To
facilitate this essential function, a ”read” wire was threaded through all the cores in a memory
matrix, one end of it being grounded and the other end connected to an amplifier circuit. A
pulse of voltage would be generated on this ”read” wire if the addressed core changed states
(from 0 to 1, or 1 to 0). In other words, to read a core’s value, you had to write either a 1 or
a 0 to that core and monitor the voltage induced on the read wire to see if the core changed.
Obviously, if the core’s state was changed, you would have to reset it back to its original state,
or else the data would have been lost. This process is known as a destructive read, because
data may be changed (destroyed) as it is read. Thus, refreshing is necessary with core memory,
although not in every case (that is, in the case of the core’s state not changing when either a 1
or a 0 was written to it).
One major advantage of core memory over delay lines and Williams Tubes was nonvolatil
ity. The ferrite cores maintained their magnetization indefinitely, with no power or refreshing
required. It was also relatively easy to build, denser, and physically more rugged than any of
466 CHAPTER 15. DIGITAL STORAGE (MEMORY)
its predecessors. Core memory was used from the 1960’s until the late 1970’s in many computer
systems, including the computers used for the Apollo space program, CNC machine tool control
computers, business (”mainframe”) computers, and industrial control systems. Despite the fact
that core memory is long obsolete, the term ”core” is still used sometimes with reference to a
computer’s RAM memory.
All the while that delay lines, Williams Tube, and core memory technologies were being
invented, the simple static RAM was being improved with smaller active component (vacuum
tube or transistor) technology. Static RAM was never totally eclipsed by its competitors: even
the old ENIAC computer of the 1950’s used vacuum tube ringcounter circuitry for data reg
isters and computation. Eventually though, smaller and smaller scale IC chip manufacturing
technology gave transistors the practical edge over other technologies, and core memory be
came a museum piece in the 1980’s.
One last attempt at a magnetic memory better than core was the bubble memory. Bubble
memory took advantage of a peculiar phenomenon in a mineral called garnet, which, when
arranged in a thin film and exposed to a constant magnetic field perpendicular to the film,
supported tiny regions of oppositelymagnetized ”bubbles” that could be nudged along the film
by prodding with other external magnetic fields. ”Tracks” could be laid on the garnet to focus
the movement of the bubbles by depositing magnetic material on the surface of the film. A
continuous track was formed on the garnet which gave the bubbles a long loop in which to
travel, and motive force was applied to the bubbles with a pair of wire coils wrapped around
the garnet and energized with a 2phase voltage. Bubbles could be created or destroyed with a
tiny coil of wire strategically placed in the bubbles’ path.
The presence of a bubble represented a binary ”1” and the absence of a bubble represented
a binary ”0.” Data could be read and written in this chain of moving magnetic bubbles as they
passed by the tiny coil of wire, much the same as the read/write ”head” in a cassette tape player,
reading the magnetization of the tape as it moves. Like core memory, bubble memory was
nonvolatile: a permanent magnet supplied the necessary background field needed to support
the bubbles when the power was turned off. Unlike core memory, however, bubble memory had
phenomenal storage density: millions of bits could be stored on a chip of garnet only a couple of
square inches in size. What killed bubble memory as a viable alternative to static and dynamic
RAM was its slow, sequential data access. Being nothing more than an incredibly long serial
shift register (ring counter), access to any particular portion of data in the serial string could
be quite slow compared to other memory technologies.
An electrostatic equivalent of the bubble memory is the ChargeCoupled Device (CCD) mem
ory, an adaptation of the CCD devices used in digital photography. Like bubble memory, the
bits are serially shifted along channels on the substrate material by clock pulses. Unlike bub
ble memory, the electrostatic charges decay and must be refreshed. CCD memory is therefore
volatile, with high storage density and sequential access. Interesting, isn’t it? The old Williams
Tube memory was adapted from CRT viewing technology, and CCD memory from video record
ing technology.
15.5 Readonly memory
Readonly memory (ROM) is similar in design to static or dynamic RAM circuits, except that
the ”latching” mechanism is made for onetime (or limited) operation. The simplest type of
15.6. MEMORY WITH MOVING PARTS: ”DRIVES” 467
ROM is that which uses tiny ”fuses” which can be selectively blown or left alone to represent
the two binary states. Obviously, once one of the little fuses is blown, it cannot be made whole
again, so the writing of such ROM circuits is onetime only. Because it can be written (pro
grammed) once, these circuits are sometimes referred to as PROMs (Programmable ReadOnly
Memory).
However, not all writing methods are as permanent as blown fuses. If a transistor latch
can be made which is resettable only with significant effort, a memory device that’s something
of a cross between a RAM and a ROM can be built. Such a device is given a rather oxy
moronic name: the EPROM (Erasable Programmable ReadOnly Memory). EPROMs come in
two basic varieties: Electricallyerasable (EEPROM) and Ultravioleterasable (UV/EPROM).
Both types of EPROMs use capacitive charge MOSFET devices to latch on or off. UV/EPROMs
are ”cleared” by longterm exposure to ultraviolet light. They are easy to identify: they have a
transparent glass window which exposes the silicon chip material to light. Once programmed,
you must cover that glass window with tape to prevent ambient light from degrading the data
over time. EPROMs are often programmed using higher signal voltages than what is used
during ”readonly” mode.
15.6 Memory with moving parts: ”Drives”
The earliest forms of digital data storage involving moving parts was that of the punched paper
card. Joseph Marie Jacquard invented a weaving loom in 1780 which automatically followed
weaving instructions set by carefully placed holes in paper cards. This same technology was
adapted to electronic computers in the 1950’s, with the cards being read mechanically (metal
tometal contact through the holes), pneumatically (air blown through the holes, the presence
of a hole sensed by air nozzle backpressure), or optically (light shining through the holes).
An improvement over paper cards is the paper tape, still used in some industrial environ
ments (notably the CNC machine tool industry), where data storage and speed demands are
low and ruggedness is highly valued. Instead of woodfiber paper, mylar material is often used,
with optical reading of the tape being the most popular method.
Magnetic tape (very similar to audio or video cassette tape) was the next logical improve
ment in storage media. It is still widely used today, as a means to store ”backup” data for
archiving and emergency restoration for other, faster methods of data storage. Like paper
tape, magnetic tape is sequential access, rather than random access. In early home computer
systems, regular audio cassette tape was used to store data in modulated form, the binary
1’s and 0’s represented by different frequencies (similar to FSK data communication). Access
speed was terribly slow (if you were reading ASCII text from the tape, you could almost keep
up with the pace of the letters appearing on the computer’s screen!), but it was cheap and fairly
reliable.
Tape suffered the disadvantage of being sequential access. To address this weak point,
magnetic storage ”drives” with disk or drumshaped media were built. An electric motor pro
vided constantspeed motion. A movable read/write coil (also known as a ”head”) was provided
which could be positioned via servomotors to various locations on the height of the drum or
the radius of the disk, giving access that is almost random (you might still have to wait for
the drum or disk to rotate to the proper position once the read/write coil has reached the right
location).
468 CHAPTER 15. DIGITAL STORAGE (MEMORY)
The disk shape lent itself best to portable media, and thus the floppy disk was born. Floppy
disks (socalled because the magnetic media is thin and flexible) were originally made in 8
inch diameter formats. Later, the 51/4 inch variety was introduced, which was made practical
by advances in media particle density. All things being equal, a larger disk has more space
upon which to write data. However, storage density can be improved by making the little
grains of ironoxide material on the disk substrate smaller. Today, the 31/2 inch floppy disk
is the preeminent format, with a capacity of 1.44 Mbytes (2.88 Mbytes on SCSI drives). Other
portable drive formats are becoming popular, with IoMega’s 100Mbyte ”ZIP” and 1 Gbyte ”JAZ”
disks appearing as original equipment on some personal computers.
Still, floppy drives have the disadvantage of being exposed to harsh environments, being
constantly removed from the drive mechanism which reads, writes, and spins the media. The
first disks were enclosed units, sealed from all dust and other particulate matter, and were
definitely not portable. Keeping the media in an enclosed environment allowed engineers to
avoid dust altogether, as well as spurious magnetic fields. This, in turn, allowed for much
closer spacing between the head and the magnetic material, resulting in amuch tighterfocused
magnetic field to write data to the magnetic material.
The following photograph shows a hard disk drive ”platter” of approximately 30 Mbytes
storage capacity. A ballpoint pen has been set near the bottom of the platter for size reference:
Modern disk drives use multiple platters made of hard material (hence the name, ”hard
drive”) with multiple read/write heads for every platter. The gap between head and platter
is much smaller than the diameter of a human hair. If the hermeticallysealed environment
inside a hard disk drive is contaminated with outside air, the hard drive will be rendered
useless. Dust will lodge between the heads and the platters, causing damage to the surface of
the media.
Here is a hard drive with four platters, although the angle of the shot only allows viewing
of the top platter. This unit is complete with drive motor, read/write heads, and associated
electronics. It has a storage capacity of 340 Mbytes, and is about the same length as the ball
point pen shown in the previous photograph:
15.6. MEMORY WITH MOVING PARTS: ”DRIVES” 469
While it is inevitable that nonmovingpart technology will replace mechanical drives in
the future, current stateoftheart electromechanical drives continue to rival ”solidstate” non
volatile memory devices in storage density, and at a lower cost. In 1998, a 250Mbyte hard drive
was announced that was approximately the size of a quarter (smaller than the metal platter
hub in the center of the last hard disk photograph)! In any case, storage density and reliability
will undoubtedly continue to improve.
An incentive for digital data storage technology advancement was the advent of digitally
encoded music. A joint venture between Sony and Phillips resulted in the release of the ”com
pact audio disc” (CD) to the public in the late 1980’s. This technology is a readonly type, the
media being a transparent plastic disc backed by a thin film of aluminum. Binary bits are en
coded as pits in the plastic which vary the path length of a lowpower laser beam. Data is read
by the lowpower laser (the beam of which can be focused more precisely than normal light)
reflecting off the aluminum to a photocell receiver.
The advantages of CDs over magnetic tape are legion. Being digital, the information is
highly resistant to corruption. Being noncontact in operation, there is no wear incurred
through playing. Being optical, they are immune to magnetic fields (which can easily corrupt
data on magnetic tape or disks). It is possible to purchase CD ”burner” drives which contain
the highpower laser necessary to write to a blank disc.
Following on the heels of the music industry, the video entertainment industry has lever
aged the technology of optical storage with the introduction of the Digital Video Disc, or DVD.
Using a similarsized plastic disc as the music CD, a DVD employs closer spacing of pits to
achieve much greater storage density. This increased density allows featurelength movies to
be encoded on DVD media, complete with trivia information about the movie, director’s notes,
and so on.
Much effort is being directed toward the development of a practical read/write optical disc
(CDW). Success has been found in using chemical substances whose color may be changed
through exposure to bright laser light, then ”read” by lowerintensity light. These optical discs
are immediately identified by their characteristically colored surfaces, as opposed to the silver
colored underside of a standard CD.
470 CHAPTER 15. DIGITAL STORAGE (MEMORY)
Chapter 16
PRINCIPLES OF DIGITAL
COMPUTING
Contents
16.1 A binary adder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 471
16.2 Lookup tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472
16.3 Finitestate machines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477
16.4 Microprocessors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 481
16.5 Microprocessor programming . . . . . . . . . . . . . . . . . . . . . . . . . . . 484
16.1 A binary adder
Suppose we wanted to build a device that could add two binary bits together. Such a device is
known as a halfadder, and its gate circuit looks like this:
A
B
Cout
Σ
The Σ symbol represents the ”sum” output of the halfadder, the sum’s least significant
bit (LSB). Cout represents the ”carry” output of the halfadder, the sum’s most significant bit
(MSB).
If we were to implement this same function in ladder (relay) logic, it would look like this:
471
472 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
L1 L2
A B
A B
A B Σ
Cout
Either circuit is capable of adding two binary digits together. The mathematical ”rules”
of how to add bits together are intrinsic to the hardwired logic of the circuits. If we wanted
to perform a different arithmetic operation with binary bits, such as multiplication, we would
have to construct another circuit. The above circuit designs will only perform one function: add
two binary bits together. To make them do something else would take rewiring, and perhaps
different componentry.
In this sense, digital arithmetic circuits aren’t much different from analog arithmetic (op
erational amplifier) circuits: they do exactly what they’re wired to do, no more and no less. We
are not, however, restricted to designing digital computer circuits in this manner. It is possi
ble to embed the mathematical ”rules” for any arithmetic operation in the form of digital data
rather than in hardwired connections between gates. The result is unparalleled flexibility in
operation, giving rise to a whole new kind of digital device: the programmable computer.
While this chapter is by no means exhaustive, it provides what I believe is a unique and
interesting look at the nature of programmable computer devices, starting with two devices
often overlooked in introductory textbooks: lookup table memories and finitestate machines.
16.2 Lookup tables
Having learned about digital memory devices in the last chapter, we know that it is possible
to store binary data within solidstate devices. Those storage ”cells” within solidstate memory
devices are easily addressed by driving the ”address” lines of the device with the proper binary
value(s). Suppose we had a ROM memory circuit written, or programmed, with certain data,
such that the address lines of the ROM served as inputs and the data lines of the ROM served
as outputs, generating the characteristic response of a particular logic function. Theoretically,
we could program this ROM chip to emulate whatever logic function we wanted without having
to alter any wire connections or gates.
Consider the following example of a 4 x 2 bit ROM memory (a very small memory!) pro
grammed with the functionality of a half adder:
16.2. LOOKUP TABLES 473
4 x 2 ROM
A0
A1
D0
D1
A
B
Σ
Cout
A B ΣCout
0 0
0 1
1 0
1 1
00
0
0
0
1
1
1
Address Data
If this ROM has been written with the above data (representing a halfadder’s truth table),
driving the A and B address inputs will cause the respective memory cells in the ROM chip to
be enabled, thus outputting the corresponding data as the Σ (Sum) and Cout bits. Unlike the
halfadder circuit built of gates or relays, this device can be set up to perform any logic function
at all with two inputs and two outputs, not just the halfadder function. To change the logic
function, all we would need to do is write a different table of data to another ROM chip. We
could even use an EPROM chip which could be rewritten at will, giving the ultimate flexibility
in function.
It is vitally important to recognize the significance of this principle as applied to digital
circuitry. Whereas the halfadder built from gates or relays processes the input bits to arrive
at a specific output, the ROM simply remembers what the outputs should be for any given
combination of inputs. This is not much different from the ”times tables” memorized in grade
school: rather than having to calculate the product of 5 times 6 (5 + 5 + 5 + 5 + 5 + 5 = 30),
schoolchildren are taught to remember that 5 x 6 = 30, and then expected to recall this product
from memory as needed. Likewise, rather than the logic function depending on the functional
arrangement of hardwired gates or relays (hardware), it depends solely on the data written
into the memory (software).
Such a simple application, with definite outputs for every input, is called a lookup table,
because the memory device simply ”looks up” what the output(s) should to be for any given
combination of inputs states.
This application of a memory device to perform logical functions is significant for several
reasons:
• Software is much easier to change than hardware.
• Software can be archived on various kinds of memory media (disk, tape), thus providing
an easy way to document and manipulate the function in a ”virtual” form; hardware can
only be ”archived” abstractly in the form of some kind of graphical drawing.
• Software can be copied from one memory device (such as the EPROM chip) to another,
allowing the ability for one device to ”learn” its function from another device.
• Software such as the logic function example can be designed to perform functions that
would be extremely difficult to emulate with discrete logic gates (or relays!).
The usefulness of a lookup table becomes more and more evident with increasing complex
ity of function. Suppose we wanted to build a 4bit adder circuit using a ROM. We’d require a
ROM with 8 address lines (two 4bit numbers to be added together), plus 4 data lines (for the
signed output):
474 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
A0
A1 D0
D1
A2
A3
A4
A5
A6
A7
D2
D3
First
4bit
number
Second
4bit
number
4bit
result
256 x 4
ROM
With 256 addressable memory locations in this ROM chip, we would have a fair amount of
programming to do, telling it what binary output to generate for each and every combination
of binary inputs. We would also run the risk of making a mistake in our programming and
have it output an incorrect sum, if we weren’t careful. However, the flexibility of being able
to configure this function (or any function) through software alone generally outweighs that
costs.
Consider some of the advanced functions we could implement with the above ”adder.” We
know that when we add two sets of numbers in 2’s complement signed notation, we risk having
the answer overflow. For instance, if we try to add 0111 (decimal 7) to 0110 (decimal 6) with
only a 4bit number field, the answer we’ll get is 1001 (decimal 7) instead of the correct value,
13 (7 + 6), which cannot be expressed using 4 signed bits. If we wanted to, we could avoid
the strange answers given in overflow conditions by programming this lookup table circuit to
output something else in conditions where we know overflow will occur (that is, in any case
where the real sum would exceed +7 or 8). One alternative might be to program the ROM to
output the quantity 0111 (the maximum positive value that can be represented with 4 signed
bits), or any other value that we determined to be more appropriate for the application than
the typical overflowed ”error” value that a regular adder circuit would output. It’s all up to the
programmer to decide what he or she wants this circuit to do, because we are no longer limited
by the constraints of logic gate functions.
The possibilities don’t stop at customized logic functions, either. By adding more address
lines to the 256 x 4 ROM chip, we can expand the lookup table to include multiple functions:
16.2. LOOKUP TABLES 475
A0
A1 D0
D1
A2
A3
A4
A5
A6
A7
D2
D3
First
4bit
number
Second
4bit
number
4bit
result
ROM
Function
control
A8
A9
1024 x 4
With two more address lines, the ROM chip will have 4 times as many addresses as before
(1024 instead of 256). This ROM could be programmed so that when A8 and A9 were both low,
the output data represented the sum of the two 4bit binary numbers input on address lines
A0 through A7, just as we had with the previous 256 x 4 ROM circuit. For the addresses A8=1
and A9=0, it could be programmed to output the difference (subtraction) between the first 4
bit binary number (A0 through A3) and the second binary number (A4 through A7). For the
addresses A8=0 and A9=1, we could program the ROM to output the difference (subtraction)
of the two numbers in reverse order (second  first rather than first  second), and finally, for
the addresses A8=1 and A9=1, the ROM could be programmed to compare the two inputs and
output an indication of equality or inequality. What we will have then is a device that can
perform four different arithmetical operations on 4bit binary numbers, all by ”looking up” the
answers programmed into it.
If we had used a ROM chip with more than two additional address lines, we could program
it with a wider variety of functions to perform on the two 4bit inputs. There are a number
of operations peculiar to binary data (such as parity check or ExclusiveORing of bits) that we
might find useful to have programmed in such a lookup table.
Devices such as this, which can perform a variety of arithmetical tasks as dictated by a
binary input code, are known as Arithmetic Logic Units (ALUs), and they comprise one of the
essential components of computer technology. Although modern ALUs are more often con
structed from very complex combinational logic (gate) circuits for reasons of speed, it should be
comforting to know that the exact same functionality may be duplicated with a ”dumb” ROM
chip programmed with the appropriate lookup table(s). In fact, this exact approach was used
by IBM engineers in 1959 with the development of the IBM 1401 and 1620 computers, which
used lookup tables to perform addition, rather than binary adder circuitry. The machine was
fondly known as the ”CADET,” which stood for ”Can’t Add, Doesn’t Even Try.”
A very common application for lookup table ROMs is in control systems where a custom
mathematical function needs to be represented. Such an application is found in computer
476 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
controlled fuel injection systems for automobile engines, where the proper air/fuel mixture
ratio for efficient and clean operation changes with several environmental and operational
variables. Tests performed on engines in research laboratories determine what these ideal
ratios are for varying conditions of engine load, ambient air temperature, and barometric air
pressure. The variables are measured with sensor transducers, their analog outputs converted
to digital signals with A/D circuitry, and those parallel digital signals used as address inputs
to a highcapacity ROM chip programmed to output the optimum digital value for air/fuel ratio
for any of these given conditions.
Sometimes, ROMs are used to provide onedimensional lookup table functions, for ”correct
ing” digitized signal values so that they more accurately represent their realworld significance.
An example of such a device is a thermocouple transmitter, which measures the millivoltage
signal generated by a junction of dissimilar metals and outputs a signal which is supposed to
directly correspond to that junction temperature. Unfortunately, thermocouple junctions do
not have perfectly linear temperature/voltage responses, and so the raw voltage signal is not
perfectly proportional to temperature. By digitizing the voltage signal (A/D conversion) and
sending that digital value to the address of a ROM programmed with the necessary correction
values, the ROM’s programming could eliminate some of the nonlinearity of the thermocouple’s
temperaturetomillivoltage relationship, so that the final output of the device would be more
accurate. The popular instrumentation term for such a lookup table is a digital characterizer.
A/D
converter
ROM
converter
D/A 420 mA
analog
signal
Another application for lookup tables is in special code translation. A 128 x 8 ROM, for
instance, could be used to translate 7bit ASCII code to 8bit EBCDIC code:
A0
A1
D0
D1
A2
A3
A4
A5
A6
D2
D3
ROM
D4
D5
D6
D7
128 x 8
ASCII
in EBCDIC
out
Again, all that is required is for the ROM chip to be properly programmed with the neces
16.3. FINITESTATE MACHINES 477
sary data so that each valid ASCII input will produce a corresponding EBCDIC output code.
16.3 Finitestate machines
Feedback is a fascinating engineering principle. It can turn a rather simple device or process
into something substantially more complex. We’ve seen the effects of feedback intentionally
integrated into circuit designs with some rather astounding effects:
• Comparator + negative feedback ———–> controllablegain amplifier
• Comparator + positive feedback ———–> comparator with hysteresis
• Combinational logic + positive feedback –> multivibrator
In the field of process instrumentation, feedback is used to transform a simple measurement
system into something capable of control:
• Measurement system + negative feedback —> closedloop control system
Feedback, both positive and negative, has the tendency to add whole new dynamics to the
operation of a device or system. Sometimes, these new dynamics find useful application, while
other times they are merely interesting. With lookup tables programmed into memory devices,
feedback from the data outputs back to the address inputs creates a whole new type of device:
the Finite State Machine, or FSM:
A0
A1
D0
D1
A2
A3
D2
D3
ROM
16 x 4
A crude Finite State Machine
Feedback
D0
D1
D2
D3
The above circuit illustrates the basic idea: the data stored at each address becomes the
next storage location that the ROM gets addressed to. The result is a specific sequence of
binary numbers (following the sequence programmed into the ROM) at the output, over time.
To avoid signal timing problems, though, we need to connect the data outputs back to the
address inputs through a 4bit Dtype flipflop, so that the sequence takes place step by step to
the beat of a controlled clock pulse:
478 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
A0
A1
D0
D1
A2
A3
D2
D3
ROM
16 x 4
Feedback
D0
D1
D2
D3
D0
D1
D2
D3
Q0
Q1
Q2
Q3
SRG
Clock
An improved Finite State Machine
An analogy for the workings of such a device might be an array of postoffice boxes, each one
with an identifying number on the door (the address), and each one containing a piece of paper
with the address of another P.O. box written on it (the data). A person, opening the first P.O.
box, would find in it the address of the next P.O. box to open. By storing a particular pattern
of addresses in the P.O. boxes, we can dictate the sequence in which each box gets opened, and
therefore the sequence of which paper gets read.
Having 16 addressable memory locations in the ROM, this Finite State Machine would have
16 different stable ”states” in which it could latch. In each of those states, the identity of the
next state would be programmed in to the ROM, awaiting the signal of the next clock pulse
to be fed back to the ROM as an address. One useful application of such an FSM would be to
generate an arbitrary count sequence, such as Gray Code:
Address > Data Gray Code count sequence:
0000 > 0001 0 0000
0001 > 0011 1 0001
0010 > 0110 2 0011
0011 > 0010 3 0010
0100 > 1100 4 0110
0101 > 0100 5 0111
0110 > 0111 6 0101
0111 > 0101 7 0100
1000 > 0000 8 1100
1001 > 1000 9 1101
1010 > 1011 10 1111
1011 > 1001 11 1110
1100 > 1101 12 1010
1101 > 1111 13 1011
16.3. FINITESTATE MACHINES 479
1110 > 1010 14 1001
1111 > 1110 15 1000
Try to follow the Gray Code count sequence as the FSM would do it: starting at 0000,
follow the data stored at that address (0001) to the next address, and so on (0011), and so on
(0010), and so on (0110), etc. The result, for the program table shown, is that the sequence of
addressing jumps around from address to address in what looks like a haphazard fashion, but
when you check each address that is accessed, you will find that it follows the correct order for
4bit Gray code. When the FSM arrives at its last programmed state (address 1000), the data
stored there is 0000, which starts the whole sequence over again at address 0000 in step with
the next clock pulse.
We could expand on the capabilities of the above circuit by using a ROM with more address
lines, and adding more programming data:
A0
A1
D0
D1
A2
A3
D2
D3
ROM
Feedback
D0
D1
D2
D3
D0
D1
D2
D3
Q0
Q1
Q2
Q3
SRG
Clock
A4Q4D4
"function control"
32 x 4
Now, just like the lookup table adder circuit that we turned into an Arithmetic Logic Unit
(+, , x, / functions) by utilizing more address lines as ”function control” inputs, this FSM
counter can be used to generate more than one count sequence, a different sequence pro
grammed for the four feedback bits (A0 through A3) for each of the two function control line
input combinations (A4 = 0 or 1).
Address > Data Address > Data
00000 > 0001 10000 > 0001
00001 > 0010 10001 > 0011
00010 > 0011 10010 > 0110
00011 > 0100 10011 > 0010
00100 > 0101 10100 > 1100
00101 > 0110 10101 > 0100
480 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
00110 > 0111 10110 > 0111
00111 > 1000 10111 > 0101
01000 > 1001 11000 > 0000
01001 > 1010 11001 > 1000
01010 > 1011 11010 > 1011
01011 > 1100 11011 > 1001
01100 > 1101 11100 > 1101
01101 > 1110 11101 > 1111
01110 > 1111 11110 > 1010
01111 > 0000 11111 > 1110
If A4 is 0, the FSM counts in binary; if A4 is 1, the FSM counts in Gray Code. In either
case, the counting sequence is arbitrary: determined by the whim of the programmer. For that
matter, the counting sequence doesn’t even have to have 16 steps, as the programmer may
decide to have the sequence recycle to 0000 at any one of the steps at all. It is a completely
flexible counting device, the behavior strictly determined by the software (programming) in the
ROM.
We can expand on the capabilities of the FSM even more by utilizing a ROM chip with
additional address input and data output lines. Take the following circuit, for example:
A0
A1
D0
D1
A2
A3
D2
D3
ROM
Feedback
D0
D1
D2
D3
D0
D1
D2
D3
Q0
Q1
Q2
Q3
SRG
Clock
A4Q4D4 D4
D5
D6
D7
A5
A6
A7
D5
D6
D7
Q5
Q6
Q7
Inputs Outputs
256 x 8
Here, the D0 through D3 data outputs are used exclusively for feedback to the A0 through
A3 address lines. Date output lines D4 through D7 can be programmed to output something
other than the FSM’s ”state” value. Being that four data output bits are being fed back to four
16.4. MICROPROCESSORS 481
address bits, this is still a 16state device. However, having the output data come from other
data output lines gives the programmer more freedom to configure functions than before. In
other words, this device can do far more than just count! The programmed output of this FSM
is dependent not only upon the state of the feedback address lines (A0 through A3), but also
the states of the input lines (A4 through A7). The Dtype flip/flop’s clock signal input does
not have to come from a pulse generator, either. To make things more interesting, the flip/flop
could be wired up to clock on some external event, so that the FSM goes to the next state only
when an input signal tells it to.
Now we have a device that better fulfills the meaning of the word ”programmable.” The data
written to the ROM is a program in the truest sense: the outputs follow a preestablished order
based on the inputs to the device and which ”step” the device is on in its sequence. This is very
close to the operating design of the Turing Machine, a theoretical computing device invented by
Alan Turing, mathematically proven to be able to solve any known arithmetic problem, given
enough memory capacity.
16.4 Microprocessors
Early computer science pioneers such as Alan Turing and John Von Neumann postulated that
for a computing device to be really useful, it not only had to be able to generate specific outputs
as dictated by programmed instructions, but it also had to be able to write data to memory, and
be able to act on that data later. Both the program steps and the processed data were to reside
in a common memory ”pool,” thus giving way to the label of the storedprogram computer.
Turing’s theoretical machine utilized a sequentialaccess tape, which would store data for a
control circuit to read, the control circuit rewriting data to the tape and/or moving the tape
to a new position to read more data. Modern computers use randomaccess memory devices
instead of sequentialaccess tapes to accomplish essentially the same thing, except with greater
capability.
A helpful illustration is that of early automatic machine tool control technology. Called
openloop, or sometimes just NC (numerical control), these control systems would direct the
motion of a machine tool such as a lathe or a mill by following instructions programmed as
holes in paper tape. The tape would be run one direction through a ”read” mechanism, and
the machine would blindly follow the instructions on the tape without regard to any other
conditions. While these devices eliminated the burden of having to have a human machinist
direct every motion of the machine tool, it was limited in usefulness. Because the machine
was blind to the real world, only following the instructions written on the tape, it could not
compensate for changing conditions such as expansion of the metal or wear of the mechanisms.
Also, the tape programmer had to be acutely aware of the sequence of previous instructions in
the machine’s program to avoid troublesome circumstances (such as telling the machine tool
to move the drill bit laterally while it is still inserted into a hole in the work), since the device
had no memory other than the tape itself, which was readonly. Upgrading from a simple tape
reader to a Finite State control design gave the device a sort of memory that could be used
to keep track of what it had already done (through feedback of some of the data bits to the
address bits), so at least the programmer could decide to have the circuit remember ”states”
that the machine tool could be in (such as ”coolant on,” or tool position). However, there was
still room for improvement.
482 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
The ultimate approach is to have the program give instructions which would include the
writing of new data to a read/write (RAM) memory, which the program could easily recall
and process. This way, the control system could record what it had done, and any sensor
detectable process changes, much in the same way that a human machinist might jot down
notes or measurements on a scratchpad for future reference in his or her work. This is what
is referred to as CNC, or Closedloop Numerical Control.
Engineers and computer scientists looked forward to the possibility of building digital de
vices that could modify their own programming, much the same as the human brain adapts
the strength of interneural connections depending on environmental experiences (that is why
memory retention improves with repeated study, and behavior is modified through consequen
tial feedback). Only if the computer’s program were stored in the same writable memory ”pool”
as the data would this be practical. It is interesting to note that the notion of a selfmodifying
program is still considered to be on the cutting edge of computer science. Most computer pro
gramming relies on rather fixed sequences of instructions, with a separate field of data being
the only information that gets altered.
To facilitate the storedprogram approach, we require a device that is much more complex
than the simple FSM, although many of the same principles apply. First, we need read/write
memory that can be easily accessed: this is easy enough to do. Static or dynamic RAM chips
do the job well, and are inexpensive. Secondly, we need some form of logic to process the data
stored in memory. Because standard and Boolean arithmetic functions are so useful, we can
use an Arithmetic Logic Unit (ALU) such as the lookup table ROM example explored earlier.
Finally, we need a device that controls how and where data flows between the memory, the
ALU, and the outside world. This socalled Control Unit is the most mysterious piece of the
puzzle yet, being comprised of tristate buffers (to direct data to and from buses) and decoding
logic which interprets certain binary codes as instructions to carry out. Sample instructions
might be something like: ”add the number stored at memory address 0010 with the number
stored at memory address 1101,” or, ”determine the parity of the data in memory address
0111.” The choice of which binary codes represent which instructions for the Control Unit to
decode is largely arbitrary, just as the choice of which binary codes to use in representing the
letters of the alphabet in the ASCII standard was largely arbitrary. ASCII, however, is now an
internationally recognized standard, whereas control unit instruction codes are almost always
manufacturerspecific.
Putting these components together (read/write memory, ALU, and control unit) results in
a digital device that is typically called a processor. If minimal memory is used, and all the
necessary components are contained on a single integrated circuit, it is called a microprocessor.
When combined with the necessary buscontrol support circuitry, it is known as a Central
Processing Unit, or CPU.
CPU operation is summed up in the socalled fetch/execute cycle. Fetch means to read
an instruction from memory for the Control Unit to decode. A small binary counter in the
CPU (known as the program counter or instruction pointer) holds the address value where
the next instruction is stored in main memory. The Control Unit sends this binary address
value to the main memory’s address lines, and the memory’s data output is read by the Control
Unit to send to another holding register. If the fetched instruction requires reading more
data from memory (for example, in adding two numbers together, we have to read both the
numbers that are to be added from main memory or from some other source), the Control Unit
appropriately addresses the location of the requested data and directs the data output to ALU
16.4. MICROPROCESSORS 483
registers. Next, the Control Unit would execute the instruction by signaling the ALU to do
whatever was requested with the two numbers, and direct the result to another register called
the accumulator. The instruction has now been ”fetched” and ”executed,” so the Control Unit
now increments the program counter to step the next instruction, and the cycle repeats itself.
Microprocessor (CPU)

 ** Program counter ** 
 (increments address value sent to 
 external memory chip(s) to fetch ==========> Address bus
 the next instruction)  (to RAM memory)

 ** Control Unit ** <=========> Control Bus
 (decodes instructions read from  (to all devices sharing
 program in memory, enables flow  address and/or data busses;
 of data to and from ALU, internal  arbitrates all bus communi
 registers, and external devices)  cations)

 ** Arithmetic Logic Unit (ALU) ** 
 (performs all mathematical 
 calculations and Boolean 
 functions) 

 ** Registers ** 
 (small read/write memories for <=========> Data Bus
 holding instruction codes,  (from RAM memory and other
 error codes, ALU data, etc;  external devices)
 includes the "accumulator") 

As one might guess, carrying out even simple instructions is a tedious process. Several steps
are necessary for the Control Unit to complete the simplest of mathematical procedures. This is
especially true for arithmetic procedures such as exponents, which involve repeated executions
(”iterations”) of simpler functions. Just imagine the sheer quantity of steps necessary within
the CPU to update the bits of information for the graphic display on a flight simulator game!
The only thing which makes such a tedious process practical is the fact that microprocessor
circuits are able to repeat the fetch/execute cycle with great speed.
In some microprocessor designs, there are minimal programs stored within a special ROM
memory internal to the device (called microcode) which handle all the substeps necessary to
carry out more complex math operations. This way, only a single instruction has to be read
from the program RAM to do the task, and the programmer doesn’t have to deal with trying
to tell the microprocessor how to do every minute step. In essence, its a processor inside of a
processor; a program running inside of a program.
484 CHAPTER 16. PRINCIPLES OF DIGITAL COMPUTING
16.5 Microprocessor programming
The ”vocabulary” of instructions which any particular microprocessor chip possesses is specific
to that model of chip. An Intel 80386, for example, uses a completely different set of binary
codes than a Motorola 68020, for designating equivalent functions. Unfortunately, there are
no standards in place for microprocessor instructions. This makes programming at the very
lowest level very confusing and specialized.
When a human programmer develops a set of instructions to directly tell a microprocessor
how to do something (like automatically control the fuel injection rate to an engine), they’re
programming in the CPU’s own ”language.” This language, which consists of the very same
binary codes which the Control Unit inside the CPU chip decodes to perform tasks, is often
referred to as machine language. While machine language software can be ”worded” in binary
notation, it is often written in hexadecimal form, because it is easier for human beings to work
with. For example, I’ll present just a few of the common instruction codes for the Intel 8080
microprocessor chip:
Hexadecimal Binary Instruction description
  
 7B 01111011 Move contents of register A to register E

 87 10000111 Add contents of register A to register D

 1C 00011100 Increment the contents of register E by 1

 D3 11010011 Output byte of data to data bus
Even with hexadecimal notation, these instructions can be easily confused and forgotten.
For this purpose, another aid for programmers exists called assembly language. With assem
bly language, two to four letter mnemonic words are used in place of the actual hex or binary
code for describing program steps. For example, the instruction 7B for the Intel 8080 would
be ”MOV A,E” in assembly language. The mnemonics, of course, are useless to the micropro
cessor, which can only understand binary codes, but it is an expedient way for programmers
to manage the writing of their programs on paper or text editor (word processor). There are
even programs written for computers called assemblers which understand these mnemonics,
translating them to the appropriate binary codes for a specified target microprocessor, so that
the programmer can write a program in the computer’s native language without ever having
to deal with strange hex or tedious binary code notation.
Once a program is developed by a person, it must be written into memory before a micropro
cessor can execute it. If the program is to be stored in ROM (which some are), this can be done
with a special machine called a ROM programmer, or (if you’re masochistic), by plugging the
ROM chip into a breadboard, powering it up with the appropriate voltages, and writing data by
making the right wire connections to the address and data lines, one at a time, for each instruc
tion. If the program is to be stored in volatile memory, such as the operating computer’s RAM
memory, there may be a way to type it in by hand through that computer’s keyboard (some
computers have a miniprogram stored in ROM which tells the microprocessor how to accept
keystrokes from a keyboard and store them as commands in RAM), even if it is too dumb to do
16.5. MICROPROCESSOR PROGRAMMING 485
anything else. Many ”hobby” computer kits work like this. If the computer to be programmed
is a fullyfunctional personal computer with an operating system, disk drives, and the whole
works, you can simply command the assembler to store your finished program onto a disk for
later retrieval. To ”run” your program, you would simply type your program’s filename at the
pro