3,951 628 60MB
English Pages 0 [488] Year 1985
THIRD EDITION
TECHNIQUES
W
D.
A.D.
COOPER
HELFRICK
5
SUM
ELECTRONIC INSTRUMENTATION AND MEASUREMENT TECHNIQUES
3rd edition
Electronic
Instrumentation
and Measurement Techniques William David Cooper Albert D. Helfrick
Prentice-Hall, Inc. /
Englewood
Cliffs,
New
Jersey
07632
Library of Congress Cataloging
Cooper, William David,
Data
in Publication
(date)
Electronic instrumentation and measurement techniques.
Includes bibliographies and index. 2. Electric meters. Electric measurements. 4. Electronic instruments. Electronic measurements. Helfrick, Albert D. II. Title. 1.
3. I.
TK275.C63
ISBN
1985 0-13-250721-8
621.3815'4
84-18185
Cover design: 20/20 Services, Inc. Manufacturing buyer: Gordon Osbourne
©
1985, 1978, 1970, by Prentice-Hall, Inc.,
Englewood
No part of this book may be reproduced, in any form or by any means, without permission in writing from the publisher. All rights reserved.
Printed in the United States of America
10
987654321
ISBN
Ul
-13-ESD7r21-fl
Prentice-Hall International, Inc.,
London
Prentice-Hall of Australia Pty. Limited, Sydney Editora Prentice-Hall do Brasil, Ltda., Rio de Janeiro Prentice-Hall
Canada
Inc.,
Toronto
Prentice-Hall of India Private Limited, Prentice-Hall of Japan, Inc., Tokyo
New
Delhi
Prentice-Hall of Southeast Asia Pte. Ltd., Singapore
Whitehall Books Limited, Wellington,
New Zealand
Cliffs,
New
Jersey 07632
8 2
CONTENTS
PREFACE Chapter
xi
MEASUREMENT AND ERROR
1
1
1-1
Definitions
1
1-2
Accuracy and Precision
2
1-3
Significant Figures
3
1-4
Types of Error
1-5
Statistical Analysis
1
-6
1-7
Probability of Errors
1
Limiting Errors
16
Questions
1
Problems
18
Chapter 2 2- 1
2-2 2-3 2-4 2-5 2-6
6 10
SYSTEMS OF UNITS OF MEASUREMENT
Fundamental and Derived Units Systems of Units Electric and Magnetic Units International System of Units Other Systems of Units Conversion of Units Problems
20 20 21
23 27 27 30 30 V
0 7
Contents
vi
Chapter 3
STANDARDS OF MEASUREMENT
34
Classification of Standards Standards for Mass, Length, and Volume Time and Frequency Standards Electrical Standards Standards of Temperature and Luminous Intensity IEEE Standards Questions and Problems
3-1
3-2 3-3 3-4
3-5 3- 6
34 36 37 39 46 47 48
DIRECT-CURRENT INDICATING INSTRUMENTS
Chapter 4
Suspension Galvanometer Torque and Deflection of the Galvanometer Permanent-Magnet Moving-Coil Mechanism
4- 1
4-2 4-3
4-5
DC Ammeters DC Voltmeters
4-6
Voltmeter Sensitivity
4-4
4-7
Voltmeter-Ammeter Method of Measuring Resistance
4-8
Series-Type
4-9
Shunt-Type Ohmmeter 1
4-12 4-13 4-14 4-15
50 51
55 61
65 68 72 74 78 80 84 85 94 97 99
Ohmmeter
4-10 Multimeter or 4-1
50
Calibration of
VOM DC Instruments
Alternating-Current Indicating Instruments
Thermoinstruments Electrodynamometers in Power Measurements Watthour Meter 4-1 6 Power-Factor Meter 4- 17 Instrument Transformers Problems
Chapter 5
BRIDGES AND THEIR APPLICATION
101
103 108
1
10
5- 1
Introduction
110
5-2
Wheatstone Bridge
1 1
5-3
Kelvin Bridge
1 1
5-4
5-6
Guarded Wheatstone Bridge AC Bridges and Their Application Comparison Bridges
5- 7
Maxwell Bridge
5-8
Hay Bridge
122 1 25 129 1 32 134 136 138
5-5
5-9
'
Schering Bridge 5-10 Unbalance Conditions
Contents
vii
5-11 5-1 2
Wien Bridge Wagner Ground Connection
5- 13 Potentiometer
Problems
ELECTRONIC INSTRUMENTS FOR MEASURING BASIC PARAMETERS
141
142 144 145
Chapter 6 6- 1
Amplified
6-2
AC
DC
Meter
1
6-3
Voltmeter Using Rectifiers True RMS-Responding Voltmeter
6-4
Electronic Multimeter
Considerations
6-6
Differential Voltmeters
6-7
Digital
Choosing an Analog Voltmeter
Voltmeters
6-8
Component Measuring Instruments
6-9
Q
Meter
6-10 Vector Impedance Meter 6- 1 1 Vector Voltmeter 6- 12 RF Power and Voltage Measurement
Problems
Chapter 7 7-
1
OSCILLOSCOPES
Introduction
7-2
Oscilloscope Block Diagram
7-3
Cathode Ray Tube
7-4
CRT
7-5
Vertical Deflection
7-6
Circuits
System
7-7
Delay Line Multiple Trace
7-6
Horizontal Deflection
7-9
Oscilloscope Probes and Transducers
System
7-10 Oscilloscope Techniques 7- 11
Special Oscilloscopes
Problems
Chapter 8 8- 1
8-2 8-3 8-4
8-5
SIGNAL GENERATION
The Sine-Wave Generator Frequency Synthesized Signal Generator Frequency Divider Generator Signal Generator Modulation Sweep-Frequency Generator
48
151 1 1
6-5
in
147
55 56
161
163 169 1 78 186 195 1 99 203 206
207 207 208 210 223 224 230 233 235 239 242 248 258
259 259 270 274 277 277
Contents
viii
8-6
8-7 8- 8
Pulse and Square-Wave Generators Function Generator Audiofrequency Signal Generation
Problems
Chapter 9
SIGNAL ANALYSIS
9- 1
Wave Analyzers
9-2
Harmonic Distortion Analyzers Spectrum Analysis Problems
9-3
TIME-INTERVAL MEASUREMENTS 10-1
Simple Frequency Counter
10-2
Measurement
10-3
Extending the Frequency Range of the Counter Automatic and Computing Counters Problems
0- 4
291
293 295
296 296 300 306 324
FREQUENCY COUNTERS AND
Chapter 10
1
282
Errors
TRANSDUCERS AS INPUT ELEMENTS 1 1 TO INSTRUMENTATION SYSTEMS
325 325 337 341
345 347
Chapter 11- 1
Classification of Transducers
11-2
Selecting a Transducer
11-3
Strain
11-4
Displacement Transducers Temperature Measurements Photosensitive Devices Problems
11-5 11- 6
Chapter 12 ACQUISITION
Gages
ANALOG AND DIGITAL DATA SYSTEMS
348 348 349 353 359 369 384 384
392
12-4
Digital-to-Analog Conversion
12-5
Analog-to-Digital Conversion
392 394 403 407 409
12-6
Multiplexing
421
12-7
Spatial
12- 1
Instrumentation Systems
12-2
Magnetic Tape Recorders
12-3
Self-Balancing Potentiometer
Encoders
424
Contents
ix
Chapter 13
COMPUTER-CONTROLLED TEST
SYSTEMS
429
13-1
Testing an Audio Amplifier
429
13-2
Testing a Radio Receiver
431
13-3
Instruments
13-4
IEEE 488
13-5
Digital Control Description
13-6
Example of Signal Timing Measurement
Used
in
Computer-Controlled Instrumen-
435 443 445
tation
Electrical Interface
in
a Microprocessor-Based
446 447
Questions
449
APPENDIX Abbreviations, Symbols,
and
450 452
Prefixes
Decibel Conversion Tables
Table
1:
Conversion of Decibels
to
Power and Voltage
Current) Ratios
(or
455
Table 2: Conversion of Voltage (or Current) and Power Ratios to Decibels
457
SELECTED ANSWERS
459
INDEX
463
PREFACE
The
third edition of Electronic Instrumentation
and Measurement Techniques
is
designed to serve as a text for students of electrical and electronic engineering at
both two- and four-year colleges and technical
serve as a refresher or
handbook
institutes.
The book can
also
for the professional engineer. This text provides
good groundwork for the basics of electrical measurement and then presents examples ranging from elementary measurements to the most sophisticated computer-controlled systems.
Several changes have been
made from the previous edition.
First, the subject
movements has been reduced and compressed from two chapters to one chapter, and the subject of bridges has been similarly compressed. This was done to make way for the addition of new material on measurement systems using newer technology. Chapter 7, on oscilloscopes, has been completely rewritten and expanded of electromechanical meter
and osciland signal analysis, and the expanded text reflects
to include descriptions of the latest technology in cathode ray tubes
loscope circuits. Chapters 8 and
comprised one chapter
9,
on
signal generation
in the previous edition,
the advances in these instruments. Chapter 13 describes computer-controlled test
equipment for automatic
test systems.
This material, which
is
entirely new,
covers a rapidly growing phase of test and measurement technology.
Those elements essential to a good textbook, such as worked-out examples and chapter-end problems and review questions, have been retained. Included in the new edition to aid the student are answers to selected questions and problems.
w. D. A. D.
COOPER HELFRICK xi
ELECTRONIC INSTRUMENTATION AND MEASUREMENT TECHNIQUES
CHAPTER
1
MEASUREMENT AND ERROR
DEFINITIONS
1-1
Measurement generally involves using an instrument
as a physical
means of
determining a quantity or variable. The instrument serves as an extension of
human
faculties
and
in
many
cases enables a person to determine the value of
an unknown quantity which his unaided
An
instrument, then,
may
human
faculties could not measure.
be defined as a device for determining the value or
magnitude of a quantity or variable. The electronic instrument, as its name is based on electrical or electronic principles for its measurement function. An electronic instrument may be a relatively uncomplicated device of simple construction such as a basic dc current meter (see Chapter 4). As technology expands, however, the demand for more elaborate and more accurate instruments implies,
increases
To
and produces new developments
in
instrument design and application.
use these instruments intelligently, one needs to understand their operating
and to appraise their suitability for the intended application. Measurement work employs a number of terms which should be defined
principles
here.
Instrument: a device for determining the value or magnitude of a quantity or variable.
Accuracy: closeness with which an instrument reading approaches the true value of the variable being measured. Precision: a
measure of the reproducibility of the measurements;
i.e.,
1
Measurement and
2
Error
Chap.
1
is a measure of the degree to from one another.
given a fixed value of a variable, precision
which successive measurements Sensitivity: the ratio of
to a
differ
output signal or response of the instrument
change of input or measured variable. Resolution: the smallest change in measured value to which the
instrument will respond. Error: deviation from the true value of the measured variable.
Several techniques
may
be used to minimize the
example, in making precision measurements,
it is
effects
of errors. For
advisable to record a series of
observations rather than rely on one observation. Alternate methods of mea-
surement, as well as the use of different instruments to perform the same experiment, provide a good technique for increasing accuracy. Although these
techniques tend to increase the precision of measurement by reducing environ-
mental or random error, they cannot account for instrumental error.* This chapter provides an introduction to different types of error in mea-
surement and to the methods generally used to express most reliable value of the measured variable.
1-2
errors, in
terms of the
ACCURACY AND PRECISION
Accuracy
refers to the degree of closeness or
conformity to the true value of the
quantity under measurement. Precision refers to the degree of agreement within
a group of measurements or instruments.
To illustrate the distinction between accuracy and precision, two voltmeters make and model may be compared. Both meters have knife-edged
of the same
and they have carefully same precision. If the value
pointers and mirror-backed scales to avoid parallax, calibrated scales.
They may
therefore be read to the
its readings may be amount. Therefore the accuracy of the two meters may be quite different. (To determine which meter is in error, a comparison measurement with a standard meter should be made.) Precision is composed of two characteristics: conformity and the number of significant figures to which a measurement may be made. Consider, for example, that a resistor, whose true resistance is 1,384,572 ft, is measured by an ohmmeter which consistently and repeatedly indicates 1.4 Mft. But can the observer "read" the true value from the scale? His estimates from the scale
of the series resistance in one meter changes considerably, in error
by a
fairly large
reading consistently yield a value of 1.4 Mft. This
is
as close to the true value
by estimation. Although there are no deviations from the observed value, the error created by the limitation of the scale reading is a as he can read the scale
Melville
B. Stout, Basic Electrical
Hall, Inc., 1960), pp. 21-26.
Measurements, 2nd
ed.
(Englewood
Cliffs, N.J.:
Prentice-
Sec. 1-3
Significant Figures
3
The example illustrates that conformity is a necessary, but not condition for precision because of the lack of significant figures obtained. Similarly, precision is a necessary, but not sufficient, condition for accuracy.
precision error. sufficient,
Too
often the beginning student
at face value.
He
is
is inclined to accept instrument readings not aware that the accuracy of a reading is not necessarily
guaranteed by its precision. In fact, good measurement technique demands continuous skepticism as to the accuracy of the results. In critical work, good practice dictates that the observer make an inde-
pendent
set
of measurements, using different instruments or different measure-
ment techniques, not
subject to the
same systematic
errors.
He must
make known
also
sure that the instruments function properly and are calibrated against a
standard, and that no outside influence affects the accuracy of his measurements.
SIGNIFICANT FIGURES
1-3
An
indication of the precision of the
measurement
of significant figures in which the result
is
actual information regarding the magnitude
a quantity.
The more
For example, resistor
if
a resistor
ft
described as 68.0
is
it is
and the measurement precision of
is
specified as having a resistance of 68
H
The
Cl, it
means
that
its
H,
its
or 69 H. If the value of the
resistance
is
closer to 68.0 fl
ft.
In 68 ft there are two significant figures; in 68.0
latter,
with more significant figures, expresses a mea-
to 67.9 fl or 68.1
there are three.
obtained from the number
significant figures, the greater the precision of measurement.
resistance should be closer to 68 Cl than to 67
than
is
expressed. Significant figures convey
surement of greater precision than the former.
number of digits may not represent measurement numbers with zeros before a decimal point are used approximate populations or amounts of money. For example, the population Often, however, the total
precision. Frequently, large for
of a city
is
reported in six figures as 380,000. This
may imply
that the true value
between 379,999 and 380,001, which is six significant figures. What is meant, however, is that the population is closer to 380,000 than to 370,000 or 390,000. Since in this case the population can be reported only of the population
to
two
significant figures,
A X
10
5 .
figures.
lies
more
how can
large
numbers be expressed?
technically correct notation uses powers of ten, 38
This indicates that the population figure Uncertainty caused by zeros to the
left
is
4
X
10 or 3.8
only accurate to two significant
of the decimal point
is
therefore
usually resolved by scientific notation using powers of ten. Reference to the
would cause no misunderstanding 5 anyone with a technical background. But 1.86 X 10 mi/s leaves no confusion. It is customary to record a measurement with all the digits of which we are sure nearest to the true value. For example, in reading a voltmeter, the voltage may be read as 117.1 V. This simply indicates that the voltage, read by
velocity of light as 186,000 mi/s, for example, to
Measurement and
4
Chap.
Error
1
is closer to 117.1 V than to 117.0 V or 117.2 expressing this result indicates the range of possible error. The voltage may be expressed as 117.1 ± 0.05 V, indicating that the value of the voltage lies between 117.05 V and 117.15 V.
the observer to best estimation,
way of
V. Another
When
a
number of independent measurements
are taken in an effort to
obtain the best possible answer (closest to the true value), the result
expressed as the arithmetic
mean of
all
usually
is
the readings, with the range of possible
error as the largest deviation from that mean. This
illustrated in
is
Exam-
ple 1-1.
Example
A
set
1-1
of independent voltage measurements taken by four observers was recorded as
117.02 V, 117.11 V, 117.08 V, and 117.03 V. Calculate (a) the average voltage, (b) the range of error.
Solution
E + E + E + EA E„ = 2
x
(a)
117.02
=
3
+
+
117.11
117.08
+
117.03
.
=
117.06
=
0.05
V
4
= E max - E =
Range
(b)
av
117.11
-
117.06
V
but also
£ - E mn = av
The average range of
117.06
-
=
117.02
0.04
V
error therefore equals
+
0.05
0.04
=
±0.045
=
±0.05
V
2
When two added, the result that
two
Example
Two
more measurements with dhTerent degrees of accuracy
or is
are
only as accurate as the least accurate measurement. Suppose
resistances are
added
in series as in
Example
1-2.
1-2
R
resistors,
x
and
R
2,
are connected in series. Individual resistance measurements,
using a Wheatstone bridge, give resistance to the appropriate
R =
18.7
l
number of
H
and
R = 2
3.624
ft.
Calculate the total
significant figures.
Solution
R =
18.7 (1 (three significant figures)
R =
3.624
{
2
H
(four significant figures)
R T = R, + R = 2
The doubtful
22.324
fi (five significant figures)
=
22.3
H
R and R 2 no value whatsoever in
figures are written in italics to indicate that in the addition of
the last three digits of the
sum
are doubtful figures. There
is
x
Sec. 1-3
Significant Figures
two
retaining the last
5
digits (the
2 and the 4) because one of the resistances
only to three significant figures or tenths of an ohm.
The
reduced to three significant figures or the nearest tenth,
The number of
i.e.,
22.3
significant figures in multiplication
Example
accurate
CI.
may
increase rapidly,
shown
in
recorded in a resistance of 35.68
ft.
but again only the appropriate figures are retained in the answer, as
Example
is
result should therefore also be
1-3.
1-3
In calculating voltage drop, a current of 3.18
A
is
Calculate the voltage drop across the resistor to the appropriate
number of
significant
figures.
Solution
E =
=
IR
(35.68)
X
(3.1(5)
-
M3.4624
=
113
V
Since there are three significant figures involved in the multiplication, the answer can be written only to a
In four;
maximum
Example
and the
1-3,
of three significant figures.
the current,
/,
has three significant figures and
R
has
result of the multiplication has only three significant figures. This
answer cannot be known to an accuracy greater than the if extra digits accumulate in the answer, they should be discarded or rounded off. In the usual practice, if the (least significant) digit in the first place to be discarded is less than five, it and the following digits are dropped from the answer. This was done in Example illustrates that the least
poorly defined of the factors. Note also that
first place to be discarded is five or greater, the previous by one. For three-digit precision, therefore, 113.46 should be rounded off to 113; and 113.74 to 114. Addition of figures with a range of doubt is illustrated in Example 1-4.
1-3. If
the digit in the
digit is increased
Example
Add
1-4
826 ±
5 to
628 ±
3.
Solution
=
826
±
5
(= ±0.605%)
N =
628
±
3
(
N, 2
Sum =
1,454
±
8
=
±0.477%)
(= ±0.55%)
Example 1-4 that the doubtful parts are added, since the ± sign means that one number may be high and the other low. The worst possible combination of range of doubt should be taken in the answer. The percentage doubt in the Note
in
original figure
N
{
and
N
2
does not
differ greatly
from the percentage doubt
in
the final result. If the
interesting
same two numbers are subtracted, as in Example 1-5, there is an comparison between addition and subtraction with respect to the
range of doubt.
Measurement and
6
Example
Error
Chap.
1
1-5
Subtract 628
±
3
from 826 ±
and express the range of doubt
5
in the
answer as a
percentage.
Solution
=
826
±
5
(= ±0.605%)
N =
628
±
3
(= ±0.477%)
=
198
±
8
(= ±4.04%)
N, 2
Difference
Again, in Example
Example numbers
1-5,
same reason as in and subtraction of the same
the doubtful parts are added for the
Comparing the Examples 1-4 and
1-4.
results of addition
when The final result after subtraction shows a large increase in percentage doubt compared to the percentage doubt after addition. The percentage doubt increases even more when the difference between the numbers is relatively small. Consider the case illustrated in Example 1-6. in
1-5,
note that the precision of the results,
expressed in percentages, differs greatly.
Example
1-6
Subtract 437
± 4 from 462 ± 4 and
express the range of doubt in the answer as a
percentage.
Solution AT,
=
N = 2
Difference
Example
=
1-6 illustrates clearly that
462
±
4
(= ±0.87%)
437 ± 4 (= ±0.92%) 25
±
8
(= ±32%)
one should avoid measurement techniques
depending on subtraction of experimental results because the range of doubt the final result
1-4
may
in
be greatly increased.
TYPES OF ERROR
No measurement
can be made with perfect accuracy, but
out what the accuracy actually
is
and how
different errors
it is
important to find
have entered into the
measurement. A study of errors is a first step in finding ways to reduce them. Such a study also allows us to determine the accuracy of the final test result. Errors may come from different sources and are usually classified under three
main headings: Gross errors: largely
human
errors,
among them misreading
of in-
struments, incorrect adjustment and improper application of instruments,
and computational mistakes.
Sec.
1
Types
-4
7
of Error
Systematic errors: shortcomings of the instruments, such as defective
worn
or
parts,
and
effects
of the environment on the equipment or the
user.
Random errors: those due to causes that cannot be directly established because of random variations in the parameter or the system of measurement.
Each of these will
classes of errors will be discussed briefly
and some methods
be suggested for their reduction or elimination.
Gross Errors
1-4.1
This class of errors mainly covers
human
mistakes in reading or using
instruments and in recording and calculating measurement results.
human
beings are involved,
some gross
As long
as
errors will inevitably be committed.
Although complete elimination of gross errors is probably impossible, one should try to anticipate and correct them. Some gross errors are easily detected; others may be very elusive. One common gross error, frequently committed by beginners in measurement work, involves the improper use of an instrument. In general, indicating instruments change conditions to some extent when connected into a complete circuit, so that the measured quantity is altered by the method employed. For example, a well-calibrated voltmeter
may
give a misleading read-
when connected across two points in a high-resistance circuit (Example 1-7). The same voltmeter, when connected in a low-resistance circuit, may give a more dependable reading (Example 1-8). These examples illustrate that the ing
voltmeter has a "loading effect" on the circuit, altering the original situation by the
measurement process.
Example
A
1-7
voltmeter, having a sensitivity of 1,000 fl/V, reads 100
connected across an unknown resistor
When known effect
the milliammeter reads 5
in series
mA,
resistor, (b) actual resistance of the
V
on
its
150-V scale when
with a milliammeter.
calculate (a) apparent resistance of the un-
unknown
resistor, (c) error
due to the loading
of the voltmeter.
Solution (a)
The
total circuit resistance equals
Kr IT
_woy_ Mkn mA 5
Neglecting the resistance of the milliammeter, the value of the
Rx = (b)
20
ka
The voltmeter
resistance equals
Ry
1,000-
X
150
V =
150
kH
unknown
resistor
is
8
Since the voltmeter
in parallel with the
is
R TR V &x = R„ — R = v T M %
(c)
error
=
actual
—
unknown 20
Measurement and
Error
we can
write
resistance,
X ™ —= 150
Example
1
kH
130
apparent
^
X 100% =
actual
=
23.05
Chap.
———- — X 100% 23.05
20
23.05
13.23%
1-8
Repeat Example 1-7 its 150-V scale.
if
the milliammeter reads 800
mA
and the voltmeter reads 40
V
on
Solution
VT
40
V
(b)
R v = 1,000^ X
(c)
%
error
=
150
——50.1
50
V =
150
kH
X 100% - 0.2%
Errors caused by the loading effect of the voltmeter can be avoided by it intelligently. For example, a low-resistance voltmeter should not be used measure voltages in a vacuum tube amplifier. In this particular measurement, a high-input impedance voltmeter (such as a VTVM or TVM) is required. A large number of gross errors can be attributed to carelessness or bad habits, such as improper reading of an instrument, recording the result differently from the actual reading taken, or adjusting the instrument incorrectly. Consider the case in which a multirange voltmeter uses a single set of scale markings
using to
with different number designations for the various voltage ranges.
It is
easy to
use a scale which does not correspond to the setting of the range selector of the voltmeter.
A
gross error
before the measurement
may is
also occur
taken; then
when
all
the instrument
the readings are
is
not set to zero
off.
Errors like these cannot be treated mathematically. They can be avoided only by taking care in reading and recording the measurement data.
Good
making more than one reading of the same quantity, preferably by a different observer. Never place complete dependence on one reading but take at least three separate readings, preferably under conditions in which inpractice requires
struments are switched off-on.
Types
Sec. 1-4
9
of Error
1-4.2 Systematic Errors This type of error
is
usually divided into
two
different categories: (1)
instrumental errors, defined as shortcomings of the instrument; (2) environmental errors, due to external conditions affecting the measurement.
Instrumental errors are errors inherent in measuring instruments because mechanical structure. For example, in the d'Arsonval movement friction
of their
in bearings of various
moving components may cause incorrect
readings. Irreg-
ular spring tension, stretching of the spring, or reduction in tension due to
improper handling or overloading of the instrument
will result in errors.
Other
instrumental errors are calibration errors, causing the instrument to read high or low along a
its
entire scale. (Failure to set the instrument to zero before
measurement has a similar effect.) There are many kinds of instrumental
errors,
making
depending on the type of
instrument used. The experimenter should always take precautions to insure that the instrument he
is
using
is
operating properly and does not contribute
may be detected by checking for erratic behavior, and stability and reproducibility of results. A quick and easy way to check an instrument is to compare it to another with the same characteristics or to one that is known to be more accurate. excessive errors for the purpose at hand. Faults in instruments
may be avoided by (1) selecting a suitable instrument measurement application; (2) applying correction factors after determining the amount of instrumental error; (3) calibrating the instrument Instrumental errors
for the particular
against a standard.
Environmental errors are due to conditions external to the measuring device, including conditions in the area surrounding the instrument, such as the effects of
changes
or electrostatic
instrument
is
in temperature,
fields.
Thus
humidity, barometric pressure, or of magnetic
a change in ambient temperature at which the
used causes a change in the elastic properties of the spring in a
moving-coil mechanism and so affects the reading of the instrument. Corrective
measures to reduce these certain
components
effects include air conditioning, hermetically sealing
in the instrument, use of
magnetic
shields,
and the
like.
Systematic errors can also be subdivided into static or dynamic errors. Static errors are
laws governing
caused by limitations of the measuring device or the physical behavior. A static error is introduced in a micrometer when
its
excessive pressure
is
applied in torquing the shaft.
by the instrument's not responding measured variable.
1-4.3
Random
fast
enough
Dynamic
errors are caused
to follow the changes in a
Errors
These errors are due to unknown causes and occur even when all systematic errors have been accounted for. In well-designed experiments, few random errors
Measurement and
10
usually occur, but they
voltage
is
become important
in high-accuracy work.
being monitored by a voltmeter which
Although the instrument
is
is
Chap.
Error
1
Suppose a
read at half-hour intervals.
operated under ideal environmental conditions and
has been accurately calibrated before the measurement,
it
will
be found that the
readings vary slightly over the period of observation. This variation cannot be corrected by any it
method of
known method
calibration or other
of control and
cannot be explained without minute investigation. The only way to
offset these
by increasing the number of readings and using statistical means to obtain the best approximation of the true value of the quantity under measureerrors
is
ment.
1-5
A
STATISTICAL ANALYSIS
statistical analysis
of measurement data
is
common
an analytical determination of the uncertainty of the of a certain measurement method
may
ments
is
allows
all
the disturbing factors.
To make
methods and interpretations meaningful, a large number of measureusually required. Also, systematic errors should be small compared
with residual or random errors, because
remove a
it
The outcome
be predicted on the basis of sample data
without having detailed information on statistical
practice because
final test result.
fixed bias contained in all the
1-5.1 Arithmetic
statistical
treatment of data cannot
measurements.
Mean
The most probable value of a measured variable is the arithmetic mean of number of readings taken. The best approximation will be made when the number of readings of the same quantity is very large. Theoretically, an infinite number of readings would give the best result, although in practice, only a finite number of measurements can be made. The arithmetic mean is given by the
the
following expression:
_ X
=
x
x
+
x2
+
x3
+
x4
4-
.
.
.
+
xn
n
x=
where
arithmetic
=
1x —
.
(1-1)
n
mean
x v x y x n = readings taken
n= Example
1-1
number of readings
showed how the arithmetic mean
1-5.2 Deviation from the Deviation
is
is
used.
Mean
the departure of a given reading from the arithmetic
of the group of readings. If the deviation of the
first
reading,
x
lt
is
mean d
called
x
,
Sec. 1-5
11
Statistical Analysis
and that of the second reading, x2 from the mean can be expressed as ,
d
=
x
x
l
—
d2
x
is
called
=
x2
d 2i and
—
so on, then the deviations
dn
x
=
—
xn
x
(1-2)
Note that the deviation from the mean may have a positive or a negative value and that the algebraic sum of all the deviations must be zero. Example 1-9 illustrates the computation of deviations. Example
A
set
12.8
of independent current measurements was taken by six observers and recorded as
mA,
mean,
1-9
12.2
mA,
12.5
mA, 13.1mA,
(b) the deviations
12.9
mA, and
12.4
mA.
Calculate (a) the arithmetic
from the mean.
Solution (a)
Using Eq.
x
we
(1-1),
=
12.8
+
mean
see that the arithmetic
12.2
+
+
12.5
13.1
+
equals
+
12.9
12.4
=
12.65
mA
6 (b)
Using Eq.
(1-2),
we
see that the deviations are
=
12.8
-
12.65
=
d2
=
12.2
-
12.65
=
d3
=
12.5
-
12.65
=
dA
=
13.1
-
12.65
=
d
5
=
12.9
-
12.65
=
d6
=
12.4
-
12.65
=
d
x
Note that the algebraic sum of 1-5.3
all
mA -0.45 mA -0.15 mA 0.45 mA 0.25 mA -0.25 mA
0.15
the deviations equals zero.
Average Deviation
The average
deviation
an indication of the precision of the instruments
is
used in making the measurements. Highly precise instruments will yield a low
By
average deviation between readings.
definition, average deviation
of the absolute values of the deviations divided by the absolute value of the deviation deviation
may
number of
is
the
readings.
sum The
the value without respect to sign. Average
is
be expressed as
D=
\d,\+\d 1
\
+ \d
i
\+
...
+K| m
i\d\ _
n
n
Example 1-10 shows how average deviation
is
calculated.
Example 1-10 Calculate the average deviation for the data given in
Example
1-9.
Measurement and
12
Error
Chap.
1
Solution
+
0.15
D =
0.45
+
0.15
+
+
0.45
+
0.25
0.25
=
mA
0.283
1-5.4 Standard Deviation In statistical analysis of
standard deviation cr
of an infinite
is
random
errors, the
a very valuable aid.
number of data
is
By
root-mean-square deviation or
definition, the standard deviation
the square root of the
deviations squared, divided by the
number of
sum
of all the individual
readings. Expressed mathemati-
cally:
In practice, of course, the possible
number of observations
standard deviation of a finite number of data
Equation (1-5)
will
is
is
4-
.
.
.
+
(1 " 5)
essentially the is
the
same
same quantity
is
the variance or
as the standard deviation except that
not extracted. Therefore (
V)
= mean
square deviation
a convenient quantity to use in
The standard
scientific results are
now
=
cr
1
many computations because
deviation, however, has the advantage of
being of the same units as the variable, making
1-6
\^d]
d\
be used in Example 1-11.
variances are additive.
Most
The
I
variance
The variance
d\
is finite.
given by
V
Another expression for mean square deviation, which the square root
+
f d\
Id]
-
is
it
easy to compare magnitudes.
stated in terms of standard deviation.
PROBABILITY OF ERRORS 1-6.1
Normal Distribution of Errors
Table
l-l
shows a tabulation of 50 voltage readings
small time intervals and recorded to the nearest 0.1 V. the measured voltage
that
were taken
at
The nominal value of
was 100.0 V. The result of this series of measurements in the form of a block diagram or histogram in
can be presented graphically
which the number of observations reading.
The histogram of
is
plotted against each observed voltage
Fig. 1-1 represents the data of
Table
1-1.
Sec. 1-6
13
Probability of Errors
TABLE
1-1
Tabulation of Voltage Readings
Voltage reading
Number
(volts)
of readings
99.7
1
99.8
4
99.9
12
100.0
19
100.1
10
100.2
3
100.3
1
50
Figure
1-1
shows that the
largest
number of readings
(19) occurs at the
central value of 100.0 V, while the other readings are placed
more or
less
symmetrically on either side of the central value. If more readings were taken at smaller increments, say
200 readings
at
0.05-V intervals, the distribution of
observations would remain approximately symmetrical about the central value
and the shape of the histogram would be about the same as before. With more and more data, taken at smaller and smaller increments, the contour of the histogram would finally become a smooth curve, as indicated by the broken line in Fig. 1-1. This bell-shaped curve is known as a Gaussian curve. The sharper and narrower the curve, the more definitely an observer may state that the most is the central value or mean reading. Normal law of error forms the basis of the analytical The Gaussian or mathematical treatment of this subject effects. Although the study of random
probable value of the true reading
Number
19
of
Observed Readings
12 io
i—1\
x oo
a>
a>
&>&>&>
Figure 1-1
o o
1-1.
o o
o o
•Volts
Histogram showing the frequency of occurrence of the 50 voltage readings
The broken curve represents the number of readings at small increments
of Table large
q d o
1
limiting case of the histogram
are taken.
when
a
Measurement and
14
is beyond the scope of on the Normal law:
(a) All
We
this text, the following qualitative statements are
Random
(c)
There
is
1
based
random
errors.
an equal probability of positive and negative random
errors.
observations include small disturbing
(b)
Chap.
Error
effects, called
errors can be positive or negative.
can therefore expect that measurement observations include plus and minus more or less equal amounts, so that the total error will be small and
errors in
the
mean The
value will be the true value of the measured variable. possibilities as to the
form of the error distribution curve can be stated
as follows:
(a)
Small errors are more probable than large errors.
(b)
Large errors are very improbable.
(c)
There
is
an equal probability of plus and minus errors so that the
probability of a given error will be symmetrical about the zero value.
The
error distribution curve of Fig. 1-2
is
based on the Normal law and shows
may
a symmetrical distribution of errors. This normal curve limiting
form of the histogram of
the true voltage
is
the
mean
Fig. 1-1 in
be regarded as the
which the most probable value of
value of 100.0 V.
1-6.2 Probable Error
The area under limits
+
oo
and
Figure 1-2
—
oo
Curve
the Gaussian probability curve of Fig. 1-2, between the ,
represents the entire
for the
probable error, where r
=
Normal
law.
±0.6745o-.
number of
The shaded portion
observations.
The area
indicates the region of
Sec. 1-6
15
Probability of Errors
under the curve between the + cr and — cr limits represents the cases that differ from the mean by no more than the standard deviation. Integration of the area under the curve within the ± or limits gives the total number of cases within these limits. For normally dispersed data, following the Gaussian distribution, approximately 68 per cent of all the cases lie between the limits of +
WW-:
R2 V 2 ¥rZero Adjust
1.5V
^22,988
^
ft'
Negative
Positive
(a)
Ohmmeter
Rx1 Range
Circuit
M
—wa
l.5ft
M38ft 21,850ft ^AAAr^^^v^AA
lOkft
Zero Adjust
~M49.5ft
I.5V
Negative,,
Positive (b)
Ohmmeter
Circuit
Rx 100 Range
--w-
11.5ft
"w—J\AAA/
21,850ft
1138ft
WA
lOkft^
WvV^r^VVvV-
22,999.5
ft'
Negative (c)
Figure 4-27
Ohmmeter
Ohmmeter
section of the
Circuit
R
x
10,000 Range
ft**
Simpson Model 260 multimeter (courtesy
Simpson Electric Company).
4-11 CALIBRATION
OF DC INSTRUMENTS
Although detailed calibration techniques are beyond the scope of this chapter, some general procedures for the calibration of basic dc instruments are given. Calibration of a dc ammeter can most easily be carried out by the arrangement of Fig. 4-28. The value of the current through the ammeter to be calibrated is determined by measuring the potential difference across a standard resistor by the potentiometer method and then calculating the current by Ohm's
Sec. 4-12
Alternating-Current Indicating Instruments
85
Rheostat
dc
Ammeter
Constant Source
Under Test
R -AAA/V
Standard Resistor
Figure 4-28
Potentiometer method of cal-
ibrating a dc
ammeter.
Potentiometer
The result of this calculation is compared to the actual reading of the ammeter under calibration and inserted in the circuit. (Voltage measurements by the potentiometer method are discussed in Sec. 5-13.) A good source of constant current is required and is usually provided by storage cells or a precision power supply. A rheostat is placed in the circuit to control the current to any desired value, so that different points on the meter scale can be calibrated. A simple method of calibrating a dc voltmeter is shown in Fig. 4-29, where the voltage across dropping resistor R is accurately measured with a potentiometer. The meter to be calibrated is connected across the same two points as the potentiometer and should therefore indicate the same voltage. A rheostat is placed in the circuit to control the amount of current and therefore the drop across the resistor, R, so that several points on the voltmeter scale can be calibrated. Voltmeters tested with the method of Fig. 4-29 can be calibrated with an accuracy of ±0.01 per cent, which is well beyond the usual accuracy of a d'Arsonval movement. The ohm meter is generally considered to be an instrument of moderate law.
accuracy and low precision.
A
rough calibration may be done by measuring a
standard resistance and noting the reading of the ohmmeter. Doing this for several points
on the ohmmeter
scale
and on several ranges allows one
to obtain
an indication of the correct operation of the instrument.
-^rtZ— Rheostat
Regulated dc Source
Figure 4-29
©Voltmeter Under
Potentiometer
Test
Potentiometer method of calibrating a dc voltmeter.
4-12 ALTERNATING-CURRENT INDICATING INSTRUMENTS The d'Arsonval movement responds to the average or dc value of the current through the moving coil. If the movement carries an alternating current with
Electromechanical Indicating Instruments
86
positive
and negative
half-cycles, the driving torque
for the positive alternation
and
If the frequency of the ac
is
is
in
one direction
in the other direction for the negative alternation.
very low, the pointer would swing back and forth
around the zero point on the meter the coil
would be
Chap. 4
scale.
At higher
frequencies, the inertia of
so great that the pointer cannot follow the rapid reversals of the
driving torque and hovers around the zero mark, vibrating slightly.
To measure
ac on a d'Arsonval movement,
some means must be devised
to obtain a unidirectional torque that does not reverse each half-cycle.
One
method involves rectification of the ac, so that the rectified current deflects the coil. Other methods use the heating effect of the alternating current to produce an indication of its magnitude. Some of these methods are discussed in this chapter.
4-12.1 Electrodynamometer
One
of the most important ac movements
is
the electrodynamometer.
It is
and ammeters, not only at the powerline frequency but also in the lower audiofrequency range. With some slight modifications, the electrodynamometer can be used as a wattmeter, a VARmeter, a power-factor meter, or a frequency meter.' The electrodynamometer movement may also serve as a transfer instrument, because it can be calibrated on dc and then used directly on ac, establishing a direct means of equating ac and dc measurements of voltage and current. Where the d'Arsonval movement uses a permanent magnet to provide the magnetic field in which the movable coil rotates, the electrodynamometer uses the current under measurement to produce the necessary field flux. Figure 4-30 shows a schematic arrangement of the parts of this movement. A fixed coil, split into two equal halves, provides the magnetic field in which the movable coil rotates. The two coil halves are connected in series with the moving coil
often used in accurate ac voltmeters
Scale
Fixed Coils Figure 4-30
Schematic diagram of an electrodynamometer movement.
Sec. 4-12
Alternating-Current Indicating Instruments
87
and are fed by current under measurement. The fixed coils are spaced far enough apart to allow passage of the shaft of the movable coil. The movable coil carries a pointer, which is balanced by counterweights. Its rotation is controlled by springs, similar to the d'Arsonval movement construction. The complete assembly is surrounded by a laminated shield to protect the instrument from stray magnetic fields which may affect its operation. Damping is provided by aluminum air vanes, moving in sector-shaped chambers. The entire movement is very solid and rigidly constructed in order to keep its mechanical dimensions stable and its calibration intact. A cutaway view of the electrodynamometer is shown in Fig. 4-31.
The operation of the instrument may be understood by returning expression for the torque developed by a coil suspended in a magnetic
to the
field.
We
previously stated, Eq. (4-1), that
T=BXA XlXN which deflects the movable coil, is directly propor(A and N), the strength of the magnetic field in moves (B), and the current through the coil (/). In the electrowhich the coil (B) depends on the current through the fixed coil dynamometer the flux density indicating that the torque,
tional to the coil constants
and
is
coil
dimensions and the number of turns on the
therefore directly proportional to the deflection current (/). Since the
Figure 4-31
ment of
fixed
coil
frame are fixed quantities
Phantom photograph of an electrodynamometer, showing the arrangeand movable coils. The rigidly constructed mechanism is surrounded
by a laminated shield to minimize the indication (courtesy
effect
Weston Instruments,
of external magnetic fields on the meter
Inc.).
88
Electromechanical Indicating Instruments
Chap. 4
any given meter, the developed torque becomes a function of the current
for
squared (I
2 ).
If the
law scale
is
electrodynamometer easily noticed, with
exclusively designed for dc use,
is
crowded
values, progressively spreading out at
developed torque squared
at
any instant
is
markings
its
square-
low current the higher current values. For ac use, the scale
at the very
proportional to the instantaneous current
2
The instantaneous value of i 2 is always positive and torque pultherefore produced. The movement, however, cannot follow the rapid
(i ).
sations are
variations of the torque
and takes up a position
in
which the average torque
balanced by the torque of the control springs. The meter deflection a function of the
mometer
is
mean
of the squared current.
The
is
is
therefore
scale of the electrodyna-
usually calibrated in terms of the square root of the average current
squared, and the meter therefore reads the rms or effective value of the ac.
The transfer properties of the electrodynamometer become apparent when we compare the effective value of alternating current and direct current in terms of their heating effect or transfer of power.
An
alternating current that produces
heat in a given resistance at the same average rate as a direct current (7) has,
by
definition, a value of
/ amperes. The average rate of producing heat by a dc R is PR watts. The average rate of producing heat
of / amperes in a resistance
by an ac of
By 1
amperes during one cycle
i
in
the
same
R
resistance
is
definition, therefore,
Jo
and
2 i
This current,
/, is
a
mark
is
=
^average
is
often referred to as the equivalent dc value.
electrodynamometer
is
calibrated with a direct current of
have an rms value of its
1
A
and
placed on the scale to indicate this 1-A dc value, then that alternating
current which causes the pointer to deflect to the same
to
2
i
then called the root-mean-square (rms) or effective value of
the alternating current and If the
dt
1
A.
We
mark on
the scale
must
can therefore "transfer" a reading made with dc
corresponding ac value and have thereby established a direct connection
between ac and
dc.
The electrodynamometer then becomes very
calibration instrument
and
is
often used for this purpose because of
useful as a its
inherent
accuracy.
is its
The electrodynamometer, however, has certain disadvantages. One of these high power consumption, a direct result of its construction. The current
under measurement must not only pass through the movable coil, but it must To get a sufficiently strong magnetic field, a high
also provide the field flux.
Sec. 4-12
mmf is
Alternating-Current Indicating Instruments
89
required and the source must supply a high current and power. In spite
of this high power consumption, the magnetic
field is
very
much weaker
than
movement because there is no iron in the path consists of air. Some instruments have been
that of a comparable d'Arsonval circuit,
i.e.,
the entire flux
designed using special laminated steel for part of the flux path, but the presence of metal introduces calibration problems caused by frequency and waveform
Typical values of electrodynamometer flux density are in the range of approximately 60 gauss. This compares very unfavorably with the high flux densities (1,000-4,000 gauss) of a good d'Arsonval movement. The low flux effects.
density of the electrodynamometer immediately affects the developed torque and therefore the sensitivity of the instrument
The addition of a
is
typically very low.
series resistor converts the
electrodynamometer into a
voltmeter, which again can be used to measure dc and ac voltages. For reasons
previously mentioned, the sensitivity of the electrodynamometer voltmeter
meter).
The reactance and
is
H/V (compare this to the 20 kfl/V of a d'Arsonval
low, approximately 10 to 30
resistance of the coils also increase with increasing
frequency, limiting the application of the electrodynamometer voltmeter to the
lower frequency ranges.
and
It is,
however, very accurate at the powerline frequencies
therefore often used as a secondary standard.
is
The electrodynamometer movement (even unshunted) may be regarded as it becomes rather difficult to design a moving coil which can carry more than approximately 100 mA. Larger current would have to be carried to the moving coil through heavy lead-in wires, which would lose their flexibility. A shunt, when used, is usually placed across the movable coil only. The fixed coils are then made of heavy wire which can carry the large total current and it is feasible to build ammeters for currents up to 20 A. Larger values of ac an ammeter, but
currents are usually measured by using a current transformer and a standard
5-A ac ammeter -
(Sec. 4-17).
4-12.2 Rectifier-type Instruments
One obvious answer
to the question of ac
measurement
is
found by using
a rectifier to convert ac into a unidirectional dc and then to use a dc
method
movement
is
very attractive, because
a dc movement generally has a higher sensitivity than mometer or the moving-iron instrument.
either the electrodyna-
to indicate the value of the rectified ac. This
PMMC movement in combisome rectifier arrangement. The rectifier element usually consists of a germanium or a silicon diode. Copper oxide and selenium rectifiers have become obsolete, because they have small inverse voltage ratings and can handle only limited amounts of current. Germanium diodes have a peak inverse voltage Rectifier-type instruments generally use a
nation with
(PIV) on the order of 300
Low-current rating
V
and a current rating of approximately 100 mA. have a PIV of up to 1,000 V and a current
silicon diode rectifiers
on the order of 500 mA.
90
Electromechanical Indicating Instruments
Rectifiers for instrument
work sometimes
Chap. 4
consist of four diodes in a bridge
shows an ac voltand a PMMC move-
configuration, providing full-wave rectification. Figure 4-32
meter circuit consisting of a multiplier, a bridge
rectifier,
ment.
The bridge rectifier produces a pulsating unidirectional current through movement over the complete cycle of the input voltage. Because of inertia of the moving coil, the meter will indicate a steady deflection pro-
the meter the
portional to the average value of the current. Since alternating currents and voltages are usually expressed in
rms
values, the
meter scale
is
calibrated in
terms of the rms value of a sinusoidal waveform.
Example 4-9 4-32(a), where the PMMC movement and requires a dc current of 1 mA for full-scale deflection. Assuming ideal diodes (zero forward resistance and infinite reverse resistance), calculate the value of the multiplier R s to obtain full-scale meter deflection with 10 V ac
An experimental ac voltmeter uses the circuit of Fig. has an internal resistance of 50
H
(rms) applied to the input terminals. Solution
For full-wave
rectification,
(b) Rectified Current
Figure 4-32
Through Meter Movement Full-wave
rectifier
ac voltmeter.
Sec. 4-12
Alternating-Current Indicating Instruments
91
and
E = dc
The
X
0.9
10
total circuit resistance, neglecting the
R,
V =
9 1
A
n -
V
mA
s
9,000
V
forward diode resistance,
= R + R
r=
9
50
n =
=
kn
9
n
8,950
nonsinusoidal waveform has an average value that
erably from the average value of a pure sine
is
wave
may
differ consid-
which the meter is calibrated) and the indicated reading may be very erroneous. The form factor relates the average value and the rms value of time- varying voltages and currents: (for
effective value of the ac
form factor
wave
average value of the ac wave
For a sinusoidal waveform: form factor
(y/2/2)E,
=
=
1.11
(4-27)
(2/ir)E n
Note that the voltmeter of Example 4-9 has a scale suitable only for sinusoidal The form factor of Eq. (4-27) is therefore also the factor by which the actual (average) dc current is multiplied to obtain the equivalent rms ac measurements. scale markings.
The
ideal rectifier element should
resistance. In practice,
however, the
the characteristic curves of Fig. 4-33. rectifier
have zero forward and
rectifier is
At low values of forward
operates in an extremely nonlinear part of
the resistance
is
large as
The lower part of the
compared
is
therefore often crowded, scale, calibrated espe-
resistance in the early part of the rectifier
Current
1
-e Keverse Current
'
f /
J.
Figure 4-33
and
to the resistance at higher current values.
ac scale of a low-range voltmeter
The high
current, the
characteristic curve,
its
and most manufacturers provide a separate low-voltage cially for this purpose.
infinite reverse
a nonlinear device, indicated by
Forward
Reverse
Forward
Voltage
Current
Current
Characteristic curves of a solid-state rectifier.
Electromechanical Indicating Instruments
92
on the
characteristics also sets a limit
sensitivity
Chap. 4
which can be obtained
in
microammeters and voltmeters.
The
resistance of the rectifying element changes with varying temperature,
one of the major drawbacks of rectifier-type ac instruments. The meter accuracy is usually satisfactory under normal operating conditions at room temperature
and is generally on the order of ± 5 per cent of full-scale reading for sinusoidal waveforms. At very much higher or lower temperatures, the resistance of the rectifier
changes the total resistance of the measuring circuit
sufficiently to cause
the meter to be gravely in error. If large temperature variations are expected, the meter should be enclosed in a temperature-controlled box.
Frequency also
affects the operation
exhibits capacitive properties
readings
may
and tends
be in error by as
much
of the rectifier elements.
The
rectifier
to bypass the higher frequencies.
as 0.5 per cent decrease for every
Meter 1-kHz
rise in frequency.
4-12.3 Typical Multimeter Circuits General rectifier-type ac voltmeters often use the arrangement shown in Fig. 4-34.
the
Two
movement
Diode
D
x
diodes are used in this circuit, forming a full-wave rectifier with so connected that
it
receives only half of the rectified current.
conducts during the positive half-cycle of the input waveform and
The draw more current
causes the meter to deflect according to the average value of this half-cycle.
meter movement
shunted by a resistance
is
through the diode
D
x
and move
its
R
in order to
sh ,
operating point into the linear portion of
the characteristic curve. In the absence of diode Z> 2 the negative half-cycle of ,
the input voltage
would apply a reverse voltage across diode
leakage current in the reverse direction.
would therefore be lower than deals with this problem.
On
it
D
x ,
causing a small
The average value of the complete
cycle
should be for half-wave rectification. Diode
the negative half-cycle,
the current through the measuring circuit, which
is
D
now
2
D
2
conducts heavily, and
in the opposite direction,
bypasses the meter movement.
The commercial multimeter its
often uses the
same
scale
markings for both
dc and ac voltage ranges. Since the dc component of a sine wave for
wave
rectification equals 0.45 times the
rms
value, a
problem
Selector
Figure 4-34
half-
arises immediately.
Typical ac voltmeter section of a commercial multimeter.
1
Sec. 4-12
Alternating-Current Indicating Instruments
93
In order to obtain the same deflection on corresponding dc and ac voltage ranges, the multiplier for the ac range must be lowered proportionately. Fig. 4-35 illustrates a solution to the
Example
in
problem and
is
The circuit of some detail
discussed in
4-10.
AAAAr
1D2 A
ac Input
Computation of the multi-
Figure 4-35
and the ac voltmeter
plier resistor
sensitiv-
ity.
Example 4-10
A
meter movement has an internal resistance of 100
R
scale deflection. Shunting resistor
H. Diodes to
have
D
x
and
D
Cl
and requires
1
mA
dc for
full-
placed across the movement, has a value of 100
sh ,
have an average forward resistance of 400
2
CL each and are assumed For a 10-V ac range, calculate (a) the the voltmeter sensitivity on the ac range.
infinite resistance in the reverse direction.
value of multiplier
R
(b)
s ,
Solution (a)
Since
Rm
and
scale deflection
R
are both 100 H, the total current the source
ih
is I,
=
2
rectified ac voltage will
mA. For
total resistance of the
is
0.45
X
instrument circuit then
R=E f This total resistance
made up
4.5
2
10
V =
V
V
a
2,250
mA
4.5
is
we are interested only in the movement receives current, we can
of several parts. Since
resistance of the circuit during the half-cycle that the
eliminate the infinite resistance of reverse-biased diode
R,
D
2
from the
R„R
= R + R D] + s
R„ + R
and
R = R + s
The value of
100
400
the multiplier therefore s
The
4-
X
100
= R + s
200
R = (b)
for full-
be
= The
must supply
half-wave rectification the equivalent dc value of the
-
2,250
sensitivity of the voltmeter
_
is
on
this
n V
450
=
10-V ac range
2,250 10
1,800 Cl
225
n/v
is
450
ft
circuit.
Therefore
94
Electromechanical Indicating Instruments
The same movement, used 1,000
in a
Chap. 4
dc voltmeter, would have given a sensitivity figure of
n/v. Section 4-10 dealt with the dc circuitry of a typical multimeter, using the
diagram of Fig. 4-24. The circuit for measuring ac volts (subfrom Fig. 4-24) is reproduced in Fig. 4-36. Resistances R 9 R n R 7 and R 6 form a chain of multipliers for the voltage ranges of 1,000 V, 250 V, 50 V, and 10 V, respectively, and their values are indicated in the diagram of Fig. 4-36. On the 2.5-V ac range, resistor R 23 acts as the multiplier and corresponds simplified circuit
tracted
to the multiplier
,
R
s
of
Example 4-10 shown
in Fig. 4-35. Resistor
,
,
R 24
is
the
meter shunt and again acts to improve the rectifier operation. Both values are unspecified in the diagram and are factory selected. A little thought, however, will
convince us that the shunt resistance could be 2,000
resistance. If the average
forward resistance of the
(a reasonable assumption), then resistance
R 2i
ft,
rectifier
equal to the meter
elements
is
500
must have a value of 1,000
This follows because the meter sensitivity on the ac ranges
is
ft ft.
given as 1,000 ft/
V; on the 2.5-V ac range, the circuit must therefore have a total resistance of ft. This value is made up of the sum of R 23 the diode forward resistance, and the combination of movement and shunt resistance, as shown in Example
2,500
,
4-10. Three-terminal Rectifier
Figure 4-36
Multirange ac voltmeter circuit of the Simpson Model 260 multimeter
(courtesy of the Simpson Electric
Company).
4-13 THERMOINSTRUMENTS 4-13.1 Thermocouple Instrument Figure 4-37 shows a combination of a thermocouple and a
ment
that can be used to
measure both ac and
a thermocouple instrument, since
thermocouple element.
When two
its
operation
dc. This is
PMMC move-
combination
is
called
based on the action of the
dissimilar metals are mutually in contact, a
Sec. 4-13
Thermoinstruments
95
PMMC
Movement
Figure 4-37
Schematic representation of a basic thermocouple instrument using
thermocouple
CDE
and a
PMMC
movement.
generated at the junction of the two dissimilar metals. This voltage
voltage
is
rises in
proportion to the temperature of the junction. In Fig. 4-37,
represent the two dissimilar metals, joined at point E, and are
CE and DE
drawn
as a light
and a heavy line, to indicate dissimilarity. The potential difference between C and D depends on the temperature of the so-called cold junction, E. A rise in temperature causes an increase in the voltage and this is used to advantage in the thermocouple. Heating element AB, which is in mechanical contact with the junction of the two metals at point E, forms part of the circuit in which the current is to be measured. AEB is called the hot junction. Heat energy generated by the current in the heating element raises the temperature of the cold junction and causes an increase in the voltage generated across terminals C and D. This potential difference causes a dc current through the PMMC-indicating instrument. The heat generated by the current is directly proportional to the current 2 squared (I R), and the temperature rise (and hence the generated dc voltage) is proportional to the square of the rms current. The deflection of the indicating instrument will therefore follow a square-law relationship, causing crowding at the lower end of the scale and spreading at the high end. The arrangement of Fig. 4-37 does not provide compensation for ambient temperature changes. The compensated thermoelement, shown schematically in Fig. 4-38, produces a thermoelectric voltage in thermocouple CED, which is directly proportional to the current through circuit AB. Since the developed couple voltage is a function of the temperature difference between its hot and cold ends, this temperature difference must be caused only by the current being measured. Therefore, for accurate measurements, points C and D must be at the mean temperature of points A and B. This is accomplished by attaching couple ends
Figure 4-38
Compensated thermocouple
measure the thermo voltage produced by current i alone. Couple terminals C and D to
are in thermal contact with heater terminals
A and
from them.
B, but are electrically insulated
PMMC
96
C
Electromechanical Indicating Instruments
D
Chap. 4
whose ends are in thermal from them. Self-contained thermoelectric instruments of the compensated type are available in the 0.5-20- A range. Higher current ranges are available, but in this case the heating element is external to the indicator. Thermoelements used for current ranges over 60 A are generally provided with air cooling fins. Current measurements in the lower ranges, from approximately 0.1-0.75 A, use a bridge-type thermoelement, shown schematically in Fig. 4-39. This arrangement does not use a separate heater: the current to be measured passes directly through the thermoelements and raises their temperature in proportion to PR. The cold junctions (marked c) are at the pins which are embedded in the insulating frame, and the hot junctions (marked h) are at splices midway between the pins. The couples are arranged as shown in Fig. 4-39, and the resultant thermal voltage generates a dc potential difference across the indicating instrument. Since the bridge arms have equal resistances, the ac voltage across the meter is 0 V, and no ac passes through the meter. The use of several thermocouples in series provides a greater output voltage and deflection than is possible with a single element, resulting in an instrument with increased and
to the center of separate
contact with
A and
copper
strips,
B, but electrically insulated
sensitivity.
c
Figure 4-39
Bridge-type thermocouple instrument.
Thermoinstruments may be converted into voltmeters using low-current series resistors. Thermocouple voltmeters are available in ranges of up to 500 V and sensitivities of approximately 100 to 500 ft/V. A major advantage of a thermocouple instrument is that its accuracy can be as high as 1 per cent, up to frequencies of approximately 50 MHz. For this thermocouples and suitable
reason,
it is
classified as
an
RF instrument.
Above 50 MHz,
the skin effect tends
to force the current to the outer surface of the conductor, increasing the effective
and reducing instrument accuracy. For small is solid and very thin. Above 3 A the made from tubing to reduce the errors due to skin effect.
resistance of the heating wire
currents (up to 3 A), the heating wire
heating element
is
Sec. 4-14
Electrodynamometers
in
Power Measurements
97
ELECTRODYNAMOMETERS IN POWER MEASUREMENTS
4-14
4-14.1 Single-phase Wattmeter
The electrodynamometer movement power.
It
may
used extensively in measuring
is
be used to indicate both dc and ac power for any waveform of
voltage and current and
it is
not restricted to sinusoidal waveforms.
As
described
electrodynamometer used as a voltmeter or an ammeter has and the movable coil connected in series, thereby reacting to the
in Sec. 4-12.1, the
the fixed coils effect
of the current squared.
When
used as a single-phase power meter, the
coils
are connected in a different arrangement (see Fig. 4-40).
Current Coil
-\JJUUUU
1
Potentiol Coil
Diagram of an electrodynamometer wattmeter connected power of a single-phase load.
Figure 4-40 the
The
fixed coils, or field coils,
shown here
connected in series and carry the total
as
to
measure
two separate elements, are The movable coil,
line current (i c ).
located in the magnetic field of the fixed coils,
is
connected in
with a
series
and carries a small current (ip ). The instantaneous value of the current in the movable coil is i p = e/R p where e is the instantaneous voltage across the power line, and R p is the total resistance of the movable coil and its series resistor. The deflection of the movable coil is proportional to the product of these two currents, ic and ip and we can write current-limiting resistor across the
power
line
,
,
for the average deflection over
one period:
0„
=k\\
T ic i
p
dt
(4-28)
Electromechanical Indicating Instruments
98
where
=
0 av
K=
average angular deflection of the coil
instrument constant
ic
=
instantaneous current in the
i
=
instantaneous current in the potential coil
p
Chap. 4
field coils
Assuming for the moment that ic is equal to the load current, i (actually, ic = + /'), and using the value for ip = e/R p we see that Eq. (4-28) reduces to i p ,
(4-29)
By
definition, the average
power
in a circuit is
P„
= if
eidt
(4-30)
which indicates that the electrodynamometer movement, connected
in the con-
figuration of Fig. 4-40, has a deflection proportional to the average power. If e
and
i
are sinusoidally varying quantities of the form e
sin (cot
±
0),
0 av
where
E
= Em
sin
at and
i
=
Im
Eq. (4-29) reduces to
= K,EI cos
0
(4-31)
and / represent the rms values of the voltage and the current, and 6
represents the phase angle between voltage and current. Equations (4-29) and (4-30)
show
that the electrodynamometer indicates the average
power delivered
to the load.
Wattmeters have one voltage terminal and one current terminal marked the marked current terminal is connected to the incoming line, and the marked voltage terminal is connected to the line side in which the current coil is connected, the meter will always read up-scale when power is connected to the load. If for any reason (as in the two- wattmeter method of measuring "
± ." When
three-phase power), the meter should read backward, the current connections (not the voltage connections) should be reversed.
of
The electrodynamometer wattmeter consumes some power for maintenance magnetic field, but this is usually so small, compared to the load power, it may be neglected. If a correct reading of the load power is required, the
its
that
current coil should carry exactly the load current, and the potential coil should
be connected across the load terminals. With the potential coil connected to point A, as in Fig. 4-40, the load voltage
through the
field coils is
high by the amount of additional the potential coil
is
is
properly metered, but the current
amount ip The wattmeter therefore reads power loss in the potential circuit. If, however,
greater by the
connected to point
.
B
in Fig. 4-40, the field coils
correct load current, but the voltage across the potential coil
amount of but
the drop across the field coils.
now by
the
amount of the
PR
The wattmeter
is
meter the
higher by the
will again read high,
losses in the field windings.
Choice of the
Sec. 4-15
Watthour Meter
99
correct connection depends tential coil at point at
B
A
on the
situation. Generally, connection of the po-
preferred for high-current, low-voltage loads; connection
is
preferred for low-current, high-voltage loads.
is
The
difficulty in placing the connection of the potential coil is overcome compensated wattmeter, shown schematically in Fig. 4-41. The current coil consists of two windings, each winding having the same number of turns. One winding uses heavy wire that carries the load current plus the current for
in the
the potential
coil.
The other winding
uses thin wire and carries only the current
to the voltage coil. This current, however,
is
in a direction opposite to the current
heavy winding, causing a flux that opposes the main flux. The effect of therefore canceled out, and the wattmeter indicates the correct power.
in the i
p
is
Current Coil
*
oFigure 4-41
Diagram of
a compensated wattmeter in which the effect of the current
canceled by the current in the compensating winding.
in the potential coil is
4-15
o
WATTHOUR METER
The watthour meter
is
not often found in a laboratory situation but
used for the commercial measurement of electrical energy. In
fact,
it is it is
widely evident
wherever a power company supplies the industrial or domestic consumer with electrical energy. Figure 4-42 shows the elements of a single-phase watthour meter
in
schematic form.
The current is
coil is
connected across the
connected in
line.
Both
series
coils are
with the
wound on
line,
and the voltage
coil
a metal frame of special
A light aluminum disk is suspended in which causes eddy currents to flow in the disk. The reaction of the eddy currents and the field of the voltage coil creates a torque (motor action) on the disk, causing it to rotate. The developed torque is proportional to the fieldstrength of the voltage coil and the eddy currents in
design, providing
two magnetic
the air gap of the current-coil
circuits.
field,
Electromechanical Indicating Instruments
100
?
Figure 4-42
Chap. 4
Line
9
Elements of a single-phase watthour meter.
the disk which are in turn a function of the fieldstrength of the current
coil.
The number of rotations of the disk is therefore proportional to the energy consumed by the load in a certain time interval, and is measured in terms of kilowatthours (kWh). The shaft that supports the aluminum disk is connected by a gear arrangement to the clock mechanism on the front of the meter, providing a decimally calibrated readout of the number of kWh. Damping of the disk is provided by two small permanent magnets located opposite each other at the rim of the disk. Whenever the disk rotates, the permanent magnets induce eddy currents in it. These eddy currents react with the magnetic fields of the small permanent magnets, damping the motion of the disk. A typical single-phase watthour meter is shown in Fig. 4-43. Calibration of the watthour meter is performed under conditions of full rated load and 10 per cent of rated load. At full load, the calibration consists of adjustment of the position of the small permanent magnets until the meter reads correctly. At very light loads, the voltage component of the field produces a torque that is not directly proportional to the load. Compensation for the error is provided by inserting a shading coil or plate over a portion of the voltage coil, with the meter operating at 10 per cent of rated load. Calibration of the meter at these two positions usually provides satisfactory readings at all other loads.
disk.
The The
the shaft
floating-shaft watthour meter uses a unique design to suspend the
magnet at each end. The upper magnet of magnet in the upper bearing, and the lower magnet attracted to a magnet in the lower bearing. The movement thus
rotating shaft has a small is
of the shaft
attracted to a is
Sec. 4-16
Power-Factor Meters
Figure 4-43
Watthour meter
101
for industrial or
domestic application (courtesy West-
inghouse Electric Corporation).
floats
without touching either bearing surface, and the only contact with the
movement
is
that of the gear connecting the shaft with the gear train.
Measurements of energy in three-phase systems are performed with polyphase watthour meters. Each phase of the watthour meter has its own magnetic circuit and its own disk, but all the disks are mounted on a common shaft. The developed torque on each disk is mechanically summed and the total number of revolutions per minute of the shaft is proportional to the total three-phase energy consumed.
4-16 POWER-FACTOR METERS The power
factor, by definition, is the cosine of the phase angle between voltage and current, and power-factor measurements usually involve the determination of this phase angle. This is demonstrated in the operation of the crossed-coil power-factor meter. The instrument is basically an electrodynamometer movement, where the moving element consists of two coils, mounted on the same
102
Electromechanical Indicating Instruments
shaft but at right angles to each other. field
provided by the
The moving
field coil that carries
coils rotate in the
Chap. 4
magnetic
the line current.
The connections
for this meter in a single-phase circuit are shown in the diagram of Fig. 4-44. The field coil is connected as usual in series with the line and carries the line current. One coil of the movable element is connected in series with a resistor across the lines and receives its current from the applied potential difference. The second coil of the movable element is connected in series with an inductor across the lines. Since no control springs are used, the circuit
balance position of the movable element depends on the resulting torque de-
veloped by the two crossed
coils.
When
the movable element
is
in a balanced
by each element must be equal but each coil is a function of the current
position, the contribution to the total torque
of opposite sign.
The developed torque
in
and therefore depends on the impedance of that coil circuit. The torque is also proportional to the mutual inductance between each part of the crossed coil and the stationary field coil. This mutual inductance depends on the angular position of the crossed-coil elements with respect to the position of the stationary field coil. When the movable element is at balance, it can be shown that its angular displacement is a function of the phase angle between through the
coil
line current (field coil)
pointer,
which
is
and
The
line voltage (crossed coils).
connected to the movable element,
indication of the
calibrated directly in
is
terms of the phase angle or power factor.
Figure 4-44
Connections for a single-phase crossed-coil power-factor meter.
The polarized-vane power-factor meter of Fig. 4-45. This instrument
because
The
its
is
shown
in the construction sketch
used primarily in three-phase power systems,
operating principle depends on the application of three-phase voltage.
outside coil
of the system. it
is
is
The
the potential coil, which
is
connected to the three phase
to act like the stator of a three-phase induction
magnetic flux. The central of the phase
lines,
lines
application of three-phase voltage to the potential coil causes
and
coil,
or current
motor
coil, is
in setting
connected
this polarizes the iron vanes.
The
up a
in series
rotating
with one
polarized vanes
move
Sec. 4-17
Instrument Transformers
103
Damping Vane
Three-phase
Field
(Potential)
Current Coil
Moving Vane
Figure 4-45
Polarized-vane power-factor meter (courtesy General Electric
Company
Limited).
and take up the position that the rotating field has maximum. This position is an indication of the phase angle and therefore the power factor. The instrument may be used in single-phase systems, provided that a phase-splitting network (similar to that used in single-phase motors) is used to set up the required rotating magnetic in a rotating
magnetic
field
at the instant that the polarizing flux is
field.
Both types of power-factor meter are limited to measurement at comparlow frequencies and are typically used at the powerline frequency Hz). Phase measurements at higher frequencies often are more accurately (60 and elegantly performed by special electronic instruments or techniques. atively
\
4-17 INSTRUMENT
TRANSFORMERS
Instrument transformers are used to measure ac at generating stations, trans-
former
stations,
and
at transmission lines, in conjunction
with ac measuring
instruments (voltmeters, ammeters, wattmeters, VARmeters,
etc.).
Instrument
transformers are classified according to their use and are referred to as current transformers (CT) and potential transformers (PT).
Instrument transformers perform two important functions: They serve to much as the shunt or the
extend the range of the ac measuring instrument,
multiplier extends the range of a dc meter; they also serve to isolate the measuring
instrument from the high-voltage power
line.
104
The range of a dc ammeter may be extended by using a shunt the current under is
Chap. 4
Electromechanical Indicating Instruments
that divides
measurement between the meter and the shunt. This method
satisfactory for dc circuits, but in ac circuits current division
depends not
only on the resistances of the meter and the shunt but also on their reactances. Since ac measurements are to obtain great accuracy. its
made over a wide frequency range,
A CT
it
becomes
difficult
provides the required range extension through
transformation ratio and in addition produces almost the same reading re-
gardless of the meter constants (reactance and resistance) or, in fact, of the
number of instruments (within Isolation of the
connected in the
limits)
circuit.
measuring instrument from the high-voltage power
line is
important when we consider that ac power systems frequently operate at voltages It would be impractical to bring the high- voltage an instrument panel in order to measure voltage or current, not only because of the safety hazards involved but also because of the insulation problems connected with high-voltage lines running closely together in a confined space. When an instrument transformer is used, only the low-voltage wires from the transformer secondary are brought to the instrument panel and only low voltages exist between these wires and ground, thereby minimizing safety hazards
of several hundred kilovolts. lines directly to
and insulation problems.
Many
textbooks develop in detail the theory underlying the operation of
transformers. Here these instrument transformers are merely described and their
use in measurement situations
is
shown.*
Figure 4-46 shows a potential transformer, Fig. 4-47 shows a current transformer.
The potential transformer (PT)
is
used to transform the high voltage
of a power line to a lower value suitable for direct connection to an ac voltmeter
The usual secondary transformer
or the potential coil of an ac wattmeter. is
voltage
120 V. Primary voltages are standardized to accommodate the usual trans-
mission line voltages which include 2,400 V, 4,160 V, 7,200 V, 13.8 kV, 44 kV,
66 kV, and 220 kV. The
PT
is
rated to deliver a certain
power
to the secondary
load or burden. Different load capacities are available to suit individual applications; a general capacity
The PT must
is
200
VA
at a
satisfy certain design
frequency of 60 Hz.
requirements that include accuracy of
the turns ratio, small leakage reactance, small magnetizing current, and minimal
we may be working with very high primary between the primary and secondary windings must be able to withstand large potential differences, and the dielectric requirements are very high. In the usual case, the high-voltage coil is of a circular pancake voltage drop. Furthermore, since voltages, the insulation
construction, shielded to avoid localized dielectric stresses.
or coils are
wound on
The assembly *For
is
fuller
The low-voltage
a paper form and assembled inside the high-voltage
thoroughly dried and
oil
coil coil.
impregnated. The core and coil as-
treatment of ac machines and circuits, consult textbooks like the following:
Michael Liwshitz-Garik and Clyde C. Whipple, AC Machines, 2nd ed. (Princeton, N.J.: D. Van Nostrand Company, Inc., 1961), chaps. 2-5. Russell M. Kerchner and George F. Corcoran, Alternating Current Circuits, 4th ed.
(New York: John Wiley
&
Sons, Inc., 1961), pp. 291-317.
sembly
is
then mounted inside a steel case, which supports the high-voltage
terminals or porcelain bushings.
Developments
The
in the synthetic
case
is
then
filled
with an insulating
oil.
rubber industry have introduced the molded
oil and porcelain bushings shows a rubber-molded 25-kV potential trans-
rubber potential transformer, replacing the insulating in
some
applications. Figure 4-46
former suitable for outdoor use. This unit is less expensive than the conventional oil-filled PT, and since the bushings are made of molded rubber, porcelain breakage is eliminated. A white polarity dot is placed on the proper bushing on
Two stud-type secondary terminals are enclosed removable conduit box. The power rating of a potential transformer is based
the front of the transformer. in a
on considerations other than load capacity, for the reasons previously outlined. A typical load rating is 200 VA at 60 Hz for a transformer having a ratio of 2,400/120 V. For most metering purposes, however, the burden will be significantly less than 200 VA. The current transformer (CT) sometimes has a primary and always has a secondary winding. If there is a primary winding, it has a small number of turns. In most cases, the primary is only one turn or a single conductor connected in series with the load whose current is to be measured. The secondary winding
Electromechanical Indicating Instruments
106
Figure 4-47
Current transformer (courtesy Westinghouse Electric Corporation).
has a larger number of turns and coil.
Chap. 4
Often the primary winding
is
is
connected to a current meter or a relay
a single conductor in the form of a heavy
copper or brass bar running through the core of the transformer. Such a called a bar-type current transformer.
The
CT
designed to deliver a secondary current of 5 A.
secondary winding
An
transformer would have 160 turns on the secondary
The primary winding of the load circuit.
When
the secondary winding
ratio) ings.
and could
easily
break
down
The secondary winding of
may
is
its
is
coil. is
connected directly
in
open-circuited, the voltage
be very high (because of the step-up
the insulation between the secondary wind-
a current transformer should therefore always
be short-circuited, or connected to a meter or relay
should never have
CT
usually
800/ 5- A bar-type current
the current transformer
developed across the open terminals
is
coil.
secondary open while the primary
A is
current transformer
carrying current;
it
Sec. 4-17
Instrument Transformers
107
should always be closed through a current meter, relay coil,
coil,
or simply a short. Failure to observe this precaution
wattmeter current
may
cause serious
damage to either equipment or operating personnel. The current transformer shown in Fig. 4-47 consists of a core with the secondary winding encased in molded-rubber insulation. The window in the core allows for the insertion of one or more turns of the current-carrying highvoltage conductor.
The nominal
A
single
conductor constitutes a one-turn primary winding.
ratio of the transformer
turns ratio (since
is
given on the nameplate; this
more than one turn can be used
indicates that a primary current of 500 5
A
when
limits,
the secondary coil
A
not the
secondary current of ammeter. Within practical
will cause a
connected to a
5-
the current in the secondary winding
is
is
is
as the primary) but only
A
determined by the primary
excitation current and not by the secondary circuit impedance. Since the primary
current
determined by the load in the ac system, the secondary current
is
related to the primary current
This
is
by approximately the inverse of the
is
turns-ratio.
true within rather wide limits of the nature of the secondary burden.
Figure 4-48 indicates the use of instrument transformers in a typical mea-
surement application. This diagram
illustrates the
connection of instrument
transformers in a three-wire three-phase circuit, including two wattmeters, two voltmeters,
phase lines
and two ammeters. The potential transformers are connected across B, and phase lines C and B; the current transformers are in
A and
Instrument transformers in a three-phase measurement application. Pomarkings of the potential and current transformers are indicated by black
Figure 4-48 larity
squares.
108
Electromechanical Indicating Instruments
phase
lines
A and
Chap. 4
D. The secondary windings of the potential transformers are
connected to the voltmeter
coils and the potential coils of the wattmeters; the current transformer secondaries feed the ammeters and the current coils of the wattmeters.
The polarity markings on the transformers, indicated by a dot at the transformer leads, aid in making the correct polarity connections to the measuring instruments. At any given instant of the ac cycle, the dot-marked terminals have the same polarity and the marked wattmeter terminals must be connected to these transformer leads as shown.
REFERENCES 1.
Bartholomew, Davis, Electrical Measurements and Instrumentation, chap. Allyn and Bacon,
2.
5.
Boston:
Inc., 1963.
Geczy, Steven, Basic Electrical Measurements. Englewood
Cliffs, N.J.:
Prentice-Hall,
Inc., 1984. 3.
Jackson, Herbert W., Introduction to Electric Circuits, 5th Cliffs, N.J.:
4.
Prensky, Sol D., and Castellucis, Richard chaps. 2 and
5.
ed.,
chap.
19.
Englewood
Prentice-Hall, Inc., 1981.
3.
Englewood
Cliffs, N.J.:
L., Electronic
Instrumentation, 3rd ed.,
Prentice-Hall, Inc., 1982.
Stout, Melville B., Basic Electrical Measurements,
2nd
ed.,
chap.
17.
Englewood
Cliffs,
N.J.: Prentice-Hall, Inc., 1960.
PROBLEMS 1.
Determine the
resistor value required to use a
resistance of 125 CL for a
0-1-V
0-1 -mA meter with an
internal
meter.
What value of shunt resistance is required for using a 50- uA meter movement, with an internal resistance of 250 H, for measuring 0-500 mA? V3. What series resistance must be used to extend the 0-200-V range of a 20,000-fl/V meter to 0-2000 V? What power rating must this resistor have? 4. What will a 5,000-0/ V meter read on a 0-5-V scale when connected to the circuit 2.
of Fig. P4-4?
400 r-
25
V— —
5.
Draw
k£l
WA
n
I00k&s}v =
—
Figure P4-4
the schematic, including values, for an Ayrton shunt for a meter
having a full-scale deflection of
6.
?
movement
mA
and an internal resistance of 500 Cl to cover the current ranges of 10, 50, 100, and 500 mA. Many electronic voltage measuring instruments have a fixed input resistance of 1 1
MCI. Which settings of the range switch of the multimeter shown in Figs. 4-23 and 4-24 would have a higher input resistance than the typical electronic instrument for dc measurements?
Problems
Chap. 4
7.
The
50-kH
resistance of a
Figs. 4-23, 4-24,
R X R X 8.
109
and
10,000 range
100 range
is
resistor
is
is
measured using the multimeter shown in power is dissipated in the resistor if the power is dissipated in the resistor if the
How much used? How much
4-27.
used?
Assume
that the zero control
is set
to
its
midpoint.
A series-type ohmmeter, designed to operate with a 6-V battery, has a circuit diagram shown in Fig. 4-21. The meter movement has an internal resistance of 2,000 fl and requires a current of 100 uA for full-scale deflection. The value of/?, is 49 kft. (a) Assuming the battery voltage has fallen to 5.9 V, calculate the value of R 2 as
required to zero the meter.
Under
(b)
to the
the conditions mentioned in part (a), an unknown resistor is connected meter causing a 60 per cent meter deflection. Calculate the value of the
unknown 9.
How
shown
section 10.
What
11.
Why
resistance.
low must the battery voltage of the 1.5-V is
in Fig. 4-27(a) fall before
a transfer instrument? sensitivity
is
(ohms per
Why
is
it is
cell in
the multimeter
ohmmeter
impossible to zero the meter?
an electrodynamometer a transfer instrument?
volt) of the ac scales of a
multimeter
less
than the dc
section? 12.
meant by a waveform error? Which ac meters are most likely form of error? What are the advantages of a thermocouple meter? What is the midscale point of a 10- A full-scale thermocouple meter?
What
by 13. 14. 15.
is
to be affected
this
The circuit diagram of Fig. 4-32 shows a full-wave rectifier ac voltmeter. The meter movement has an internal resistance of 250 fl and requires 1 mA for full deflection. The diodes each have a forward resistance of 50 fl and infinite reverse resistance. Calculate (a) the series resistance required for full-scale meter deflection when 25 V rms
is
applied to the meter terminals, and (b) the ohms-per-volt rating of this ac
voltmeter. 16. Calculate the indication of the
a peak value of 20 17. If
V
is
an electrodynamometer
W, what
is
meter
in Prob. 15
when
a triangular
waveform with
applied to the meter terminals. is
used to measure power with a full-scale reading of 100
the one-quarter scale reading?
CHAPTER 5
BRIDGES AND THEIR APPLICATION
5-1
INTRODUCTION
Bridge circuits are extensively used for measuring component values, such as resistance, inductance, or capacitance,
and of other
circuit
parameters directly
derived from component values, such as frequency, phase angle, and temperature. Since the bridge circuit merely compares the value of an to that of
an accurately known component
curacy can be very high indeed. This
is
unknown component
(a standard), its
measurement
ac-
so because the readout of this comparison
measurement, based on a null indication at bridge balance, is essentially independent of the characteristics of the null detector. The measurement accuracy is
therefore directly related to the accuracy of the bridge components, not to
that of the null indicator
itself.
This chapter introduces some of the basic dc bridges. Starting with the portable test instruments,
we present
the Wheatstone bridge for the measurement
of dc resistance, the Kelvin bridge for low-resistance measurements, and the test set for resistance testing field,
we
measurement of very high
5-2
of cables. In the high-precision test and calibration
introduce the principle of the guarded Wheatstone bridge and the resistances.
WHEATSTONE BRIDGE 5-2.1 Basic Operation Figure 5- 1(a)
is
bridge. Its operation
110
a photograph of a portable, self-contained Wheatstone
is
based on the fundamental diagram of Fig.
5- 1(b).
The
Figure 5-1
Laboratory-type Wheatstone bridge used for the precision measurement
of resistances ranging from fractions of an switches the ratio arms in decade steps. resistance of the standard
arm
(courtesy
ohm
to several megohms. The ratio control The remaining four step switches set the Beckman Instruments, Inc., Cedar Grove
Operations).
bridge has four resistive arms, together with a source of
emf
(a battery)
and a
null detector, usually a galvanometer or other sensitive current meter.
The
current through the galvanometer depends on the potential difference between points c
and
d.
The bridge
is
said to be balanced
when
the potential difference
Bridges and Their Application
112
across the galvanometer
V
0
is
the voltage from point
d
so that there
when
nometer. This condition occurs
is
Chap. 5
no current through the galva-
the voltage from point c to point a equals
to point a; or
by referring to the other battery terminal,
when
the voltage from point c to point b equals the voltage from point
point
b.
Hence
the bridge
I If the
galvanometer current
d
to
balanced when
is
X
R =
I2 R 2
X
(5-1)
zero, the following conditions also exist:
is
=
7 >
= r^tt,
(5 " 2)
=
(5 " 3)
and
h= Combining Eqs.
and
(5-1), (5-2),
(5-3)
R + R x
*7T^
and simplifying, we obtain
3
(5-4)
R + R4 2
from which
Equation (5-5)
is
the well-known expression for balance of the Wheatstone bridge.
from Eq.
(5-5).
Hence,
if
may
have known values, the fourth
If three of the resistances
R4
is
unknown
the
be determined
Rx
resistor, its resistance
can be
expressed in terms of the remaining resistors as follows:
Rx = * Resistor
^
3
is
called the standard
3
arm of
^
(5-6)
the bridge, and resistors
R
2
and
R
x
are called the ratio arms.
The measurement of
the
unknown
resistance
Rx
is
independent of the
characteristics or the calibration of the null-deflecting galvanometer, provided
that the null detector has sufficient sensitivity to indicate the balance position
of the bridge with the required degree of precision.
5-2.2
Measurement Errors
The Wheatstone bridge
is widely used for precision measurement of refrom approximately 1 Cl to the low megohm range. The main source of measurement error is found in the limiting errors of the three known resistors. Other errors may include the following:
sistance
(a)
Insufficient sensitivity of the null detector. This
more
fully in Sec. 5-2.3.
problem
is
discussed
Wheatstone Bridge
Sec. 5-2
(b)
Changes
113
arms due
in resistance of the bridge
to the heating effect of
the current through the resistors. Heating effect
(PR)
of the bridge
arm currents may change the resistance of the resistor in question. The rise in temperature not only affects the resistance during the actual measurement, but excessive currents may cause a permanent change in resistance values. This may not be discovered in time and subsequent measurements could well be erroneous. The power dissipation in the bridge arms must therefore be computed in advance, particularly when low-resistance values are to be measured, and the current
must be
limited to a safe value. (c)
Thermal emfs also cause
in the bridge circuit or the
problems when low-value
galvanometer circuit can
resistors are being
measured.
To
prevent thermal emfs, the more sensitive galvanometers sometimes have
copper
coils
and copper suspension systems to avoid having dissimilar
metals in contact with one another and generating thermal emfs. (d) Errors
due to the resistance of leads and contacts exterior to the actual
bridge circuit play a role in the measurement of very low-resistance values.
These errors may be reduced by using a Kelvin bridge
(see
Sec. 5-3).
5-2.3 Thevenin Equivalent Circuit
To determine whether to detect
or not the galvanometer has the required sensitivity
an unbalance condition,
it is
necessary to calculate the galvanometer
current. Different galvanometers not only
may
may have
deflection (current sensitivity), but they also sistance. It
make
is
require different currents per unit
a different internal re-
impossible to say, without prior computation, which galvanometer
more
an unbalance condition. This sensitivity can be calculated by "solving" the bridge circuit for a small unbalance. The solution is approached by converting the Wheatstone bridge of Fig. 5-1 to its Thevenin equivalent. will
Since
the bridge circuit
we
are interested in the current through the galvanometer, the
Thevenin equivalent c and d in Fig. 5-1.
The
first
sensitive to
determined by looking into galvanometer terminals steps must be taken to find the Thevenin equivalent:
circuit
Two
is
step involves finding the equivalent voltage appearing at terminals c
and d when the galvanometer
is
removed from the
circuit.
The second
involves finding the equivalent resistance looking into terminals c and the battery replaced by Fig. 5- 1(b) is
redrawn
its
For convenience, the
internal resistance.
d,
step
with
circuit of
in Fig. 5-2(a).
The Thevenin, or and we can write
open-circuit, voltage
E = E — Ead = cd
ac
is
I
]
found by referring to Fig.
R — ]
I2 R 2
5-2(a),
'
114
Bridges and Their Application
Chap. 5
1
I
WAr
AAAA/
1
od
Co-
i
'WW
.
b
AVA
—
R4
(b)
R TH AAAAr TH Figure
(g)r
5-2
Application
of
Thevenin's
theorem to the Wheatstone bridge, (a) Wheatstone bridge configuration, (b) Thev-
(
enin resistance looking into terminals c and d. (c)
Complete Thevenin
circuit,
with the
galvanometer connected to terminals c and
(0
d.
where
I
x
— R
E = — —— ~\~
i
R$
and I2
E
= R
2
H~
R4
Therefore
This
is
the voltage of the Thevenin generator.
The
Thevenin equivalent circuit is found by looking back and d and replacing the battery by its internal resistance. The 5-2(b) represents the Thevenin resistance. Notice that the internal
resistance of the
into terminals c circuit of Fig.
resistance,
Rb
,
of the battery has been included in Fig. 5-2(b). Converting this
Wheatstone Bridge
Sec. 5-2
115
more convenient form requires use of the delta-wye transformation theorem. Readers interested in this approach should consult texts on circuit circuit into a
analysis
where
this
theorem
derived and applied.* In most cases, however,
is
the extremely low internal resistance of the battery can be neglected and this simplifies the reduction of Fig. 5 -2(a) to its Thevenin equivalent considerably.
we
Referring to Fig. 5-2(b),
a and b
when
see that a short circuit exists between points
the internal resistance of the battery
Thevenin resistance, looking into terminals c and
assumed
is
to be 0 ft.
*™ = srrk + The Thevenin
by Eq.
sistance given
emf
This
(5-8).
the null detector
Thevenin equivalent
is
Example
5-1
is
described by Eq. (5-7) and an internal re-
shown
in the circuit of Fig. 5-2(c).
now connected
is
circuit, the
/
where Ig
(5 - 8 >
equivalent of the Wheatstone bridge circuit therefore reduces to
a Thevenin generator with an
When
The
then becomes
d,
to the output terminals of the
galvanometer current
=
is
found to be
^Z*
the galvanometer current and
Rg
(5.9)
its
resistance.
Figure 5-3(a) shows the schematic diagram of a Wheatstone bridge with values of the bridge elements as shown.
The galvanometer has
The
battery voltage
is
5
a current sensitivity of 10
V and its internal mm/jLiA and an
resistance negligible. internal resistance of
100ft. Calculate the deflection of the galvanometer caused by the 5-ft unbalance in
arm BC. Solution
Bridge balance occurs
BC as
if
arm
a resistance of 2,005
BC
has a resistance of 2,000
ft.
The diagram shows arm ft). The first step
representing a small unbalance (< 2,000
ft,
in the solution consists of converting the bridge circuit into its circuit.
Since
equivalent difference voltage.
we
is determined with respect to galvanometer terminals B and D. The potential from B to D, with the galvanometer removed from the circuit, is the Thevenin
Using Eq.
Eth
(5-7),
we
obtain
— EAD
= ,
The second
s
100 5Vx(,100 100 + 200 2.77
1,000
1,000
B
Herbert
and D, and replacing the battery with
W.
Jackson, Introduction
Prentice-Hall, Inc., 1981), pp. 448
jf.
+
2,005.
mV
step of the solution involves finding the equivalent
into terminals
*
Thevenin equivalent
are interested in finding the current in the galvanometer, the Thevenin
its
to Electric Circuits,
Thevenin
resistance, looking
internal resistance. Since the
5th ed. (Englewood
Cliffs,
N.J.
—
|
«
116
Bridges and Their Application
Chap. 5
lOOOft
5V
—
D
(a)
Wheatstone Bridge
(b)
Calculation of the
2005 SI
m— —vwv— —iooon 100ft
A i
I
Thevenin
Resistance
—vwv —VWv i
!
= 730 n
R TH AAAA/
o I
g
=3.34^A (c)
Thevenin Equivalent Circuit
2.77
Rg =
mV^"
I00&
Galvanometer Figure 5-3
from which we
is
the simplified Thevenin approach.
0
fi,
the circuit
is
represented by the configuration of Fig. 5-3(b)
find
100
R th = The Thevenin equivalent
—X
200
300
circuit
is
+
1,000 X 2,005 ^rzrz 3,005
given in Fig. 5-2(c).
connected to the output terminals of the equivalent
galvanometer
= IOmm/^.A
Calculation of galvanometer deflection caused by a small unbalance in
arm BC, using battery resistance
sensitivity
=
730
When
H
the galvanometer
circuit, the current
is
E TH 8
The galvanometer
R T„ +
deflection
d
2.77
= R.
730
3.34 llA
X
mV
a +
100
_
*
a
is
=
10
mm =
uA
33.4
mm
is
now
through the
Sec. 5-3
Kelvin Bridge
At
117
this point the merit of the
Thevenin equivalent
circuit for the solution
of an unbalanced bridge becomes evident. If a different galvanometer
is
used
(with a different current sensitivity and internal resistance), the computation of its
deflection
is
very simple, as
galvanometer sensitivity
is
clear
from
Fig.
5-3(c).
Conversely,
the
if
we can solve for the unbalance voltage needed mm). This value is of interest when we want to
given,
is
to give a unit deflection (say
1
determine the sensitivity of the bridge to unbalance, or in response to the question: "Is the galvanometer selected capable of detecting a certain small unbalance?"
The Thevenin method cases
is
Example
is
used to find the galvanometer response, which in most
of prime interest. 5-2
The galvanometer of Example
5-1
is
replaced by one with an internal resistance of 500
and a current sensitivity of 1 observed on the galvanometer
mm/oA.
detecting the 5-Cl unbalance in
arm
Assuming
that a deflection of
mm
1
fl
can be
new galvanometer
is
capable of
Since the bridge constants have not been changed, the equivalent circuit
is
again rep-
scale,
determine
BC of Fig.
if this
5-3(a).
Solution
resented by a Thevenin generator of 2.77
new galvanometer
is
now connected
mV
and a Thevenin resistance of 730
to the output terminals, resulting in a
The
fl.
galvanometer
current 2.77
730
The galvanometer
mA
n +
500
deflection therefore equals 2.25
n u-A
=
uA
2.25
X
1
mm/u-A =
2.25
mm,
indicating that this galvanometer produces a deflection that can be easily observed.
The Wheatstone bridge is limited to the measurement of resistances ranging from a few ohms to several megohms. The upper limit is set by the reduction in sensitivity to unbalance, caused by high resistance values, because in this case the equivalent Thevenin resistance of Fig. 5-3(c) becomes high, thus reducing the galvanometer current. The lower limit is set by the resistance of the connecting leads and the contact resistance at the binding posts. The resistance of the leads could be calculated or measured, and the final result modified, but contact resistance is very hard to compute or measure. For low-resistance measurements, therefore, the Kelvin bridge
5-3 KELVIN
is
generally the preferred instrument.
BRIDGE
5-3.1 Effects of Connecting Leads
The Kelvin bridge
is
a modification of the Wheatstone bridge and provides
greatly increased accuracy in the
measurement of low-value
resistances, generally
118
below
1
ft.
Bridges and Their Application
Chap. 5
Ry
represents
Consider the bridge circuit shown in Fig.
R
the resistance of the connecting lead from
nections are possible, to point
m
to
3
or to point
connected to point m, the resistance
Ry
n.
5-4,
where
R x Two .
When
galvanometer con-
the galvanometer
of the connecting lead
is
added
is
to the
unknown R x resulting in too high an indication for R x When connection is made to point n, R y is added to bridge arm R and the resulting measurement of R x will be lower than it should be, because now the actual value of R is .
,
3
3
nominal value by resistance R y If the galvanometer is connected to a point p, in between the two points m and n, in such a way that the ratio of the resistances from n to p and from m to p equals the ratio of resistors R and R 2 we can write higher than
its
.
,
,
The balance equation
for the bridge yields
R x + R np =
^(* + 3
Substituting Eq. (5-10) into Eq. (5-11),
we
R mp )
(5-11)
obtain
(5-12)
which reduces
to
Rx = Equation (5-13) bridge and
it
from point
m
is
R
the usual balance equation developed for the Wheatstone
indicates that the effect of the resistance of the connecting lead to point n has been eliminated
to the intermediate position p.
Figure
5-4
Wheatstone
showing resistance point
(5-13)
m
to point n.
Ry
bridge
circuit,
of the lead from
by connecting the galvanometer
Sec. 5-3
119
Kelvin Bridge
Figure 5-5
Basic
Kelvin double bridge
circuit.
This development forms the basis for construction of the Kelvin double bridge,
commonly known
as the Kelvin bridge.
5-3.2 Kelvin Double Bridge
The term double bridge is used because the circuit contains a second shown in the schematic diagram of Fig. 5-5. This second
of ratio arms, as
set
set
of arms, labeled a and b in the diagram, connects the galvanometer to a point
p
at the appropriate potential
the yoke resistance
R y An .
and b is the same as the The galvanometer indication
*1+ *2
when
and
n,
initially established
ratio of a
the potential at p, or
m
between
E = kl
ratio of will
R
*J
-f
yields
R
2
(a
2
+
b
+
yields
Mi + £
£
2
(a
2
+
^+ R 6
bR y
y)
(a
+
+
R,)
{a
+
b
(fl
+
H
+
so that
R Using the
+
'~-rT
(a
+
b
condition that
initially established
(5 -' 6)
{T2
a/b
b)
= R /R l
2
,
we
see that Eq.
(5-16) reduces to the well-known relationship
Equation (5-17)
the usual working equation for the Kelvin bridge.
is
that the resistance of the yoke has
no
effect
It
indicates
on the measurement, provided that
two sets of ratio arms have equal resistance ratios. The Kelvin bridge is used for measuring very low resistances, from approximately 1 ft to as low as 0.00001 ft. Figure 5-6 shows the simplified circuit diagram of a commercial Kelvin bridge capable of measuring resistances from 10 ft to 0.00001 ft. In this bridge, resistance R 3 of Eq. (5-17) is represented by the variable standard resistor in Fig. 5-6. The ratio arms (R^ and R 2 ) can usually the
be switched in a number of decade steps. Contact potential drops in the measuring circuit
and
may
cause large errors
to reduce this effect the standard resistor consists of nine steps of 0.001 ft
each plus a calibrated manganin bar of 0.0011 total resistance of the
R arm 3
ft
with a sliding contact. The
amounts to 0.0101 of 0.0011 ft by the sliding
therefore
steps of 0.001 ft plus fractions
ft
and
is
contact.
variable in
When
both
contacts are switched to select the suitable value of standard resistor, the voltage
drop between the ratio-arm connection points is changed, but the total resistance around the battery circuit is unchanged. This arrangement places any contact resistance in series with the relatively high-resistance values of the ratio arms,
and the contact resistance has
The
ratio
R /R x
standard resistance
unknown
resistance
nificant figures,
is
negligible effect.
should be so selected that a relatively large part of the
2
used in the measuring
Rx
is
circuit.
In this
way
the value of
determined with the largest possible number of
and the measurement accuracy
is
improved.
sig-
Bridges and Their Application
122
5-4
Chap. 5
GUARDED WHEATSTONE BRIDGE 5-4.1
Guard Circuits
The measurement of extremely high
resistances,
such as the insulation
on the order beyond the capability of the ordinary dc Wheatstone bridge. One of the major problems in high-resistance measurements is the leakage that occurs over and around the component or specimen being measured, or over the binding posts by which the component is attached to the instrument, or within the instrument itself. These leakage currents are undesired because they can enter the measuring circuit and affect the measurement accuracy to a considerable extent. Leakage currents, whether inside the instrument itself or associated with the test specimen and its mounting, are particularly noticeable in high-resistance measurements where high voltages are often necessary to resistance of a cable or the leakage resistance of a capacitor (often
of several thousands of megohms),
is
obtain sufficient deflection sensitivity. Also, leakage effects are generally variable
from day to day, depending on the humidity of the atmosphere. The effects of leakage paths on the measurement are usually removed by some form of guard circuit. The principle of a simple guard circuit in the R x arm of a Wheatstone bridge is explained with the aid of Fig. 5-7. Without a along the insulated surface of the binding post guard circuit, leakage current adds to current Ix through the component under measurement to produce a total circuit current that can be considerably larger than the actual device current. A guard wire, completely surrounding the surface of the insulated post, intercepts this leakage current and returns it to the battery. The guard must be carefully
Return to Battery
From Bridge
Figure 5-7
^
-
JJ
^
Leakage Current Intercepted by Guard Wire
Circuit
Simple guard wire on the
eliminates surface leakage.
Rx
terminal of a guarded Wheatstone bridge
Guarded Wheatstone Bridge
Sec. 5-4
123
Figure
Guarded
5-8
terminal
returns
leakage current to the battery.
placed so that the leakage current always meets some portion of the guard wire
and
is
prevented from entering the bridge
circuit.
In the schematic diagram of Fig. 5-8 the guard around the
Rx
binding
by a small circle around the terminal, does not touch any part of the bridge circuitry and is connected directly to the battery terminal. The principle of the guard wire on the binding post can be applied to any internal part of the bridge circuit where leakage affects the measurement; we then speak post, indicated
of a guarded Wheatstone bridge.
5-4.2 Three-terminal Resistance
To
avoid the effects of leakage currents external to the bridge circuitry,
the junction of ratio arms
RA
and
RB
is
usually brought out as a separate guard
terminal on the front panel of the instrument. This guard terminal can be used
shown
The high mounted on two insulating posts that are fastened to a metal plate. The two main terminals of the resistor are connected to the R x terminals of the bridge in the usual manner. The third terminal of the resistor is the common point of resistances R and R 2 which represent the leakage paths from the main terminals along the insulating posts to the metal plate, or guard. The guard is to connect a so-called three-terminal resistance, as
resistance
in Fig. 5-9.
is
,
x
connected to the guard terminal on the front panel of the bridge, as indicated in the
RA
,
schematic of Fig.
but since R^
is
very
This connection puts
5-9.
much
Similarly, leakage resistance
resistance of
R
2
effect is a slight
is
so
much
R
larger than 2
is
RA
in parallel
,
its
R
{
in parallel
with ratio arm
shunting effect
is
negligible.
with the galvanometer, but the
higher than that of the galvanometer that the only
reduction in galvanometer sensitivity.
The
effects
of external
leakage paths are therefore removed by using the guard circuit on the threeterminal resistance.
R and R 2 would be and the measured value of R x would be considerably in error. Assuming, for example, that the unknown is 100 Mft and that the leakage resistance from each terminal to the guard is also 100 Mft, resistance R x would be measured as 67 Mft, an error of approximately 33 per cent. If the
guard circuit were not used, leakage resistance
directly across
Rx
x
124
Bridges and Their Application
Three-terminal resistance
(a)
Three-terminal Resistance
(b)
Resistance Multiplier
Guarded bridge
circuit
Three-terminal resistance, connected to a guarded high-voltage
Figure 5-9
Chap. 5
megohm
bridge.
Megohm
5-4.3
A
Bridge
-
commercial high- voltage
megohm
bridge
the various controls can easily be identified.
instrument
is
the variable ratio
arm
RB
dial to the right of the large ratio dial
The
is
shown
in Fig. 5-10,
of Fig. 5-9.
The
resistance multiplier
corresponds to standard resistor
the circuit diagram and provides for multiplication of the ratio in a
decade 10
V
steps.
The dc supply is
is
in
adjustable over several increments from
made
to connect
an external generator. The
basically a dc amplifier with output meter
sensitivity to detect small is
is
Rc
number of
and the necessary R A and when measuring brought out as a front panel guard terminal, to be used
null detector
RB
voltage
to 1,000 V, while provision
where
large dial in the center of the
unbalance voltages. The junction of ratio arms
a three-terminal resistance.
The high-voltage megohm bridge high-resistance measurements. Other
is only one of the instruments used for methods may include the use of the well-
AC
Sec. 5-5
Bridges and Their Application
known megger
to
direct deflection
method
5-5
125
measure the insulation resistance of
method of
electrical
testing insulation samples,
machinery, the
and the
loss-of-charge
for checking the leakage resistance of capacitors.*
AC BRIDGES AND THEIR APPLICATION 5-5.1 Conditions for Bridge Balance
The
ac bridge
is
a natural outgrowth of the dc bridge and in
consists of four bridge arms, a source of excitation,
and a
its
basic
form
null detector.
The
power source supplies an ac voltage to the bridge at the desired frequency. For measurements at low frequencies, the power line may serve as the source of excitation; at higher frequencies,
Figure 5-10
an oscillator generally supplies the excitation
Commercial high-voltage megohm
bridge, used for the
resistances in the terra-ohm range (courtesy General
* Cf.
Melville B. Stout, Basic Electrical Measurement,
Prentice-Hall, Inc., 1960), pp. 126-33.
measurement of
Radio Company).
2nd
ed.
(Englewood
Cliffs,
N.J.:
Bridges and Their Application
126
Chap. 5
fa Z4 Figure
form
General
5-11
of
the
ac
bridge.
voltage.
The
null detector
must respond
and
to ac unbalance currents
in its
cheapest (but very effective) form consists of a pair of headphones. In other applications, the null detector
may
consist of an ac amplifier with an output
meter, or an electron ray tube (tuning eye) indicator.
The
general form of an ac bridge
Z Z
is
shown
in Fig. 5-11.
The
four bridge
and Z 4 are indicated as unspecified impedances and the detector is represented by headphones. As in the case of the Wheatstone bridge for dc measurements, the balance condition in this ac bridge is reached when the
arms Z„
2,
3
,
detector response null response
The
is
is
zero, or indicates a null. Balance adjustment to obtain a
made by varying one
or
more of
general equation for bridge balance
the bridge arms.
is
obtained by using complex
notation for the impedances of the bridge circuit. (Boldface type indicate quantities in
The condition
or admittances as well as voltages or currents. requires that the potential difference from
be the case
B
to C, in
when
is
used to
complex notation.) These quantities may be impedances
the voltage drop from
A B
to
C
in Fig. 5-11
A
to
for bridge balance
be zero. This will
equals the voltage drop from
both magnitude and phase. In complex notation we can write
Eba
—
or
IjZj
—
I2
Z
(5-18)
2
For zero detector current (the balance condition), the currents are
=
I,
Z,
+ Z
(5-19) 3
and
E
=
I2
Z,
+ Z
(5-20) 4
— —"
Substitution of Eqs. (5-19) and (5-20) into Eq. (5-18) yields. j
or
when
'
)
*ifiS(^ (5-21)
using admittances instead of impedances.
Y,Y 4
= YY 2
3
(5-22)
AC
Sec. 5-5
127
Bridges and Their Application
Equation (5-21)
is
the most convenient form in most cases and
is
the general
equation for balance of the ac bridge. Equation (5-22) can be used to advantage
when
dealing with parallel components in bridge arms.
Equation (5-21)
states that the
product of impedances of one pair of
opposite arms must equal the product of impedances of the other pair of opposite
arms, with the impedances expressed in complex notation. If the impedance written in the form
Z = Z
Z0, where
Z
phase angle of the complex impedance, Eq. (5-21) can be rewritten (Z, Z0.)
(Z4 Z0 4 )
is
represents the magnitude and 0 the
= (Z
in the
Z0 2 )(Z 3 Z0 3 )
2
form
(5-23)
Since in multiplication of complex numbers the magnitudes are multiplied and the phase angles added, Eq. (5-23) can also be written as
Z,Z4
+
Z(0,
0 4)
= ZZ 2
3
Z(0 2
+
0 3)
(5-24)
Equation (5-24) shows that two conditions must be met simultaneously when The first condition is that the magnitudes of the imped-
balancing an ac bridge.
ances satisfy the relationship
_____
.
Z,Z4 = or, in
ZZ 2
(5-25)
3
words:
The products of the magnitudes of the opposite arms must be
The second condition
equal.
requires that the phase angles of the impedances satisfy
the relationship
,
A Z0,
+
Z0 4
=
Z0 2
+
Z0 3
I
ge
j
(5-26)
*
Again, in words:
The sum of the phase angles of the opposite arms must be equal
5-5.2 Application of tho Balanco Equations
The two balance conditions expressed in Eqs. (5-25) and (5-26) can be when the impedances of the bridge arms are given in polar form, with both magnitude and phase angle. In the usual case, however, the component applied
values of the bridge arms are given, and the problem is solved by writing the balance equation in complex notation. The following examples illustrate the
procedure.
Example 5-3
The impedances of
the basic ac bridge of Fig. 5-11 are given as follows:
=
100
H
Z 80° (inductive impedance)
Z =
250
H
(pure resistance)
Z,
2
128
Bridges and Their Application
=
Zj
Chap. 5
c
400
fl
Z30 (inductive impedance)
Z 4 = unknown Determine the constants of the unknown arm.
Solution
The
first
condition for bridge balance requires that
= ZZ
Z,Z4
The second condition
(5-25)
3
known components and
Substituting the magnitudes of the
z< =
2
¥ — Z,Z,
X
250
=
400
i«r-
solving for
Z
4,
we
obtain
„ = U)0OQ ,
for bridge balance requires that the
sums of the phase angles of
opposite arms be equal or
e Substituting the
known phase 04
Hence
the
=
02
+
x
+
=
e,
03
-
=
0,
Z4 we
e3
(5-26)
angles and solving for 04
unknown impedance Z 4 can be
indicating that
+
e2
+
0
30
,
we
obtain
- 80= -
50°
written in polar form as
1,000
0
L -50°
are dealing with a capacitive element, possibly consisting of a series
combination of a resistor and a capacitor.
The problem becomes
slightly
more complex when
the
component values
of the bridge arms are specified and the impedances are to be expressed in
complex notation. In
this case, the inductive or capacitive reactances
be calculated when the frequency of the excitation voltage
is
known,
as
can only
Example
5-4 shows.
Example 5-4
The ac bridge of Fig. 5-11 is in balance with the following constants: arm AB, R = 450 CL: arm BC, R = 300 Cl in series with C = 0.265 uE; arm CD, unknown; arm DA, R = 200 H in series with L = 15.9 mH. The oscillator frequency is 1 kHz. Find the constants of arm CD. Solution
The general equation
for bridge balance states that
Z,Z 4 Z,
= R =
Z = R 2
450
= Z,Z
(5-21)
3
n
j/aiC
=
(300
-
;600)
f)
Comparison Bridges
Sec. 5-6
129
Z = R + jvL = 3
(200
+ ylOO)
£1
Z4 = unknown Substituting the
known
values in Eq. (5-21) and solving for the
z> _ This result indicates that Z4 at
a frequency of
obtain
5-6
L =
23.9
1
450
x
(200
+;i00)
_
(30o-y6oo)
unknown
yields
+jma
a pure inductance with an inductive reactance of 150
is
kHz. Since the inductive reactance
XL =
27r/L,
we
solve for
Cl
L and
mH.
COMPARISON BRIDGES Comparison Bridge
5-6.1 Capacitance
form the ac bridge can be used for the measurement of an or capacitance by comparing it with a known inductance or capacitance. A basic capacitance comparison bridge is shown in Fig. 5-12. The ratio arms are both resistive and are represented by and R 2 The standard arm consists of capacitor Cs in series with resistor R s where Cs is a high-quality standard capacitor and R sl variable resistor. Cx represents the unknown capacitance and R x is the leakage resistance of the capacitor. To write the balance equation, we first express the impedances of the four bridge arms in complex notation and we find that In
its
basic
unknown inductance
.
,
s
R
2;
Z = 3
R,
-
J
= Rx -
a)C/
a)Cx
Substituting these impedances in Eq. (5-21), the general equation for bridge balance,
we
obtain
0^
Unknown Figure 5-12
Capacitance comparison bridge.
Bridges and Their Application
130
Chap. 5
which can be expanded to
=
Two complex numbers
- Ri-^
R2R,
when both
are equal
(5-28)
R,R X
= RR 2
Rx =
or
S
and
their real terms
we
imaginary terms are equal. Equating the real terms of Eq. (5-28),
their
obtain
r£
(5-29)
Equating the imaginary terms of Eq. (5-28), we obtain
(5-30)
Equations (5-29) and (5-30) describe the two balance conditions met simultaneously and they also show the two unknowns Cx and
known
terms of the
in
To
satisfy
elements in
Rx
must be
expressed
bridge components.
both balance conditions, the bridge must contain two variable configuration.
its
that
Any two
chosen, although in practice capacitor
of fixed value and
is
equations shows that
of the available four elements could be
C
s
is
a high-precision standard capacitor
not available for adjustment. Inspection of the balance
R
s
does not appear in the expression for
eliminate any interaction between the two balance controls,
choice as a variable element.
We
further accept that
R
l
is
R
s
Cx is
Hence
.
to
an obvious
the second variable
element, as indicated in Fig. 5-12. Since
we
are measuring an
be very small, the
R
{
is
first
unknown
capacitor whose resistive effects could
adjustment should be
made
minimum sound
in the
therefore adjusted for
for the capacitive term,
and
headphones. In most cases,
the sound will not altogether disappear, because the second balance condition
has not yet been met. Hence the sound the
two
is
made
resistors
is
R
s
is
adjusted for balance of the resistive term and
to decrease further. It
achieve the true balance condition. clear
when we
is
found that alternate adjustment of
necessary to produce zero output in the headphones and to
realize that
The need
any change
in
for alternate adjustment
R
x
becomes
not only affects the capacitive
balance equation but also the resistive balance equation, since
R
x
appears in
both expressions.
The process of alternate manipulation of R and {
balancing procedure for ac bridges and
is
R
s
is
typical of the general
said to cause convergence of the balance
point. It should also be noted that the frequency of the voltage source does not
enter either of the balance equations and the bridge
independent of the frequency of the applied voltage.
is
therefore said to be
Comparison Bridges
Sec. 5-6
131
E
Chap. 6
Reference Section
^Meter
Section-^
dc Input
0-I000 V
Feedback
Range Selector
Block diagram of the dc standard /differential voltmeter
Figure 6-17 of operation.
The meter
in the differential
mode
section indicates the voltage balance between the reference section and
the dc amplifier section.
The range
selector
on the front panel of the instrument controls both the feedback
voltage and the voltage that
output in such a
way
is
applied in opposition to the reference divider
that the 1-V capability of the reference supply
is
never
exceeded.
In the third mode of operation, the instrument is connected as a voltmeter and the dc amplifier acts as a buffer stage to provide high-input impedance to the unknown voltage source. The input voltage is amplified, and the dc output voltage is applied directly to the meter circuit. The meter circuit incorporates a feedback-controlled amplifier and allows selection of its sensitivity by adjustment of the feedback loop through a front panel control, marked sensitivity. This feature provides for extreme sensitivity of the meter circuit, often on the order of 1 fxV full-scale deflection. Meaningful measurements at the very high
Sec. 6-7
Digital
sensitivities,
169
Voltmeters
however, are
difficult to
make because
of the problems of noise
generation and pickup.
An
ac-to-dc converter can be incorporated in the instrument to provide
the capability of ac voltage
6-7 DIGITAL
measurement by potentiometric methods.
VOLTMETERS
6-7.1 General Characteristics
(DVM) displays measurements of dc or ac voltages numerals instead of a pointer deflection on a continuous scale as in analog devices. Numerical readout is advantageous in many applications because The
digital voltmeter
as discrete
it
reduces
human
reading and interpolation errors, eliminates parallax error,
increases reading speed,
and often provides outputs
in digital
form
suitable for
further processing or recording.
The
DVM
is
a versatile and accurate instrument that can be used in
many
laboratory measurement applications. Since the development and perfection of
power requirements, and cost of the DVM have been drastically reduced so that DVMs can actively compete with conventional analog instruments, both in portability and price. The DVM's outstanding qualities can best be illustrated by quoting some typical operating and performance characteristics. The following specifications do not all apply to one particular instrument, but they do represent valid information on the present state of the art: integrated circuit (IC) modules, the size,
(a)
Input range: from ±1.000000
V
to
±1,000.000 V, with automatic
range selection and overload indication (b)
Absolute accuracy: as high as ±0.005 per cent of the reading
(c)
Stability: short-term,
0.002 per cent of the reading for a 24-hr period;
long-term, 0.008 per cent of the reading for a 6-month period (d) Resolution: (e)
1
part in 10
(1
u.V can be read on the 1-V input range)
Input characteristics: input resistance typically 10 Mft; input capacitance typically 40
(f)
6
pF
Calibration: internal calibration standard allows calibration independ-
ent of the measuring circuit; derived from stabilized reference source (g)
Output
signals: print
command
allows output to printer;
BCD (binary-
coded-decimal) output for digital processing or recording
Optional features
and voltage transducers.
may
ratios.
include additional circuitry to measure current, resistance, Other physical variables may be measured by using suitable
Electronic Instruments for Measuring Basic Parameters
170
Chap. 6
Digital voltmeters can be classified according to the following broad categories:
(b)
DVM Integrating DVM
(c)
Continuous-balance
(a)
Ramp-type
DVM
(d) Successive-approximation
6-7.2 Ramp-type
The operating ment of the time
it
DVM
DVM
principle of the ramp-type
takes for a linear
ramp
DVM
is
based on the measure-
voltage to rise from 0
V
to the level
of the input voltage, or to decrease from the level of the input voltage to zero.
measured with an electronic time-interval counter, and the number of digits on electronic indicating tubes. Conversion from a voltage to a time interval is illustrated by the waveform diagram of Fig. 6-18. At the start of the measurement cycle, a ramp voltage is This time interval
count
is
is
displayed as a
+ 12
V
—J— o
v
--12
v
Voltage Beingf
Measured
1
Gating
Time
Interval
Clock Pulses to
Counter
Figure 6-18
initiated; this voltage
ramp, shown
At
voltage.
Voltage-to-time conversion using gated clock pulses.
can be positive-going or negative-going. The negative-going
in Fig. 6-18, is
unknown input unknown voltage, a
continuously compared with the
the instant that the
ramp
voltage equals the
coincidence circuit, or comparator, generates a pulse which opens a gate. This gate
is
shown
in the
block diagram of Fig. 6-19. The
to decrease with time until
it
finally reaches
0
V
(or
ramp
voltage continues
ground potential) and a
second comparator generates an output pulse which closes the gate.
An
oscillator generates clock pulses
the gate to a
number of decade counting
which are allowed to pass through (DCUs) which totalize the number
units
Sec. 6-7
dc
Digital
171
Voltmeters
Input
Voltoge
Input
Comparator
Start Pulse
Gate
Oscillator
Counter
Stop Pulse
Readout
Ground Comparator
Figure 6-19
Block diagram of a ramp-type
The decimal number, displayed by the indicator
of pulses passed through the gate. tubes associated with the
digital voltmeter.
DCUs,
is
a measure of the magnitude of the input
voltage.
The sample-rate multivibrator determines the rate at which the measurecycles are initiated. The oscillation of this multivibrator can usually be adjusted by a front panel control, marked rate, from a few cycles per second to as high as 1,000 or more. The sample-rate circuit provides an initiating pulse for the ramp generator to start its next ramp voltage. At the same time, a reset ment
pulse
is
generated which returns
all
the
DCUs
to their
0
state,
removing the
display momentarily from the indicator tubes.
6-7.3 Staircase-ramp
The staircase-ramp is
DVM
DVM given in DVM but is
a variation of the ramp-type
is
block diagram form in Fig. 6-20.
somewhat simpler
It
in overall design,
resulting in a moderately priced general-purpose instrument that can be used in the laboratory,
on production
test-stands, in repair shops,
and
at inspection
stations.
DVM
This makes voltage measurements by comparing the input voltage an internally generated staircase-ramp voltage. The instrument shown in Fig. 6-20 contains a 10-MH input attenuator, providing five input ranges from 100
to
mV
to 1,000 V full scale. The dc amplifier, with a fixed gain of 100, delivers 10 V to the comparator at any of the full-scale voltage settings of the input divider.
The comparator
senses coincidence between the amplified input voltage
staircase-ramp voltage which its
cycle.
is
and the
generated as the measurement proceeds through
Sec. 6-7
Voltmeters
Digital
When
173
the measurement cycle
is first initiated,
ation oscillator) provides pulses to three
DCUs
the clock (a 4.5-kHz relax-
The
in cascade.
units counter
provides a carry pulse to the tens decade at every tenth input pulse.
The
tens
decade counts the carry pulses from the units decade and provides its own carry pulse after it has counted ten carry pulses. This carry pulse is fed to the hundreds decade which provides a carry pulse to an overrange circuit causes a front panel indicator to light up,
circuit.
The overrange
warning the operator that the
input capacity of the instrument has been exceeded.
The operator should then
switch to the next higher setting on the input attenuator.
Each decade counter unit is connected to a digital-to-analog (D/A) conThe outputs of the D/A converters are connected in parallel and provide an output current proportional to the current count of the DCUs. The staircase amplifier converts the D/A current into a staircase voltage which is applied to verter.
the comparator.
and the
When
the comparator senses coincidence of the input voltage
staircase voltage,
it
provides a trigger pulse to stop the oscillator.
current content of the counter
is
The
then proportional to the magnitude of the input
voltage.
The sample cillator triggers
second.
The
rate
and
is
controlled by a simple relaxation oscillator. This os-
resets the transfer amplifier at a rate of
two samples per
transfer amplifier provides a pulse that transfers the information
The
stored in the decade counters to the front panel display unit.
trailing
edge
of this pulse triggers the reset amplifier which sets the three decade counters to zero and initiates a
new measurement
cycle by starting the master oscillator or
clock.
The
display circuits store each reading until a
new reading
is
completed,
eliminating any blinking or counting during the computation.
6-7.4 Integrating
The
integrating
DVM
DVM measures the true average of the input voltage over DVM which samples the
a fixed measuring period, in contrast to the ramp-type voltage at the end of a measuring cycle.
A widely used technique to accomplish
integration employs a voltage-to-frequency
(V/F)
converter.
The
V/F converter
functions as a feedback control system that governs the rate of pulse generation in proportion to the
6-21.
magnitude of the input voltage.
The simplified block diagram of an integrating The dc voltage under test is applied to the input
meter circuitry from the
The attenuated input
test circuit
signal
is
DVM stage
is
given in Fig.
which
isolates the
and provides the necessary input attenuation.
applied to the
V/F converter.
This circuit consists
of an integrating amplifier, a level detector {comparator circuit), and a pulse generator.
The
integrating amplifier produces an output voltage proportional to
"O
c
'
O
CJ
1-
2§ CO
T3 co
O N OX
(J
to
"O
< o Q
O
O
Si -5
CP
>»•-
oo
o
c7)0
«
0)
.>
_
_
o 2
^
Q
N
-
?nera
o
-WW
«_
-I
o
5 i
— o CL
C
o i
en
O c
< 174
Sec. 6-7
Voltmeters
Digital
175
the input voltage, related to the input and feedback elements by the equation
(6-5)
If the input voltage is constant, the
output
is
a linear
ramp
following the equation
(6-6)
When
ramp reaches
the
a certain negative voltage level, the level detector
to the summing The sum of the input voltage and the pulse voltage is negative, causing the ramp to reverse its direction. This "retrace'* is very rapid since the pulse is large in amplitude compared to the input voltage. When the now positive-going ramp reaches 0 V, the level detector generates a reset trigger to the pulse generator. The negative pulse is removed from the summing junction of the integrating amplifier and only the original input voltage is left. The amplifier then produces a negative-going ramp again and the protriggers the pulse generator,
which applies a negative voltage step
junction of the integrating amplifier.
cedure repeats.
The voltage.
rate of pulse generation
A
is
governed by the magnitude of the dc input
larger input voltage causes a steeper
ramp and
therefore a higher
pulse repetition rate (PRR).
The major advantage of this system of A/D conversion is its ability to measure accurately in the presence of large amounts of superimposed noise, since the input
The
is
integrated.
level-detector output pulse controls the signal gate allowing the decimal
counters to accumulate a count provided by the crystal oscillator circuitry.
remainder of the circuit
is
essentially identical to
The
any conventional counter and
needs no further elaboration.
6-7.5 Continuous-balance
DVM
The continuous-balance DVM is a low-cost instrument that provides exThe accuracy of this voltmeter is usually on the order of
cellent performance.
0.1 per cent of its input range. It has
an input impedance of about 10
MH
and
acceptable resolution.
The block diagram of a servo-driven continuous-balance DVM is given in The dc input voltage is applied to an input attenuator that provides suitable range switching. The input attenuator is a front panel control that also causes a decimal point indicator to move on the display area in accordance with Fig. 6-22.
the input range selected. After passing through an overvoltage protection circuit
and ac rejection
filter,
the input voltage
is
applied to one side of a mechanical
176
Electronic Instruments for Measuring Basic Parameters
Chap. 6
Precision
Potentiometer
Range Switch Overload Protection and ac Rejection
Input
Attenuator
Filter
1
3 4
Readout
Digital
Figure 6-22
7
Functional block diagram of a servo-balancing potentiometer-type digital
voltmeter.
chopper comparator. The other side of the comparator
arm of supply.
is
connected to the wiper
the motor-driven precision potentiometer, connected across a reference
The output of
the chopper comparator, which
voltage and vibrates at the line frequency rate,
amplitude of the square wave
is
is
is
driven by the line
a square-wave signal.
polarity of the dc voltages connected to the opposite sides of the chopper.
square-wave signal fed to a
power
is
The
a function of the difference in magnitude and
The
amplified by a high-impedance, low-noise preamplifier and
amplifier.
This amplifier has special damping to minimize
overshoot and hunting at the null position.
The
servo motor, on receiving the
amplified square-wave difference signal, drives the
arm of
the precision poten-
tiometer in the direction required to cancel the difference voltage across the
chopper comparator. The servo motor also drives a drum-type mechanical indicator that has the digits 0 to 9 imprinted about the periphery of its drum segments.
The
position of the servo
motor
amount of
shaft corresponds to the
feedback voltage required to null the chopper input, and this position
is
indicated
by the drum-type indicator. The position of the shaft therefore is an indication of the magnitude of the input voltage. It is clear that this instrument does not "sample" the unknown dc voltage at regular intervals, as is the case
with more sophisticated instruments, but
continuously seeks to balance the input voltage against the internally generated reference.
Because of the different mechanical movements involved in the mech-
anism, such as the positioning of the potentiometer indicator mechanism, the average reading time
of design and low cost, however,
when extreme accuracy
is
make
not required.
this
is
arm and
the rotation of the
approximately 2
s.
Simplicity
instrument a very attractive choice
Sec. 6-7
Digital
Voltmeters
177
6-7.6 Successive-approximation
DVM
Digital voltmeters capable of 1,000 readings per second or
more
are
now
commercially available. These instruments generally use successive-approximation converters to perform the digitization (analog-to-digital conversion). A simplified block
diagram of such a
DVM
is
At the beginning of the measurement start-stop multivibrator. This sets a
control register and a 0 in register, its
all bits
1
shown
in the
most
is
significant bit
applied to the
(MSB)
of the
of less significance. Assuming an 8-bit control
reading would then be 1000 0000. This
register causes the output of the
in Fig. 6-23.
cycle, a start pulse
D/A
initial setting
of the control
converter to be one-half the reference
supply voltage (T V). The converter output is compared to the unknown input by the comparator. If the input voltage is larger than the converter reference voltage, the comparator produces an output that causes the control register to retain the 1 setting in its MSB, and the converter continues to supply its reference
output voltage of \ V.
The
ring counter next advances one count, shifting a
of the control register, and
its
1
in the
second
MSB D/
reading becomes 1100 0000. This causes the
A converter to increase its reference output by one increment, to y V + V, and another comparison with the unknown input voltage takes place. If in this case, the accumulated reference voltage exceeds the unknown voltage, the corn-J-
Input
SH
Voltage
cct
D/A Converter
Ref.
Supply
Comparator \C Gate Digital
Control Register
Delay Ring
Set
MSB
Counter
Start/
Stop
Stop Start Figure 6-23 meter.
o-
Simplified block diagram of a successive-apprxoximation digital volt-
Readout
Electronic Instruments for Measuring Basic Parameters
178
parator produces an output that causes the control register to reset
MSB
to 0.
The converter output then
returns to
Chap. 6
its
second
previous level of j
its
V
the ring counter advances another count, the third is -J-
MSB
and
When
awaits another input from the control register for the next approximation.
of the control register
and the converter output rises by the next increment, to y V + V. The measurement cycle thus proceeds through a series of successive apset to
1
shown in Fig. 6-24, retaining or rejecting the converter output manner described. Finally, when the ring counter reaches its last count, the measurement cycle stops, and the digital output of the control register represents the final approximation of the unknown input voltage.
proximations, as in the
Reject
Converter Voltage
Final
Reading
11.250 V (16 Tries)
Level
Time
Number Figure 6-24
of
Comparisons
Successive approximations are used to
make an
version. Converter reference voltages are switched to the
sequence and are rejected
if
analog-to-digital con-
comparator
in
an 8-4-2-1
the accumulated converter output exceeds the input
voltage.
With input voltages other than zation and decisions
dc, the input level
made during conversion
conversion error, a sample-and-hold (SH) circuit
following the input attenuator and amplifier, as
form, the
SH
6-25. In the
circuit
changes during
are not consistent. is
To
placed in the input, directly
shown
in Fig. 6-22. In its simplest
can be represented by a switch and a capacitor, as
sample mode the switch
is
digiti-
avoid this
in Fig.
closed and the capacitor charges to the
mode the switch is opened and the capacitor holds the voltage that it had at the instant the switch was opened. If the switch drive is synchronous with the ring counter pulse, the actual measurement and conversion take place when the SH circuit is in the hold mode. In a practical circuit the simple switch of Fig. 6-25 is replaced by fastacting transistor switches, and an operational amplifier is added to increase the instantaneous value of the input voltage. In the hold
charging current into the capacitor.
6-8
COMPONENT MEASURING INSTRUMENTS
Bridges for measuring component values of resistance, inductance, and capacitance were discussed in Chapter
5.
Bridges are potentially very accurate and
)
Component Measuring Instruments
Sec. 6-8
V; n
179
O
-TLTLTLTL Switch Drive
(a)
0"
Simple sample-and-hold circuit Figure 6-25
A
(b)
Reconstruction of the waveform
sample-and-hold circuit freezes the input voltage during digitization
so that voltage levels do not change during the successive-approximation process.
component measurements using measuring frequencies to the low megahertz region. They have some disadvantages in that they involve a variable inductor, resistor, or capacitor, depending on the type of bridge, and this usually reliable for
makes it difficult to automate or commeasurement since an actual mechanical movement is required. For
involves an operator. This adjustment puterize the
manual measurements, this tends to slow down the measurements, but computer interface, this tends to make the task nearly impossible.
6-8.1 All-Electronic
for
Component Measurements
Chapter 5 discussed the Wheatstone bridge for resistance measurements, and the simple ohmmeter was discussed in Chapter 4. This is an example of a bridge and an all-electronic instrument for measuring resistance. (In the case of the moving-coil meter, the actual meter movement is mechanical, but this could be replaced with a digital readout, making the resistance measurement allelectronic.)
There are several methods of performing an capacitance measurement where the measurement parison, as
is
all-electronic inductance or is
not performed by a com-
the case with a bridge. Figure 6-26 shows one possible
Voltage Source Figure 6-26
This circuit can measure the
value of a capacitor by measuring the current through the capacitor with a
applied voltage.
known
v
method of
v
/n
180
Electronic Instruments for Measuring Basic Parameters
Chap. 6
measuring the value of a capacitor, where a voltage is applied to the capacitor and the current through the capacitor can be measured. The relationship between the current through a capacitor and the voltage applied to the capacitor
=
Ic
where
V is
tance.
The meter
the applied voltage, is
£=
/ is
it
(6-7)
the applied frequency, and
C is
the capaci-
simply calibrated in capacitance because of the linear
lationship between the capacitance a useful circuit,
V{27TfC)
is
and the current. Although
re-
in theory this
is
not practical because of the typical values of capacitors
is
encountered in the electronics industry. Capacitors of a few picofarads are not unusual, and these capacitors typically could have working voltages of less than
RF
25 V.
current measuring devices, essentially thermocouple instruments, are
not available for currents of
less
than a few hundred milliamperes, and thus the If, as an
current expected must be greater than a few hundred milliamperes.
example, a capacitor of 10
V
applied voltage of 10
pF were
to
produce a current of 100
mA,
with an
rms, which would be safe for a 25- V capacitor, the
frequency would have to be higher than 1,600
MHz. At
this frequency,
most
capacitors have ceased to behave as capacitors and lead inductance, dissipation resistance,
and other
generator, which
impedances will dominate the measurement. In measurement is dependent on the frequency of the
parasitic
addition, the accuracy of the
would be
difficult to
control at 1,600
MHz.
Therefore, smaller
currents must be used for capacitance measurements.
An
is shown in Fig. 6-27. In this example the current sampled across a known resistance and the resultant voltage is amplified and measured. The amplifier provides the necessary gain so that the current through the capacitor can be quite small and within practicality. The voltage across the resistor can be expressed as
alternative
method
through the capacitor
is
(6-8)
R is the resistance, Vm is the generator voltage, V is the voltage across the resistor, Cis the capacitance of the unknown capacitor, and /is the frequency where
AC
Figure 6-27
and an
Amplifier
Measuring the current through a capacitor using a sampling
amplifier.
resistor
Component Measuring Instruments
Sec. 6-8
of the generator. If of the
unknown
Vm
,
f,
and
capacitance.
R
The
181
are kept constant, the voltage scale
would have
V is
a function
to be calibrated in a nonlinear
An applied frequency of a few megahertz can provide a practical system using this technique. The actual movement of the meter depends on not only the constants mentioned above, but on the gain of the amplifier. It can be difficult to maintain a constant gain in an amplifier at several megahertz, especially for the large dynamic range encounfashion because of the relationship of Eq. (6-8).
An alternative approach example the phase angle between the applied
tered while measuring capacitance using this system. is
shown
in Fig. 6-28. In this
Limiting
Phase
Amplifier
Detector
voltmeter
n-^>—i Figure 6-28
Capacitance measuring meter using the phase
of an
shift characteristics
RC circuit. voltage and the voltage across the capacitor
is
measured.
An
amplifier
is
used
scheme except that the gain of the amplifier is not a factor in the measurement. Typically, a limiting amplifier such as that found in an FM receiver would be used. The phase angle can be expressed as in this
=
0
The
angle, 0, will be read
tan-
1
A - tan- (27r/RC) I
by the meter
calibrated in capacitance since this angle tance. This
would
(6-9)
Xc
in this circuit is
and the meter can be
a function of the
unknown
capaci-
result in a nonlinear but useful display.
Using the Taylor expansion, the expression for the angle 6 can be rewritten: 0
=
=
1
tan" (27ryKC)
(Itt/RC)
-
^(27t/KC)
3
+
^(2tt/KC)
5 .
.
.
(6-10)
As can be seen from the Taylor expansion, the value of the arctangent approach the angle, in radians, if the value of (lirfRC) is small. To gain an idea of how small the arctangent must be so that just one term of the Taylor expansion may be used, that is, the first term, consider an arctangent of less than 0.1. The value of the Taylor expansion using the first term only is, of will
The actual value of the arctangent is 0.0996687, which is only 0.3% than the actual angle, in radians. If the meter in this technique were calibrated directly in capacitance and the phase angle were restricted to less than 0. 1 rad,
course, 0.1. less
the error due to
0
=
{Itt/RC) for
this less
approximation would not exceed 0.3%. Therefore,
than
0.1.
Electronic Instruments for Measuring Basic Parameters
182
The capacitance meter based on
Chap. 6
the circuit of Fig. 6-28 could be configured
by changing the value of R, such that the full-scale As an example, assume that it is desired that the lowest range
to cover several ranges
reading
rad.
is 0. 1
cover from 0 to 100 at
MHz
1
pF
1 MHz. Therefore, pF must be 0.1 rad or
with a source frequency of
full scale,
the phase shift of the resistance, R, and 100
=
0.1
(IttR
X
100 pf)
(6-11)
Solving for R,
* To
cover from 10 to 1,000
for 10
pF
scale,
and
full scale,
1
pF
ft for
6.28
(6 " 12)
x\o- w
full-scale meter, the resistors
30 pF, 159
H
for 100 pF, 47.7
could be 1,590
H
for 300
pF
H
full
pF full scale. It is difficult to measure capacitors pF using the 1-MHz source because the impedance of a capacitor
15.9 fl for 1,000
greater than 100 at
477
=
Mhz
is
too low to achieve an accurate measurement with this type of
instrument.
6-8.2 Sources of Error
The accuracy of low-capacitance measurements
is
limited by the distributed
capacitance of the measuring circuits. Figure 6-29 shows the basic measuring
Generator Resistance Limiting Amplifier
Phase Detector
Voltmeter
Amplifier Input Capacitance
Figure 6-29 ance,
Capacitance measuring meter, showing the parasitic inductance,
circuit with the parasitic capacitances added. series
resist-
and capacitance.
The
series resistance, R,
has some
inductance and the input of the amplifier will have a certain amount of
input capacitance. Primarily, the amplifier input capacitance will have the great-
on the accuracy of the measurement. It would be difficult to design an amplifier with an input capacitance low enough to allow measurements of capacitors below 10 pF without some form of compensation. Figure 6-30 shows
est effect
a modified measuring circuit allowing the effects of the input capacitance of the amplifier to be nulled out. In this example, the resistor has been placed at the
183
W 184
Electronic Instruments for Measuring Basic Parameters
amplifier input,
and the
out-of-phase component. injecting for the
some of
trimming
signal source
The
is
Chap. 6
applied to a transformer to create an
of the input capacitance are nulled out by
effects
the out-of-phase signal through a variable capacitor. Except circuits, the
operation of this capacitance measuring system
is
similar to the previous example.
Another source of error
The phase
shift
of the
is
the harmonic distortion of the signal source.
RC circuit, which is the heart of this capacitance measuring
system, will satisfy the equations presented only sine function without
any harmonic
distortion.
if
the signal source
is
a pure
For an accuracy of 0.3 per
cent,
which was the theoretical limit for the linear approximation using the Taylor expansion, the harmonic content of the signal source must be better than 50 dB down from the nominal level. A crystal oscillator is capable of supplying a signal purity of this magnitude only if the output is carefully coupled from the oscillator. In addition to the coupling point, the signal should be passed through a lowpass
filter.
By sistance.
far the largest source of error
The
to the resistance of the circuit, but the
to the capacitance but to the point join, as
is
the equivalent series or parallel re-
series resistance called equivalent series resistance, or
shown
in Fig. 6-31.
phase measurement
where the
ESR
is
and the
made
ESR, adds relative not
circuit resistance
This causes an error because the phase
shift is
not
Durce
Figure 6-31
Effects of equivalent series resistance
on capacitor measurement.
being measured accurately. Likewise, an equivalent parallel resistance, which
is
an erroneous reading because it changes the equivalent resistance as seen by the capacitor and hence changes the phase shift. This capacitance measuring method is not suitable for measuring capacitors
due to leakage
resistance, will cause
with high dissipation factors or high ESR. Corrections can be dissipation factor or
ESR
is
made
if
the actual
known, but capacitance and dissipation factor can
both be measured in a capacitance bridge. Generally, the quality of capacitors
measured by this instrument is very good with and dissipation factors, and the errors caused by these resist-
in the region of capacitance
insignificant
ESR
ances are negligible.
The same
basic system can be used to measure inductance. Figure 6-32
1
Component Measuring Instruments
Sec. 6-8
185
Limiting
R Signal
,
Source
^
~
Amplifier
Phase
r\
Detector
.
Modification of the capacitance measuring circuit described previously,
Figure 6-32
allowing the measurement of inductance.
shows a modification of the capacitance measuring instrument previously cussed to measure inductance. In this circuit the phase shift is
= tan-'^
Figure 7-6
uniform
Force
/
electric field.
on
Cathode Ray Tube
Sec. 7-3
213
we
Substituting Eq. (7-2) into Eq. (7-5),
a
=
where a
obtain
mm
= — =
acceleration of the electron, in
/= m = When
2
(7-6)'
m/s 2
N
force on the electron, in
mass of the
m/s
kg
electron, in
the motion of an electron in an electric field
is
discussed
it is
usually
customary Cartesian axes, as shown in Fig. 7-7. In discussing the concepts which follow, we shall use subscript notation for the vector components of velocity, field intensity, and acceleration. For example, the specified in respect to the
velocity
component along the
X
axis will be written vx (m/s); the
component The motion of an electron
of the force along the
Y
in a given electric field
cannot be determined unless the
axis
is
fy
written
and displacement are known. The term
(N), etc.
initial
values of velocity
initial represents the
value of velocity
or displacement at the time of observation, or time
be used to indicate these
ponent along the
X
Figure 7-7
Consider
axis
=
0.
For example, the
initial values. is
t
written as v0jc
The
subscript 0 will
initial velocity
com-
.
Cartesian coordinate system.
now an
electric field of constant intensity
with the lines of force
Y direction, shown in Fig. 7-8. An electron entering this field in the positive X direction with an initial velocity v0x will experience a force. Since the field acts only along the Y axis, there will be no force along either the X axis or the Z axis, and the acceleration of the electron along these pointing in the negative
axes must be zero. Zero acceleration means constant velocity, and since the electron enters the field in the positive it
will continue to travel
along the
Z
axis
electron along the
along the
was zero
Z
at
time
X t
X
direction with an initial velocity v0x
=
0,
there will be no
Figure 7-8
movement of
axis. ly
Path of a moving electron
a uniform electric
field.
in
,
axis at that velocity. Since the velocity
,
the
Chap. 7
Oscilloscopes
214
Newton's second law of motion, applied to the force on the electron acting in the
Y
direction, yields
f = ma
= — ee =
mm
a= —/
or
v
constant
(7-7)
Equation (7-7) indicates that the electron moves with a constant acceleration in Y direction of the uniform electric field. To find the displacement of the
the
electron due to this accelerating force, velocity
we
use the well-known expressions for
and displacement: v
=
x
=
+
v0
+
x0
(m/s)
at
+
v0 t
j at
2
(m)
(velocity)
(7-8)
(displacement)
(7-9)
Subject to the initial condition of zero velocity in the
Y
direction (v0y
=
0) Eq.
(7-8) yields vy
which, after substitution of Eq.
=
(7-7), results in
= The displacement of which
yields,
=
(m/s)
m
the electron in the
applying the
zero velocity (v0y
(m/s)
ay t
initial
Y
(7-10)
direction follows
from Eq.
conditions of zero displacement (y0
=
(7-9),
0)
and
0),
y
=
\ay t 2
(m)
which, after substitution of Eq. (7-7), results in
y
The
X
(m)
(7-11)
by the electron in the time interval and we can write, again using Eq. (7-9),
=
x which, after applying the 0),
^2m
distance, traveled
initial velocity v0x
=
=
x0
initial
+
v0x t
+
\a x t 2
conditions for the
t,
depends on the
(m)
X
direction (x 0
=
0 and a x
becomes
x
=
v0x t
or
Substituting Eq. (7-12) into Eq. (7-11),
t
we
=
—
(s)
(7-12)
obtain an expression of the vertical
deflection as a function of the horizontal distance traveled
by the electron:
Cathode Ray Tube
Sec. 7-3
215
Equation (7-13) shows that the path of an electron, traveling through an of constant intensity and entering the
field
flux, is
parabolic in the
X-Y
field at right
electric
angles to the lines of
plane.
In Fig. 7-9 two parallel plates, called deflection plates, are placed a distance
d
apart and are connected to a source of potential difference
electric field € exists
between the
plates.
The
^a
(V/m)
Ed
,
so that an
intensity of this electric field
is
given by
€
An
=
(7-14)
electron entering the field with an initial velocity v0x
is
deflected toward the
positive plate following the parabolic path of Eq. (7-13), as indicated in Fig. 7-9.
When
the electron leaves the region of the deflection plates, the deflecting
no longer exists, and the electron travels in a straight line toward point P' a point on the fluorescent screen. The slope of the parabola at a distance x = ld where the electron leaves the influence of the electric field, is defined as force y
,
=
tan 0
where y
is
=
x
ld
— dx
=
= —
The
straight line of travel of the electron
and
this tangent intersects the
X
V — m\r
(7-16)
0x
is
tangent to the parabola at x
The
axis at point O'.
=
ldi
location of this apparent
given by Eq. (7-13) and Eq. (7-16) since
m< = i ^-0 =
tan
ety ld
2
x
(m) v
. 17) (7 y
'
'
p'
Straight
Parabolic
/Path
/Path
T
y
!
x and
yields
tan 0
is
O7 " 15 )
given by Eq. (7-13). Differentiating Eq. (7-13) with respect to
substituting
origin 0'
dy
4dx
_u
Vo
~4(-
r-
_
L
-
21
—ror '
-xt
J
2
2
~
!
1
Figure 7-9
Deflection of the cathode ray beam.
p
Oscilloscopes
216
The apparent
origin O'
L from
distance
The
is
Chap. 7
therefore at the center of the deflection plates and a
the fluorescent screen.
deflection
on the screen
is
given by
D = L Substituting Eq. (7-16) for tan 0,
we
(m)
tan 6
(7-18)
obtain e
D = L -^4 The
(m)
(7-19)
kinetic energy of the electron entering the area
with an
between the deflection plates
initial velocity v0x is
eEa where
E
a is
(7-20)
the accelerating voltage in the electron gun. Rearranging (7-20),
obtain
vL
=
^
(7-21)
m
Substituting Eq. (7-14) for the field intensity ey
of the electron in the
X
where
D = L = — d = Ed — E = ld
a
deflection
,
and Eq. (7-21)
direction v0x into Eq. (7-19),
mvL
we
2dEn
distance from center of deflection plates to screen (meters) effective length of the deflection plates (meters)
distance between the deflection plates (meters) deflection voltage (volts)
accelerating voltage (volts)
Equation (7-22) indicates that for a given accelerating voltage is
for the velocity
obtain
on the fluorescent screen (meters)
particular dimensions of the
screen
we
CRT,
directly proportional to the deflection voltage
tionality indicates that the
E
the deflection of the electron
Ed
.
a
and
for the
beam on
the
This direct propor-
CRT may be used as a linear voltage-indicating device. Ed was a fixed dc voltage. However, the deflection
This discussion assumed that voltage usually
is
a varying quantity and the image on the screen follows the
variations of the deflection voltage in a linear manner, according to Eq. (7-22).
The
deflection sensitivity
S
of a
CRT
is
defined as the deflection on the
screen (in meters) per volt of deflection voltage.
S = where
S =
deflection sensitivity
Wd = E
^W 2dE
(m/V)
a
(
By
m/v >
definition, therefore
(7 " 23)
Cathode Ray Tube
Sec. 7-3
The
deflection factor
S and
is
217
G of a CRT,
by
definition,
G=l all
the reciprocal of the sensitivity
2
=
lt
s with
is
expressed as
(V/m)
(7 " 24)
terms denned as for Eqs. (7-22) and (7-23). The expressions for deflection
sensitivity
S and
deflection factor
G
CRT
indicate that the sensitivity of a
is
independent of the deflection voltage but varies linearly with the accelerating potential. High accelerating voltages therefore produce an electron beam that requires a high deflection potential for a given excursion on the screen. A highly
beam possesses more kinetic energy and therefore produces a brighter image on the CRT screen, but this beam is also more difficult to deflect and we sometimes speak of a hard beam. Typical values of deflection factors range from accelerated
10 V/cm to 100 V/cm, mm/V, respectively.
corresponding to
sensitivities of 1.0
mm/V
to 0.1
7-3.3 Postdeflection Acceleration
The amount of light given off by the phosphor screen depends on the amount of energy that is transferred to the phosphor by the electron beam. If the electron beam is to be deflected at a rapid rate, allowing the oscilloscope to respond to fast occurring events, the velocity of the electron beam must be great; otherwise, the light output will drop to accelerate the electron
beam
other hand the greater electron deflect the It
off.
Thus, for a
to the greatest
beam
fast oscilloscope
amount
velocity will
it is
desirable
possible, while
make
it
more
on the
difficult to
beam.
can be seen that the greater the accelerating potential the more
difficult
beam. This would require higher deflection voltages, but, more important, because the voltage is higher the time change of voltage, that is, dV/dt, is also greater. This would require not only higher voltage for it is
to deflect the electron
deflection but higher currents to charge the capacitance of the deflection plates.
This becomes a very significant problem for high-frequency oscilloscopes with frequency responses greater than 100
MHz. Modern
cathode ray tubes use a
two-step acceleration to eliminate this problem. First, the electron
beam
is
accelerated to a relatively low velocity through a potential of a few thousand volts.
The beam
is
is
further accelerated to
amount of
acceleration after the
then deflected and, after deflection,
the desired final velocity. In this fashion the
deflection does not affect the deflection sensitivity. This type of cathode ray tube is
called the postdeflection acceleration tube.
Figure 7-10 shows a diagram of a postdeflection acceleration cathode ray
beam scan. manner beam is further
tube using a mesh that further increases the amount of the electron In this example the electron similar to the previous
beam
is
accelerated and deflected in a
example of the simple tube. However, the V or more, after the
accelerated through a very high potential of 10,000
deflection,
218
Chap. 7
High Voltage Acceleration Electrode
A
Figure 7-10
postdeflection acceleration oscilloscope tube using a scan expansion
mesh.
on the deflection sensitivity. A metallic mesh is beam, and acts as a magnifying lens that causes the deflection to be further increased, which improves the deflection sensitivity. With this technique, deflection sensitivity can remain on the order of 5 to 50 V/ cm even though the total electron beam acceleration is more than 10,000 V. There are several disadvantages to the mesh type of postdeflection acceleration cathode ray tube. First, the mesh tends to defocus the electron beam and make the spot broader than it would be without the mesh interfering with so
it
does not have an
suspended
effect
in the electron
the beam. Second, the
mesh conducts some of the electron beam away from the beam current and thus reduced spot intensity.
screen. This results in a reduced
Another problem with the postdeflection acceleration cathode ray tube, and this problem is not unique to the mesh, is that the electron beam tends to be defocused in the vicinity of the deflection plates owing to repulsion from charge distributions within the tube. Several recent advances in cathode ray tube design have eliminated the
mesh and
alleviated these problems, thus producing a high-perform-
ance electron gun for use in high-frequency cathode ray tubes. Figure 7-11 shows the electron gun for the meshless cathode ray tube. The electron
beam
is
generated from a conventional heated cathode surrounded by
The first accelerating anode and two focus electrodes follow and provide focus, as well as the first accelerating voltage. These focus electrodes differ from the cylindrical elements used in the conventional tube in that they are constructed from individual metal wafers with noncylindrical holes in the the control grid.
center, as can be seen in Fig. 7-12. This allows for a different focusing characteristic in the horizontal plane
one plane while being convergent
and the
vertical plane, typically divergent in
in the other.
The
holes in the center of the
Cathode Ray Tube
Sec. 7-3
219 Quadrupole Scan Expansion
Focus Element
Hrst Grid
Anode
\
\
Cathode Vertical
Defection Plates
Horizontal Deflection Plates
High-Voltage Acceleration Electrode
Figure 7-11
Diagram of
a meshless scan expansion postdefiection acceleration cath-
ode ray tube.
metal wafers can be formed with greater precision than in a formed cylinder,
and thus greater tolerances can be achieved at a lower cost. After the two focusing electrodes, the beam passes through the vertical deflection plates. The beam at this point is not fully focused, which decreases the amount of beam distortion due to the internal charge distributions. The
beam
will
be further focused after deflection to provide a fine spot.
After vertical deflection, the
then deflected in
passes through a scan expansion lens
amount of beam bending
that increases the
lens,
beam
in the vertical plane.
The beam
is
the horizontal direction and passed through another electron
which provides additional focusing.
The beam
Figure 7-12 lenses.
is
accelerated to the final velocity by a quadrupole lens, which
Modern
oscilloscope tube electron
gun showing the quadrupole electron
220
Oscilloscopes
Chap. 7
provides not only an increase in electron velocity, but adds to the scan angle (scan expansion, which
is
similar to the
distorting or defocusing the electron
The
V/cm
result of this design
mesh
an increased deflection
is
example) without
in the previous
beam. sensitivity, typically 2.3
and 3.7 V/ cm in the horizontal direction. The and horizontal deflection sensitivities is due to
for the vertical deflection
difference
between the
vertical
the fact that the vertical deflection occurs at a lower
beam
velocity.
Because the
horizontal deflection of the oscilloscope involves only a time linear sweep, while
more
the vertical deflection requires complex waveforms, the
sensitive deflection
should be reserved for the vertical direction.
Using the meshless electron gun, 100-MHz plus oscilloscopes can be conV or even less for deflection. The meshless tube, being considerably shorter, results in smaller and lighter oscilloscopes for laboratory and portable use. structed with integrated circuits using only 40 or 50
7-3.4 Screens for
When
the electron
CRTs
beam
strikes the screen of the
CRT,
produced. The screen material on the inner surface of the
a spot of light
CRT
is
that produces
this effect is the phosphor. The phosphor absorbs the kinetic energy of the bombarding electrons and reemits energy at a lower frequency in the visual spectrum. The property of some crystalline materials, such as phosphor or zinc oxide, to emit light when stimulated by radiation is called fluorescence. Fluorescent materials have a second characteristic, called phosphorescence, which
refers to the property of the material to continue light emission
even after the
source of excitation (in this case the electron beam)
The length of
is
time during which phosphorescence, or afterglow, occurs of the phosphor. Persistence for the
CRT
is
cut is
off.
called the persistence
usually measured in terms of the time required
image to decay to a certain percentage (usually 10 per cent) of the
original light output.
The
intensity of the light emitted
depends on several
from the
CRT
factors. First, the light intensity
is
screen, called luminance,
controlled by the
number beam
of bombarding electrons striking the screen per second. If this so-called current
is
increased, or the
same amount of beam current
smaller area by reducing the spot
size,
is
concentrated in a
the luminance will increase. Second,
luminance depends on the energy with which the bombarding electrons strike the screen, and this, in turn, is determined by the accelerating potential. An increase in accelerating potential will yield an increase in luminance. Third,
luminance
is
a function of the time the
beam
strikes a given area of the
therefore sweep speed will affect the luminance.
And
function of the physical characteristics of the phosphor
finally, itself.
phosphor;
luminance
Almost
all
is
a
man-
ufacturers provide their customers with a choice of phosphor materials. Table 7-1
summarizes the characteristics of some of the commonly used phosphors. As Table 7-1 shows, a number of factors must be considered in selecting
1
a
e
§
.s
1-5 C- T3
P
c
CD
c
o
3
0
_
o
If I £ 2 OnC fc
0
£
i
£ 8
3
c
t **
'P
—o
a 2
,o rrt
•c
o o
interrupted and
current charges the target in the negative direction.
The
target
voltage therefore decreases (becomes less positive), and the secondary-emission
The
ratio changes, following the curve of Fig. 7-46(b).
rate of charge decreases
A
on the curve. At this point, the secondary-emission current equals the primary beam current, and the net chargas the target voltage approaches point
ing rate
is
zero.
emission ratio
At point A,
is
the target voltage
and the
one,
called the lower stable point,
is
slightly negative, the secondary-
A
target has reached a stable condition. Point
and the target
is
is
considered to be in the erased
condition. If the initial or starting voltage of the target is to the right of crossover
+ 100 V
point C, say at
in Fig. 7-46(b), the secondary-emission ratio
than one. This means that Is
is
greater than Ip
net electron flow leaving the target surface.
When
switch S
target continues to emit secondary electrons, so that
more
positive.
Hence the secondary-emission
point
B where
the rate of discharge
is
is
greater
and there must therefore be a
ratio
it
is
now
opened, the
discharges and becomes
moves up along the curve
to
once again zero and the target obtains a
At this so-called upper stable point the secondary-emission ratio and the target is considered to be in the written condition. As long as the primary gun is on and primary electrons bombard the target, the target will always be at a stable point, upper or lower, depending on the initial voltage of the target. Crossover point C on the curve is uniquely unstable in the sense that the target voltage will always move up to point B or down to point A, depending on which way the target voltage is first shifted by stable condition.
is
one,
noise.
The
CRT of Fig.
7-46
is
an elementary bistable storage device.
Its
condition
can be interrogated by measuring the target voltage. If the target voltage "high," the target
The tube
is
written;
if
the target voltage
therefore has an electrical readout
is
and
"low," the target its
is
storage condition
is
erased. is
not
visible.
Figure 7-47(a) shows the principle of a bistable storage tube capable of
and erasing an image. This storage tube differs from the one in Fig. 7-46(a) in two aspects: It has a multiple-target area, and it has a second electron gun. The second electron gun is called the flood gun; it emits lowvelocity primary electrons that flood the entire target area. The distinguishing feature of the flood gun is that it floods the target at all times and not just intermittently as does the writing gun. The cathode of the flood gun is at ground
writing, storing,
potential, so that the target voltage will follow the secondary-emission curve
indicated in Fig. 7-47(b).
negative with respect to
+ 200
The lower
stable point of the target
the flood gun cathode, and the upper
V, the collector voltage.
The cathode of
is
a few volts
stable point
is
at
the writing gun, however,
is
at
Sec. 7-1
Special Oscilloscopes
1
253 Collector
Mesh
(a)
Storage tube with multiple targets and two electron guns
Writing
Secondary
Gun Action
Emission Ratio
Flood Gun Action
=
1
= 0
-2000
V
Writing
Target
Gun Cathode
Voltage
(b)
Figure 7-47
— 2,000 curve.
the
V, and
It is
sum
its
Storage
CRT
Secondary emissions
with multiple targets and two electron guns.
secondary-emission curve
found that the combined
effect
is
superimposed on the flood gun
of writing gun and flood gun
of the individual effects of each electron
The
flood
gun
is
on
at all times.
stable point, the erased condition.
When
Assume
beam by
that the target
the writing gun
is
is
gated on,
electrons arrive at the target with a potential of 2,000 V,
is
simply
itself.
at its its
lower
primary
which causes high
secondary emission from the target. The target voltage therefore leaves the lower stable point and starts to increase. The flood gun electrons, however, attempt to maintain the target in voltage. If the writing
its
gun
stable condition
is
and oppose the increase
in target
switched on long enough to carry the target past
gun electrons will aid the writing gun electrons and carry the target all the way to the upper stable point, so that the target is written. Even if the writing gun is now switched off, the target will be held in its upper stable condition by the flood gun electrons, thereby storing the information delivered by the writing gun. When the writing gun is not switched on the crossover point, the flood
254
Oscilloscopes
Chap. 7
long enough to carry the target past the crossover point, the flood gun electrons will
simply
move
the target back to
its
lower stable condition, and storage does
not occur.
Erasing the target simply means restoring the target voltage to the lower
can be accomplished by pulsing the collector negative, so that momentarily repels the secondary-emission electrons and reflects them back
stable point. This it
into the target. This reduces the collector current Is ratio
and the secondary-emission
drops below one. The target then collects primary electrons from the flood
gun (remember that the writing gun voltage decreases until
it
is
off)
and charges negative. The
target
reaches the lower stable point where the charging
ceases, and the target is in the erased condition. After erasure, the collector must be returned to its original positive voltage ( + 200 V in this case), and the erase pulse must therefore be returned to zero. As indicated in Fig. 7-47(a), this must happen gradually, so that the target is not accidentally driven past the crossover point and becomes written again.
The
target area of the storage tube in Fig. 7-47(a) consists of a
number
of small individual metal targets electrically separated from one another and
numbered from 1 to 5. The flood gun is of simple construction, without deflection plates, and it emits low-velocity electrons that cover all the individual targets. When the writing gun is gated on, a focused beam of high-velocity electrons is directed at one small target (number 3 in this case). This one target then charges positive and is written to the upper stable point. When the writing gun is turned off again, the flood electrons
hold target 3 at
its
upper stable point
(store). All
the other targets are held at their lower stable points (erase).
The
last step in
our development of the bistable direct-viewing storage
tube consists of replacing the individual metal targets with a single dielectric sheet, as in the typical tube of Fig. 7-48. This dielectric storage sheet consists
of a layer of scattered phosphor particles capable of having any portion of
its
surface area written and held positive or erased and held negative without affecting the adjacent areas
on the surface of the
deposited on a conductive-coated glass faceplate. the storage target backplate, and
it
sheet. This dielectric sheet
The conductive coating
is
is
called
the collector of secondary-emission elec-
is
gun and its deflection plate assembly, this has two flood guns and a number of collimation electrodes that
trons. In addition to the writing
storage
CRT
form an electron lens
to distribute the flood electrons evenly over the entire
surface area of the storage target.
After the write gun has written a charge image on the storage target, the flood guns will store the image.
bombarded by form of visible
The
written portions of the target are being
flood electrons that transfer energy to the light.
phosphor layer
in the
This light pattern can be viewed through the glass faceplate.
Since the storage target areas are either positive or negative, the light output
produced by the flood electrons There is no gray scale
brightness.
is
either at full brightness or at
in
between.
minimum
Sec. 7-1
Special Oscilloscopes
1
255
Col limating Electrodes to Shape Flood Beams
Faceplate
Layer
Storage Target
Backplate Figure 7-48
Schematic view of a bistable storage tube (courtesy Tektronix,
Inc.).
7-11.2 Sampling Oscilloscope
When
the frequency of the vertical deflection signal increases, the writing
speed of the electron speed
is
beam
increases.
The immediate
a reduction in image intensity on the
sufficient
image
brilliance, the electron
more
CRT
result of higher writing
screen. In order to obtain
beam must be
accelerated to a higher
and normal image brightness is maintained. An increase in electron beam velocity is easily achieved by raising the voltage on the accelerating anodes. A beam velocity so that
kinetic energy
is
available for transfer to the screen
with higher velocity also needs a greater deflection potential to maintain the deflection sensitivity. This immediately places higher
demands on the
vertical
amplifier.
The sampling
oscilloscope uses a different approach to improve high-
frequency performance. In the sampling oscilloscope the input waveform reconstructed from
many samples
is
taken during recurrent cycles of the input
waveform and so circumvents the bandwidth limitations of conventional CRTs and amplifiers. The technique is illustrated by the waveforms indicated in Fig. 7-49.
In reconstructing the waveform, the sampling pulse turns the sampling circuit
instant
on for an extremely short time is
measured. The
CRT
spot
is
interval.
sponding voltage input. The next sample the input
waveform
The waveform
voltage at that
then positioned vertically to the correis
taken during a subsequent cycle of
at a slightly later position.
zontally over a very short distance and
is
The
CRT
spot
is
moved
repositioned vertically to the
hori-
new
256
Sec. 7-1
1
Special Oscilloscopes
257
value of the input voltage. In this
by point, using as
many
way
the oscilloscope plots the
waveform point
as 1,000 samples to reconstruct the original
The sample frequency may be
as
waveform. low as one-hundredth of the input signal
frequency. If the input signal has a frequency of 1,000
MHz,
the required
bandwidth of the amplifier would be only 10 MHz, a very reasonable figure. A simplified block diagram of the sampling circuitry is given in Fig. 7-50. The input waveform, which must be repetitive, is applied to the sampling gate. Sampling pulses momentarily bias the diodes of the balanced sampling gate in the forward direction, thereby briefly connecting the gate input capacitance to the test point. These capacitances are slightly charged toward the voltage level of the input circuit. The capacitor voltage is amplified by the vertical amplifier and applied to the vertical deflection plates. Since the sampling must be synchronized with the input signal frequency, the signal amplifier, allowing the
a trigger pulse it
is
sweep triggering
delayed in the vertical
is
done by the input
to be
signal.
When
received, the avalanche blocking oscillator (so called because
uses avalanche transistors) starts an exactly linear
applied to a voltage comparator.
The
ramp
voltage,
which
voltage comparator compares the
voltage to the output voltage of a staircase generator. are equal in amplitude, the staircase generator
is
When
is
ramp
the two voltages
allowed to advance one step
and simultaneously a sampling pulse is applied to the sampling gate. At this moment, a sample of the input voltage is taken, amplified, and applied to the vertical deflection plates.
Input
Sampling
Vertical
Vertical
Signal
Gate
Amplifier
Signal
Sampling Pulse
P
Scanning
*
Trigger
Blocking
Input
Oscillator
Ramp
Voltage
Staircase
Generator
Comparator
Generator
Horizontal Signal
Time Scale />
Magn.
Attenuator
Figure 7-50
Simplified block diagram of the sampling circuitry (courtesy Hewlett-Packard
Com-
pany).
The
real-time horizontal sweep
is
shown
in Fig. 7-50, indicating the hor-
beam. Notice that the horizontal displacement of the beam is synchronized with the trigger pulses which also determine the moment of sampling. The resolution of the final image on the CRT screen is determined izontal deflection rate of the
by the
size of the steps of the staircase generator.
greater the horizontal distance between the trace.
CRT
The
larger these steps, the
spots that reconstitute the
Chap. 7
Oscilloscopes
258
REFERENCES Prensky, Sol D., and Castellucis, Richard L., Electronic Instrumentation, 3rd
1.
chap. 10. Englewood
Cliffs, N.J.:
van Erk, Rien, Oscilloscopes: Functional Operation and Measuring Examples.
2.
ed.,
Prentice-Hall, Inc., 1982.
New
York: McGraw-Hill Book Company, 1983.
PROBLEMS 1.
What
are the major blocks of the oscilloscope,
2.
What
are the major components of a cathode ray tube?
(3/
How
4.
the electron
is
What
beam focused
How much
voltage
is
What
7.
Why
on the
1° if
the accelerating potential {6)
on the face of the cathode ray tube?
deflection sensitivity?
required across two deflection plates separated by
an electron beam
deflect
to a fine spot
does increasing the writing rate of an oscilloscope by increasing the
effect
accelerating potential have 5.
and what does each do?
is
the effective length of the deflection plates
1,000
is
1
2
cm to cm and
V?
the velocity of electrons that have been accelerated through a potential of
is
2,000 V? are the operating voltages of a cathode ray tube arranged so that the deflection
plates are nearly
($}
How
is
ground potential?
the vertical axis of an oscilloscope deflected?
How
does this
differ
from the
horizontal axis? 9.
What
is
oscilloscope probe compensation!
noted when the compensation
is
How
this adjusted?
is
11.
Why Why
12.
What
are the advantages of dual trace over dual
How
does alternate sweep compare with chopped sweep?
10.
13.
What
effects are
not correctly adjusted?
is
an attenuator probe used?
is
a delay line used in the vertical section of the oscilloscope?
beam
for multiple-trace oscilloscopes?
When would
one method
be chosen over the other?
When
14.
What
is
15.
What
are the advantages of using an active voltage probe?
16.
How
delayed sweep?
is it
used?
are the effects of direct current
on the
flux density of the current
probe min-
imized? 17.
What is
18>
is
the relationship between the period of a
How
does the digital storage oscilloscope
differ
cilloscope using a storage cathode ray tube? 19.
waveform and
its
frequency?
How
from the conventional storage
What
What
os-
are the advantages of each?
does the sampling oscilloscope increase the apparent frequency response of an
oscilloscope? 20.
How
an oscilloscope used to determine frequency?
precautions must be taken
when
using a sampling oscilloscope?
8
CHAPTER
SIGNAL GENERATION
The generation of signals is an important facet of electronic troubleshooting and development. The signal generator is used to provide known test conditions for the performance evaluation of various electronic systems and for replacing missing signals in systems being analyzed for repair. There are various types of signal generators, but several characteristics are
common
to all types. First, the fre-
quency of the signal should be well known and stable. Second, the amplitude should be controllable from very small to relatively large values. Finally, the signal should be free of distortion.
There are many variations of these requirements, especially for specialized and sweep generators, etc.,
signal generators such as function generators, pulse
and these requirements should be considered
8-1
as generalizations.
THE SINE-WAVE GENERATOR
Because of the importance of the sine function, the sine-wave generator represents the largest single category of signal generators. This instrument covers the
many
frequency range from a few hertz to the sine-wave generator
is
as
shown
gigahertz, but in
its
simplest form
in Fig. 8-1.
The simple sine-wave generator consists of two basic blocks, an oscillator and an attenuator. The generator's performance depends on the success of these two main parts. The frequency accuracy and stability and freedom from distortion depend on the design of the oscillator, while the amplitude accuracy depends on the design of the attenuator. 259
Signal Generation
260
Chap. 8
Set Level
Set Frequency
/
/
Altenuator
RF
Output Block diagram of a simple
Figure 8-1
sine-wave generator.
Oscillator
8-1.1 Inductor-capacitor
There
is
Tuned
Oscillators
a broad class of oscillators that use the resonant characteristics
A
of an inductor-capacitor, LC, circuit to generate a stable frequency.
block
diagram of an oscillator is shown in Fig. 8-2. The oscillator consists of an amplifier and a feedback network such that the total gain of the loop, that is, the gain of the amplifier divided by the loss of the feedback network,
equal to one, and the total phase shift around the loop
is
is
exactly
zero. Oscillators are
designed such that these characteristics are met at only one frequency. This can
be achieved by using various combinations of inductors, capacitors, and
resistors.
Feedback Network Output Figure 8-2
Block diagram of an
oscillator,
showing the amplifier and feedback
net-
work.
The resonant frequency of
a circuit
is
given by
1
/=
(8-1)
2ttJZc where L is the circuit inductance in henrys, C is the circuit capacitance in farads, and / is the resonant frequency in hertz. When a resonant circuit is used in the feedback of an oscillator, the oscillation frequency is the resonant frequency of the circuit.
Figure 8-3 shows the actual circuit of a Hartley oscillator and the equivalent
showing the amplifier and feedback components. Because a common-
circuit
emitter amplifier
used as the active element of the oscillator,
is
that the circuit has a phase shift
operating frequency.
phase
shift
oscillator
be clear
is
apparent
The feedback network,
that
is,
the resonant circuit, has a
of 180° at resonance. Therefore, the phase shift requirement for the
can be met
how
at the resonant
frequency of the tuned
circuit. It
may
not
the loop gain can be equal to one, especially since the gain of a
transistor amplifier can be quite high,
For an
it
to the amplifier of 180° regardless of the
due
and there
is
no
loss in the
tuned
oscillator to sustain oscillations, the gain of the active element
reduced, and this
is
circuit.
must be
accomplished by automatic adjusting of the operating charthrough self-bias. The amplitude of the ac voltages
acteristics of the transistor
The Sine-Wave Generator
Sec. 8-1
261
+V
Noninverting Amplifier Figure 8-3
Hartley oscillator using a bipolar junction transistor.
in the oscillator build until the effective gain of the transistor is
the total loop gain
is
equal to one. This
is
accomplished
in
most
reduced so that
oscillator circuits
by increasing the transistor bias voltages so that the gain of the device is reduced. This usually results in large amplitude and distorted voltages and currents associated with the active device, which suggests that care should be taken
when
choosing the point from which to couple the oscillator output.
A
circuit similar to the Hartley oscillator
shown
is
in Fig. 8-4;
it is
called
the Colpitts oscillator. Instead of the tapped inductor, the Colpitts oscillator uses a tapped capacitance to achieve the required 180° phase
operation
is
identical. In fact, all simple transistor
shift.
LC oscillators
Otherwise, the are practically
identical.
These two basic
circuits, as well as other
simple oscillator circuits, are
used as the signal source for most radio-frequency, RF, generators from tens of 1 GHz and greater. There are practical problems with constructing an oscillator of the simple sort for frequencies above 1 GHz using these circuits,
kilohertz to
and most signal generators
for
microwave frequencies use specialized
oscillators.
Likewise, for lower frequencies the size of the inductors required for the tuned circuit
become
are used.
prohibitive,
and
oscillators using other than
LC-tuned
circuits
262
Chap. 8
Signal Generation
+V
1
X
1 Figure 8-4 (^olpitts oscillator using a
bi-
polar junction transistor.
Because both the inductance and capacitance have a similar control on the operating frequency of the oscillator, both elements can be used to set the
frequency of the oscillator. In practice, the inductor while the capacitor
is
is
changed with a switch,
used for the tuning of the oscillator. This
accomplished by switching the inductor
in
bands while the capacitor
is
is
usually
connected
to the signal generator dial.
The second is
part of the sine generator
to supply signals of
known
known amplitude
is
the attenuator.
known
as well as
The
signal generator
frequencies. If a signal
were applied to the input of an attenuator, the known as long as the attenuator were accurate. Signal generators are often used to supply known signal levels at very low levels for testing and evaluating receivers. It is not possible to measure and calibrate a signal at a very low level, and thus low-level signals are generated by feeding an attenuator with a higher-level signal for which the amplitude is easily measured and calibrating the attenuator steps. An attenuator is a device that will of a
fixed amplitude
output signal level would be
reduce the power level of a signal by a fixed amount. The attenuator should terminate with a fixed impedance, relative to either the input or output, regardless of the value of attenuation.
The attenuator reduces
the
input power to the output power
power of an input such is
expressed as the log of the input to
that the ratio of the
The reduction in power can be output power ratio by the following rela-
a constant.
tionship:
A
(dB)
=
10 log -i
(8-2)
o
is the attenuation in decibels, P0 is the power output, and P is power input of the attenuator. If a signal is passed through two attenuators in cascade as shown in Fig. 8-5, the output of the first attenuator is reduced by the ratio P,/P0 while the signal is further reduced by the ratio of the second
Where A (dB)
t
the
,
j
The Sine-Wave Generator
Sec. 8-1
263
Attenuator
Two
Figure 8-5
P'/P0 The
attenuator,
'.
Po
Pi
Attenuator
attenuators cascaded for increased attenuation.
total reduction
is
the product of the two attenuations,
or
A
(dB)
=
10 log
=
(J)
£+
10 log
10 log
g
Replacing each attenuation ratio with the corresponding decibel representation yields
A where A
,
and
^4
2
=
(dB)
A,
+ A
(8-3)
2
are the attenuations of each attenuator. Therefore, the total
two cascaded attenuators
attenuation, in decibels, of
attenuation of each attenuator.
is
the
sum
of the decibel
not difficult to be convinced that this can
It is
be extended to more than two attenuators to derive the general rule that the attenuation, in decibels, of any
the decibel attenuations of
number of cascaded
the attenuators.
all
It
attenuators
is
sum
the
of
should also be clear that the
order of the cascaded attenuators will not affect the end result.
The
decibel notation
convenient for a variety of reasons but needs a
is
slight modification so that
it
can represent an absolute
equation were written as
dBr whee dBr
is
W,
=
P
10
log-
a decibel notation referenced to
of decibels above or below the equation
level.
If the decibel
some
P
r
,
^
^
faA
(g
"
4)
then dBr represents the number
reference power,
P
r
.
For example,
if
P
r
were
1
would read
dBw =
10 log
-4t w l
dBw, which is a standard notation, describes an absolute power level referenced to 1 W. Another important power level is the dBm, which is referenced to 1
mW across 500 uAV.
50
dBm
CI. is
vast majority of
For example,
+3 dBm
is
2
mW,
while
—3 dBm
is
\
mW or
convenient for a 50-fl system impedance, which includes the
equipment operating
at frequencies greater
than
1
MHz.
Various attenuator types can be used in signal generators. The pi attenuator,
named
for the
Greek
one of the more
letter,
which the schematic representation resembles, is versatile types. Three resistors are required for
common and
the pi attenuator, as
shown
The
The pi attenuator can be fabricated with dB and for frequencies to about 100 MHz.
in Fig. 8-6.
standard components up to about 20
.
S,f+
B
sin o) 2 t)
3
=
K,(A 3
sin
3
o> x t
+ B
3
sin
3
a) 2 t
(9-6) sin
2
n
-
2IP
dBm, Ip is the power power of the two input
the level of the third-order product in
the third-order intercept in
To determine
the
dBm, and P
in
is
the
(9-7) level of signals.
dynamic range of the spectrum analyzer, the third-order
314
Signal Analysis
Figure 9-12
Third-order products as a function of the level of two input signals.
intermodulation products must be the same as the seen by the analyzer, that
the signal that
is,
minimum
just visible
is
is
level.
equal to the noise
Therefore, the equation can set the third-order intermodulation products
equal to the
minimum
detectable signal:
P = 3P - 21p = MDS where
MDS is the minimum detectable signal,
noise level, in
- MDS) =
The dynamic range of
minimum
signal equal to the
2(IP
- MDS)
the spectrum analyzer
is
(9-9)
the difference in level
detectable signal and the input that produces a spurious
MDS,
or
Pm ~ MDS =
is
spectrum analyzer
(9-8),
3(Pln
Example
essentially the
dBm.
Rewriting Eq.
between the
(9-8)
in
3
What
signal that can be
above the noise
For the sake of simplicity, assume that the intermodulation level.
Chap. 9
\V
P
- ME>S)
(9-10)
9-1
dynamic range of a spectrum analyzer with a third-order intercept point of and a noise level of —85 dBm?
the
+ 25 dBm
Spectrum Analysis
Sec. 9-3
315
Solution
Using the formula 2
dynamic range
= -(/,- MDS) =
and substituting the given data, the dynamic range
2
- [25
is
-
(-85)]
=
73
found to be 73 dB.
The minimum detectable signal or the noise level of the spectrum analyzer is determined by two characteristics, the bandwidth of the IF filter in use and the noise figure of the analyzer. The noise figure of the analyzer is set by the design of the front end of the unit, while the IF
from a
later stage of the analyzer.
The
filter
bandwidth
is
a parameter
noise level of the spectrum analyzer can
be related to the noise figure and the IF bandwidth by the following:
MDS = where
BW
is
-114 dBm
4-
10 log
(BW/1 MHz) + NF
the 3-dB bandwidth in megahertz of the IF
filter,
and
(9-11)
NF
is
the
noise figure in decibels.
Example 9-2
What is the minimum detectable signal dB and using a 1-kHz, 3-dB filter?
of a spectrum analyzer with a noise figure of 20
Solution
-114 dBm
The
ability of the
+10
log
1
kHz/1
MHz +
20
= -124 dBm
spectrum analyzer to separate signals
the second IF bandwidth.
To
resolve
two
is
a function of
signals that are close in frequency, a
narrow IF filter is required. In addition, signals that are close in frequency and are at two different amplitudes are even more difficult to resolve. Consider, as an example, two signals that are the same amplitude but are separated by 10 kHz. These signals could be resolved using an IF filter with a 3-dB bandwidth of 10 kHz, as shown in Fig. 9-13. The dip in the spectrum display is only 3 dB but it is clearly visible. On the other hand, if the two signals were separated not only by 10 kHz but by 10 dB, they would not be resolvable with the 3-dB, 10kHz filter. The resolution of a spectrum analyzer is defined as the 6-dB bandwidth of the second IF It
may
filter.
appear that a
filter
with sharper skirts would solve the problem of
The ratio of the 6-dB point of a filter to the 60-dB point is an indication of the steepness of the skirts of the filter. Thus, it would appear that a filter with a lower shape factor would resolve signals close in frequency, and to a certain extent it would. However, there are significant disadvantages to sharp-skirted filters in spectrum analyzers. resolving closely situated signals.
The reader should be
when
a modulated signal
to pass the entire
is
familiar with the nature of distortions introduced
passed through a
filter
with a bandwidth too narrow
modulation bandwidth. Not only
will the high-frequency
316
Chap. 9
Signal Analysis
—H
Figure 9-13
Two
signals 10
kHz
[—10
kHz
apart as displayed with an IF
bandwidth of
filter
10 kHz.
components be reduced, but the high-Q circuits of the filter will introduce ringing. Even though the signal being analyzed with the spectrum analyzer may not be modulated, the fact that the signal sweeps through the causes ringing in the
local oscillator. If
of the
first
second IF
is
bandwidth
is
filter
seen, the
cause
filters
shape called Gaussian causes the
least
rate as a function of the
given by the following equation:
maximum sweep As can be
CW input signal
too great, the amplitude of the signal out of the
amount of distortion. The maximum permissible sweep filter
center frequency
be reduced and possibly distorted. Sharp-skirted
the most distortion, and a special
Gaussian
the
is
local oscillator,
filter will
filter
filter,
modulated as a function of the sweep speed of the the sweep speed, that is, the megahertz per second rate
of the spectrum analyzer first
Relative to the second IF
filter.
rate
=
2.3
maximum sweep
(bandwidth) rate for a
can be quite slow, and usually a spectrum analyzer
is
2
Hz/s
(9-12)
narrow-bandwidth
filter
equipped with a storage
display.
9-3.1
Spectrum Analyzers
for Higher
Frequencies Spectrum analysis
at frequencies higher
than about 100
MHz
is
a very
important tool for the development of circuits and systems at these higher frequencies.
are
no
With the exception of a few higher-frequency
tools for the analysis of signals at frequencies
oscilloscopes, there
above a few hundred
Spectrum Analysis
Sec. 9-3
317
megahertz. Most signal analysis
done with the oscilloscope
is
for lower fre-
The
quencies, such as the determination of amplitude, phase, and distortion.
spectrum analyzer provides a sensitive instrument to investigate these parameters at
higher frequencies.
The frequency of the VCO for a spectrum analyzer is required to extend from a frequency higher than the highest input frequency to a frequency at least twice the highest input frequency. For spectrum analyzers operating above 1,000 MHz, this implies an oscillator from at least 1,000 to 2,000 MHz and, in practical designs, more on the order of 2,500 to 3,500 MHz. This frequency range usually requires an oscillator with a tuned circuit other than the typical coil and capacitor found in lower-frequency oscillators. An oscillator circuit suitable for this frequency range is the YIG-tuned oscillator.
YIG, yttrium
iron garnet,
is
a ferromagnetic material that has
very useful properties at microwave frequencies.
YIG,
like
many
some
ferromagnetics,
moments
has the property that the molecules of the garnet have magnetic
that
normally are randomly aligned. The magnetic moments can be aligned in one field. The moments to
direction by the application of a static magnetic
application of an
alternating magnetic field will cause the magnetic
precess
a toy top.
YIG
The
precession frequency
material and the strength of the applied static magnetic
field.
amplitude of the precession occurs when the applied alternating to the precession frequency of the
be used to create oscillators and
Q
the gigahertz region, and the
YIG about 0.25
YIG
filters.
of a
much
like
a function of the type and size of the
is
crystal. Therefore, this
The resonance frequency
YIG
The
greatest
field is
equal
resonance can is
typically in
resonator can be quite high.
made from highly polished spheres of YIG The sphere is placed in a static magnetic field of
resonators are typically
mm
field intensity
to the static
in diameter.
H, as shown in Fig. 9-14.
magnetic
and out of the
YIG
be added, which
is
field
and
sphere. In
is
A pickup coil is arranged at right
angles
used as the method of coupling energy into
some applications a second coupling coil would static field and the other coupling
orthogonal to both the
coil.
The equivalent
circuit of the
YIG
resonator
circuit with a small fixed series inductance.
is
essentially a parallel-tuned
The resonant frequency of the
parallel
can be electronically tuned by varying the current through the magnetic coils. Unlike the typical electronically tuned oscillator using a varactor,
circuit field
where the resonant frequency tuned
circuit, the
YIG
inductance. This allows for a in
is
varied by changing only the capacitor of the
resonator tunes both the equivalent capacitance and
more constant impedance of
the resonant circuit
an oscillator and also allows for a tuning range of several octaves, rather
than just the two that are typical of a varactor-tuned
oscillator.
The YIG resonant circuit can be used in an oscillator as the frequencydetermining element, as shown in Fig. 9-15. In this example the resonant circuit is
placed in the emitter, while positive feedback
the base lead.
is
introduced by the choke in
318
Signal Analysis
YIG
Figure 9-14
sphere and the associated coupling coil and static magnetic
The frequency of
Chap. 9
field.
by varying the same fashion as the
this circuit is controlled electronically
current through the static magnetic
field coils
much
in the
voltage across a varactor diode would be used to tune a conventional oscillator.
There are some
significant differences
between the YIG-tuned oscillator and the
maximum-to-minimum frequency recommended limit for varactor-tuned
varactor-tuned oscillator. First, the ratio of the
can be
much
oscillators.
greater than
2,
which
Q
Second, the high
is
the
of the
YIG
oscillator brings
improved phase
noise performance for spectrum analyzers and sweep generators.
The frequency range of
the spectrum analyzer can be extended without
resorting to a higher-frequency local oscillator by a technique called harmonic
mixing.
A
difference
mixer
will convert
between the local
also convert
an input signal to an IF by taking the sum or and the input signal. Many mixers will
oscillator
an input signal with harmonics of the local
oscillator.
RF Out
YIG Sphere
"1
+ 10 V
-10 V Figure 9-15
Oscillator circuit using a
YIG
resonator.
Spectrum Analysis
Sec. 9-3
319
An example of a simple harmonic mixer is shown in Fig. 9-16. In this example a single diode is used to mix an input RF signal with the third harmonic of the local oscillator. If the level of the local oscillator is sufficiently high, the diode can be thought of as a switch being switched at the rate of the local oscillator.
Mixing
is
essentially multiplying
two
signals together,
action of the diode can be thought of as multiplying a square
and the switch wave with an
amplitude of 1 with the input waveform. Because a square wave is made of the summation of the fundamental and all the odd harmonics of the base frequency, it would be expected that the simple diode mixer would not only mix the RF input with the local oscillator but with practical circuit, because the duty cycle just the
input
the odd harmonics as well. In a not an exact 50 per cent, more than
all
is
odd harmonics are
RF
signal with all
present, and the example diode mixer harmonics of the local oscillator.
will
mix the
To Local Oscillator
IF Output
Figure 9-16
Simple
series
diode mixer capable of mixing by harmonics.
In the spectrum analyzer described previously, the possibility of generating any spurious inputs with the harmonics of the local oscillator was eliminated by the input low-pass filter. If this low-pass filter were eliminated, or if a bandpass filter were placed at the input of the spectrum analyzer, certain harmonics
of the local oscillator could be used to extend the range of the spectrum analyzer.
As an example will
of
how
this
might work, the previous spectrum analyzer example
be used.
The
MHz. MHz,
local oscillator covers
If the
from 400 to 700
MHz
and the
first
IF
is
400
second harmonic of the local oscillator were used, 800 to 1,400
the second harmonic minus the first IF would give an input range from 400 to 1,000 MHz, while the sum would yield 1,200 to 1,800 MHz. The third harmonic of the local oscillator, 1,200 to 2,100 MHz, would allow the conversion of 800 to 1,700 MHz and 1,600 to 2,500 MHz for difference and sum, respectively. Other harmonics can be used for extending the frequency range further. One other method of gaining yet another frequency range from the same mixer and local oscillator is to simply use the sum of the first IF and the local oscillator frequency that covers the frequency range of 800 to 1,100 MHz.
320
It
should be noticed that, although the range of the spectrum analyzer
can be extended by that
is
321
Spectrum Analysis
Sec. 9-3
range of frequencies in this example from 300 to 400 MHz. When complete frequency required, the second IF is often used in lieu of the first IF to provide this technique, there is a
not covered and that
coverage
is
is
the required frequency coverage.
When harmonic
mixing
spectrum analyzer display.
is
used, several corrections are required to the
First,
when
the harmonic
mix
used, the center
is
frequency dial of the analyzer must have the correct frequency calibrations. This is
usually handled by having a mechanical dial arrangement that simply displays
the correct frequencies. Electronic dials can manipulate the ically.
Second, because a harmonic of the local oscillator
change of frequency,
relative to the
Nth harmonic, per
is
volt
numbers
electron-
used, the rate of is
N times
that at
the fundamental, so therefore the spectrum analyzer display for this. This
is
corrected in a simple fashion;
the local oscillator tuning voltage efficiency of the less
mixer
at
is
if
the
Mh
must be corrected harmonic is being used,
simply divided by N. Finally, the mixing
harmonics, especially the higher-order harmonics,
is
than at the fundamental. Therefore, the display will have to be corrected
for this loss of signal. This
number of
accomplished by simply offsetting the display by
is
and the harmonic mixer loss. A block diagram of a spectrum analyzer with harmonic mixing is shown in Fig. 9-17 with all the required correcting circuits. The chief problem in using the harmonic mixing spectrum analyzer is that the input low-pass filter is removed and all the possible harmonic mixing ranges are present at the input of the spectrum analyzer. Therefore, there is considerable ambiguity in the display as some signals can appear at more than one point on the display. Various signal-identifying techniques can discern between the correct and incorrect signals, but the best technique is to place an external band-pass filter between the system being tested and the spectrum analyzer, which will eliminate many of the spurious signals. An example of a spectrum analyzer with the
decibel difference between the fundamental mixer loss
harmonic mixing capability
is
shown
in Fig. 9-18.
9-3.2 Applications of the Spectrum Analyzer
The spectrum analyzer those
is
many applications. To applications may not be readily
a powerful tool and has
who have never used the instrument, these To illustrate some of the applications,
apparent.
descriptions,
and the spectrum analyzer
display, as
the following signals, their
shown
in Fig. 9-19, will be
presented. (a)
Pure sinusoid with no modulation or harmonic is
persion of the spectrum analyzer (b)
distortion.
This signal
characterized by a single spectral line regardless of what the disis
or the IF
filter
bandwidth.
Amplitude modulation. When a carrier is modulated with amplitude modulation, two sidebands are generated, one above the carrier fre-
322
Signal Analysis
Figure 9-18
up
to
40
Example of
GHz
Chap. 9
a harmonic-mixing spectrum analyzer covering frequencies
(courtesy Polarad Electronics, Inc.).
quency and a second below the carrier frequency. The separation frequency between the carrier and the sidebands
is
equal to the
in
mod-
The power contained in the sidebands is dependent on the percentage of modulation. One hundred per cent modulation produces sidebands that are 6 dB below the carrier. The amplitude of the carrier, on the other hand, does not change, regardless of the ulation frequency.
percentage of modulation. (c)
Frequency modulation. Frequency modulating a carrier produces
side-
bands that are centered around the carrier as in the case of amplitude modulation, except that more than one sideband is generated. The
number of sidebands and the amplitude of those sidebands is described by complex formulas based on the Bessel functions. The sidebands are all multiples of the modulating frequency, and the amplitude of the is affected by the amount of modulation supplied. The precise amount of frequency modulation can be determined if the modulation
carrier
is
adjusted so that the amplitude of the carrier or other sidebands goes
to zero. (d)
Asymmetrical
spectra.
The generation of
metrical about the carrier
is
a spectrum that
is
not sym-
usually an indication that both frequency
and amplitude modulation are occurring simultaneously. This could occur in an FM system where the passband of an amplifier is not flat and the frequency modulation is introducing amplitude modulation.
Suit Sec. 9-3
Spectrum Analysis
-4-
B /
\
i
! 1
j
i
(a)
(e)
(f
Figure 9-19
)
Applications of the spectrum analyzer.
Likewise, amplitude modulation applied to a carrier that also causes
frequency
instabilities,
which
is
a
common problem
with phase-locked
loops, will cause a similar spectrum. (e)
Harmonic
distortion.
Harmonics appear
spectrum analyzer display often required that the
the order of 60 or
as additional signals in the
at multiples of the carrier frequency. It is
harmonic content of a
more dB below
signal be kept low,
the carrier.
As an example,
on this
may be required so that a transmitter operating at an assigned frequency will not interfere
with other radio services at twice the assigned
fre-
quency that may be located near the transmitter. (f)
Pulse modulation. Examining pulse modulation was the
first
application
of the spectrum analyzer. Determining the pulse modulation of radar transmitters
was a
difficult task in the early
development of radar, and
324
Chap. 9
Signal Analysis
the spectrum analyzer
was used
to evaluate the quality of the pulse
modulation. The spectrum of a rectangular amplitude pulse in Fig. 9- 19(f).
The
structure of the sidebands shows the rise-
is
shown
and
fall-
times of the pulse modulation, and the symmetry indicates the presence
or absence of frequency modulation, which
is
a problem with modulated
such as those used with high-power radar transmitters.
oscillators
REFERENCES 1.
Engleson, Morris, and Tewlewski, Fred, Spectrum Analyzer Theory and Applications.
Dedham, Mass.: 2.
Hay ward, W.
ARTECH
House, 1974.
H., Introduction to
Radio Frequency Design, chap.
6.
Englewood
Cliffs,
N.J.: Prentice-Hall, Inc., 1982. 3.
Krauss, Herbert
L., Bostian,
Engineering, chaps. 2 and
7.
Charles W., and Raab, Frederick H., Solid State Radio
New
York: John Wiley
&
Sons, Inc., 1980.
PROBLEMS 1.
What
is
dynamic range of a spectrum analyzer
the
—80 dBm and two — 10-dBm
equal to
signals
if
the noise level of the display
is
produce third-order intermodulation
products that just appear above the noise? 2.
3.
What
is
of 30
kHz?
sideband this
if
is
amplitude modulation with only one sideband and no
modulation look
What would device
like displayed
What
be the third-order intermodulation products relative to the input of a
two input
signals of
— 10 dBm
were applied
to a device with a third-order
intercept of
+15 dBm?
What
dynamic range of a spectrum analyzer with a 30-kHz, 3-dB bandwidth,
is
the
How What
+25 dBm?
does placing a fixed attenuator ahead of a spectrum analyzer affect
order intercept, (b) the dynamic range, and 8.
carrier.
on a spectrum analyzer?
a noise figure of 15 dB, and a third-order intercept of 7.
with a 3-dB bandwidth
is the maximum sweep rate in kilohertz per second that could be used with a spectrum analyzer without introducing distortion with a 3-kHz Gaussian filter?
would
6.
filter
What
4. Single
5.
the resolution of a spectrum analyzer using an IF
(a) the third-
(c) the noise figure?
frequency ranges could be covered with a spectrum analyzer having a
of 2,050
MHz and an input of 0 to
harmonic?
1,000
MHz using harmonic
first
IF
mixing up to the third
CHAPTER
10
FREQUENCY COUNTERS
AND
TIME-INTERVAL
MEASUREMENTS
10-1 SIMPLE
FREQUENCY COUNTER
Standards of time and frequency (time and frequency being essentially the same standard) are unique in that they to another without the actual
may
be transmitted by radio from one location
movement of the
standard. Therefore,
it is
possible
have traceability to the primary standard without difficulty. Additionally, the primary standard is related to the structure of matter, and primary standards to
can be easily duplicated throughout the world to allow high-accuracy measure-
ments anywhere. Because of the
relative ease with
which frequency and time
can be measured to great accuracies, electronics systems have developed around this capability. Consider, as
an example, the tolerance expected of radio-trans-
mission equipment. The spectrum required by a voice-modulated two-way radio transmitter using frequency modulation that
if
is
on the order of
a communications channel could be assigned every 15 efficient
5
kHz. This implies
kHz and make
the most
use of the radio spectrum. Because accurate measurement techniques
are available and standards can be are assigned every 20
kHz
in the
made
available, the
communications channels
UHF (450 MHz) band.
frequency accuracy and stability of only 5 kHz, which per cent, which
is
easily achieved with
ment techniques. Although relatively
many
1
the frequency of the transmitter carrier could be held to absolute precision
task.
modern frequency
approximately 0.001 control and measure-
stable frequency standards have been available for
years, precise frequency
surement
This requires a carrier is
measurement has not always been an easy mea-
Early frequency measurement required precision standards,
fre-
325
— Frequency Counters and Time-Interval Measurements
326
Chap. 10
quency comparators and interpolation oscillators, as well as a lot of operator skill. This came to an abrupt end with the introduction of digital logic and the development of the frequency counter. Figure 10-1 shows the block diagram of a simple frequency counter. AlDisplay
Strobe
Memory Input
Input
Decade
Signal Processor 1
Counters
Reset
Time Base
1
Basic block diagram of a frequency counter.
Figure 10-1
though referred to as "simple," if
1
this basic
the parts are constructed properly.
counter
is
capable of great precision
The frequency counter operates on
the
principle of gating the input frequency into the counter for a predetermined
As an example,
time.
an exact
1
second
(s),
an unknown frequency were gated into the counter for number of counts allowed into the counter would be
if
the
precisely the frequency of the input.
an
AND
or an
OR
gate
is
to be accumulated. Figure 10-2
This example shows an similar circuit. to
A
AND gate
AND
is
AND the
AND
gate
is
zero. Thus, exactly
allowed at the output of the
input pulses
is
these pulses
and display the
AND
output
s
1,
applied
the output
pulse returns
of
unknown
gate. It is necessary to
result.
Jinjijijiririr^ |~
Gate
1
s is
1
As long as the 1-s pulse is a logic same as the unknown input. When the 1-s gate.
to logic 0, the output of the
^
shows the waveforms associated with this action. gate; however, an OR gate could be used in a
positive-going pulse having a period of exactly
one input of the
of the
The term gated stems from the fact that unknown input into the counter
used to allow the
|
_TLnjijirTJirir^^
(Figure 10-2
Waveforms
associated with the gating function of a frequency counter.
count
Simple Frequency Counter
Sec. 10-1
If the gate
is
open for exactly
average frequency of the gate was open for 10 in 0.1
Hz. Likewise,
unknown
327
1
s,
the count accumulated
input in hertz (Hz).
the accumulated count
if
the gate were open for 0.1
When
as
is
equal to the
an example, the
would be the average frequency s, the count would be the a frequency counter has more than
s,
average frequency in tens of hertz.
If,
one gate time interval available, the decimal point of the display
is
switched
with the gate time selector switch to correct the frequency display.
10-1.1 Display Counters
The
actual counting circuits are, in practice, constructed
circuit counters, but
it is
from integrated
constructive to understand the internal operation of a
digital counter.
The
heart of a frequency counter
constructed from four flip-flops and an
is
the decade counter, which can be
AND
gate, as
shown
in Fig. 10-3. This
Outputs
J
Input
Q
c K
Figure 10-3
Ripple binary coded decimal counter.
form of decade counter is called a ripple counter owing to the fact that the clock of one flip-flop is derived from the output of the previous flip-flop, which requires that the clock pulses ripple through the counter from the first stage to the last stage. The last stage, however, derives its clock from the first stage, which reduces the propagation delay to a certain degree.
A
method of constructing a counter is to use a synchronous shown in Fig. 10-4, requires that all the flip-flop clocks be connected together, which greatly reduces the propagation delay and allows superior
counter. This circuit,
higher counting speeds.
The output of the decade counter follows the sequence shown in Fig. and is called binary coded decimal (BCD), which implies that the normal binary code is used except that each digit is defined only for values between 0 and 9. As an example, the decimal number 138 is 0001 0111 1000 in BCD. Each BCD counter allows one decade of counting and thus the BCD counters must be cascaded. For example, three cascaded BCD counters are 10-5
328
->
O
-3
O
*
329
Simple Frequency Counter
Sec. 10-1
Clock
Counter State
1
2 3
u
c
Q D
0 0
0
0
1
1
0
0 0 0 0 0
4 5
6 7
8 9
1 1
0
IO
0 0 1
1
1
A
1
1
0 0
0
1
1
0
1
1
1
0 0 0
0 0 0
0 Figure 10-5
1
0
Binary coded decimal count-
ing sequence.
required to count from 0 to 999. There are two methods of cascading
BCD
and synchronous. Ripple cascading is usually reserved counters and, unfortunately, makes the slow ripple counter even slower.
counters, ripple cascading for ripple
With the exception of low-frequency counters, the serious frequency-measuring equipment.
The
ripple counter
is
not used in
ripple connection requires the last
output of the least significant counter to drive the clock input of the next more significant counter, as
must respond weight of
8,
shown
10-6.
in Fig.
The clock input
to the negative edge of the clock as the last bit,
goes low at the transition from 9 to
A B C D
Figure 10-6
0.
A B C D
A B C D Clock
Clock
Clock
to the next stage
which has a binary
Cascading ripple counters.
The synchronous counter has a terminal count or carry output for the purpose of cascading counters, as shown in Fig. 10-7. This output goes to a logic is
1
after the clock that
changes the state of the counter to
9.
This output
used to enable the following counter to be incremented on the next clock
pulse. This insures that the state of the next counter
is
coincident with the clock
and preserves the synchronous counter operation when the counters are cascaded. When more than two counters are cascaded, the requirement for any one counter to change state is that all of the less significant counters must be at 9. Some integrated-circuit counters have internal cascading logic that propagates the
A B C D TC
PE A B C D TC
PE A B C D ye
pE
Clock
Clock
Clock
Figure 10-7
Cascaded synchronous counters.
Frequency Counters and Time- Interval Measurements
330
"nine" state from the least significant digit through
most
to the
When
significant digit.
there
is
a large
all
Chap. 10
the intervening counters
number of cascaded
counters,
the delay can limit the count frequency of the counter. Therefore, other tech-
niques called look ahead or carry forward are used to reduce the amount of
propagation delay.
The
BCD
information available at the output of the counter must be
converted to some form of visible display. The conversion depends on the type of display desired. For example, conversion from
segment display requires a
shows a
4-bit
BCD to the very popular seven-
single, inexpensive integrated circuit.
Figure 10-8
counter including the seven-segment code conversion. Counter
Decoder/Driver
7490
7447
Display
a
Clock
b
A
c
B C D
d e f
g
Figure 10-8
Block diagram of a decade counter interfaced with a seven-segment
display.
It is
desirable in a frequency counter to display the count continuously.
Since the counter
is
reset to zero
and allowed
during this time the output of the counter
is
to count during the gate period,
would appear as a the end of the measurement period is
of the counter cannot be displayed during this period as
meaningless blur. Therefore, the count at stored in a simple
memory and
which the next count
is
required to store only 4
and
is
the entire
typically a simple 4-bit latch,
clocked together, with each
it
displayed during the next counting period, after
stored in the
bits,
The output
constantly changing.
memory and
displayed. This
memory
is
BCD word, for each decade of the counter
which consists of four D-type
flip-flop storing
1
bit
flip-flops all
of data.
Digital logic usually cannot supply the required current for driving a
Even those displays
minimal amounts of current, such as which are not readily available from the decoder output. Therefore, a display driver is included between the decade counter and the displays. For counters requiring a large number of digits, typically 10 or more, there are various techniques to reduce the required hardware, one of which is shown in Fig. 10-9. This technique is called display multiplexing and reduces the number of drivers and decoders required to implement large counters. In this example a common decoder and driver are shared between all the display digits. A multiplexer selects the BCD data from one of the latches and routes these data to the input of the seven-segment decoder. The decoded seven-segment information is applied to the proper display. The entire process is driven by an
display.
that require
liquid crystals, require special signals,
Simple Frequency Counter
Sec. 10-1
331
Scan Counter
Digit Driver
Scan Oscillator
Displays
Clock
Figure 10-9
oscillator is
done
Block diagram of a multiplexed display used
and a counter called the scan
at a rapid rate, the display
oscillator
and
in a
counter.
this
is
When
appears constant to the eye.
that the inclusion of the multiplexer, scan oscillator,
drivers
frequency counter.
It
this process
would seem
and multiplexed display
hardly worth the aggravation to save a few simple decoders. However,
technique has significant advantages
when
the frequency counter circuits
are integrated into a single silicon chip.
Consider, as an example, a 10-digit frequency counter. This scale of frequency counter could be integrated onto a single silicon chip except that 70 outputs would be required for the readouts alone if they are of the seven-segment
332
type.
Frequency Counters and Time-Interval Measurements
Add
to this the
power and ground,
and other inputs
a time-base input,
required for the frequency counter and the net result
is
Chap. 10
80 or more pins, which
does not allow for inexpensive packaging. The readout output could be multiplexed with seven outputs for the segments and a 4-bit binary output for selecting
each
digit,
which
results in only
1 1
output pins for the display interface. Adding
the other required pins results in a package size that can be handled with
conventional packaging technology.
Time Base
(10-1.2
The sequence of
events within the frequency counter
is
controlled by the
time base, which must provide the timing for the following events: resetting the counter, opening the count gate, closing the count gate, and storing the counted
frequency in the latch. The resetting of the counter and storing of the count are not critical events as long as they occur before and after the gate period, respectively.
The opening and
closing of the count gate,
on the other hand,
determine the accuracy of the frequency counter and are very
critical in its
timing.
Since the accuracy of the frequency counter depends directly on the ac-
curacy of the time-base signals, the time base
is
driven from an accurate crystal-
controlled oscillator. This element of the time base
compensated crystal
is
typically a temperature-
oscillator operating at several megahertz.
A
crystal
oven
could be used to supply a similar accuracy, except that the oven requires a relatively long period after the initial application of
power, up to 24 hours, to
The temperature-compensated oscillator does not require the appliof power to provide the correct frequency and is available for use im-
stabilize.
cation
mediately after power-on.
Figure
10-10
shows a simplified diagram of a
Varactor Diode
RFC
Temperature Sensor (Thermistor)
o
Oscillator
— RF
i
Correction
Network
Block diagram of a temperature-compensated crystal
Figure 10-10
temperature-compensated cry stal
Oul
oscillator.
A
oscillator.
conventional crystal oscillator
is
used as the basic building block of the compensated oscillator, except that a varactor diode
is
placed across the crystal. The varactor allows the frequency
of the oscillator to be changed by minute amounts.
quency error
is
The
crystal oscillator fre-
characterized over the desired operating temperature, and the
error characteristic
is
stored in the correction network. This can either be a
L 333
Simple Frequency Counter
Sec. 10-1
digital storage
technique or an analog circuit with nonlinear characteristics. The
ambient temperature
is
fed to the correction network,
which adjusts the
oscillator
frequency by varying the varactor voltage as a function of temperature.
Aside from the temperature variation of frequency of a crystal
oscillator,
quartz crystals tend to age and change frequency over a period of time. This
undesired frequency change can be reduced by special crystal fabrication techniques, but
still
it
can be as high as
5
X
10
compensated for by periodic recalibration. Many temperature-compensated crystal
"7
parts per year. This
oscillators
must be
have the capability of
being electronically adjusted. If the frequency counter has a standard frequency
output that can be compared to one of the available broadcast frequency standards, the frequency of the time-base oscillator in a frequency counter can be set to within
1
part in 10
9 .
Three outputs are required from the time base: a reset pulse, the gating pulse, and a strobe pulse, in that order. Figure 10-11 shows a simple circuit for generating the three required pulses without overlap.
The
crystal oscillator
divided by powers of ten, as the period of the frequency of the crystal shorter than the desired gate time.
counter that has 16 states.
The zero
The
final digital divider is a 4-bit
state of the counter
is
binary
decoded to provide
The 2 state is decoded to provide the was not used so as to provide a delay after the
the reset pulse for the frequency counter. gate open pulse.
The
1
state
°p en Gate
—n
Reset
4-bit Binary Counter
-_J~"|__ Close gate 1
10 Hz 100Hz1 Hz
sec
0.
1
sec
0 .01 sec |_J~~ Store
Figure 10-11
is
much
is
Logic diagram of a time base for a frequency counter.
Frequency Counters and Time-Interval Measurements
334
reset pulse to allow the counters to
be fully recovered from the
reset.
Chap. 10
The
gate
remains open for exactly 10 clock pulses, and thus the 12 state of the counter
decoded to provide the gate close pulse. The 13 state of the counter is not decoded so as to provide a delay period before the counter is stored in the latch during the 14 state. The 15 state is not decoded and provides the necessary nonoverlap between the store and reset pulses, which occur immediately after is
the 15 state of the counter.
important that the propagation delay from the input clock to the
It is
edges of the open and close pulses be the same for each so that the gate exactly equal to the correct
number of clock
pulses. This requires fast logic
is
and
careful design.
Most frequency counters have
several available gate time intervals that
can be selected by a switch. As shown in Fig. 10-11, the input of the binary counter can be selected from a choice of
These frequencies provide gate times of
10,
1
Hz, 10 Hz, 100 Hz, and 1,
0.1,
and 0.01
1
kHz.
respectively.
s,
10-1.3 Input Signal Processing
The unknown frequency input level to drive the
is
not guaranteed to be of the correct logic
frequency counter, and a processing circuit
is
required.
Typ-
an amplifier to increase the signal level, an attenuator to adjust for variations in input amplitudes, and a comparator so that the slow risetime ically, this is
of the input waveforms can be reduced to provide reliable operation of the internal logic circuits. circuit
is
shown
A schematic diagram of a typical frequency counter input
in Fig.
10-12.
Amplitudes of a few
millivolts
can be used to
trigger the frequency counter using this circuit.
Comparator
—WA
To Counter Amplifier I
VWV
Input
©
H Figure 10-12
Input circuits for a simple frequency counter.
— Simple Frequency Counter
Sec. 10-1
10-1.4 Period Measurement
.
335
^^
*
' '
If two input signals were substituted for the open and close gate signals, and one of the internal clock signals, that is, one of the available frequencies that are powers of 10 Hz, is supplied to the count gate, the time interval between the two input signals could be measured. The arrangement of this period mea-
suring
is
shown
in Fig. 10-13.
The input
measurement.
A
second identical circuit
must be processed
signals
fashion as the count input signal, and the
same
will
same
in the
can be used for period
circuit
have to be supplied for the period
measurement.
IS
Input
Signal
A
Signal
Start
Processor Counter
1
1
Input
Signal B-
Signal
Processor
IS Stop
Resolution 1
ms
o \00fxs lO/zs
1
mHz
I00 kHz
10
kHz
I
kHz
From Time Base Circuit arrangement for
Figure 10-13
making period measurements.
Another period measurement can be made using a
single input. This
would
be useful for determining the period of pulses and other signals. In this
mode
of operation, the gating signal
is
the input, and the internal frequency clocks
To measure
are used as timing sources.
the period of a pulse waveform,
it
is
necessary to open the count gate at the rising edge of the pulse and to close the gate at the falling edge of the pulse. In the case of a negative-going pulse, this
procedure would be reversed, that
is,
opening the gate on the negative edge and
closing the gate at the positive edge. If the risetimes
compared
pulse are short,
actual trigger point
is
not
and
falltimes of the input
to the resolution of the period
critical.
A
measurement, the
sophisticated frequency counter will have
independent control over the voltage level of both the rising and falling edges, as
shown
in Fig. 10-14.
measurements,
Although
this type of
method of viewing the
this results in the
most
and accurate skill and a Because most
flexible
frequency counter requires operator
trigger points, such as
period measurements involve pulses with fast
an oscilloscope. rise-
and
falltimes, a simple al-
s
Frequency Counters and Time-Interval Measurements
Chap. 10
Vref
Open
SI Frequency counter input
Figure 10-14 falling
circuits
showing the
ability to set rising
and
edges individually.
I
WW+ 5
V
© B
5 V
© 0 V
+ 2.5
Out
_L
r\ r
V
— 2.5 V
5 V
0 V Figure 10-15
waveforms.
Zero-crossing detector for a frequency counter and the associated
Measurement
Sec. 10-2
Errors
337
ternative is to ac couple the input signal and open and close the count gate at the zero crossings of the ac-coupled signal. Figure 10-15 shows a typical pulse waveform input and the resulting trigger points after ac coupling.
One
very important period measurement
is
the period measurement to
determine frequency. This measurement is not made from rising edge to falling edge but from a point in an input cycle to the same point in the next cycle,
which
is
the period of the input signal. In this case, the gate
a point of the input cycle.
This
is
waveform and closed
at precisely the
is
opened
to be
at
in the next
accomplished in the following fashion. The input signal
coupled, and a zero crossing detector triggers a crossing
is
same point
is
ac
The following zero flip-flop. The next zero
flip-flop.
of the opposite slope and does not trigger the
crossing, however, occurs after a time period equal to the period of the input
waveform and toggles the
flip-flop,
which provides a gate time exactly equal
the period of the input waveform, as
shown
to
in Fig. 10-16.
J
To
Q
frequency Counter
c
K
Figure 10-16
The
Input circuit configuration for measuring the period of a waveform.
typical laboratory counter, such as that
shown
in Fig. 10-17,
has both
input period measurement and independent control of risetime and falltime triggering selectable
10-2
from a front-panel switch.
MEASUREMENT ERRORS 10-2.1 Gating Error Frequency and time measurements made by an electronic counter are
subject to several inaccuracies inherent in the instrument
instrumental error
is
itself.
One very common
the gating error, which occurs whenever frequency and
period measurements are made. For frequency measurement the main gate
is
opened and closed by the oscillator output pulse. This allows the input signal to pass through the gate and be counted by the decade counters. The gating pulse is not synchronized with the input signal; they are, in fact, two totally unrelated signals.
Frequency Counters and Time-Interval Measurements
338
Figure 10-17
Chap. 10
Microprocessor-controlled computing counter (courtesy of Racal-Dana
Instruments, Inc.).
In Fig. 10-18 the gating interval (a)
and
is
indicated by
(b) represent the input signal in different
to the gating signal. Clearly, in
one
±
(c).
Waveforms
case, six pulses will be counted; in the other
case, only five pulses are allowed to pass
a
waveform
phase relationships with respect
through the
gate.
We
have therefore
count ambiguity in the measurement. In measuring low frequencies, the
1
gating error
may have an
appreciable effect on the results. Take, for example,
the case where a frequency of 10
equals
1
count of 10
Hz
reasonable assumption).
s (a
±
1
is
to be
measured and the gating time
The decade counters would
indicate a
count, an inaccuracy of 10 per cent. Period measurements are
therefore to be preferred over frequency measurements at the lower frequencies.
The
dividing line between frequency and period measurements
determined as follows: Let
f =
crystal (or clock) frequency of the instrument
fx =
frequency of the
c
unknown
input signal
w
JuuuLrui_rLJo_n_
«
_n_JtTLJuiJuua_n_
w
/ r«
N Gate Open
Figure 10-18
Gating
error.
»•{
may
be
Measurement
Sec. 10-2
Errors
In a period measurement the
339
number of
pulses counted equals
N = Ji
(10-1)
P
f,
In a frequency measurement with a 1-s gate time the
number of
pulses counted
is
Nf = f which N = Nf
(10-2)
x
The
crossover frequency (f0) at
p
7 =L
or
f = Q
is
(10-3)
yjfc
Jo
Signals with a frequency lower than
f
0
should therefore be measured in the
"period" mode; signals of frequencies above f0 should be measured in the "fre-
quency" mode
The
in order to
minimize the
accuracy degradation at
effect
caused by the
f
0
of the
±
1
±
1
count gating error.
count gating error
is
100/
yjfc per cent.
10-2.2 Time-base Error Inaccuracies in the time base also cause errors in the measurement. In
frequency measurements the time base determines the opening and closing of the signal gate, and
it
provides the pulses to be counted. Time-base errors consist
of oscillator calibration errors, short-term crystal stability errors, and long-term crystal stability errors.
Several methods of crystal calibration are in simplest calibration techniques
is
common
use.
One
of the
to zero-beat the crystal oscillator against the
standard frequency transmitted by a standards radio station, such as
This method gives reliable results with accuracy on the order of
1
WWV.
part in 10
6 ,
which corresponds to 1 cycle of a 1-MHz crystal oscillator. If the zero-beating is done with visual (rather than audible) means, for example, by using an 7 oscilloscope, the calibration accuracy can usually be improved to 1 part in 10 .
Several very low frequency
(VLF) radio
stations cover the
North American
continent with precise signals in the 16-20-kHz range. Low-frequency receivers are available with automatic servo-controlled tuning that can be slaved to the signal of
one of these
stations.
The
error between the local crystal oscillator and
on a strip-chart recorder. A simplified diagram of this procedure is given in Fig. 10-19. Improved calibration accuracy can be obtained by using VLF stations rather than HF stations because the transmission paths for very low frequencies is shorter than for high-frequency the incoming signal can then be recorded
transmissions.
Short-term crystal stability errors are caused by momentary frequency variations
due to voltage
transients,
shock and vibration, cycling of the crystal
oven, electrical interference, etc. These errors can be minimized by taking fre-
Frequency Counters and Time-Interval Measurements
340
Chap. 10
Phase or
Time Error
Receiver
Frequency
Phase
Multiplier
Detector
To Recorder
O
Local
Frequency Source Figure 10-19
Servomotor or Manual Adjustment
Calibration of a local frequency source.
quency measurements over long gate times (10 s to 100 s) and multiple-periodaverage measurements. A reasonable figure for short-term stability of a standard 7 crystal-oven combination is on the order of 1 or 2 parts in 10 .
Long-term
stability errors are the
more subtle contributors to the inaccuracy
of a frequency or time measurement. Long-term stability
and deterioration of the in
crystal.
As
the crystal
is
is
a function of aging
temperature-cycled and kept
continuous oscillation, internal stresses induced during manufacture are
lieved,
ness.
and minute
particles adhering to the surface are shed reducing
Generally, these
phenomena
will
its
re-
thick-
cause an increase in the oscillator
frequency.
A The
typical curve of frequency
change versus time
may
of change of crystal frequency
initial rate
is
shown
in Fig. 10-20.
be on the order of
6
10 per day. This rate will decrease, provided that the crystal
is
1
part in
maintained
at
its
operating temperature, normally about 50° to 60°C, with ultimate stabilities
of
1
part in 10
9 .
If,
however, the instrument containing the crystal
from the power source appreciably, a
new
into operation. It
slope of aging will ensue
is
is
unplugged
for a period of time sufficient to allow the crystal to cool
when
the instrument
is
put back
possible that the actual frequency of oscillation after cool
Initial
Slope
Time (Weeks) Figure 10-20
Frequency change versus time
for an oven-controlled crystal.
Extending the Frequency Range of the Counter
Sec. 10-3
341
vary by several cycles and that the original frequency will not again be
off will
reached unless calibration
To show
done.
is
the effect of long-term stability on the absolute accuracy of the
measurement, assume that the oscillator was calibrated to within 1 part in 10 9 and that a long-term stability of 1 part in 10 8 per day was reached. Assume further that calibration
time
then
is
X
1
1(T
is
60
10~ 8
X
maximum
be seen therefore that exact calibration
was done 60 days ago. The guaranteed accuracy
+
9
performed a
=
6.01
X
10~ 7 or 6 parts in 10 7 ,
.
at this It
absolute accuracy can be achieved only
relatively short time before the
can
if
an
measurement
is
taken.
10-2.3 Trigger Level Error
(j6W^n
In time-interval and period measurements the signal gate closed by the input signal. closed signal
The accuracy with which
the gate
is is
opened and opened and
a function of the trigger level error. In the usual application the input
is
amplified and shaped, and then
is
that supplies the gate with
a certain
its
it is
applied to a Schmitt trigger circuit
control pulses. Usually the input signal contains
amount of unwanted components or
noise,
which
is
amplified along
with the signal. The time at which triggering of the Schmitt circuit occurs function of the input signal amplification and of general,
we can
plitudes
and
its
is
a
signal-to-noise ratio. In
say that trigger time errors are reduced with large signal
am-
fast risetimes.
Maximum
ac curacy can be obtained
if
the following suggestions are fol-
lowed:
(a)
The
effect
of the one-count gating error can be minimized by making
frequency measurements above y[fc and period measurements below is the clock frequency of the counter. yjfci where
f
(b) Since
long-term stability has a cumulative
surement
is
effect,
the accuracy of mea-
mostly a function of the time since the
last calibration
against a primary or secondary standard. (c)
The accuracy of time measurements
is
greatly affected by the slope of
the incoming signal controlling the signal gate. Large signal amplitude
and fast risetime assure maximum accuracy.
1
0-3
EXTENDING THE FREQUENCY RANGE OF THE
COUNTER Using the
fastest logic
and the most sophisticated carry
frequency counter shown in Fig. 10-1 speed.
is
limited to about
To increase the frequency range of the counter, One technique is to use a prescaler as shown in
be used.
circuits, the
simple
100-MHz counting
several techniques can Fig. 10-21.
A prescaler
p
342
Frequency Counters and Time- Interval Measurements
Chap. 10
1MHz-IOMHz
^, 41
IO-IOO MHz Figure 10-21
is
Frequency
-10
>
Counter
Using a prescaler to extend the range of a frequency counter.
a fast digital counter that, typically, divides the input frequency by
prescaler does not drive a display,
is
10.
The
not gated, nor are the output data strobed
into the storage latch. Therefore, the propagation delay of the prescaler
is
not
important as long as the prescaler can operate at the desired frequency. If a divide-by- 10 prescaler were used ahead of a
10-MHz
counter, the counter fre-
quency would be increased by a factor of 10 and the system would be capable of counting to 100 MHz. Prescalers are available for frequencies up to 1 GHz with divisions of 10 or 100, which can extend the range of the example 10-MHz counter to
1
There
GHz. is
a penalty to be paid for the use of the prescaler.
of the frequency counter
example,
if
a
last digit,
if
10, all the digits
the counter
when
The
resolution
reduced by the same factor as the prescaler. As an
10-MHz counter were used with a prescaler,
would be multiplied by implies that
is
the frequency displayed
including the least significant. This
had a resolution of
1
Hz, which
is
the value of the
multiplied by 10 the resolution would be reduced to 10 Hz.
This can be overcome by simply using a longer time base and restoring the
become a practical problem if the prescaler has a large and very accurate frequency measurements are to be made. For example, if the divide-by- 100 prescaler were used to extend the frequency range of the 10-MHz counter to 1 GHz, and a measurement of 1-Hz resolution were desired, the gate time would be 100 s, which could be a significant problem. Typically, frequency measurements with resolution of better than 1 kHz at 1 Ghz are rare. The prescaler, as effective as it can be, is limited to frequencies below about 1.5 GHz with the current state of technology. For making frequency counter measurements at higher frequencies, heterodyning techniques are used. Figure 10-22 shows a heterodyning converter for a frequency counter. This resolution. This can
division
Input Signal
100
MHz
\
Harmonic
Tuned
Low -pass
Generator
Cavity
Filter
Figure 10-22
Manually tuned heterodyning frequency converter
frequency range of frequency counters.
To Counter
for extending the
Extending the Frequency Range of the Counter
Sec. 10-3
converter
is
used with a
50-MHz
343
counter, which requires that the converter
MHz or less, which it does with mixing Because both the sum and the differences are used, the converter frequency never exceeds 50 MHz. A 100-MHz source, which is reduce the input frequency to 50
frequencies every 100
MHz.
derived from the frequency counter's time base, feeds a harmonic generator using a step recovery diode. characteristic in that
it
The step recovery diode has a unique reverse recovery
stops conducting very abruptly, which generates har-
monics of the driving waveform to several gigahertz. The harmonic content of 5-GHz region. Harmonics from the fundamental at 100 MHz to 5 GHz are selected by a tuned cavity that tunes one of the harmonics. It is necessary to know which of the 50 harmonics is being tuned, and a calibrated dial is provided as a tuning meter to peak the desired signal. The setting of the harmonic tuner dial does not affect the accuracy of the measurement unless the incorrect harmonic is tuned. The 50 harmonics represent a 2 per cent resolution, which can be easily achieved with a mechanical the diode generator extends well into the
assembly.
The
selected
amplified,
and fed
harmonic
mixed with the input and the difference is filtered, Because there is a harmonic available every 100 MHz, the input signal is never more than 50 MHz from one of the harmonics. To select the correct harmonic, the input frequency must be known to within 10 MHz or so, which can be done with another measurement technique such as a wavemeter or spectrum analyzer. Since either the sum or difference between the selected harmonic and the input signal
may
is
to the counter.
be counted, the operator
is
required to
make
the necessary
calculations to determine the actual frequency. This involves adding or subtracting,
depending on whether the sum or difference
frequency that
is
read from the harmonic tuner
is
counted, the harmonic
dial.
Modern frequency counters are capable of tuning the harmonic and making the necessary calculation automatically. Figure 10-23
shows a block diagram of
an automatic heterodyning unit for converting frequencies up to 4 GHz to extend the range of a 500-MHz counter. A 100-MHz signal from the frequency counter
is
multiplied using a bipolar transistor frequency multiplier to 500
This signal multiplier.
is
MHz.
amplified and used to drive a step recovery diode frequency
The output of the
MHz
step recovery diode multiplier
is filtered
to recover
and 3.5 GHz. The input signal is fed to an amplifier, which feeds the mixer and a level detector. When the presence of an input signal is detected with the level detector, the six possible mixing frequencies, that is, 1, 1.5, 2, 2.5, 3, and 3.5 GHz, are electronically sequenced in ascending order, while the presence of an output signal below 500 MHz is determined by a level detector at the mixer output. When it has been determined that a difference exists below 500 MHz, the selected mixing frequency is transmitted to the frequency counter and added to the counted frequency. Because there is a mixing frequency every 500 MHz, and these frequencies are sequenced from the lowest to the highest, the first detection signals at 1,000
and
1.5, 2, 2.5, 3.0,
Frequency Counters and Time-Interval Measurements
344 1
1.5
Chap.
10,
GHz
GHz Switch
Harmonic 2.0
Generator r+-
GHz
x5
a
2.5 GHz
IOO MHz from Frequency Counter
3.0
GHz
3.5
GHz
Input
500 MHz
4 GHz Level Detector
Figure 10-23
Automatic heterodyning unit
frequency counters to 4
for extending the frequency range of
GHz.
of an output from the mixer less than 500
MHz represents the difference between
the input frequency and the selected mixing frequency.
an output using the next-higher mixing frequency, but the
first
It is
this is
mixing frequency to supply an output below 500
It is
possible to obtain
avoided by selecting
MHz.
informative to calculate the effects on the accuracy of both the pre-
and the heterodyning methods of frequency extension. For the case of the prescaler, assuming that the prescaler does not miss counts, and this is generally true, the output frequency is simply the input frequency divided by N, the prescaler ratio. The displayed frequency is the input frequency to the counter times the gate time, which is scaler
displayed frequency
Because
N
is
fin
(10-4)
N
a constant, the accuracy of the display
is
simply a function of
the gate time. Thus, the accuracy of the counter with a prescaler
same
is
t,
exactly the
as the accuracy of the counter without a prescaler.
Consider the case when using the heterodyning frequency converter where the mixing signal
is
derived from the same clock as
base within the counter.
The
gate time
is
is used to derive the time an integer number of cycles of the
time-base clock, or gate time
= ? Jc
(10-5)
Automatic and Computing Counters
Sec. 10-4
Q
where
is
345
the division of the time base and fc
The mixing frequency
in the converter
output frequency of the converter
A
is
the time-base clock frequency.
is
derived from the same source, and the
is
= A' ± Nf
(10-6)
c
where fm is the frequency into the counter, fj is the frequency into the converter, is the multiplication between the internal time-base clock and the hetand
N
erodyning
signal.
The displayed frequency of the counter is times the gate time, which
displayed
The is
the input frequency of the counter
is
frequency
= fm = *t£. (|)
relationship between the actual input frequency
a function of
f
c
QN
+
(10-7)
and the displayed frequency QN, is a
only, as the second term of relationship (10-7),
constant.
Therefore, neither the heterodyning nor the prescaling
method of increasing
the frequency range of a counter affects the accuracy.
10-4
AUTOMATIC AND COMPUTING COUNTERS
The frequency
counter, being an intensely digital machine,
didate for automating and computerizing.
One
is
an excellent can-
measurement that can be handled by a calculating counter is the measurement of low frequencies with accuracy. One significant problem with the frequency counter is the measurement of low frequencies. If a signal of less than 1 Hz was to be measured with a resolution of 0.01 Hz, the time required would be 100 s if the conventional gatecontrolled counter were used. An alternative measurement technique is to measure the period of the input waveform and calculate the frequency from the excellent
relationship:
frequency
=
—— \
(10-8)
-
period
The time required
to display the frequency
is
the period of the
unknown
input plus the computation time. For the example of a frequency on the order is on the order of 1 ms any waveform can be measured within the time of one period plus a small increase for the computation. However, the determination of frequency from a single period measurement has a statistical
Hz, the period
of
1
or
less.
is
1
s,
while the computation time
Essentially, the frequency of
is very great. A second frequency calculation made from a second period measurement would improve the probability of error, while a third calculation would further improve the error. The calculating frequency
probability of error that
Frequency Counters and Time-Interval Measurements
346
counter would continue to
make frequency
calculations
Chap. 10
from the period of the
input as long as the input were present and display the arithmetic
mean
of the
calculations.
Not only ability of a
are low-frequency measurements improved from the calculating
frequency counter; the measurement of pulsed carriers can be im-
proved by the calculating counter.
often necessary to determine the frequency
It is
of bursts of energy that do not last for long periods of time. consider a
1-jlls
burst of a
1-GHz
carrier.
To measure
As an example,
the frequency of the burst,
only 1,000 complete cycles are available that can be counted.
counter has an ambiguity of
an error of is
1
+
or
—
one count which in
part in 1,000, or 0.1 per cent. If the accuracy of the
measurement
more than one burst has to be counted and ultimately frequency calculation. The calculating frequency counter can make
to be better than this,
used for the
several measurements, average the result of each measurement, statistically
and display a
determined frequency.
In the section on frequency counter accuracy, is
The frequency
this case represents
it
was discussed that there
a point where measuring the period of an input with a certain clock frequency
produces improved accuracy over the measurement of the input frequency for a fixed gate time.
A block
capability of automatically
diagram of an automated frequency counter with the making period or input frequency measurement and
then performing the necessary mathematics to display the correct frequency
is
shown in Fig. 10-24. In this counter, rather than a conventional gate, there are two gated counters. One counter is used to accumulate the input frequency, while the second counter accumulates a precision clock. Both counters are gated Matched Gates Input
>-
A Counter
van
>
B Counter
Gate
Time Base
Computer
Precision
Clock
Period/Frequency Display
Figure 10-24
Precision computing counter using dual counters.
347
Problems
Chap. 10
simultaneously, such that the
counter B.
number of input
A while a precision clock,
cycles has been accumulated in
or the elapsed time
The frequency of the input can be determined from input frequency
count in
=
:
count in
The opening and
is
accumulated
counter
in
the following relationship:
A (10-9)J
B
closing of the gate are controlled from either the input
if the gate is controlled by measurement will be made, and if the gate is controlled by the input signal, a period measurement is being made. As previously explained, the frequency where the accuracy changes from a period measurement to a frequency measurement is (fc ) l/2 where fc is the clock frequency from which the time base is derived and the clock used for the period measurement. In this example the precision clock used for the period measurement is 500 MHz, which places the changeover point at 22 MHz. From the setting of the input switches, which can select the number of significant digits and the resolution, the automatic frequency counter will select the method of measure-
signal or the internal precision clock. Essentially,
the internal clock a conventional frequency
,
ment.
REFERENCES 1.
Prensky, Sol D., and Castellucis, Richard
Englewood 2.
Cliffs, N.J.:
Tocci, Ronald
Englewood
J.,
L.,
Digital Systems: Principles
Cliffs, N.J.:
Electronic Instrumentation,
3rd ed.
Prentice-Hall, Inc., 1982.
and
Applications, chaps. 4,
5,
and
7.
Prentice-Hall, Inc., 1980.
PROBLEMS 1.
A
frequency counter capable of measuring an
unknown frequency
to within
1
Hz
by measuring frequency rather than period would require what minimum gate time? (2}
To what
accuracy can a frequency counter determine an
kHz, using a 3.
How many
1-s
unknown frequency
of 450
time base and a time-base accuracy of 0.01 per cent?
displays (total decades) should a frequency counter have
and resolution are
if its
accuracy
to be 0.001 per cent?
(5plf the internal time base of a frequency counter is 10.000 MHz, what frequency range is best measured by a period measurement, and what frequency range is best
measured by a conventional frequency measurement? 5.
What
effects
on accuracy, resolution, etc., does the addition of a on a frequency counter?
fixed
modulus
prescaler have
^pWhat method can be used to increase the frequency range of a frequency counter? How can this be achieved without degrading the accuracy of the counter?r 7.
What problems
are associated with the
measurement of pulsed
signals?
^4
CHAPTER
1 1
TRANSDUCERS AS INPUT ELEMENTS TO INSTRUMENTATION SYSTEMS
11-1 CLASSIFICATION
An
OF TRANSDUCERS
number of components which measurement and record the result. An instrumentation system generally consists of three major elements: an input device, a signal-conditioning or processing device, and an output device. The input device receives the quantity under measurement and delivers a proportional electrical signal to the signal-conditioning device. Here the signal is amplified, filtered, or electronic instrumentation system consists of a
together are used to perform
a'
otherwise modified to a format acceptable to the output device. device
may
The output
be a simple indicating meter, an oscilloscope, or a chart recorder
for visual display. It
may be a magnetic tape recorder for temporary or permanent it may be a digital computer for data manipulation
storage of the input data, or
or process control.
how
The kind of system depends on what
the measurement result
The input quantity
for
is
is
to be
measured and
to be presented.
most instrumentation systems
is
nonelectrical. In
order to use electrical methods and techniques for measurement, manipulation, or control, the nonelectrical quantity device called a transducer.
One
when actuated by energy
in
same form or transmission
in
may
is
converted into an electrical signal by a
definition states "a transducer
is
a device which,
one transmission system, supplies energy
in the
another form to a second transmission system." This energy be
electrical,
mechanical, chemical, optical (radiant), or ther-
mal.
This broad definition of a transducer includes, for example, devices that convert mechanical force or displacement into an electrical signal. These devices
348
Sec. 11-2
Selecting a Transducer
349
form a very large and important group of transducers commonly found in the and the instrumentation engineer is primarily concerned with this type of energy conversion. Many other physical parameters industrial instrumentation area,
(such as heat, light intensity, humidity) may also be converted into electrical energy by means of transducers. These transducers provide an output signal when stimulated by a nonmechanical input: a thermistor reacts to temperature variations, a photocell to changes in light intensity, effects,
and so on. In
all cases,
an electron beam to magnetic
however, the electrical output
measured by
is
standard methods, yielding the magnitude of the input quantity in terms of an analog electrical measure.
Transducers may be classified according to their application, method of energy conversion, nature of the output signal, and so on. All these classifications usually result in overlapping areas. A sharp distinction between, and classification types of transducers is difficult. Table 11-1 shows a classification of transducers according to the electrical principles involved. The first part of the table lists transducers that require external power. These are the gassive transducers, proof,
ducing a variation in some electrical parameter, such as resistance, capacitance, and so on, which can be measured as a voltage or current variation. The second category are transducers of the self-generating type, producing an analog voltage or current when stimulated by some physical form of energy. The self-generating transducers do not require external power. Although it would be almost impossible to classify all sensors and measurements, the devices listed in Table 1 1- 1 .represent a good cross section of commercially available transducers for application in instrumentation engineering.
Some
of the more
common
trans-
ducers and their application are discussed in the following sections.
11-2 SELECTING
A TRANSDUCER
In a measurement system the transducer
function of transforming
signal. Selection of the appropriate
most important step
is
the input element with the critical
some physical quantity transducer
is
to a proportional electrical
therefore the
in obtaining accurate results.
A
first
and perhaps
number of elementary
questions should be asked before a transducer can be selected, for example,
(a)
4/* J (b) (c)
The
first
What is the physical quantity to be measured? Which transducer principle can best be used to measure What accuracy is required for this measurement?
this quantity?
question can be answered by determining the type and range of the
measurand.
An
appropriate answer to the second question requires that the
input and output characteristic of the transducer be compatible with the recording
or measurement system. In most cases, these two questions can be answered readily,
implying that the proper transducer
is
selected simply
by the addition
1
"
C £ u a
a
H &
3 £
>
£ 3
.ti
>
«J
.3
-a
o o
'3
H
U
—
0)
1
§
5
£
o ^ C C D
c/5
o 13
"o
G U
O
Si
c 3
g 1)
£?
o .2 a >
si
^
4>
2 ° '5 s
ed
2 *? D C
« £
~
^
i3 6JD
c
I
"S
g
> * £ o 03
C •s.
2
5
C
8 o c
«
I
00
| 4= 00
3
> g o '-3
.2
s
«
3
-
2
-a
3 43 a C Cu O 03 c O _ 43 o a E « 2
T3
w c 5/5
a.
°
TZ
^ o
3 C O
T3
-a
o
S5g 3
T3 t« (S
03
O
§ .2 8 2 3 O 03
Comparator Outputs
Figure 12-13
and j V;
if all
than } V. In
1
Analog
Reference Voltage
V
Chap.
^V {V
to-^V to
|v
Simultaneous analog-to-digital converter.
the comparators are on, the analog voltage must be greater
four different output conditions
total,
parators on to all comparators on.
may
The analog input
exist:
from no com-
voltage can therefore be
resolved in four equal steps. These four output conditions can be coded to give
two binary
bits
of information. This
is
shown
in the table alongside the
diagram
of Fig. 12-13. Seven comparators would give three binary bits of information,
comparators would give four bits, etc. The advantage of the simultaneous system of A/D conversion is its simplicity and speed of operation, especially when low resolution is required. For a high-resolution system (a large number of bits), this method requires so many fifteen
comparators that the system becomes bulky and very
12-5.3 Counter-type
A/D
If the reference voltage to variable, the
Converter
which the analog input
number of comparators could be reduced
the reference voltage
is
costly.
is
to be
to only one.
a linearly increasing voltage (a ramp) that
compared were If,
is
for instance,
continuously
applied to the comparator input, coincidence of the reference voltage and the
unknown voltage could be determined in terms of the time elapsed since ramp started. But a digitally controlled variable reference already exists in form of the simple
D/A
converter of Fig. 12-12. This
used to convert a digital number in
which can be compared
to the
its
DAC
unknown analog
D/A
register into
the the
converter can be
an analog voltage
input by a comparator circuit.
Sec. 12-5
If the
Analog-to-Digital Conversion
411
two voltages are not equal, the digital number in the DAC register is its output is again compared. This is exactly the operation of the
modified and
circuit of Fig. 12-14.
The
A/D
generalized
converter of Fig. 12-14
is
feedback system, where the main components are the and some control logic circuitry.
D/A
the comparator,
may be used to control the conversion that takes place One of the simplest ways is to start the DAC at zero
Various methods in the
actually a closed-loop
DAC,
converter.
and count the number of input pulses required to give an output voltage that equals the analog input. Analog Input
Reference Supply
DAC Feedback Loop
Digital
Output
Gating
and
Flip-flop Register
Control
D/A Converter Figure 12-14
A/D
converter using a
DAC
to provide the
comparison voltage. The
contents of the flip-flop register provide the digital output.
The
counter-type
A/D
converter of Fig. 12-15 contains a
D/A
converter
(DAC), the reference supply, the DAC register of Fig. 12-14. The
section consisting of the resistive divider network
and a six-stage counter which replaces comparator again receives the unknown analog input the generated
DAC
for
comparison against
output voltage. The control circuitry consists of a pulse
generator or clock, a signal gate that steers the clock pulses to the counter, and a control flip-flop for starting and stopping the conversion.
When
a start signal
start-stop flip-flop
is reset.
is
given, all the counter flip-flops are cleared
This
flip-flop
and the
provides a gating level (positive logic)
to the signal gate, allowing the clock pulses to be applied to the counter register.
The clock
pulses are propagated through the counter, and the
DAC
output increases in steps toward the top of the reference voltage. divider output
is
the
equal to the analog input, the comparator switches, delivering
an output signal to the start-stop
flip-flop.
This
flip-flop sets
to zero, blocking the clock pulses at the signal gate.
stores the
divider
When
number of clock
At
and
its
output drops
this instant, the
counter
pulses that were required to raise the reference
voltage to the level of the analog input voltage.
then the binary equivalent of the analog input.
The contents of
the counter are
412
Sec. 12-5
Analog-to-Digital Conversion
The
413
measured from the moment a request is given to is available. For the counter-type A/D converter the conversion time depends on the magnitude of the analog voltage and is conversion time
moment
the
is
a digital output
therefore not constant. If the input signal
know when
is
variable,
uncertainty of this time measure
is
is
also important to
The
called the aperture time (sometimes also
window or sample time). The aperture occurs shown in the waveform diagram of Fig. 12-16.
called as
it
the input signal had the value given by the digital output.
at the
end of the conversion,
Volts
Analog Voltage
Figure 12-16 of coincidence
The bits (extra
Waveform diagram of analog when readout occurs.
input and divider output showing points
resolution of the counter-type converter
is
improved by adding extra
counter stages). This addition can be done at small extra cost. The
conversion time, however, increases rapidly with the number of bits used, since
an
jV-bit
converter needs time for 2
A
counts to accumulate. Higher resolution
therefore obtained at the cost of increased conversion time.
is
One method into sections.
of
five bits each.
the counter
of decreasing the conversion time
For instance, a is
At
10-bit converter
to divide the counter
the start of the conversion, the least significant section of
preset to all ones
significant section.
is
could be divided into two sections
When
and counts are inserted only into the most
the comparator indicates that the analog input level
has been exceeded, the least significant section of the counter the
until the correct value
complete a conversion
is
6
)
cleared, reducing
reached.
The maximum number of
steps required to
5
is
2 for the most significant counter and 2
significant counter, giving a total of 2
(2
is
DAC divider output. Pulses are then inserted into the least significant section
versus 1,024 counts (2
10
)
6
steps.
This
is
a
for the standard counter.
maximum
5
for the least
of 64 counts
Analog and
414
The
section counter technique
where the output
is
is
Digital
Data Acquisition Systems
Chap. 12
frequently used in digital voltmeters,
to be in decimal notation.
Each
section of the counter then
represents a sectional digit.
12-5.4 Continuous
The
A/D
Converter
big disadvantage of the counter converter
is
that the entire comparison
process starts from the beginning each time a coincidence has been detected by the comparator. This
A
means low resolution and low speed. counter method involves replacing the simple
slight modification of the
counter with a reversible counter, or up-down counter. This allows the converter to continuously follow the analog input voltage this voltage changes.
Once the converter
whatever the direction
in
which
starts running, the digital equivalent
of the input voltage can be sampled at any time, and an extremely rapid readout is
possible.
The
simplified logic block
converter.
The
D/A converter,
(b) the
diagram of
Fig. 12-17 represents the continuous
up-down counter,
illustration contains four basic parts: (a) the (c)
the comparator, (d) the synchronization and control
logic.
An
ordinary binary counter counts in the forward direction (up)
trigger input of the succeeding binary
is
connected to the
1
when
the
output of the
preceding binary. The count will proceed in the reverse direction (down)
if
the
made instead to the 0 output of the preceding binary. The two methods may be used simultaneously to produce an up-down counter. In Fig. coupling
is
12-17 additional
make
AND
nization of the count
D/A
The Fig.
gates are used in the trigger circuits of the binaries to
moment; synchro-
sure that counts are accumulated only at the desired
and the comparison must take
converter section
is
place.
identical to the basic resistive divider of
12-11; the reference supply provides the required precision voltage for
accurate conversion.
DAC,
but
it
The 0 outputs of the
binaries are connected directly to the
should be understood that appropriate level conversion takes place
between the binaries and the
DAC
input terminals.
The comparator again compares
the analog input voltage with the
output voltage, providing two possible output voltage is larger than the feedback voltage comparator output terminal is connected to the
input voltage
a gating element. Similarly,
when
(DAC
When
levels.
output), the appropriate
set input
of the up flip-flop via
the analog input voltage
is
feedback voltage, the comparator provides an output voltage at
which
The flops
is
then connected, via a gate, to the
is
smaller than the its
terminal of the
actual transfer of the comparator output signals to the
other terminal
down
flip-flop.
up and down
flip-
controlled by gating pulses from the clock, which controls the synchro-
nization of the entire flops are exclusive
both
set
DAC
the analog
measurement
OR 'd
cycle.
together to
make
The outputs of
the up and
down flipwhen
sure that no count takes place
flip-flops are set (a safety precaution).
r
Digital
Outputs
Up-down Counter
D/A Conversion
° FF
0
t
0
1
FF
—
0
FF
FF
1
1
c
i
0
T.
1
FF
1
Down
Up Synchronization /Control
Comparator
re
Feedback Voltage
0
1
C
3
1
Analog Input o
2 Figure 12-17
Down Flip-Flop
Up Flip-Flop]
0
1
q
s
i
Delay
n
Clock
Simplified logic block diagram of the continuous converter.
415
Analog and
416
At the
measuring
start of a
Data Acquisition Systems
Digital
cycle,
when
all
Chap. 12
the flip-flops are cleared, the
clock generates a pulse that samples the comparator output. If the analog input is
it usually is when a measurement first The delayed clock pulse is then allowed to trigger the same time, it conditions the trigger gate of the
larger than the feedback voltage (as
starts),
the
the up flip-flop
first
is set.
binary, while at
succeeding binary. The 0 output of the
and the resulting
DAC output
analog input
If the
is still
is
first
compared
binary
and also
at
The procedure
DAC
output
is
this at the
pulse
up
flip-flop again,
binary back to
original
its
used for comparison against the analog
repeats until the feedback voltage equals the analog input,
which time the comparator output If the
by the comparator.
The count then has advanced by one
to trigger the next binary.
and the corresponding input.
first
DAC,
connected to the
larger, the next clock pulse sets the
allowing the delayed clock pulse to trigger the state
is
to the analog input
is
zero and the count
stopped.
is
analog input changes to a lower value, the next clock pulse detects
comparator output and
sets the
down
Now
flip-flop.
the delayed clock
allowed to enter the binary counter at the trigger input of the
is
binary, but the count
is
first
carried from stage to stage at the 0-output side of the
binaries of Fig. 12-17 so that the contents of the counter are reduced by one.
DAC
The
output then also drops the appropriate amount and the next com-
parison determines whether the up flip-flop or the
down
flip-flop will
be
set.
The
counter therefore continuously follows the analog input voltage.
The waveform diagrams of Fig. The aperture is the time
converter.
12-18 illustrate the action of the continuous
The assumption more than ± 1 LSB (the
for the last step.
that the analog input voltage does not change
increment of the the
DAC)
between conversion
steps.
To meet
maximum range of change of the input voltage must
rate of
made
smallest
this requirement,
not exceed the
maximum
change of the converter.
12-5.5 Successive-approximation
The successive-approximation to a
is
DAC
reference voltage that
illustrated in Fig. 12-19,
where a
A/D
is
A
Converter
converter compares the analog input
repeatedly divided in half.
four-digit binary
the full reference supply voltage V,
corresponding to - V.
A/D
is
number
The process
divided in half (binary
comparison between
is
(1000), representing
number
this reference voltage {\
100),
V) and
is made. If the result of this comparison shows that this first approximation was too small (y V is smaller than the analog input), then the next comparison will be made against j V (binary number 1 10). If the comparison showed that the first approximation was too large ( T V larger than the analog
the analog input
input), then the next
comparison
will
be
made
against \
After four successive approximations, the digital
number
will
V
number
(binary
is
number
resolved.
A
010).
six-digit
be resolved in six successive approximations. This compares very 6 with a conventional ) comparisons needed
favorably with the sixty-four (2
counter-type converter.
Final
Compa rison Third
Comparison Second Comparison First
HE
Comparison
1
f V
RTO
Ton 1
\ V
10
fTOQ
oo
oTTl [011
oTol
iV 1010 DAC
001
Output Voltage
ov
ooo
Contents of Control Register Figure 12-19
Operation of the successive-approximation
A/D
converter.
417
Analog and
418
Digital
The successive-approximation method previous methods since
it
is
a
little
more
Chap. 12
elaborate than the
requires a special control register to gate pulses to the
bit, and so on. The additional cost of the control and the converter can handle continuous and diswith large and small resolutions at moderate speed and
first bit,
then to the second
register,
however,
is
continuous signals
moderate
Data Acquisition Systems
small,
cost.
The generalized block diagram of Fig. 12-20 shows the basic successive approximation converter. The converter uses a digital control register with gateable
and 0
1
comparison
tribution register
which step
inputs, a digital-to-analog converter with reference supply, a
timing loop, and a distribution register. The dis-
circuit, a control
is
is
like a ring
counter with a single
1
circulating in
it
to determine
taking place.
Analog Input
9 Digital
Reference Supply
-to-Analog Converter
Digital
Output Control Register
Time Delay Set
MSB
Startstop Distribution Register
Flip-flop Start
o
End Figure 12-20
of
Conversion
Simplified block diagram of the successive-approximation
A/D
con-
verter.
At the beginning of the conversion distribution register are set with a in all bits of less significance.
The
1
.
.
.
,
both the control register and the
most significant
bit
(MSB) and
a 0
distribution register therefore registers that
the cycle has started and that the process
which now reads 1000
cycle,
in the
is
in its first phase.
The
control register,
causes an output voltage at the digital-to-analog
converter section of one-half of the reference supply. At the same instant, a pulse enters the timing delay chain.
By
the time the
D/A
converter and the
comparator have settled, this delayed pulse is gated with the comparator output. When the next most significant bit is set in the control register by the action of
Sec. 12-5
Analog-to-Digital Conversion
419
the timing chain, the most significant bit either remains in the
1
-state or
it is
depending on the comparator output. The single 1 in the distribution register is shifted to the next position and keeps track of the number of comparisons made. reset to the 0-state,
This procedure repeats, following the diagram of Fig. 12-19, until the
final
approximation has been corrected and the distribution register indicates the end of the conversion. Synchronization
is
not required in this system because the
comparator controls only one flip-flop at a time. For a successive-approximation converter, the
some value
digital
output corresponds
had during the conversion. Thus the aperture time is equal to the total conversion time. This is illustrated in the waveform reconstruction of Fig. 12-21. Aperture time of this converter can be reduced by using redundancy techniques or by using a sample-and-hold circuit. to
that the analog input
Analog
Read Read Read Read Read Read Read Read Read Read Read Read Read Out 1
Out 2
Out
Out
3
4
Out 5
Out
Out
Out
Out
6
7
8
9
Out 10
Out II
Out 12
Voltage
Figure 12-21
proximation
Waveform diagrams
A/D
converter.
illustrating the operation of the successive-ap-
Out 13
Analog and
420
Digital
Data Acquisition Systems
Chap. 12
12-5.6 Sample-and-hold Circuit
A
sample-and-hold circuit
used with an
is
necessary to convert a high-frequency signal that
The sample-and-hold
an accurate conversion.
is
A/D
circuit
amplifier that charges a capacitor during the sample
converter
is
it
is
basically an operational
mode and
retains the charge
The sample-and-hold
of the capacitor during the hold mode.
when
varying too rapidly to allow
circuit
can be
represented by the simple switch and capacitor of Fig. 12-22.
(b)
(a) Circuit
Input
waveform
Hold
Sample
/
/
Al
1
1
I
I
1
i
1
I
i
1
t
(c)
Figure 12-22
When
the switch
Output waveform
Operation of the sample-and-hold
is first
circuit.
closed, the capacitor charges to the value of the
input voltage and then follows the input (assuming a low driving source impedance).
When
at the
time the switch was opened (assuming a high-impedance load).
The
the switch
is
opened, the capacitor holds the voltage that
acquisition time of the sample-and-hold
is
The aperture time
and the uncertainty time
is
in the
is
is first
the time required for the switch to change state
time that this change of state occurs. The holding
the length of time the circuit can hold the charge without dropping
than a specified percentage of It is
had
the time required for the
capacitor to charge up to the value of the input signal after the switch shorted.
it
its initial
possible to build a sample-and-hold circuit exactly as
12-22. Often, however, the circuit
is
more
value.
built
shown
in Fig.
with fast-acting transistor switches
and an operational amplifier to increase the available driving current into the capacitor or to isolate the capacitor from an external load on the output. However the sample-and-hold circuit is built, it always acts as the simple switch and capacitor shown.
Sec. 12-6
421
Multiplexing
—VWV-
V\AAr-
|Okn
Input
9.5
11%
1
kH
1
kn
7
-
20
dB
-
20
dB
2.884
X
10
X
10
=
100
X
100
dB
Number
9.2
dB
2.884
288.4
8.318
8.318
X
=
of decibels negative (-).
number of
to the given
=
decibels until the
sum
83180
Add + falls
20 decibels successively
within the range of Table
1
For the voltage ratio, divide the value from the left-hand voltage-ratio column by 10 for each time 20 dB was added. For the power ratio, divide the value from the left-hand power-ratio column by 100 for each time 20 dB was added. Example Given: -49.2
dB -49.2 dB
Voltage
ratio:
-9.2 dB
—»
+
ratio:
-9.2 dB
—>
dB + 20 dB = -9.2 dB
0.3467
0.3467
Power
20
x
X
i-
= 000 34 67
0.1202
0.1202
X
—X—= 100
0.00001202
100
453
Voltage Ratios to Decibels: Table
For ratios smaller than those by 10 successively
by
Table
2.
Multiply the given ratio
product can be found in Table
until the
of decibels thus found, subtract tiplied
in
457
2, p.
+20
2.
From
number was mul-
the
decibels for each time the ratio
10.
Example Given: Voltage ratio
=
0.0131
X
0.0131
From Table
2:
1.31
—»
2.345
10
=
1.31
dB
dB -
2.345
X
10
20
dB - 20 dB
For ratios greater than those
in
=
-37.655 dB
Table
2.
Divide the given ratio by
10 successively until the remainder can be found in Table decibels thus found,
add
+20 dB
for each time the ratio
2.
Example Given: Voltage ratio
=
712 712
From Table
2:
7.12
—
>
17.050
17.050
454
x
To
x To
=
7 12
dB
=
-
dB
dB + 20 dB
+
20
57.050
To
the
number of
was divided by
dB
10.
1
1
TABLE To account (a)
Conversion of Decibels to Power and Voltage (or Current) Ratios
1
for the sign of the decibel:
For positive
+ ) dB
(
values use the
two right-hand columns. The voltage and power
ratios
are greater than unity. (b)
For negative
(
— dB )
values use the two left-hand columns.
The
voltage and
power
ratios
are less than unity.
-dB+ Voltage
Hal 10 1.0000 .9886 .9772 .9661
.yoou .9441
.9333 .9226 .9120 .yui o
flower Ratio
1.0000 .9772 .9550 .9333 9 1 20
dB 0 .1
.2 .3
A
.8913 .8710
.5 .6
.8511
.7 .8
.8318
Q .y
.8913 .8810 .8710 .8610 851
.7943 .7762 .7586 .7413
.8414 .8318 .8222 .S12K .SUoO
.7079 .6918
.7943 .7852 .7762 .7674
o8b
.6310 .6166 .6026 .5888 .0/ 04
.7499 .7413 .7328 .7244 .7161
.5623 .5495 .5370 .5248 .5129
2.5 2.6 2.7 2.8 2.9
.7079
.5012 .4898 .4786 .4677
3.0
.
/
.6998 .6918 .6839 .6761
7244
.6761 .6607 KA ^7
-dB+ Voltage Ratio
Power
1.000 1.012 1.023 1.035
1.000
1
047
1.059 1.072 1.084 1.096 1
1
no
Voltage Ratio
Power Ratio
3.41
10.655
dB
(voltage) X
Voltage Ratio 1.0 1.1
1.2 1.3 1.4
.00 .
000
.828 1.584
2.279 2.923
.02
.01
.086 .906 1.656 2.345 2.984
.03
.172 .984 1.727
=
dB X -
10.655
dB (power)
5.328
.05
.04
.06
.07
.424 1.214 1.938 2.607 3.227
.506 1.289 2.007 2.671 3.287
.588 1.364 2.076 2.734 3.346
.668 1.438 2.144 2.798 3.405
2.212 2.860 3.464
3.637 4.190 4.711 5.201 5.666
3.694 4.244 4.761 5.249
3.750 4.297 4.811 5.296 5.756
3.807 4.350 4.861 5.343 5.801
3.862 4.402 4.910 5.390 5.845
3.918 4.454 4.959 5.437 5.889
3.973 4.506 5.00S 5.483 5.933
4.028 4.558 5.057 5.529 5.977
6.319 6.729 7.121 7.495 7.854
6.361 6.769 7.159 7.532 7.889
6.403 6.809 7.197 7.568 7.924
8.199 8,530 8.850 9.158 9.455
8.232 8.563 8.881 9.188 9.484
8.266 8.595 8.912 9.218 9.513
9.771 10.049 10.317 10.578 10.832
9.799 10.076 10.344 10.604 10.857
2.41
3.046
5.621
6.064 6.486 6.888 7.272 7.640
6.
107 6.527 6.927 7.310 7.676
6.
150 6.568 6.966 7.347 7.712
6.
2.2 2.3 2.4
6.021 6.444 6.848 7.235 7.604
193 6.608 7.008 7.384 7.748
6.235 6.649 7.044 7.421 7.783
6.277 6.689 7.082 7.458 7.819
2.5 2.6 2.7 2.8 2.9
7.959 8.299 8.627 8.943 9.248
7.993 8.333 8.659 8.974 9.278
8.028 8.366
8.062 8.399 8.723 9.036 9.337
8.097 8.432 8.755 9.066 9.367
8.131 8.465 8.787 9.097 9.396
8.165 8.498 8.818 9.127 9.426
3.0
9.542
9.939 10.211 10.475 10.731
9.686 9.966 10.238 10.501 10.756
9.743
9.883 10.157 10.423 10.681
9.629 9.911 10.184 10.449 10.706
9.714
9.827 10.103 10.370 10.630
9.571 9.855 10.130 10.397 10.655
9.657
3.1
9.994 10.264 10.527 10.782
10.021 10.291 10.553 10.807
10.931 1 1 174
10.955 1 1.198
10.980
1.41
1K434
1L457
1
1.481
1L504
.5
1.6 1.7 1
.8
1.9
to 2.1
3.2 3.3 3.4
3.580 4.137 4.660 5.154
8.691 9.005 9.308
9.600
5.71
.053 1.293
11.078
1
11
1L527
1L573
11.754 11.976
11.550 11.777 11.998
12.770 12.967
12.171 12.382 12.588 12.790 12.987
12.192 12.403 12.609 12.810 13.006
12.213 12.424 12.629 12.829 13.026
12.234 12.444 12.649 12.849 13.045
13.141 13.330 13.516 13.697 13.875
13.160 13.349 13.534 13.715 13.892
13.179 13.368 13.552 13.733 13.910
13.198 13.386 13.570 13.751 13.927
13.217 13.405 13.589 13.768 13.945
13.236 13.423 13.607 13.786 13.962
14.031 14.202 14.370 14.535 14.696
14.049 14.219 14.387 14.551 14.712
14.066 14.236 14.403 14.567 14.728
14.083 14.253 14.420 14.583 14.744
14.100 14.270 14.436 14.599 14.760
14.117 14.287 14.453 14.616 14.776
14.134 14.303 14.469 14.632
14.855 15.010 15.163 15.313 15.461
14.870 15.026 15.178 15.328 15.476
14.886 15.041 15.193 15.343 15.490
14.902 15.056 15.208 15.358 15.505
14.917 15.072 15.224 15.373 15.519
14.933 15.087 15.239 15.388 15.534
14.948 15.102 15.254 15.402 15.549
11.596 11.821
11.387 11.618 11.844
11.641 11.866
11.664 11.888
11.687 11.910
11.709 11.932
11.732 11.954
4.9
12.041 12.256 12.465 12.669 12.869
12.063 12.277 12.486 12.690 12.889
12.085 12.298 12.506 12.710 12.908
12.106 12.319 12.527 12.730 12.928
12.128 12.340 12.547 12.750 12.948
12.149
4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9
13.064 13.255 13.442 13.625 13.804
13.084 13.274 13.460 13.643 13.822
13.103 13.293 13.479 13.661 13.839
13.122 13.312 13.497 13.679 13.857
5.0
13.079
5.1
5.2 5.3 5.4
14.151 14.320 14.486 14.648
13.997 14.168 14.337 14.502 14.664
14.014 14.185 14.353 14.518 14.680
5.5 5.6 5.7 5.8 5.9
14.807 14.964 15.117 15.269 15.417
14.823 14.979 15.133 15.284 15.432
14.839 14.995 15.148 15.298 15.446
4.1
1 1
1
50
1
1
11.102
1 1
1
10.906
1L364
3 6
.749 1.511
1 1
10.881 1 1.126
3.7 3.8 3.9
3.5
.09
.341 1.138 1.868 2.542 3. 167
3.522 4.082 4.609 5. 105 5.575
1
.08
.257 1.062 1.798 2.477 3. 107
1 1
.222
11.005 1 1 .246
12.361 12.568
.029 1.270
317
1
1.341
11.799 12.019
14.791
457
TABLE
2
Conversion of Voltage (or Current) and Power Ratios to Decibels (cont.)
T
V olX(XQt Ratio
.00
.01
.02
.03
.04
.05
.06
.07
.08
.09
6.0
15.563 15.707 15.848 15.987 16.124
15.577
15.592 15.735 15.876 16.014
15.606 15.749
15.635
778
15.649 15.792
15.918 16.055 16.191
15.931 16.069 16.205
15.664 15.806 15.945 16.083 16.218
15.678 15*820 15.959 16^096 16.232
15.692 15.834 15.973 16^1 10 16.245
16.258 16.391 16.521 16.650 16.777
7.2 7.3 7.4
16.151
16.028 16.164
15.621 15.763 15.904 16.042 16.178
16.272 16.404 16.534 16.663 16.790
16.285 16.417 16.547 16.676 16.802
16.29S 16.430 16.560 16.6S8 16.815
16.312 16.443 l6.57o 16.701 16.827
16.325 16.456 16.586 16.714 16.840
16.338 16.468 16.598 16.726 16.852
16.351 16.483 16.612 16.739 16.865
16.365 16.496 16.625 16.752 16.877
16.378 16.509 16 637 16.764 16.890
16.902 17.025 17.147 17.266 17.385
16.914
16.927 17.050
16.939 17.062 17.183
17.290 17.408
17.302 17.420
16.951 17.074 17.195 17.314 17.431
16.964 17.086 17.207 17.326 17.443
16.976 17.098 17.219 17.338 17.455
16.988 17.110
17.171
17.231 17.349 17.466
17.001 17.122 17.243 17.361 17.478
17.013 17.135 17.255 17.373 17.490
7.5 7.6 7.7 7.8 7.9
17.501 17.616 17.730 17.842 17.953
17.513 17.628 17.741 17.853 17.964
17.524 17.639 17.752 17.864 17.975
17.536 17.650 17.764 17.875 17.985
17.547 17.662
17.570 17.685
17.886 17.996
17.559 17.673 17.786 17.897 18.007
17.908 18.018
17.582 17.696 17.808 17.919 18.029
17.593 17.707 17.820 17.931 18.040
17.605 17.719 17.831 17.942 18.051
8.0
18.062 18.170 18.276 18.382 18.486
18.073 18.180 18.287 18.392 18.496
18.083 18.191 18.297 18.402 18.506
18.094 18.202 18.308 18.413 18.517
18.105 18.212 18.319 18.423 18.527
18.116 18.223 18.329 18.434 1H.537
18.127 18.234 18.340 18.444 18.547
18.137 18.244 18.350 18.455 18.558
18.148 18.255
18.159 18.266
8.2 8.3 8.4
18.361 18.465 18.568
18.371 18.475 18.578
8.5 8.6 8.7 8.8 8.9
18.588 18.690 18.790 18.890 18.988
18.599
18.609 18.710 18.810 18.909 19.007
18.619 18.720 18.820 18.919 19.017
18.629 18.730 18.830 18.929 19.027
18.639 18.740 18.840 18.939 19.036
18.649 18.750 18.850 lh.949 19.046
18.660 18.760 18.860 18.958 19.056
18.670 18.770 18.870 18.968 19.066
18.680 18.780 18.880 18.978 19.075
.0
19.085
9.1
19.181 19.276 19.370 19.463
19.094 19.190 19.285 19.379 19.472
19.104 19.200 19.295 19.388 19.481
19.114 19.209 19.304 19.398 19.490
19.123 19.219 19.313 19.407 19.499
19.133 19.228 19.323 19.416 19.509
19.143 19.238 19.332 19.426 19.518
19.152 19.247 19.342 19.435 19.527
19.162 19.257
9.2 9.3 9.4
19.351 19.444 19.536
19.171 19.226 19.360 19.453 19.545
9.5 9.6 9.7 9.8 9.9
19.554 19.645 19.735 19.825 19.913
19.564 19.654 19.744 19.833 19.921
19.573 19.064 19.753 19.842 19.930
19.582 19.673 19.762 19.851 19.939
19.591 19.682 19.771
19.600
19.609 19.700 19.789 19.878 19.965
19.618 19.709 19.798 19.886 19.974
19.627 19.718 19.807 19.895 19.983
19.636 19.726 19.816 19.904 19.991
6.1
6.2 6.3 6.4
6.5 6.6 6.7
6.8 6.9
7.0 7.1
8.1
Voltage Ratio
15.721 15.862 16.001 16.137
17.037 17.159 17.278 17.396
IN. 700
18.K00 1K.900 18.998
15 890
17 775
19.860 19.948
15
17 797
19.691
19.780 19.869 19.956
0
1
2
3
4
5
6
7
8
9
20 30 40
20.000 26.021 29.542 32.041
20.828 26.444 29.827 32.256
21.584 26.848 30.103 32.465
22.279 27.235 30.370 32.669
22.923 27.604 30.630 32.869
23.522 27.959 30.881 33.064
24.082 28.299 31.120 33.255
24.609 28.627 31.364 33.442
25.105 28.943 31.596 33.625
25.575 29.248 31.821 33.804
50 60 70 80 90
33.979 35.563 36.902 38.062 39.085
34.151 35.707 37.025 38.170 39.181
34.320 35.848 37.147 38.276 39.276
34.486 35.987 37.266 38.382 39.370
34.648 36.124 37.385 38.486 39.463
34.807 36.258 37.501 3S.588 39.554
34.964 36.391 37.616 38.690 39.645
35.117 36.521 37.730 38.790 39.735
35.269 36.650 37.842 38.890 39.825
35.417 36.777 37.953 38.988 39.913
100
40.000
10
To the
458
'
convert voltage and power ratios outside the range of this table, use
method
illustrated
on page 454 of the introduction to these
tables.
SELECTED ANSWERS
CHAPTER
1
mV
2.
I
4.
75.0
6.
82
8.
(a) 147.5 ft, (b) 0.21 ft, (c) 0.3 ft, (d) 0.2 ft
±
jLtF
0.1
mV
10. (a) (c)
36
± ±
ft
24.32
1.8 ft,
75
ft
±
3.75
ft,
(b) 111
±
5.55
ft,
111
± 5%
3.65 ft
3.7% ±7.55%, ±0.57%
12. (a) 435.3 ft, (b) 14. (a)
CHAPTER
2
GHz,
1.
1.5
3.
X 10~ 2.85 X 10 4.6
5.
180
9.
35.7
19
cm m/s
X 200 V
10
15. 4.6875
X
11. 3.6
13.
12,500 Hz, 0.125 jnH, 346,400 V, 0.0053 A, 5,000 9 13 12 hr, 14 X 10" /lis J, 0.0014 ms, 8.89 X 10~
17. (a) 8,930
6
10
15
kg/m
3 ,
(b)
557 lb/ft
3
mH,
ft,
Selected Answers
460
CHAPTER 6.
10.
3
0.999993 1.0190
fl
V
CHAPTER 4 n
1.
875
3.
36
MCI
6.
50
V
7.
(a)
0.094
9.
1.25
15.
900
17. 25
and higher
mW,
(b) 4.29
mW
V
n/v
W
CHAPTER 5 i.
o.oi
3.
6
v 5. 7.
X
n 10- 7
R = 34.3 ft, L = (a) R = 1,000 O, s
29 (b)
mH R = s
250
H
CHAPTER 6 mV,
2.
15
4.
26.6
6.
2+ pF
CHAPTER 6.
2.65
10,000
m,
X
H/V
2.66 kn, 266
O
7 10
7
m/s
CHAPTER 8 1.
2.65
3.
+35 dBm, -30 dBm, +26 dBw,
VtV+ 15.
15.9
Hz
1
N-
\
1
V,
+3 dBw, -17 dBw,
0.22
Selected Answers
461
CHAPTER 9 dB
1.
70
5.
-60 dBm
7.
(a)
(b) (c)
amount equal to the attenuation. dynamic range. Increases the noise figure by an amount equal to the attenuation. Increases the third-order intercept by an
Does not
affect the
CHAPTER 10 1.
Is
3.
Five digits
CHAPTER 4.
%
694
11
kg/cm
2
25 iiV
6,
4.16
7.
2.5
mm X
10" 3
mm
INDEX
463
1
464
AC
Index
Digital /analog conversion, 407
voltmeter, 151
ATE, 429
Direct frequency synthesis, 273
ATN, 434 Accuracy,
Distortion analyzer, 300
Distributed-parameter delay
1
Acquisition time, 420
Droop
Alternate sweep, 234
Dynaloy, 356
(pulse),
line,
232
284
Alternating-current instruments, 85
Analog /digital conversion, 409 Arithmetic mean, 4, 10 Atomic definition, 38 Atomic standard, 38
EDI, 435
Attenuator, 262
Electrostatic deflection, 212
piston,
266
Electrodynamometer, 86 Electromagnetic system, 22 Electronic multimeter, 156
Electrostatic system, 22
Average deviation, Ayrton shunt, 63
1
English system, 27
Environmental
errors, 9
Errors, 2
CRDX,
55
Campbell standard, 46 Capacitance comparison bridge, 129
FM
Capacitance standards, 45
Fluorescence, 220
Cathode ray
Form
tube,
210
Chopped sweep, 234
recording, 401
Falltime, 283
factor, 91
Chopper-stabilized amplifier, 149
Frequency divider signal generator, 274 Function generator, 291
Colpitts oscillator, 261
Fundamental-suppression analyzer, 303
Compensated attenuator, 225
Fundamental
Component measuring instrument, 178 Computing counter, 345 Constantan, 355
A/D
Continuous
DVM,
175
Conversion time, 413
Counter
Gating
error,
337
Gauss, 25 Gilbert, 25
Giorgi system, 22
327
Graticule, 222
frequency, 326
Guarantee
synchronous, 328
Counter-type Critical
Galvanometer, 50 Gas-filled phototube, 386
converter, 414
Continuous-balance
BCD,
units, 21
A/D
damping,
Guarded
errors, 16
bridge, 122
converter, 410
53, 55
Current transformer, 104
Hard beam, 217 Harmonic analyzer heterodyne, 302
DAV,
tuned
433
Delay
line,
Derived
230
units,
circuit,
Harmonic
Damping, 54 20
Detector, crystal, 281
Hay
301
converter, 342
bridge, 134
Heterodyne wave analyzer, 297 Horizontal deflection, 235
Deviation from the mean, 10 Differential voltmeter, 163
IEEE, 488, 433, 443
Digital recording, 402
IEEE
Digital voltmeter, 169
IFC, 435
standards, 47
Index
IM
465
Potential transformer, 104
distortion
second-order, 312
Potentiometer, 144
third-order, 312
automatic, 376
Indirect frequency synthesis, 270
Power-factor meters, 101
Inductance comparison bridge, 46
Precision,
Inductance standards, 46
Prescaler, 341
2
1,
Inductor-capacitor oscillator, 260
Primary standard, 35
Input resistance, 161
Probe, oscilloscope, 239
Instrument,
Pulse generator, 282
1
Instrument transformer, 103
Q
Instrumental errors, 9 Integrating
DVM,
meter, 186
173
International standard, 35
REN,
435
RF
power measurement, 203 Ramp-type DVM, 170
Josephson junction, 42
Random Kelvin bridge, 117
errors, 9, 186
Reciprocal mixing, 308 Rectifier-type instruments, 89
LORAN-C, LVDT, 362
39
Repetition rate, 284
Resistance thermometer, 369
Limiting errors, 16
Resolution, 2
Lumped-parameter delay
MKSA
line,
231
system, 22, 27
Risetime, 283
SI system, 22, 27
Magnetic tape recorder, 394
SINAD,
Maxwell
bridge, 132
SRQ, 435
Megohm
bridge, 124
Sag
Modulation, signal generator, 277 Multimeter, 80, 92
431
(pulse),
284
Sample-and-hold
circuit,
420
Sampling oscilloscope, 255
Multiplexing, 330, 421
Saturated
Multiplier phototubes, 386 Multiplier resistor, 65
cell,
42
Schering bridge, 136 Scientific notation, 3
Secondary standard, 35
NDAC, NRFD,
434
Self-balancing potentiometer, 403
433
Sensitivity, 2
Nichrome V, 355 Normal distribution,
voltmeter, 42 12
Significant figures, 3
Ohmmeter, 74
Spatial encoder, 424 Spectrum analyzer, 306 Square-wave generator, 282, 285
Oscilloscope, 208
Stability
Ohm,
25
long-term, 340
PIN
diode, 268
short-term, 340
Passive transducer, 349
Stabiloy, 356
Phosphorescence, 220
Staircase-ramp
Photoconductive
cell,
388
DVM,
171
Standard deviation, 12
Pi attenuator, 263
Statampere, 24
Postdeflection acceleration, 217
Storage oscilloscope, 248
466
Index
Transducer (cont.)
Strain gage, 353
Successive-approximation
A/D
oscillation,
converter, 416
Successive-approximation
365
photoelectric, 366
DVM,
177
Sweep-frequency generator, 277 Systematic errors, 9
piezoelectric, 367
potentiometric, 368
Trigger level error, 341
True
RMS
voltmeter, 155
Taut-band suspension, 59
Temperature compensation (meters), 60 Thermistor, 377 Thermocouple, 374 Thermoinstruments, 94
Vacuum
Third-order intercept point, 313
Vertical deflection, 224
phototube, 384
Variable differential transducer, 362
Vector impedance meter, 195 Vector voltmeter, 199
Three-terminal resistance, 123
Time Time
base, 332
solar,
Wagner ground, 142 Watthour meter, 99
37
universal, 37
Transducer displacement, 359
Wattmeter, 97
Wave
analyzer, 296
Weston
cell,
42
THIRD EDITION
ELECTRONIC INSTRUMENTATION
AND MEASUREMENT TECHNIQUES WD COOPER A.D HELFRICK
This new Third Edition traces test and measurement systems from the simplest basic parameter measurement through the latest computercontrolled test system. Mathematical derivations are given for basic measurement to help you understand these measurements.
Among • •
• •
this
book's many special features:
improved material on digital measurements. expanded discussion of spectrum analysis. entire chapter on computer-controlled measurements. updated material on oscilloscopes and on recent advances
in
CRT
construction.
Of special interest: In-depth discussion of basic measurements systems; worked-out examples and review questions; broad range of measurement systems.
PRENTICE-HALL,
INC.,
Englewood
Cliffs, N.J.
07632
ISBN 0-13-250721-5