Electrical: Electronic
Electrical: Electronic
Electrical: Electronic
. . ·. .
. .
By
A. K. SAWHNEY
M.Sc. (E11gg.)
Professor of Electrical E11gineeri11g
Tbapar Engineering College,
PATIA LA
@All rights rese,frcd by, the author. No parts of this publication may be reproduced', stored in re;rieval .$ystem., ot·
transmitted, ibany.form or by a~y ~eans, elect~onic, mecha6ical, photocopying, recording, or o herwise, wi~h~~t
the prior wri~ten permissio.1 of the autho~.anJ the publisher. . , ·. ·.
"·
By tbe Same ~uthor
1. A Course in Electrical Machine Design.
2. A Course in Mechanical Measurements and Instrumentation.
Pl,lblished by : J.C. Kapur', fqr Dhanpat Rai & Sons, Delbi"l10006 (H.o. JuUundur) ..
ctmposed by : Karan P1ibtio1·Sr.cv'ice, Subhash Park, Shahdara, Delhi·l1Qo3f.
Printed at 1. D.R., Prilltloj Service, Panch.sheeJ Garden, ShahdAta, DeJhH 10032/
Preface to the· Fourtn 1:arc1on
. 'The author i~ pleased to brfogi out the Fourth Editiori of the book a.ncf is _fhankful t~· both·
teachers and students for their affecfi-0nate and warm receptio:n to the thlfd ed1t~n of the boo~
whJ9h has been sold out in a period of less than one year. The present edit.ion retai~s essentially
the same subjectmatter as the third edition. However, the typographical and (\ther .errors,
which had crept in earlier edition, have been corre.cted.
The book is intended as a stand~rd text for students studying for their first de8ree i•.
Electrical, Electronics and lhstrumentation Engineering at ~ndian Universities (}nd abroad, qd·
also for those appearing for A.M.l.E. section Band other profession~l examinations. The book
is equally useful for postgraduate students as well as practising engineers involved in ihe iel4A
of Measurements and Instrumentation. 1
· •
There have been significant changes in curriculum of almost all the upiversities in recent
years. Electrical and Electronic In,strumentation is now offered· as separate paper in maay
universitks. This has been necessitated on account of latest technological advances which put
greater emphasis and reliance on Electronic Instrumentation. . .
The contents of the book have been· drastically modified, .re-arrange~ an9 updated t9 ·
acquaint the reader of modern trends in the field: pf_ Measurements and Instrumeµtation. _ .
, The bobk has b.een divided into two parts. -i Part I d~aJs with Mea.surements .and Measuriq
Instruments and Part II takes care of the Instrmnentat1on. There 1s an extensive coverage 4f
Electrical and Electronic Instrumentation in this edition of the book as compared with the earlier
ones. The coverage of Instrumentation portion is about 500 pages · out of a total of nearly
1200 pages. The subject of ·Instrumentation has been developed in logical steps. Separate ahapteu
are devoted to topics like Generalized Measurement Systems, Transducers, Signal Conditioning, Data
Transmission and Telemetry, Display Devices and Recorders, Measurement of Non· electrical,
Quantities and Data Acquisition Systems. Also there are additional chapters on static and dynamic
characteristics of Measurement Systems. There is an increased emphasis on digital instrumftlnts and
instrumentation which is constant with the present trends. · .·
There are .three Appendices in the book. Appendix A deals with Number Systems,
Appendix B with Logic gates and Appendix C with conversions from various system~ of units to
S.I. The significant additions to the measurements· portion are Transformer ratio bridges (wliich
are fast replacing the conventional four arm a.c. bridges), spectrum analyzer, vector impedance
meter, vector meter, digital maximum demand indicator and Hall effect multiplier to name a few .
.The book ,in fact covers a very wide spectrum of the_Jield of Electrical and Electro1io
Measurements ?.nd Instrumentation and .is a conlplete reference in. itself.
Another outstanding feature of the book is the i~clusion Qf over 400 1solveQproblems which·
ill addition to linking the theory with actual applications gives an insight of the industrial practieo.
Also about 300 unsolved problems (with answers) have been included to give the students practiee
in solving problems.
. The author considers the inclusion of problems on Instrumentation (both· ''solved and uB.l
~~ved) as a speciality of this book. This is a pioneering effort which is the outcome of cuthor's,
experience of teaching the subject for almost two decades.
The book though voluminous, covers two papers, i.e., first on Electrical and Ele~tronie:
Measurements and Instruments and second on Electrical and Electronic Instrull_1entat\Qn .arid thus
fully justifies its volume. SI units have been uniformly used in .the book throughout. 1
·
The autho'r will feel highly obliged to·all the readers for their constructive suggestions aQd
healthy criticism of the book which will go a long way in the improvement of the text. - ,·
The author is thankful to his wife, Chander J for render,ing assistance in the compilation Qd
editing of the work. · · ·· . . ' ·
The author is grateful to his brothers, Ravish a~d Ajay, for their constant help durinatlae:
p~eparatiQn of tqe text. - ·· ·
I To. . . aut~or;s parents who have ~~en~ s~urce of encourascment and guidance.
PATIALA 26-9-1.984 ' ~. A..K. S.A.WQlf
-· -- _.,.,,.,, ••>"'
' '
<"
DEDICATED
TO
Anuradha and Puneet
4',
• ,I
.,
Chapt
'
,\
\,'
Contents
PART I
ELECTRICAL. AND ELECTRONIC MEASUREMENTS ANO
MEASURING INSTRUMENTS
Chapter 1-Measurements and Measuring Systems ll-8
Measurements. Significance of Measurements Methods of Measurement.
Direct Methods. Indirec,t Methods .. Instruments. Mechanical, Electrical and ,
Electronic Jnstruments. Mec4anical .Instruments. Electrical . Instruments. -
Electronic Instruments.. Classification of Instruments: Analog and· Digital.
Modes ofc Operation. i Functions of Instruments and Measurement Sy.stems.
Applications of Measurement Systems. Elements of a Generalized Measure-
ment Syste~. · · · '
Chapter 2 - Characteristics of Instruments and Measurement Systems 9-34
Measurement System J,>erformance. §tatic Calibration. Static Characteristics.
Errors in Measurements. True Value. Static Error. Static Correction.
Scale Range and Scale Span. Error Ca.libration Curve. Reproducibility
and Drift. RepeatabiJity. Noise. Accuracy and Precision. Indications
of Precision. Significant Figures. Range of Doubt or Possible Errors and
Doubtful Figures_. Static Sensitivity. Instrument Efficiency, ·Index·Scale
and Index Number. Linearity. Hysteresis. Threshold, Dead Time. Dead
Zone. Resolution or Discrimination. Loading , Effe~ts. Loading Effects
due to Shunt connected Instruments. Loading Effects due to s~ries connec·
ted Instruments. · Impedance Matching and Maximum Power Transfer.
Dynamic Response Measuiing Lag. Standard Signals, Overshoot. Unsol·
ved Problems.
Chapter 3-Errors in Measurements and their Statistical Analysis 35-'4
Limiting Errors (Guara'otee Errors). Relative (Fractional) Limiting ErrorL
Combination of Quantities with Limiting Errors · Known Errors.Types of
Errors. Gross Errors. Systematib Brrc,rs. In'strumental Errors. Environ·
mental Errors. Observational Errors. Random (Residual) Errors.. Central
,\/.'.
L~~r Value. Statistical , Treatment of Data. Histogram. Arithmetic Mean.
Measure of Dispersion from the Mean. Range., Deviation. Average Devi·
· ation. Standard Deviation (S.D.), Variance. Normal or Gaussian Curve
of Errors. Precision fodex. Probable Error. Average Deviation for the
Normal Curve. Standard Deviatian for the Normal Curve. Probable
Error of a Finite Number of Readings. Standard Deviation of Mean.
Standard Deviation of Standard Deviation. Specifying ODDS. · Specifying
Measurement Data. Variance and Standard Deviations of Combination of
Components. Probable Error of Combination of Components. lJ'ncer·
tainty Analysis and Treatment of Single Sample Data : Propagation of
Uncertainties. Unsolved Probems.
Chapter 4-Units, Systems, Dimensions and Standards 65-q 101
Introduction. Unit. Absolute Units. Fundamental and Derkre.4 · Plcltts•.
Dime11sions. D!mepsi9ns ·of Me~hanical. 9. u~. ntities: ~istoric~l :Q.. a.pk.f. &r?,µp~ .
of System of Units. COS System .of Umts: Electromagnetic U~it( (~~¥P·
.··~
./
(it)
for Mass and Length. Atomic Frequency and Time Standards. Temperature
Standards. Luminons Intensity Standards. Electrical Standards. Emf.·
Standards : Primary Standard of Emf. Secondary Standn,rd of Bmf LaQo- ·
ratory Standards·of Emf. Higher Voltage Zener Sources and Temperature
Effects... Primary Standards of Resistance.. Current Standards. Inf. ilctanoo
Standards. Capacitance Standards. . Unsolved Problems.
Chapter 5 -Circuit Components (Rolstors, Inductors and Capacitors) and their Residues
Residues. Resistors. Resistance Materials. Spools (Formers) for ·coils
Resistance Wires. Ageing. Annealing. Resistance Standards. Resistance
Standards for. D.C. Low Resistance Standards. Stan·dard Resistances for
A.C. Circuits Frequency Errors of Resistors. Methods of Reducing Residual
.Inductance. Resistance Boxes. Thin Film Resistors. Composition Resistors
Shielded Resistors. ·
INDUCTORS
Standards of Inductance. Formers for Inductance Coils. Coils of Induc·
tance Coils. Standards of Mutual lnouctance. Standards of Self-Inductance.
Vax:iable foduct1nces. Inductors for High Frequency Work. Inductors for
Low Frequency Work. Frequency Errors in Inductors.
CAPACITORS
Loss Angle and Power. Factor. Distributed Capacitance. Capacitance
Standards. Vacuum and Gas-filled Capacitors. Solid Dielectric Capacitors
Variable Capacitors. Decade Capacitance Boxes. Frequency Errors in
Capacitors. Unsolved Problems.
Chapter 6-Analog (pointer) Iostrnments
Analog Instruments. Classification of Analog Instruments. Principles of
Operation.
ELECTROMECHANICAL INDICATING
.
INSTRUMENTS
'
t .
'i ··~
(iv)
its Demagnetization. Current Transformers for High Frequencies. Potential
Transformers. Difference between C. T. and P. T. Theory. Errors. R,educ·
tion of Errors. Construction of Potential Transformers. High Voltage
Potential Transformers. Capacitive Potential Transformers Characteristics
of Potential Transformers. Testing of Instrument Transformers. Current
Transformer Testing. Potential Transformer.Testing. Unsolved Problems. ..1.
'
Methods of Measurement of M.edium Resistances. Ammeter Voltmeter
Method. Substitution Method. Wheatstone· Bridge. Sensitivity of · Wheat-
stone Bridge. Galvanometer Current. Precisfon Measurement. of Medium
Resistances with Wheatston~ Bridge. Carey-Foster Slide-wire Bridge. Kelvin
Varley Slide. Limitations of Wheatstone Bridge.
MEASUREMENT OF LOW RESISTANCES
Introduction. Construction of Low Resistances. Methods for Measurement
of Low Resistance. Ammeter Vol1meter Method ..~Kelvin Double Btidge.
Kelvin Bridge Ohmmeter. · Unbalanced Kelvin Bridge.
MEASUREMENT OF HIGH RESISTANCES
Introduction. Difficulties· in Measurement of fligh Resistances~ Use of
Guard.Circuit. Methods for
Measurement of High Resistance. Direct Deftec·
tion M,ethod. Loss of Charge Method. · Megohm Bridge Method. Measure-
ment of Insulation Resistance with Power On. Unsolved Problems.
~b1»1er 14-:-Potentlometers
. D.C. POTENfIOMETERS
Introduction. Basic Potentiometer Circuit. Laborhtory Type (Cro~pton'$)
Potebt~ometer. Multiple·Range Potentiometer. Cor.structional Details of
Potentiometers. Precision Type Potentiometers. Vernier Potentiometer.
Standard Cell Dial. True Zero. Brooks Deffectional Potentiometer. Volt-
Ratio Box.. Application of D.C. Potentiometers: Protection of Standard
Cell and Galvanometer. Self-Balancing Potentiometers.
A.C. POTENTIOMETERS
Introduction. Standardidng of A C. Pote11tiometers · and Use t>f Transfer
Instruments. Types of A C. Potentiometers. Drysdale Polar Po(.Cniiometu.
Gall-Tinsley (co-ordinate type) A.C. Potentiometer. Quadrature Adjustments,
of Currents. Campbell· Larsen Potentiometer. Applications of A.C. Poten-
tiometers Unsolved Problems,
Chapter 15-Power System Measurements . 481-495
. MEASUREMENT OF EARTH RESJSTANC~
Necessity" of Earth. Biectrode. Necessity of Measurement of Resistante of
Earth Electrode. Factors Affecting Earth. Resistance. Methods of ~asuring.
Earth Resistance. ,
LOCALIZATION OF CABLE FAULTS
Types of Faults Methods Used for Localizing Faults Methods· Used., for
Localizing Ground and sho1t Circuit Faults. Murray L0op Test. Varfoy
J.oop Test. Location of Open Circuit Faults in Cables.
SYMMETRIC{\L COMPONENTS ANO- THErR MBASUREMENT
"'
Introduction to Symmctncal Comp'onen •.>. Ca;vul~tion of Symmetrical
_Conipon~nts M:easureinent of symmetrical tompon~nts; Measuremeat of
r·.-..,
•,,
(vi)
Positive Sequence Component of Current. Measurement of Negative Sequ·
ie!iJlce Component of Current. · Measurement of Zero Sriquence Component
of Current. Measurement of Positive·Sequence Cemponent of Voltag@.
Measnrement of Negative Sequence Components of Voltago. Measurement
of Zero Sequence Component of Voltage. Vector Meter. Unsolved
Problems. .. (~
~pteff 11©-A.C. Bd~gts 4~539
Introduction. Si1urces and Detectors. General Equation for Bridge Balaooo.
GcnernJ form of an A.C. Bridge.
MEASUREMENT OF SELF INDUCTANCE
Maxwell's Inductance Bridge. Maxwell's Inductance-:-Capacitance Bridse•.
Hay's Bridge. Anderson's Bridge. Owen's Bridge. Measurements of ·
Incremental Inductance.
MEASUREMENT OF CAPACITANCE
De Sauty'-s Bridge; Sobering Bridge. High Voltage Sobering Bridge.
Measurement of Relative Permittivity with Sobering Bridge.
MEASUREMENT' OF MUTUAL INDUCTANCE
Uses of Mutual Inductance in Bridge Circuits. Mutual Inductance Measured
as Self· Inductance. Heavisjde Mutual Inductance Bridge. Campbell's
Modification of Heaviside Bridge, Heaviside Campbell Equal Ratio Bridge.
Carey Foster Bridge ; Heydweiller Bridge. . Campbell's Bridge.
MEASUREMENT OF FREQUENCY
Wien's Bridge. Universal Impedaace Bridge. Som·ces of Errors in Bridge
Circuits. P:recalations and Techniques used for Reducing Errors. Stray
Electromagnetic and Electrostatic Couplings. SLiclding of Bridge Elements.
Wagner Earthing Device. Sereeried·and-balanced Transformers. Difference
and Substitution Methods. Transformer Ratio Bridges Applications and
Features of Ratio Transformers. Measurement of Resistance. Measurement
of Capaciumce. Measurement of Phase Angle. Transformer Dou.blo hQio
Bridge§. Measurement of Inductance. Measu.rements of Components in
'SITU,. Unsolved Problems
Chapter 17-Hiah Voltage Measurements and Testing
Types of Tests. Testing Apparatus. High Voltage Transformers. Voltago
Control.Voltage Control by Variation of Alternator Field Current. Voltage
Control ·by Resistance Potential Divider. Reactance Voltage . Control.
Voltage Control by Tapped Transformer. Voltage Control by Variac.
Voltage Control by Induction Regulators. Control Gear and· Protective
Devices. Equipment for Voltage Measurement. Measurement of R.M.S.
Values of Voltage; Measurement of Peak Values of Voltages. Measurement
of D.C. Voltages. Measure11u~nt of Irrntantaneomi Voltage. Low Frequency
H.V. Tests. High Voltage D.C. 'Testing. Cockcroft Waiton Cfrcuit High I
Voltage D.C. Testing of Cables. Equivalence of D.C. and A.C. Test Voltages.
Localization of Faults in High Voltage Cables. High F.-equency Tests.
,S\1rge (Impluse) Testing, Basic Impluse Generator Circuit. Sinale Stage
}
Impluac Generators. Multi·stage Impulse Generator. Impulse~: Testing of
Transformers. Testing of Insulating Materials. High Voltage Testing of
Cables. High Voltage Tests o~ Porcclaiti' ,fo~ulaf<irs, .Testri~;of Eloo,trle
'· 'trengtb of Insulating Oils. .· '
(vii)
Chapter 18-:"'Msgnetic Measuremenlt 570-609
Introduction. Types of Tests.
BALL!SfIC TESTS
Measurement of Flux Density. Measurement of Value of Magnetising
·Force (H). · Magnetic :Potentiometer. Testing of Ring Specimens.· Deteru
mination of BwH Curve. Determination of Hysteresis Loop. Testing of Bar
Specimens. Permeameters. Hopkinson Permeameter (Bar and Yoke method).·
Ewing Double Bar Permeameter. lUiovici Permeameter. Burrows Perrneap
. meter. Fahy's Simplex Permeameter. The National Physical Laboratory
(Engiand) form of Permeameter. Measurement of Leakage Factor with
Flux Meter · -
ALTERNATING CURRENT MAGNETIC TESTIN'G
I Iron Loss Curves. Sepamtion of· Iron Losses. Methods of Iron Loss
'II
Measurement. Wattmeter Method. Bridge Methods. A.C. Potentiometer
Method. Oscillographic· Method. Methods of Measurements of Air Gap
Flux. Testing of Permanent Magnets. Magnetostriction. Methods used in
Modern Instrumentation Systems. Unsolved Problems.
Chapter 19-lllomination
Physks of Light. Light. f lectroniagnetic Spechuw. International Lumi-
nosity Curve. .Photon. Energy and Power. Intensity. Effect of Temperaw
ture on Radiation. Full Radiator or Black Body. Primary Standard.
Calibratio,a of Standard Lamps with Primary Standard. Secondary Standards ·
and Substandards. Laboratory Standards. Terms, Definitions and Symb'.>ls.
Laws of Illuminance (Illumination). Visual Photometry. Measurements of
Luminous Intensity. Photometer Head. Distribution of Luminous lntensity.
Determination of Mean Horizontal Luminous Jntensity and Polar Curve.
Determination of Mean Sp~erica! Luminous Intensity. Calculation with a
Distributed Source. Physical Phofometry. Photo·emissive Cells. _Photo-
multipliers. Photoconductive Transducers. Photoconductive Celis. Photo
diodes. Phototransiswrs Noise in Photoconductor Transducers. · Photo ..
voltaic Cell. Photocells in Physical Photometry. Measurement of Luminance
Unsolved Problems.
\ '
Chapter 2~Electrooic Instrlliments 639-6S
Electronic Voltmeters. Advantages of Electronic Voltmeters. Vacuum Tube.
Voltmeters (VTVMs). Average reading Diode Vacuum Tube V()ltmeters.
Peak reading Diode Vacuum Tube Voltmeters. Triode Vacuum Tube
Voltmeters. Balanced Bridge Triode Vacuum Tube Voltmeter. EJectrometer
Type VTVM. Transistor Voltmeters (TVMs). TVM ·with Cascaded Tran-
sistors. TVMs Using PET as Input Stage. Chopper Amplifier Type Voltmeter.
Balanced Bridge TVM. Differential Voltmeter. AC. Voltage Measurements.
J
R.M.S. Reading Voltmeter. Current Measurements Using Electronic
Instruments. Instruments for Measurement of Power at Higher Frequencies.
Chapter 21-Catbode Ray Oscilloscope (CRO) 'S5....;67i
lntroductjon. Cathode Ray Tube , (CRT). Electmn · .Gun.. Blectro~tatic ·
Focusing. Electrostatic Dcftection .. · .Effect <?f Beam Transit Time and Fr"°
quericy Limitations. Deflection Pl.ates. Sere.ens for CRTs . CRT Graticule.
·rime Base Generators. ·Basic CRO Circyits. O~servadon ·of Waveform cm
CRO. Measurement of Voltages and Cqrrents~ MeHuremmw oC P~-
(viii)
and Frequ.ency JLiss11j9?s:J>_~t~_~n_~. Spe~ial Purpose Oscilloscopes. Multiple.
Beam Osc1lloscopes. Multiple Trace Osc1Jloscopes. Sampling Oscilloscopes
Im~uise Waveform Os7illoscopes. Scanning Oscilloscopes. Storage Type
Osc1lloscope. Accessories of Cathode Ray Oscilloscopes. Calibrators.
Probes. Cameras. Electronic Switch. Unsolved Problems.
hspter 22-Irmstrnmeimts for Generation and Analysis of Waveform 677-697
OSCILLATORS
Introduction. Classification of Oscillato~s. Feedback Oscillators. Types
of Feedback Oscillators. RC Oscillators. Wien Bridge Oscillator. Practical
Circuit of l Wien Bridge Oscillator Phase Shift Oscillator. L.C. Oscillators ·
Armstrong Oscillator. Hartley Oscillator. Colpitts Oscillator. Crystal
Oscillators. Beat Frequency Oscillator (B.F.O). Negative. Resistance
Oscillators.
SIGNAL GENERATORS
Standard Signal Gern~!ator; ··~-:- i
'f 1
l I
MEA$l[REMENT OF PRESSURE
Intro~uction; Types ?f Pres~ure Measurement Devicl!s. Measurement of
Pressure Usmg Electncal Transducers as Secondary Transducers. Force
Summing Devices. Secondary TrlJ.nsducers.
MEASUREMENT OF LOW PRESSURE
(V~cuum Measurements)
MEASUREMENT OF VIBRATIONS
Necessity for Meas~rement of Vibrations. Nature of Vibrations. Quaptities
involved in Vibratio'n Measurements. Seismic Transducer. Types of AcceJe..
rometers. Potentiometric Type Accelerometer. LVDT Accelerorooters.
Piezo-electric Accelerometers.
MEASUREMENT OF TEMPERATURE .
Temperature Measuring Devices. Electrical Resistance Thermometer. Plati·
num Resistance Thermometer. Measurement. of Resistance of Thermometers.
Salient Features of Resistance Wire Thermometers. Semi-conductor Thermo-
meters.
Thermistors. Characteristics of Thermistors. Measurement of Temperature with
Thermistors. Salient Features of Thermistors. Thermocouples. Thermocouple
Constr~tion. Measurement of Thermocouple .Output. Compensating
Circuits.,' Reference junction Compensation. Lead Compensation. Advan-
tages and Disadvantages of Thermocouples. Quartz Crystal Thermometer.
Radiation Pyrometers. Principles Used for Radiation Temperature Measur-
ing Devices. Radiation Receiving Elements~ Total Radiation Pyrometers
Infrared Pyrometer. Optical Pyrometers.
MEASUREMENT OF FLOW
Rate of Flow. Turbine Meters. Electromagnetic Flow Meteu. Hot Wire
Anemometers. FJow Meter Using Thermistors.. Electrical Methods : Rrsis·
tive ti.lethod. Inductive Methods. Capacitive·· Methods. Measurement of
Liquiti' Level with Gamma Rays. Ultrasonic' M~thods. M~asuremeQt of
(xiii)
NUMBER SYSTEMS
Number Systems. Decimal Number System. A General Number System.
Binary Number System. Binary to Decimal Conversion. Decimal to Bfoary
Conversion, Arithmetic Processes in a ninary System. : Binary Addition.
Binary Subtraction.. Multiplication of Binary Numbers. Division of Binary ·
Numbers. Binary Coded Decimal Sys(ems (BCD). The 8-4-2-1 Code.
Other BCD Codes Gray Code. Conversion from Binary to Gray Code.
Gray Code to Binary Conve1·sion. .
1010-1073
LOGlC GATES
Introduotion. Gate. OR Gate. Truth Table. Logic Conversion. AND
Gate. NOT Gate. NOR Gate. NAND Gate.
'I
'
11
,i I
'-~.J
I
PART I
ELECTRICAL AND ELECTRONIC MEASUREMENTS
A'ND.
MEASURING INSTRUMENTS
.,J _..)'
1
-~·
,l
Measurements and Mea~u~ement Systems
1'1. Measurements
. The .measurement of a given quantity is essentially an act or the result of comparison between
the quantity (whose magnitude is unknown) and a predefined standard. Since two quantities are
compared the result. is expressed in numerical values. .. .
In order that the results of the measurement are meaningful, there are two basic requirements :
(i) Th.e standard used for comparison purposes must be accurately defined and should be
commonly acc_epted, ·
\ .-. and({;) The apparatus used and the method adopted must be provable.
1·2. · Significance of Measurements
The advancement of Science and Technology is dependent upon a parallel progress in
measurement techniques. It can be safely said that the quickest way to assess ·a nation's progress in
Science and Technology is to examine the type of measurements that are being made and the way in
which the data is acquired by measurements and is processed.
The reasons for this are abvious. As Science and Technology move ahead, new phenom~na
and rela~ionships are discovered an~ these.~dvances make new types of measurements imperative. New
discov~rtes are not of any practical utihty unless they are backed by actual measurements. The
measurements, no doubt, confirm the validity Of·a hypothesis but also add to its understanding. This
results in an unending chain which leads to new discoveries that require more, new ·and sophisticated
measurement techniques. Hence modern Science and Technology are associated with sophisticated
methods of measurement while elementary Scieuce and Technology require only ordinary methods
of measurement.
"
There are two major functions of all branches of engineering :
vJ Design of equipment and processes,
1·3·2~ Imliiled Methods. Measurement by direct methods are not always possible, feasible
and pracdcable. These methods in most of the cases, are inaccurate because they involve human
factors. They are also less sen~itive. Hence direct methods are not preferred and are rarely used.
Jn engineering applications Measurement Systems are used. These measurement systems use
indirect methods for measurement purposes.
A meaemrement system consists . of a transducing element which converts the quantity to be
measured in an analogous form.\ . The analogous signal is then processed by some intermediate means
and is. then fed to the end devices\which present the results of the measurement. ·
1·4. Instromeimt
Measurement generally involves using an in.strume~t as a physical means of determining a
quantity or variable. The. instrument serves as an· ex~ension of human faculties and enables the · man
to determine the value of an unknown quantity which his unaided human faculties cannot measure.·
An instrument may be de.fined as a device for determining the· value or magnitude of a
quantity or variable.
1·s. Mecbanicnl, Electrica' and Electronic Instruments t - -
The first instruments were mechanical in nature and the principles on which these instruments
worked are even in vogue today. The earliest scientific instrumpnts used the same three essential
elements as our modern instruments do. These elem~nt~ are,: .
(i) a detector, (ii) an in.termediate transfer device, and (iii) an indicator, recorder or a storage
device.
The history of development of in~truments ~ncompasses three phases of instruments, viz. :
{l) mechanical instrume~ts, (ii) electrical instrum~nts ~I;ld (i(i) tilCQtronic instruments.
1'5'1. M~niml IMtnmeutl. ·These instruments are very reliable for static and stable
conditions. But they suffer from a very major disadvantage. · They. are unable to respond rapidly to
measurements of dynamic and transient conditions. This.fS-<due to the fact that th~se instruments
have moving parts that are rigid, heavy and bulky and oonseqmmtly have a large mass. Mass presents
inertia pr-0bmll)s aod hence. these instruments cannot faithfuUy foUow the rapid changes which are
involved in dynamio 1measurements. Thus it would be virtually impossible to measure a 50 Hz voltage
by a mechanica] method, but it is r~latively .e~_sy to, meapu~,e a. ~lqwJy v.arying pressure. Anotqer
disadvantage of mechanical insttuments i! ehat 'most of them are ~ potential.sour~e of poise and cause
pollution of silence.
rs~2. Electrleill Instrumetm. B~ctrfoal nttthods. of indicating the output of detectors are
more rapid 'than m(iebanioal methods. It ii'. 1unfortunatt that~- eleetrioal system normally depends
upon a mechanicru meter movement.as indicating. device.· This: mechanical movement has som~ inertia
and therefore these instruments have a limited time (and hence, frequency): response. For.· example,
some electrical recorders can give full scale response in 0·2 s, the majo1'itY of industrial reoorde,rs have
responses of 0·5 to 24 s. Some g~lvan,otn~ters ~aq follow 50 Hz variationst bu.t ~en these are too
slow for present day requirements of fast meastu:~ment. .
1·5·3. ElectroDic lnstrnmentll.. .These -days m0.$t of tie scientific and ind\lstrial measurements
require very. fast· responses. The ·mechanical and ~l~ctr1:~ i.nstr~ts and systems cannot copo up
with these requirements. The necessity to step UP, response;! · ~p~ ~nd also the detection of dynamic.
changes in certain parameters,. which require ..the . m.onitorin2 tim~ of tbe order of ms and many a
times, i.t8" ha.ve led to the desisn of today's el~~.troniQ in,~ruttl.~ts and their .~ssociated ~ircuitey . These
1
instrutruints ~equire vacuum tubes or s"mi,cmnducto.r .devices.. R,ecent l>~ctice is to use· se.mi~ondu~tor
devices as they have _many ad'1antages over th~ir vaC\UJm tub~ ,cc;~nterparts.. $in~.in elyctronic
~~vices, the only mov~nt~I)t iQvoJV'ed is that of "f~tion~ -'~h~ ·se~p9n~.~ tjJAe i$ e'tr~e~y small o~
MEt\SUREMENfS AND MEASUREMENT SYSTEMS
account of very small inertia of electrons. P\>r example, a C.R.O. is capable ·of following dynamic
and transient changes of the order of' a few ns (l0- 9 s).
Electronically controlled power supplies are used to provide stable voltages for studies in the
field of chemical reactions and nuclear instrumentation. Electronic instruments are steadily becoming
more reliable on acco.unt of improvements in design and manufacturing processes of semi-conductor
devfoes. Another a~va'ntage of using electronic devices is that very weak signals can be detected by
using pre-amplifiers and amplifiers. The foremost importance of the electronic instruments is the
power amplification provided by the electronic amplifiers. Additional power may be fed into the system
to provide an increased power output beyond that of the input. This. has been only possible through
the use of electronic amplifiers, which have no important mechanical counterpart This is particularly
important where the data presentation devices use stylus type recorders, galvanometers, cathode ray
oscilloscopes and magnetic tape recorJers.
It is a fact that hydraulic and pneumatic systems may be used for power amplification of
signals. However, their use is limited tJ slow acting control applications like servo-systems, chemical
processes and power systems. Electronic instruments find extensive use in detection of electro.
magnetically produced signals such as radio, video, and microwave. Electrical and electronic
instruments are particularly useful in the intermediate signal modifying stage. Electronic instruments
are light compact and have a high degree of reliability. Their power consumption is very low.
Communications is a field which is entirely dependent upon the electronic instruments and
associated apparatus. Space commuriications, especially, makes use of air borne transmitters and
receivers and job of interpreting the signals is left entirely to the electronic instruments. .
Electronic instruments make it possible to build analog and digital eomputers without which
the modern developments in science and technology are virtually impossible. Computers require a
very fast time response and it is only possible with use of electronic instruments.
l '6, Classification of Instruments
There are many ways in which instruments can be classified. Broadly, instruments are classified
into two categories : -
(1) Absolute Instruments, and (2) Secondary Instruments.
1. Absolute Instruments. These instruments give the. magnitude of .the quantity under
measurements in terms of physical constants of the instrument. The examples of this class of
instruments are Tangent Galvanometer and Rayleigh's current balance. .
2. Secondary Instruments. These instruments are SO' constructed that the quantity being
measured can only be measured by observing the output indicated by the instrument. These instruments
are calibrated by comparison with an absolute instrument or another secondary instrument which has
already been calibrated against an absolute instrument.
Working with absolute instruments for routine work is time consuming since every time a
measurement is made, it takes a lot of time to compute the magnitude of the quantity under measure-
ment. Therefore secondary instruments are most co~monly used. Absolute instruments are seldom
used except in standards institutions while secondary instruments find usage almost in every sphere of
measurement. A voltmeter, a glass thermometer and a pressure gauge are typicalexample.s of secondary
instruments.
1·1. Analog an~.Digital Modes of Operation. Secondary instruments work in two modes :
(i) Analog . mode, and (ii) Digital mode.
Signals that vary in a continuous fashion and take on an infinity of values in any given range
are called analog signals ..The devices which produce ihese signals· are called analog. 4~vi~el!I.
In contrast, "the signals which vary in discrete steps and th1i~ 1~ke up ordy ~oite different
vaiues in a given range are called digital signals. The devfoes the.t producQ such ·si$nals are called·
di~ital devices. · ·
,, I
4 ELECJRICAb MBASURBMBNTS AND MEASURING, lN5TRUMBNTf
Let us elabo.rate further on Digital and Analog instruments and systems. In an analogue
system the function varies continuously. A typical exam~le. of 10
I.
variation is shown in Fig .. l' 1. On the other hand the d1g1tal .
values are discrete and vary in equal steps. Each digital 9
number is a fixed sum of equal steps which is defined by the
number.
t.
8
-- -- -- -- [/"' ~
quantity into a digital ~
-7 "0.
' In ord'1. to convert an analog 1 -- ---
number, the vertical displacements must be divided into equal CJ 6
parts. For exam pl~ in Fig. 1·1, the vertical quantities are ·~ -- >--; ~
divided into l 0 equal parts and each part has a length of 1 > S ~.,,
unit. When dealing with digital numbers, a quantity between 'E -
o to o·s is O while a quantity between 0·5 to 1'5 is 1 and a ~ 4
quantity between 1·5 to 2'5 is 2. For example a point A on ; 3 -
the analogue curve is 5·5 from the origin but in digital system
it would be read as 5. From A to B is 6 and from B to C is · 2
!' .•
7. It apparentl~ seems t~at if we ado~t digit~l ~ystem, the 1
errors involved will be considerable. But tf we d1v1de each of I
the 10 steps into 10 equal parts, we get 100 steps instead of 10.
And if these 100 steps are fUrther divided into 10 parts each, 'f 2 3 4 5 6 7 s 9 10
we will hav~ 1000 steps. This gives much better accuracy in lnd~penderrt variable _.,...
converting ~~a~ogue quantities ·into digit.al numbe!s· We can Fig. l '1. Representation of {ln
go on subd1v1ding further and furth~rt1ll ~he destred ~c?uracy . analogue quantity. .
is achieved. But it should be kept m mmd that a d1g1tal number 1s still a sum of equal umts.
In a digital system, magnitudes lying within one of these steps lose their identity and are all
defined by the same number. For example, if we have ten steps, numbers lying between 2'5 to 3·5 i.e.,
2·6, 2·7t 2·s, 2·9, 3·0, 3·1, 3·2, 3·3, 3'4, would ~Ube read as 3. ·
From the above discussion we conclude that the difference between analog and digital
information is .that the analog output is' a continuous function while the digital output is a discrete
number of units. The last digit of any digital number is rounded to ±0'5 of the last digit. It should
also be marked that the magnitude of the digital quantity is measured only at the instant the reading
is taken. One reading persists till another reading is taken (unlike the analogue quantity which
is a continuo\ls function).
The majority of present day instruments are analogue type. The impNtance of digital instru-
ments is increasing1 mainly because of the increasing use of digital computers in both data
reduction a~d automatic control systems. Since digital computer works only with digital signals, any
information supplied to it must be in digital form. The computer's output is also in digital form.
Thu,s working with a digital computer at either the input or the outp~t, we must use digital.signals.
However, most of our present day· measurement and control apparatus produces signals of
analog · nature, it is thus necessary to have b9th Analog to 'Digital (A/D) Converters at the input to
the computer and Digital to Analog (D//1) Converters at the output of the computer.
1·s. Functiom of IMtnme!lltS ud Measurement Systems
· . There is another way in which instruments or measurement systems may be classified. This
classification is based upon the functions they perform. The three main functions are explained
below: '
' 1. Indicating Function. Instruments and systems use different kinds of methods for supplying
information con~rning the variable quantity under measurement. M-0st of the time this information
r~---
is obtained as a deflection of a pointer of a measuring instrument. In this way the instrument per-
forms a function which is commonly known as indicating function. For example, the deflection of
pointer of a ~pe~dometer indicates the speed of the automobile at that moment. . A pressu~ gauge is
used for md1~t1ng pressure'. , ·
2. R~ording Function. In many cases the instrument·makes a written record, usually on paper,
of the value ofthe quantity under measurement· against time or ~gainst some other variable. Jhus the
MEASUREMENtS AND MEllSUREMBNT SYStEMS 5
instrument performs a recording function. F011 example, a potentiometric type of recorder used fo ~
monitoring temperature records the inst~ntaneous temperatures on a strip chart recorder.
3. Controlling Function. This is one of the most important functions especially in the field
of industrial control processes. In 1his case, the information is used by the instrument or the system
to control the original measured quantity.
Thus there are three main groups of instruments. The largest group has the indicating function.
Next in line is the group of instruments which have both indicating and or recording functions. The ·
1
last group falls into a special category and performs all the three functions, I.e., indicating, recording ·
and controlling.
In this text, main emphasis is laid upon instruments whose functions are mainly indicating
auu recording, especially those instruments which are used for engineering analysis purposes. The
control function will be analyzed in those cases where controlling enters as an integral part of the
indicating and recording functions of instrumentation.·
The examples of controlling instruments are thermostats for temperature control and floats
for liquid level control. '
1'9. Appli~ations of Me!lsurement Systems
In order to build u·p background for our later detailed study of measuring instruments and
systems and their characteristics, it is useful to discuss, in general, the various ways these instruments
are put in use. The way the instruments and measurement systems are used for different applications
are as under :
I. Monitoring of processes and operations. 2. c(~drol of processes and operations.
3. Experimental Engineering analysis.
1. Monitoring of Processes· and Operations. There are certain applications of measuring
instruments that have essentially a monitoring function. They simply indicate the value or condition
of parameter under study and their readings do not serve any control functions. For example, an
ammeter or a voltmeter indicates the value of current or voltage being monitored (measured) at a
particular instant. Similarly, ·water and electric energy meters installed in homes keep· track of
commodity used so that later on its cost may be computed to be realized from the user.
2. Control of Processel] Hd Operations. A very useful application of instruments is in
automatic control systems. There has been a very strong association between measurement and
control.
In order that process variables like temperature, pressure, humidity, etc. may be controlled,
the prerequisite is that they can be measured at the desired location in the individual plants. Same is
true of servo-systems, i.e., systems connected with measurement of position, velocity and acceleration.
A block diagram of a. simple control systept is shown in Fig. 1·2. Let us assume that the
output variable to be controlled is non-electrical a~d the control action is through electrical means,
I
Ou1put
Input Error signal Feedforward 1--~ Aduator 1---411--.,...--.....,~
·-+'- elements Conirolled
R~fefence (Amplified quant\tv'
(Oesirfld ou1put
Measuring
lnstrumen1s ~--~....---'
- or
Transducer
Fig. 1'2. Block diagram of a simple control system.
/.
6 ~LECTRICAL MEASUREMENTS AND MEA5URING INSTRUMENIS
The input is reference which corresponds to the desired value of the output. The input is com-
pared with the 'output with the help of a comparato'r. The output is a non-electrical quantity and
'is converted into a corresponding electrical form by a transducer connected in the feedback
loop. In case the input and output differ, there is a resultant error signal. This error signal is
amplified and then fed to an actutator, which produces power to drive the controlled circuitry.
The corrective action goes on till the outp.ut is at the same level as the input which corresponds
to the desired output. At this stage, there is no error signal and hence there is no input to the
actuator and the contrnl action stops.
Examples of this type of application are numerous .. A common one is the typical refrigera·
tion system which employes a thermostatic control. A temperature measuring device (often a
bimetallic element) senses the room temperature, thus providing the information necessary for proper
functioning of the control system.
3. Expedme1mtal EJl]gioecring Analysis. For solution of engineering problems, tfieoretical
and experimental methods are available. ·Many applications require application of both the methods.
The relative affacability of the method depends upon the nature of the problem. Experimental
engineering analysis has many uses and some are listed below :
1. Testing the validity of theoretical predictions.
2. Formulations of generalized. empirical relationships in cases where no proper theoretical
backing exists. . · , · ·
3. Determination of system parameters, variables artd performance indices.
4. For development in important spheres of study where there is ample scope of study.
5. Solutions of mathematical relationships with the help of analosies ..
1'10. Ekmemts of a Generalized Measurement System
It is important to have a systematic organization and analysis of measurement systems. An
instrument may be· defined as a device or a system which is designed to maintain a functional
relationship between prescribed properties of physical variables and must include ways and means
of communication to a human observer. The functional relationships remain valid only as long as
the static calibration of sy&tem remains constant. On the other hand, the performance of a
measurement system can be described in terms of static and dynamic characteristics.
It is possible and desirable to describe the operation of a measuring instrument or a system in
a generalized manner without resorting to intricate details of the physical aspects of a specific
instrument or a system. The whole operation can be described in terms of functional elements.
Most of the measurement systems contain three main functional elements. They are :
1. Primary Sensing Element, 2. Variable Conversion Element, and ,
3. . Data Presentation Element. '
Each functional element is made up of a distinct component or groups of components which
perfo,rm required and definite steps in the measurement. These may be taken as basic elements,
whose scope is determined by their functioning rather than their construction.
t. \Primary Sensing Element. The quantity under measurement makes its first contact with
the pri,mary sensing element of a measurement system. In other words the measurand is first detec·
ted by primary sensor. This act is then immediately followed by the conversion of measurand into
an analogous electrical signal. 1 his is done by a transducer. A transducer in general; is defined
as a device which cotnerts energy from one form to another: But in measurement systems, this
definition is limited in scope. A transducer is defined as a device which converts a physical quantity
into an electrical quantity. The physical quantity to be measured, in the first place is sensed
and dete.cted by an element which gives the output in a different analogous form This output
is then converted into an electrical signal by a transducer. This is true of most ;br the cases but
,is not.true for all. In many .cases the physical quantity is directly con. ver!ed . i9'fo an elect:r!cal·
quant1ty by a transducer. However, tbe first stage of a measurement system is knpwq as a ~~tedpr
transducer stage. ·. . · · / 1 • /
MBASUlU!MENTS ANl> MB}...SUREM~NT S\'STEMS 7
2. Varil!!bJe ·Conversio~ Elemtmt. The output of the primary sensing element may be any
kind of electrical signal. It may be a voltage, a frequency or some other electrical parameter. Some·
times tbiS" output is not suited to the system. For the instrument to perform the desired function,
it may be necessary to convert this output to some other suitable form while preserving the in-
form~tion content of the original signal. We may cite an example. Suppose output is in analogue
form and the next stage of the system accepts input signal only in digital form. Therefore we will
haye to use an A/D converter.
Many instruments do not need any variable conversion element, while others need more .
than one element.
. Variable Mmwpulation Element. The . function of this element is to manipulate the signal
presented to it preservinjZ the origiµal nature of the signal. Manipulation here means a change in
numerical value of the signal. For example, an electronic amplifier accepts a small voitage signal
Oat a
,_ ___.... prqsentatlon
elvlllflnt
.
Fig. 1'5, Schematic diagram of a bourdon tube pressure gauge.
When a control device is used for the final measurement stage, it is necessary. to apply ·some
feedback to the input signal to accomplish the control objectives. The .control stage compares the
signal representing the measured variable with a reference signal of the same form. This reference
signal has a value the measured signal should have and is presented by a con.troller. If the measured
signal agrees with the reference value, the controller does nothing. However, if there is a difference
between the measured value and the reference vnlue, an error signal is generated. Thus the controller
sends a signal to a device which acts to alter the value of the measured signal. Suppose the measured
variable is flow of a liquid, then the control device is a motorized valve placed in the flow system. In
case the measured flow .rate is too low than the preset flow rate, then the controller would cause the
valve to open, thereby increasing the flow rate. If on the other band, the flc;>w rate were too high, the
valves are closed. The operation of closing or opt..iing of valve will cease when the output flow rate
is equal to preset value of flow rate.
2-
Characteristics of Instruments and
Measurement Systems
Actually all working instruments, t.e., those instruments which are actually m~ed for measu~
ment work m:ust be calibrated against some refereace instruments which have a higher accuracy. Ttli '
reference instruments in turn must be calibrated against instrument of still higher grade of accurac
or against primary standard, or against other standards of known accuracy. It is essential that a:
measurement made·must ultimately be traceable to· the relevant primary standards.
2·3, Static Characteristics
, The main static characteristics discussed here are :
(i) Accuracy (ii) Sensitivity, (iii) Reproducibility
(iv) Drift (v) Static error, and (vi) Dead· Zone
'l
The qualities (i), (ii) and (iii) ai:e desifable; while qualities (/v), ( v) and (vi) are undesiral
· In addition to above characteristics, definitions of many other quantities have been given;
It must be stated; however, that sf.ere are many definitions 'of the above characteristics, anc
9
10 El13CTRICAL MBASU.t\BMBNTS AND .MEASti.lUNO iNSf.RUMENd
some cases the definitions are unrelated. Care has been taken to .select the most ge11erally accepted
definitions so as to avoid confusion.
2·4, Errnrsin Measurements
Measurements done in a laboratory or at some other place always involve errors. No ~.·
measurement is free from errorn. If the precision of the equipment is adequate, no matter what its
accuracy is, a discrepancy will always be observed between two measured results. · 1
In ordinary usage, the word err()r may have certain unpleasant connotations. It may imply I
a mistake, a moral offence, or possibly a belief in something untrue. In its extreme, it may be a \,
blunder. But errors are to be there in measurements and therefore there is nothing shameful· about
them as it should be understood that no measurement is free from errors.
Since errors are a must in any measurement, it is imperative to interpret. the results of a
quantitative measurement in an intelligent manner. An understanding and thorough,e'.rvaluation of
the errors is essential.
2·s. True Value
The true value of quantity to be measured may be defined as tb.e average ofan infinite
number of measured values when the average deviation due to the various contributing factors tends
to zero. ·Such an ideal situation is impossible to realise in:practice and hence it is not possible to
determine the ~'True ValueH of a quantity by experimental means. The rea;son for this is that the
positive deviations from the true \'.alue do not equal the negative deviations and hence do not cancel
each other.
· Thus, r.ormally an experimenter would never know that the value or quJntity being measured
by experimental means is the "True Value" of the quanti~ or not: ·
In fact in practice, tbe term, ''Tr!ro Value", then, refers to a value that would be obtained
if the quantity under consideration were measured by an "E111mpl1r Methed",. that is a method
agreed upon by experts as being sufficiently accurate for the purposes to which the data will ultimately
be put to use. ·
2'6, Static Eno.Ii ·
The most important characteristic of an instrument qr measurement system is its accuracy, l
which is the agreement of the instrument reading with the trµe value of quantity being measured.
Tbe accuracy of an instrument is measured in terms of its error.
We have mentioned earlier that it isimp-0ssible to tneasure the true value of a quantity. An
approximation of the "true va~ue" obtained by sufficiently extended series of measurements and also
taking into account parameters and conditions to which corrections may be applied, we obtain, what
is called the best meatmred v~h1e ·of the quantity. WbiJe it is·never possible to measure the true or
exact value of a quantity, it is nearly always possible to give· a best measured value. Static error is
defined as the difference between the measured :value and the· true. value of the 'quantity. Then :
. 8A. =Am~At ...(2• l)
where 8A = error,
Am =measme4 value of quantity,
and Ai=true value <:>f~quantity. ·
~A is also called the absolute static error of qua~tity A.. ·
We have Eo=8A ... (2•2) .
vhere Eo=absolute static error of quantity A (under measurement).
The ab~olute ·value of ~A- does not indicate precisely the . accuracy of measurements. As an
:xample, an error of +. 2 A is negligible when the current being·meas~d is of the order of 1000 A
vhile the same error of ±2 A may be regarded as intolerable when the current under measurement
i 10 A or so. Thus the quality of measurement is provided by the relative static error, i.e., the ratio
CHARACJiRElICS-OF INSTRUMENIS AND MEAStJRlMBNT SYSTEMS 11
of absolute static error 8A to the true value At of the quantity under measurement. Therefore, the
relative static error Er is given by :
E =' abs· lute error = SA = _io _
... (2'3)
r tr e value · At At
Percentage static error %E-r =Er X 1CO - .. (2'4)
We have At=Am-SA (see Eqn. 2'1)
=Am-Eo=Am-ErAt (see Eqns. 2·2, 2'3)
Am
= l+Er~ ... (2'5)
However, when the absolute static error Eo =8A is small, which means that the di1ltrence
between measured and true values is small.
fr~l-
:. Eqn. 2'5 may be written as Ai=Am(l-Er)
2·1. Static Correction
It is the differ~ncebetween the true value and the measured value of the quantity, or 1
This ind~cates that the noise has a · magnitude 'which is. about 3'75 times that of signal and
hence the signal will be completely lost in the noise.
2'13. Accuracy an,~ Precision
. In ordinary' usage, the distinction between words ~'Accuracy" 'and ''Precision'' is usually
very vague. In fact even the diction~ties invariably link the definition of one \\ ith the other. But
1
as far as field of measurements is concerned, there is a. big difference between the two terms as they r
have sharp differ,eilces in !meanings. In the fiel~,of measurements, the two terms may be defined as :
~ccuracy. It is the closeness with which· an instrument reading approaches the true value
of th~ quantity b~ing measured: Thus accuracy of a· measurement means confirmity to truth. ·.·
.. i>recjsion:. It is a measure of the reprodl1cibility of the measurements, i.e. given' a fixed
va.lue of a quantity,·precision is. a measure ·of·t~e degree of agreement within. a group of measure~
· merits:. The term 'Precis~': rµe~_ns cleaily or s_harplf defined. .As an example of the difference in
tneani:tig of the two terms, suppose.that we have an amtµefer which possesses high degree of precision
by virtue bf its clearly legible, finely· divide9, ·.distinct scale and a knife edge pointer with mirror
arrangements to remove parallax. Let us say that its, readings can be taken to 1/100 Of an ampere.
At the same time, its zero adjustment is w'rrihg. Noi,y every time we take a reading, the ammeter
i.s as precise as ever, we can take readings down to 1/100 of an: ampere, and the readings are consistent · ·~.,;
a1~1.;i."dear.ly defiried'' . . However, the.readings taken with thfa ammeter· are not accurate, since they
<h nofconfirm to turth on accouutof,its faulty- zero ,adjustment. ' . .
· . ·.Let us .. cite anot her exainpte. . Consider th~ measurement of a known voltage of 100. V
with a.meter. Fivereadingsare·takeji, andthe~indicated values ar~ 104, 103, 105, 103 and lOSV.
From the~e values it is seen·· that the instruhi.ent. cannot pe depended on for · an accuracy better .·
than 5% (5 V in thi~ .case), while a precision of'.± I%, is indicated sine~ the maximum deviation from ·
the ifl.ean reading of 104 V is only .1 ·o. V~ Thus we find that the instrument can be calibrated so
that. it could b.e used to read ± 1 V depen9ably.. This example illustrates that accuracy can, be
improved OP.on ~ut not the precision of the instrument by. calibration. Another point which is evident
from ab,ove is tba~ although th~ readj~g& are ·close. t.ogeWer thfY . have ~, small scatter (or dis~e:sion) ·
and thµs have a h1gh degree of pr~c~ston ,but ~he result~ are ·Jar from accurate. The precmon of
an instrument ~s usually dependentupon man,y factors and' requires many sophisticated techniques of
. 8,,n.alysis. , ·
Thus we say fhat .a. set of readings sho\vs precision if the results agree among themselves.
Agreement, however, 'is no guarantee, as there may be some systematic disturbing effect that causes
all the values to be in error. . ''
2'14 Indication~ of Precision
· J>recisivnis comp'bsed of two characteristics :
· {i) Conformity and (ii)· N\linher of significant figures.
Precision is used in measurements to d~scribe the consistency1 or the reprodudbility of results.
A quantity catted predsioo ir>dex defined in chapter 3 describes the spread, or dispersion of repeated .
result about some central value. High precision means a tight cluster of repeated results while low:·
precision indicates a broad scattering of r~sults. · But this may not lead us to the misconception
that high preci~ion indicates high degree of accuracy since all the repetitions in result may be biased
in the sarne way by some systematic effect that produces saCPe deviation of results from the true value.
For example, a sp'ring scale used with a spring designed for a different spring scale may repeatedly·
show the same value of weight. Thus the rea·Hngs may display exceptional agreement between them-
selv~s, but they all would be inaccurat~ ·values of weight since use of a wrong spring introduces a
systematic shift of all readings, . · · " . .
. 'we ·m~y well cite a~othe,r ¢~llmple to illµstr~t~ another aspect of precision.' ' ...:. " .'
Consider, for ex~01ple, th~f { re~i~t0r ~hose true resistance i~ 1,185,692 n, L'1:~'ein'g measured.
by an Ohmmeter. The obmmetercon·sistently and repe~t_edly indicates the true value. ·nut the obser"'
~er cannot read this value f~om t~e:~9ale: The observer~s estimates from the scale reading consistently
. . •. '.1.'· • . . ' .
CHARACIERISTICS OF INSTRUMENTS AND MEASURBMENTSYSTEMS 15
yield a value of 1·4 megohm (l '4 MO). This is as ·close to the true value as he can read the· scale by
estimation. Although there are no deviations from the observed value, the error creakd by the limi-
tation of the scale reading is a precision error.
The above example illustrates that conformity is necessary, but not sufficient condition for
precision because of lack of significant figures obtained. Similarly, precision is a necessary, but not
sufficient condition for accuracy. , · ·
•. In critical work, good pr~ctice requires an independent set of measurements, using differen~
'instruments or different measurement techniques which are not subject to the same systematic errors.
Where this is not possible, the experimenter must take steps that insure proper functioning of instrn·
ments and to discover and eliminate any systematic disturbing factor. Calibration against a known
standard may be resorted to in order to achieve the above l?urpose.
2'15. Signfficnnt Figures, .
· · An indicaticm of the precision of the measurement is obtained from the number of significant
figures in which it js·expressed. Significant figures convey actual· information regarding the magni·
tude and the measurement precision Qf a. quantity.· The more the significant figures, the greater the
precision of measurement. 1
' , ,. · , . .·
Let us take an example. If a voltage is specified as 256 V its value should be taken as closer to
256 V than to either 257 Wot 255 V. If the value .of voltage is described as 256'0 V it means that the
voltage is closer to 256'0 V than it is to 256' l V or 255'9 V. In 256 there are three significant figures
whHe in 25~'0 there are four. The latter, with more sig~ificant figures, expresses a measurement of
greater precision than the former.
Frequently iarge numbers with zeros before a decimal point are used for approximate popula~
tions or amounts of money. For example the population of a City is reported as 490,000. This
may due to misconception, imply that the true value lies between 489,999 and 490,001, which is six
significant figures. But in fact, what is meant, however, is that the population is closer to 4·,0,000
!ban to 480,000 or 500,000, Since the population can be reported only to two significant figures.
How else large numbers be expressed ? A more technically correct notation uses powers of ten like
49 x 104 or 4·9 x Ios. This indicates that the population figure is only accurate to two significant
figures. Thus. reference to populatiOil of a city as 3,000,COO would be interpreted automatically as
an approximate number while reference to velocity of light as 300,000,000 metre per second creates
·no confusion to a person with a technical background. Uncertainty caused by zeros to the left of
decimal point is therefore usually resolved 'by scientific notation using powers of ten.
Exan,Df)le 2'6. Stat'' the number Of significant figures in each of the followinj! numn?.r.Q ;
(a) 302 A(b) 302'10 V (c) 0'00030 !l (d) 0'0000300 Hl (e) 5·01x10 4 (/) sorno.
Solution : (a}. The number is 302. This means'that it is more close to 302 that to either 30 l or.
303~ Thus this number.has 3 significant figures. ' .
(b) The number inv9lved h .302' IO, TheFefore it is .more close to 302' 10 than either to 301 ·09
or to l02'I l. Thus it has 5 significant figures. · -. . ,
· (c) The. number is 0'00030. This means tb~t it is more close to o·ooo 30 than either to
0'00029 or 0'00031. Thus it has 5 sigoificaoe figures.,
(cl) The resistance is 0'00003 MO. . This can be written as 30 !l Thus it is m.ore close to
l{) thu to ~itbor 29 or 31. Therefore it bas 2 significant figures. The zeros to the left of 3 are due
Ci> large size of unit.
(e) The number under consideration is S'OJ x 104. Hence it.is more close to s·o 1 x 104 than to
either s·oo x 10' or 5·02x104. Thus it bas 3 significant figures.
( f) ·1;be number is 50100. This is· a larg: n'uml?er, and a situation may arise that causes uncer·
taint) Strictly ,S0,100 means thafthe.,iun>bet'is more close to 50,100. th!ln either to. 50,099, or .50; 101.
This means ihat the number has 5 $~gnificant figur~s. ij.ow~ver, if this number is po~mlation of a
town it could be written as 501x102• Under.this' situation we ·can say that the population is close to
x
~01 x J0 2 than to ~ither soox 102 Of $02 IQ'. This·mQ&ns that nu~ber bas 3 significant figures.
'
16 ELECTRICAL MERSUREMENTS AND MBASURINO INSTRUMBN'fS
. Examp1e·_2·1. .A set' of independent current measurements were recorded as 10'03, 10· 10~ 10· 11
and 10'08 A. Calculat~·(a) the· average current, and (b) the range of error.
Solution :
(a). Avera,i~ cµrrent Ia~ 1i+~2tla+h .=!0'0~+101_0~1Q.1l+to·os =l0.'08 A.
'··:', ,·'
·< ·.. '..>' . . •( . ., " . .
(b) ·:'Ma·?tiri:J~m value of current.Imaai= 10· 11 A
,.... " ,(·.•:· .... .. .. "· . '. .
· · ~ang~ ~i~aai;..:.lav=10'11~10'08=0'03'A~
. :·.'':,1 ~ ':• ..,. ·1·.• "' :i ' •, '
· · .MiniI.Jlum ·.value bf l~ngth /mt"~ 10·03 A.
. . ·lav.....:lmin~jo~os-10·03 .:::o·os A.
~· ' - .
I . ~" ;· ,: . . ·0.03+0'05 ·
· Therefor~ •v~rage range.of ,error is . · =±0'04 A.
2
The ·null)ber of significant figures in a quantity is one measure of precision, though not as
definite as a percentage statement. Suppose the range. of dou~t in a 101 0 resistor (the value has 3
signifioaqtitfigures) is t n. This means that the ·value lies between 100·5 Q and 101'5 n. This range
. of doubt i.n p~rcentage terms is 1 perceJ?.t. The same range of doubt of I n in a 999 0 .resistor (again
the value has·~ ·sigQ.ific1n.t figure's) creates a percentage range of doubt as only o· 1 per cent;
. . Therefore three significant .figure~ may cover a percentage range of doubt of o· 1 to 1 per cent
and hence is indefinite. as ameasure of precision.
-·~upefn:uous fi~ures are so~~~imes allowed ·to accumulate in ordinary arith~etic processes of
1
,. I
Rs= 0'612 Q,
R= 15'64)0.
' ' .
The result cannot be expressed as 65'642 .0 as even the figure in'theterith place i.e., 6 is in
doubt. Therefore the resultant resistance is 65'6 n witJ1~6 as first doubtful figure.
Exampl«J 2·9. The ~oltage and current recorded in a d.c. circuit are respectively 12'16 V and
1·34 A. calculate the power;
Solution: Power-12'16Xl'34 W.
Multiply by long hand, we have :
12'16
1'34
4864
3648
1216 ...;
16'2944
· The power is expressed as 16'2 Was 2 is the first. doubtful figure. It would obv1ousJy be
absurd to write the answer with the entire product obtained ..
·When two or more measurements with different ·degrees o(, accuracy are added, the result
is only as accurate as the least 'accurate measurement. ·This is illustrated by the following ex'\mple : ·
Example 2·10. Tw() resistors. Ri and R2 are con~ect~d in serie~ Y'ith 'R1=28'7 0 and
R2 = 3'624 n. Calculate the total resistance to the appropriate number 'Of s1gmficant figures.
Solution :
Ri=28'7 0 (three significant lfii'es)
R2=3'624 0 (four significantigures}
+
Total .resistance R=R1 R2
=132'324 !l (five significant ftgures) ·
=32'3 0 (three significant figures)
"
I
~,,.,I
This is done because one of the. resistances is accurate only to three significant figures (or
tenth of an ohm in this case) and therefore the' result should also be reduced to three significant figures
(or the nearest tenth of an ohm in this case) and hence the value 32'3 (l.
The number of significant figures in IQUltiplication may increase rapidly. but only the appro·
priate figures are retained in the answer as is illustrated by the following example. · · ·
' <
Example 2·11. In calculating voltage drop; a current of 4'37 Ais. recorded in a resistance
of 31 '27 n. Calculate the voltage drop across the resistor to the appropriate number of significant
figures. ·
Solution:
Current /=4'37 A (three significant figures)
Resistance R= 31 '27 n . I (four significant figures) .
Voltage-drop E=/R •..,4·31x31'27=l36'6499 V
(seven sisnificant fisures)
'.\ ~,+·
18 ELECflUCAL MEASUREMENTS AND MEASURING INSTRUMENTS
Since there are three significant figures involved in the multiplication, the answer can te
written orly to a maximum of three significant figures.
:. E~137 V.
t 17. Static Sensitivity
The static sensitivity of an inst! ument or an 'instrumentation system is the . ratio of the
magnitude of the output signal or response to the magnitude of input. signal or the quantity being
measured. Its units are millimetre per micro-ampere, counts per volt etc. depending upon the type 1
Sometimes the static seinsitivity is expressed as the ratio of the magnitude ~f the measured
quantity to the magnitude of the response. Thus the. sensitivi.ty expresse d this way_ has the units of
1
micro-ampere per millimetre or volt per count etc. as the case may be depending upon nature of input
and output. Thus it is reciprocal of the sensitivity as detined above. This ratio is defined as deflection
· :factor or invrrse sensitivity. Many manufacturus define the sensitivity of /their instruments in terms of
inverse sensitivity and still call it sensitivity. 1
When a calibration cui:ve is linear as in Fig. 2· 3 (a) the sensitivity. of the instrument can be
defined as in slope of the caHbratiou curve. For this case the sensitivity is constant over the
entire rnnge of ~he instr~me~~t.: However, if the :c.u~ve }s no~ nominall~/ a straight line the sensitivity
varies with tbe mput as m Fig.:2'3 (b). The sens1ttv1tv m this case vart~s ·
i. '
i
i
I ,•
Output
f
----A~z··
I
-·-·
t
Output
I
,/
.
' i :.o.q,o
I
lnput,q\- lnput,q.t -
I
(a) (b)
Fig. 2'3. Definition of sensitivit,v.
In general, the static sensitivity at the operating point is defined as. :
St tic sensitivit _ ~l}fip.~!~.s~m_al change ~n ~utput..,. 6qo ... (2' 11)
a · Y mfimtes1mal change 1ll mput b. q,
Similarly,
\.
CHAR.ACTERiSTICS OF INSfR.UMaNTS AND.MSASUREMBNr SYSfBMS . 19
..
Inverse ;. . '"} f: magnitude of input 1·00 .0
sens1hv1ty or sea e actor=;: · 't d f · t · t .
., · magm u e o ou pu · resp~nse 3'000 mm
= 2;33. O/mtl1. .
· Example 2'13. A mercury thennometer has a·capillary\ube of 0·25 mm diameter. If the
-.,,, bulb is made of a zero expansion matedal what volume must it have if a sensitivity of 2'5 mm/°C is
desired ? Assume that the operating temperature is 20°C and the c<,>-efficient of volumetric expansion
oflmercury is 0'181x10- 3/°C.
Solution : Let :
Lc=length of capillary tube which would be occupied by mercury contained in the bulb
when it is not heated ; mm,
Le+ 6 l.c=length of capillary tube which would be occ~pied by mercury contained in the bulb
when heated; mm,
Ae=area of capillary tube; mm 2,
~v=co-efficient of volumetric expansion ; J11m 3/mm3·°C,
and 6T=change in temperature; QC, · . .
It should be noted that there will be only a change in length of mercury column since it is
given that the bulb has a zero expansion. material and hence there will be no changes in its area and
length. This is true of capillary tube as well.
Sensitivity S· = ~:: ·=(Le+~)- Le ~f~ =2'5 mmrc.
Now Ac (Le+ l:,Lc)=Ac (Le+ iv Le/::.T) ·
· Length of capillary tub~
- l ·!::.Le _ _l -~ X2'5-13'8X103
L c- -13"8
~~ 6.T.= 0'181x10-s . . ~ mm- m.
TC ,
Hence area of bulb Ah=AcLc=4-(0'25)2 x 13'8X 103=680 mms.
that the resistance varies as shown by a curved dotted. line; in that ·case there is deviHticvo fron}
linearity. This deviation from linearity ·may be expressed by a term ''Percent Linearity".
Percent linearity may be defined as
. .· .. ·(maximum resistance deviation·) X r. •..' (215)
Percent 1mear1ty= ---- 10 v
. · full scale deviation
for this particular case. In general percent linearity is :
_:.(maximum displacemen~eviation) x 100 ... (2' 16)
-- fuli scale deviation
It is desirable to keep the percent linearity as small as possible as it would, in thttt ca1c, rcsuh
in small errors in ·the read out system. For example if a self-balancing potentiometer has a percent
.. linearity i>fO'l per cent, its accuracy would be 1 part in 1000, With a 1000 count digital encoder
connected to the shaft of the potentiometer, it would produce an error of 1 count in 1000 counts
. . ' : . . 'Example 216. A 10,000 .a variable resistance has a linearity of o· 1%and the movement of
contact arm is 320~. (a) Determine the maximum position deviation in degrees !ind the resistance de via-
~. tion'in ohm. (b) If this instrument is to be used as a pqtentiometer with a linear scale of 0 to l 6 V.
determine the maximum voltage error.
Solution: (a) From Eqn. 2'16,
percent linearity x fuH scale d~via~!~!l
Maximum displacement deviation
100
=O'I x320. 0.32•
• 100 I
. . · maximum
.. '. · d'
.1spIacemen t· = o· 1x 10,000
---w IO iJ
Srmtlarl~, resistance
0-:----· =
'(b) A qisplacement 320° corresf>onds to 1'6 V and therefore 0·32· corresponds to a
voltage of;
.. Q:32 x1·1 = l '6 x 10.:.3 v
'320 '
:. Maximum voltage·error=1'6X 10"'."3 V=l'6 mV.
2·21 .. Hysteresis. Hysteresis effects show up in any physical, chemical or electric01l phe1rnme-
non. Hysteresis is a\phenomenon which depicts different output effects when loading and unl~C1.<iin..,
whether-it is a mechanical system or an electrical system and for that matter any system. Hystere.\i~
is non-incedence o~.'1.oading'·and unloading curves, Hysteresis, in a system, arises due to th~ fact that
all the energy put info the stressed parts when. loading is not I recoverable upon unloadin~. This is
·because the secondlaw of thermodynamics rules out any perfectly reversible process in the world .
. ii·' . . . ' .
. ·Hysteresis effects are there in electrical phenomena. Orie of the examples .is the relationship
:Mtwun ou'tpu~ voltage and field current in a d.c; generator. This fa due to magnetic hystw~5'is. This
~, ctirYe is of the shape shown in Fig. 2·5 (a). · · ·
In mechanical parts of a system, there may be internal friction, external sliding friction and
··coulomb friction. There may be a (ree PlilY or loDseness in the mechanism. In a giycn instrument a
number of causes, such as listed above, combine to give an overall effect which may result in output~
-'input relationshipsuch as shown in Fig. 2·5 (b).
2·21: T~ahold. His c!e~r from above that if th.e instrument input is incr~~sed very gradually
from zero there wdl be- some m101mum value below which no output c~ange can be detected. This
22 EtrctRICAL MEASUREMENTS AND MEASURlNO INStROMtlNTS
tJutputf
OUfput f
'°"
Max.output
Hysteresls
...,
I
I.
(a) (b)
Fig. 2·5, Hysteresis effects.
minimum value defines the threshold of the instrument. In specifying tbres~old, the first deteclible
output change is often described as being any ''noticea~le measurable change ·'
2·23, Dead Time. Dead time is defined as the time required by a measurement. system to · ..... : i
begin to respond to a change in the rneasuraod. Fig. 2'6 sho•vs the measured quantity and its value .
as indicated by an instrument. Dead time, in fact,:,is the time before · 100 ..-----,--r-.----r----.
the instrument begins to respond after the measured quantity bas ' Dynamic
error
been changed.
2'24. Dead Zone. It is defined as the largest change of input
quantity for which there i5 no output of the instrument (Fig. 2'6).
For example the input applied to the instrument may not be
sufficient to overcome the friction and will, in that case, not move
at all. It will only move when the input is such that it produces a
driving force which can overcome friction forces. As stated in Art. ·
2·20 other factors which produce dead zone are backlash or bystere·
sis in the instrument.
The term "dead zone" is sometimes used interchangeably
with term hysteresis. Howe\'er, it may be defined as the total range Time t _
of input values possible for a given output and may thus be num·
ericalJy twice the hysteresis defined in Fig. 2'5 (b). Fig. 2'6, Dead time and dead zone.
Example 2·17. The dead zone in a certain pyrometer is 0'125 percent of span. The
calibration is 400°C to 1000°C. What temperature change might occur before it is detected.
Solution: Span=l000-400=600°C.
0'125 ,,.; 1
Solution : . . . = 200.= ·2 v
1 seaIe div1s10n
100
~
I
~"'";'
oltoga ~ o
ource -
_J
A
f Eo -r EL
A IL
I
9·
the same. However, tJ1e load impedance is 'riot infinite and therefore when a voltmeter' with an input
impedance ZL is connected across A an_d B, a current IL flows. This causes a voltage drop ILZo.
;'. Qutput voltage under- loaded conditfons is :
EL=~-lr.Zo=ILZL or Eo=IL(ZL+Zo)
.. . : . Ratio of actual voltage appearing across the load (when the instrument is connected) to
4 11
I
Thus the voltage which is measured is modified both in phase and magnitude. This means
that the original voltage signal is distorted on account of con~.ection of measuring instrument across it.
It is clear from Eqn. 2'18 that in order that the original signal Eo should remain undistorted
the value of input impedance of the instrument, ZL, should be infinite (or the value of output ........
l
impedance of the source, Zo should be equal to Iero whichi is not' attained in practice).
In order to obtain as less distortion as possib1e ~he value of Zr., the input impedance of
instrument, should be very high as compared with Zo, the outp..:.t impedance of the source.
· To illustrate the loading effects of shunt devjpes a few examples are given below. First we will
give examples connected with d.c. and then go ovefto a.c. applications.
Example 2·20. A multimeter having a sensitivity of ·2 ,000 O./V is used to measure the voltage
across a circuit having an out~ut resistance of 10 kO.. The open circuit voltage of the circuit is ~ V.
Find the reading of the multimeter when it is set to its 10 V scale. Find the percentage error. //
Solution: Input impedance of,voltnieter ZL=20,000x10 n .... 20 kfi
Output impedance of circuit Zo= 10 k!l
Open circuit voltage of circuit under measurement Eo=6 V.
From Eqn. 6' 18, reading .of voltmeter is
Eo 6
Er.= 14 ~fZL =I+ l0/20 =
4V
. -. .\
4 6
:. Percentage errol"'in \'.Oltage reading x100=-33% .or 33% low.
6
The loadh1g problem given .in Example 2·20 is typical of cases where a common type of
voltmeter such as a multimeter whose input impedance is.comparable with that of the impedance of
the circuit under test is used for measuring voltage. Thus voltmeters having output impedance
comparable to that of circuit under test should not be used as they seriously modify the value of test
. voltages. Such types of voltmeters are unsuitable for communication and electronic circuits where the
impedance levels are high but they can be'used for low impedance circuits giving a good accuracy.
If we wish to achieve 99% accuracy in voltage measurement, th~ input resistance of the
voltmeter should be greater 100 times the output re&istance. _ For an accuracy of 95% . the input
resistance should be atleast 20 times the output resistanc'e. . .
:Example 2'2l. Suppose the voltmeter of Example 2·20 is used for measurement of voltage of
circuit having an output impedance of I000 n and an open circuit voltage of 6 V at its 10 V scale.
find the error in measurement. ·
I .
J · io 50
EL=1+z--;1z~ = i +-ioo11000 = 45 5 v.
.
.
Loadmg ~rror=
45'5-50.=....,.. 9%.o = 9%o 1'ow.
50
Fig. 2'9 Accuracy= 100-% Joading error= 1.00-9 ==91 %.
Operation with A° C. The anaJysis of loading effects with a.c. is
not so easy. It should be .borne in mind that both Zo and ZL are
dependent upon frequency. .Therefore .the indicated volt~ge will depend upon . ~he frequency of
operation. On accou~t of the inp?t capacitance effects ~r the 1~struo;ient, the v~Iue of mput impedance
ZL becomes low at .high frequencies with the result the mput signal 1s substantially attenuated at· high
I
I ,
frequencies. ·. . ..,. .
It is not only the magriitupe of tlie signal that is effected)ut alsl) its· phase. Worst still; as is
consequence of the shunt capacitance, the.non-sinusoidal signals are distorted in waveform a]so.
· The rnagnit~de of th.e measured signal becomes-'substantiaJly smaller with increase in freqi1ency
i?<S shownin Fig. 2·10. .) .
EL
t
i _;t _,
""' ,;
..
'. (.
_." • .. 1
The sharply changing non-sinusoidal waveforms are rounded off because of the finite .time. it ·
takes to charge a capacitor. This is shown in Fig. 2·11. ,'
The effect of frequency on magnitude and phase shift of a signal is shown in the following
example. . '
Example 2'23. An oscilloscope having an input resistance of 1 MO shunted by 50 pF capacitance
is connected across a circuit having an effeGtive output resistance of 10 kO. If the open circuit voltage
has 1·o V peak for a 100 kHz sine wave, what will be the voltage indicated on the oscilloscope when
the frequency is (i) 100 kHz and (ii) 1 MHz ?
The equivalent circuit for the measurement system is Oscilloscope
shown in Fig. 2· 12. 1Qk!l
When frequency= 100 kHz :
The value of capacitivereactance at 100 kHz is
Volt age ""
1 1 source lMO. Opf
Xo= 2efC = 27t->aoo-x 1000-x·scfx 10-12 =32000 !~ to
The input impedan.ce of the oscilloscope is :
R(--jXo) _ 106 x(-j32xl03) ·. · Fig. 2·12
Zr,= ·R=fx~- - -106 j32x-10~- ~ -]32x10a .a
=32x10 3 L ....;.90°0.
, · . Eo
fhe voltage across the Joad is EL=-----:--L·-·--:
· . · 1+Zo/iL
=l'OL0°X'--7
1
10xl~3L0° -1+0·3~3L90° l+j~'313
+Tfx10 3 L-90"
· 0·954 L-17· 4° V (peak) .
.This means that the magnitude of voltage indicated by the ! oscilloscope-is 0·954 of its original
value. I · ·
:.' The error is (1-0'954)X 100=-=4'6 per cent. Also the voltage under loaded conditions
lags the voltage under open circuit conditions by an angle of 17'4°. · !
=l'OLO"X ---·--·- :1 - - =
. ·-----I -------··· -
IOX 1osL~o 0
1+3'13L90?
l+ 3·2x193L-90°
=0·304L-n·3° v (peak).·
In this case the measured value is only 0'304 of its original value and the phase shift 4s 72·3•,
Thus the output is considerably attenuated and is Jess than one third of its original value.
,.
,·
. ' ·.~,.. ·
..
'! '
i
CHARACfElUSTICS OF INS1RtJMENtS ANb MEASUREMENT
.
$YStbMS ~
21
, This indicates. the effect of distortion of signal on account of increased shunting effect due to
increase in frequency. · ·
2,'28. Loading Effects due to Series Connected Instruments
. . . Consider a network represented 'by a voltage source· having a voltage Eo and an output
r 1rupedance Zo. The output terminals are A and Bas shown in Fig. 2'13(a).
A
A.
lo
Ca) (b)
Fig. 2'13, Loading effect of ammeters.
{
. The value of current flowing between [terminals A and B under ideal ~onditions is 10• It is
the current·that flows wben_terminals A and Bare shorted. ·
Io=~
1
or Eo=IoZo.
Zo
However, when we· actually .measure the 'current, a· current measuring device has to be
· introduced between terminals A and ·B. It is ·usually an ammeter. When an ammeter is pJaced
between output terminalls, it adds to the impedance of cir~uit. This added impedance modifies the
value of the current. · · · .
Suppose ZL . input impedance of ammeter..
. \
From Fig. 2'13 (/;)
... (2'19)
In order that the measured value of .current, be equal to the actual value of the current, lo,
the value of Zo~ZL. This means that the input impedance .of the ammeter should be very small as
compared with the output impedance of the source. . ·
. We can express the relationships of currents under loaded and unloaded conditions in terms
of. admittances; · ·
From Eqn; 2·19;
... (2'20)
lo
:::o---"--- ... (2'20
I+Yo/YL
· . . In oth¢r wofd~ tl,ie input admittance of the series elements should be very large as compared
with the o~tptit, admittance 'of the source fa order to reduce ·loading effect.
' ·. . ' ,; ' - • . :· :, I • • ' i' ; ,. I
for achieving 99%. accuracy in measur.ements the output ·resista.nce should be at least 100
times the resistance of Che meter. Jn order to. have 95% accuracy the output resistance should be at
.]east 20 times the .resistance of metet
L
28 ELECTRICAL MEA~UREMENTS At'D MEAStJkJNG JNSTRUMBN't!i,
Ex·ampic 2'24. It is desired to measure
the value of current in the 500 n resistor as
Amnn~rer
shown in Fig. 2·14 by connecting a JOO n
---i
·ammeter. rlnd :
(a) the actual value of curi;e11t,
(b) me<J.sured value of current, and
(c) the percentage error in measurement
and the accuracy.
e
Fig. 2'14
1 . .7 1000..
Solution. (a) Let us reduce the actual circuit to an equivalent Thevenin's source.
Open circuit voltage as applying at terminals. A and B is :
. ' 10 .
Eo=IO- ·. - X 1000=5 V.
2000
Output imp~d ance of source as looking iflto terminals A and B is.:
' . .
1000x1000 . .
~0=1000-+-Tooo-+soo= moo !l
The Thevenin equivale.nt circuit is as shown in Fig. 2' 15.
Z0 ::'1000fi
~--0A
tl_ ..
·TEo'°sv . .. ·
·----00
fig;.2'15
Eo .:. ._; 5 · · · .
Actual value ofcurrent lo:~
.
-z; - -;- 1000 A= 5 mA.
. .
(b) When the amm~ter is introduced into the.circuit the y~Ju.e of current is. modified.
Example 2'26. A voltage source has an open circuit voltage of 20 V and has an output
impedance of 0·5 +j 1 0. The voltage souree is connected to the load t.hrough a transmission net·
work having an impedance of 1·s+j40. At what load will 11Daximum power transfer be.realized'/
Calculate the maximum power .. Also calculate the losses in the voltage source and the transmission
network. ·What is the efficiency under maximum ,power transfer conditions ?
Solution. The transmission network is connected in series with the voltage source,
.·. Output impedance of source and the transmission network is :
_=(O'Sfj o+o·s+; 4)=2+j s n. ·
For maximum power transfer the load impedance should be complex conjugate of the·above
impedance.
... Impedance of load for maximum power transfer i.-e. ZL = 2-j 5 n
Eo2 (20)2
Maximum power Pina:1 R X -·50 W.
4 0 4 2 -
20 '
Load current IL=(l+)5)f(f-=Js)=5 A ·
Power loss. in source-(5)2 x0'5=12'5 W. ·
Power loss in transmission network=(5)2 x1'5=37'5 W
Total losses=J2'5+37 5=50 .W.
output x 100 50 ' I,
Bfficie~cy
output+ losses so+so xrno. so%. j'
1
'I
CH ARAC'!ElUSTICS OE INSTRUMENTS AND MEASUREMENT SYSTEMS 31
Example 2·21. The output voltage of. an audio amplifier is 10 V and 4 V when delivering
powers, 25 Wand 16 W respectively. Find the open circuit voltage and output resistance of the
amplifier. What is the maximum power that the amplifier can give ? ·
Solution : Let Ro and RL be the output resistaace and load resistance re:spectively.
. '. EL1 2 (10)2
· Lo ad resistance in the first case
1
R_Ll = p;- = 25 =4 0
. d res1s
Loa • the second case RL2= p-;:
• t ance m EL22 (4)2 1. un
= 4=
Eo RL
Voltage across load EL= Rof RL
instrument or the measuremen~ system to settle to its fi;ial · steady state conditions. Under such
conditions the study of behaviour of the system under transient state,. called 'transient re'spoose' is
not of much of importance ; only steady state resp.onse of the system need be considered.
However, in many areas of measurement system . appli<.-ations it becomes necessary to study
the response of the system under both transient as well as steady .state conditions. In many applica~
tions, the tnuisient response of the system i.e., the way system settles down to its final steady state
conditions is more important than the steady state response.
It bas been pointed out earlier that the instruments and measuring systems do not respond
to the input imm~diately~ · '.fbis is on· account of the presence of energy storage elements in the sys-
tem, These energy storage elements are electrical inductance and capacitance, mass, fluid aml
thermal capacitances etc. The systems exhibit a characteristic sluggishness on account of presence
of these elements. Furthermore pure delay in .time is encountered when a syst~m "waits'' for some
specific changes and reactions. to take place, . ·
Invariably measurement ~yetemsp 'especially in industrial, aecospace, a~ biological applications
are subjected to inputs which sre not static· but d1namic in nat:ure. i.e. the inputs vary with time
Since the input varies from instant to instant, so does the output. The behaviour of the system
under such conditions is described by the dyumic response of the system.
The dy1flmic chi.tracterhltics of any measurement system are :
(I) Speed of response (ii) Lag
(iii) Fidelity (iv) Dynamic error
The qualities Ulto4)m.t the left side are .desirable fo a. dynamic. system while those on the tight.
are undesira~le. 1
32 ELEC!RiCAt MEASUREMENTS AND MEASURING, INSIRUMENTs .
>.
'O~t-----
c;
:~ 0 '
c; ;J
:>CT
0 0
Time __..
f~ l I bl
r . The above signals are used for studying dynamic behaviour in the time domain. For studies
in frequency domain, steady state response to a_ sinusoidal input signal yields a great deal of informata
ion. This is because all actual inp~ts can be thought of c~nsisting of a band offrequencies ranging
from zero onwards (All types of signals c1n be broken mto sum of a series of sinusoidal signals
according to Fourier series). · !1 \
,\ .
··When system stu~ies are carried out in time ,domain, the dynamic behavi~ut of the system
depends upon the system pole~ and not on t~pe · of input. Therefore the system behaviour to any
kind of inputs can .be predicted by studying its ·.
response to one of! the, standard signals. T.he I
·standard input .·chosen for this _purpose is a step
input; ' ·· Final st12ady
p0s1tion
The step, ramp and sinusoidal signals are J5 ,
shown in Fig. 2·20
2'3 3. Overshoot. Moving parts of instru~
ments have mass and·· thus possess iriertia. When
an input is applied to instruments, the. pointer does
ndt inimediat~ly come to rest at its .steady state
(or final deflected).7positiQn·1>U! goes beyond. i~ or
in other words "overslioots" its steady position.
The overshoot is defined as the maximum amount
by which moving system moves beyond the steady
, state position. (See Fig. 2'2C, · In many instru- Fig. 2'21. Overshoot.
merits,. especially galvanometers it is desirable to have a little overshoot but an exc.essive overshooO
is undesirable. ·
ExaJDple 2·2s. ·. A step Input of 5 A is applied to an ammeter. The p9inter swings.to a voltage
· of 5• 18 A and finally comes toi re;st at 5.02 A. (a) Determine the overshoot of the reading in ampere
and in percentage of final reading'. (b)'Determine the percentage error in the instrument.
Solution. (a) Overshoot=5'J8-5'02=0'16 A ·
. . h.. 0'16% 0
Percentage overs oot.- . · x 100=3'2 o ..
5 02
· . . . s·o2-s·o
(b) Percentage error, - ~ x 100~04%.
,. o
5 \
UNSOLVED PROBLEMS
1. An
ammeter reads 6'7 A and the true value of current is 6'54 A. Determine the error an(tbe correction for
this instrument. · , · _ [Ans. 0·1~ A, -0'16 AJ
2. A voltmeter reads 109.5 V. The error taken from an. error curve is -0'37 V. Determine .the true ¥oltage;
. , / . [Ans. 109'87 VJ
3, The measured value of a voltage is 11 i V while its true is 110 V. Calculate the relative error. (An~. 0'91%1
4. A 0-100 Vvoltmeter has 200 scale divi:.;i,ms which can be read .to 1/2 divisiOn. Determine tbe resoJution
of the meterin volt. · · . (A11s; o·~- VJ
· 5. A diaphragm type pressure m·easuring instrument is calibrale~ .for absolute pressures of 6 to 76() mni ~f
mer~ry. ·It has an acc~racy of ± 1%. Calculate the scale range, scale spen 1md maximum S~l}tic error,
. , [Ans. 760 mm, 754 mm, t 1·s4 mm]
6. State the number of significant figures 'in each of the.fQllowing : .
. ~) ~2.A, (b).l~65 V,. <.c.> ~t:is•\\r,;(d~-~QJll.{t) 4~~~.Jlll0.~2.. (,.). 0'34Sk.Q .
. ... ,.- .: :. , . ; . , , , '-· . , . . ·.•~ .. <aH.
. .
·\
t1'H. (c:).. 4, (d> Sa .(f}i; t /). J, (l)S]
-,
~ ..
l ).
34 ELECTRICAL MEASUREMENTS AND MEASURING INSTRU'MB:N'IS.
7. Four resistors are conneeted in series. The values of resistors are 28'4 .Q; 4.'25 Q, 56'605 Q,_and 0'75 Q
with a~ uncertainty of one unit in the last digit in each. Calculate the total series resistance, giving only significant
figures in the answer. . [Ans. 90 Q].
. S. A curri!!nt of 2'56 A is flowing in a re5istance of 45'73 n. Each quantity has an uncertainty of one unit
m the last digit. Find the value of voltiige drop to the appropriate number of significant figures. . [Ans. 117 V]
9. Determine the lineacity of a potentiometer to obtain an error not to exceed 1 part in 10,000. [Ans. 0'01%1
h 10. A recording instrument requires a current of 0'05 A to overcome initial friction and produce motion of
t e movement. Define this effect and list factors which produce it. (Ans. Dead zone=O'OS AJ
11. The dead zone of a certain pyraometer is 0'125 per cent of the span. The calibration is 800°C to 1800°C.
What temperflturn change must occur before it is detected? . . [Ane. 12'S'C]
12. Wb!lt i,IJ the true value of.voltage ~ross the SOO k{l resistor connected between terminals A and B as
shown in I ig, 1'22 7 What would a voltmeter with a sensitivity of 20 kQ/V read on tte foltowiog ranges : 50, 15, 5
volt ?
~M~O A
r-~~
..:.. WV 500k0.
B~
-l-.~~
t
I~_J_'o_k_°'_~·,...,1o:_j
Fig. 2·22 fig. 2'23
13. What is the true value of cuaent in the 15 kQ resistor of Fig 2'23. If an ammeter of 2 kQ resistance is
i1iied to measure the current in the 1S k(J resistor, what will it read? If a loading accuracy of 99 percent is desired
in measuring the current, what should the arome1er resistance be ? [Ans. 100 µ.A, 92'6 µA, 250 O]
14. A volt.Ilg~ source has an open circuit voltage of 100 V and ao output impedance of lO+J 20 n. It is
connectea Hl n Joad through a transmission network. The impedance of tbe load is 20+J XL Q and that of
transmis~ion r1etworlt Rrr:-J 15 o. Spe.cify tbe value of Rrr and XL for maximum power transfer to the load. If
the variation in load resistance were permitted, what will t.e its value for maximum power transfer?
. .· · . . (Ans. 10 0., 5 0. capacitive, 10 fl]
15. A step Volta~ of 20 V is applied to a voltmeter and it swings to a maximum value of 28'8 V and finally
s1&ttJes to 20 V. Find tho percentage overshoot. . . [Ans. 4 0%]
. 16. A self-balancing potentiometer is connected to spatial encoder required to read 2000. The lin~arity is
O'OS%, find the accuracy of the read out system. (Ans. 1 part 10 2000]
.~J
. I
3"
Errors in Measurements and .th.air Statistical Anaf y.si~
3•1 Limiting Errors (Guarantee Errors)
; I
The accuracy and precision of an instrument depends upon its design, ·the material used
and the workmanship that goes into· making the/ instrument. The.. choice of an instrument for a
particular application depends upon the accuracy desired. If on'ly a fair degree of accuracy is desired,
it is not economical to use expensive materials/and skill into the manufacture of the instrument.
·But an instrument used for an application req,uiring ~ high degree of accuracy has to use expensivt(
material and a highly skill~d workmanship. The· economical production of any instrument requires
the proper choice of material, design and skill. In. order ·to assur.e the purchaser of the quality of
the instrument, the manufacturer guar aQtees a· certain accuracy. In most instruments the accu~
racy is guaranteed to. be within a certain. percentage of full scale reading. Cbmponents are guar~nteed
to be within a certam percentage of the rated value; Thus the manufacturer has to specify the
deviations from the nominal value of a particular quantity. The limits of these deviations. from
the specified value are defined ~s Limiting Errors or Guarantee Errors. · . · ,
We can say that. the manufacturer guarantees or' promises that the error inthe· item he is
selling is no greater than the limit set. . ·
· The magnitude of a quantity having a nomfoal value As and a maximum error or limiting
error of ±6A must have a magnitude Aa between the limits As-8A and As+8A or · · · .
Actual value Aa =Aa±8A . ...(3;1)
, for example, the nominal magnitude of a resistor is 100 0 with a 'limiting error of ±10 il
The magnitude of the resistor will be between the limits : . ·
. A=lOO±lO Q ·or A)90 Q and A<;llO 0
In other words the manufacturer guarantees that the value of resistance of the resistor lies
between 90 n and 110 n.
3·2. Relative (Fractional) Limiting Error
. . . The relative (fractional) error is defined . as the.· ratio of ~he error to' the specified (nominal),
magnitude of a quantity .. Therefore.,
· · BA.
relative limitillg error Er= A;= A:E ·.;.(1'2)
1
• . .. Aa-As
Re]attve 11m1tmg error, E,. = Aa ... (3'7)
Example 3'1. Th~ value· of capacitance of a' capacitor is specified as 1 ~·F:b5% by the
manufacturer.' !find the'fiinits between which the value of the capacitance is guaranteed. •
. Solution': <~:T4tjuarariteed·value·of the cap~eit4'11c,e' lio within:;the limits:"
. · · · A·~ Aa(l ±Er)=i= 1x (l ±0·0~)..:..0·9·5 to t,·os µF. ·. · .· " · ; . '
Note; The. same. idea (Jf a guarante~ ·11mlting the worst possible case applies t~ electrical
measurements. The meas.urements may involve several ~omponents, each of which may be delimited by a
guarantee error. Thu8:>,Jhe same treatment is to be followed for quantities under measurement as is
followed for specified quantities. '. . · .. : . : ·
Example 3;2. A 0-150 V voltmeter bas a guaranteed accuracy. of 1 percent c>f full scale
·.reading. The voltage measured by this instrument is 75 V. Calculate the limiting error in percent..
Comment upon .the result. .. · ·, ·
Solution : The magnitµde of limiting error .6finstrutl\~,t,
. 3A ErAs! · ·. ·
· 8A=O'Ol X150=1'5 V.
Th~ magnitude of the voltage being measured is 75 V.
< '
Comments. ~tis important to note that this .meter is guaranteed to have an error of less than
1 percent offull scale or t~e l_i.t;niting err?r is I percent at the fall sc~IeAe.flecHori of ~50 V. But w~en
the meter reads 75 V, the hm1tmg error is 2 percent. The percent hm1t10g error wlll be greater 1f a
smaller voltage is measured. If' the meter reads ,37'S V~ the percentage limitieg error is
:::1' 5- x 100 · 4'0 percent. . ·
. 37' 5 . .
This increase in the percentage limiting error as small voltages. are 'measured occurs ,because
the magnitude of the limiting error 8A is based upon the fuIJ scale reading of the meter and is a fixed
quantity, while the actual voltage readio,gs can be of·any Illagnitude frorir Qto ISO V. We infer from
here that. the percentage ~rror increases as the voltage. b,eiog mea$P,red decreases. _. ·
. - . . . ·.,···.. r . :. . . :- .1 / •
.. Thus while sel~cting iristru.~e~ts, pa~tipular c~re sh~uld ~e t~Jcen ·a~-f~gar~s. the range.' The .
values to be meas~re~ shoU;ld n~t lie m the Jo'tV~r third ofrtl1e f:ange., ,Tbts tS, partlcu1artv imnortaot
if the.meter accuracy 1s specified 10 tenns of the full sc11Ie defiecttonf.s.d. (another name for f.s.d. is .
' )
ERRORS IN MEASUREMENTS ANO THEIR STATISTICAL ANAtYSis 31
fudicial value) as considerable error, (as a percentage ?factual value), may occur as is·.· Step in.
Example 3·2. Thus mete~s which read well up their scale ~hould be selected as far as possible·. ·
Another example is given below which is illustrative of the .comments given above::
Example 3·3, A wattmeter having a range 1000 Whas an error of± l %of full scale deflection.
If the true power is 100 Vf, what would be the range of readings ? · .· ·· . ·
Suppose the err'or is specified as percentage of true value, what would ~ . the range·of the
readings.· · ·
Solution : When the error is specified as a percentage of full scale deflection, ·the magnitude
of limiting error at full. scale ·
=±1~0 x 1000~±10 w.
. . . . Thus the wattmeter reading when the true ~eading i~ 100 W may be 100±10 vJ, i~e., between
1
·
90 to 110 w. ,. . . . .. .. .
+10
The relative error = lOO x 100- ± 10 %.
-'Now ·suppose the error is specified as percentage of true value;.~
. 1 .
The magnitude of error=± X100=±1 W. ·
100
Therefore the meter may read 100± 1 W or between 99 to- 101 W.
3'3. Combination of Quantities with Limiting Errors. Wh~~ two or more· quantilfos, each
having a limiting error, are combined, it is advantageous to be able to compute the limiting error of
the combination. The limiting error can be easily found by considering the relative increment of the
function if the final result is in the form of an algebraic equation.
(i) Sum of two Quantities. Let y be the final result which is the sum of measured. quantities
u and v. ·
y=u+v.
The relative increment of the function is given by
.dY = d(u+v) =du +dv.
y y y y
Expressing the result in terms of relative increment of the component quabtities
' . dy =.!:!__ .duu +~ ..dv. ·,.
y y ·y v
If the err9rs i~ t~e co~ponent quantities are represent~d. by ±8u and ·±Bv then co~respondiilg
limiting error 8y m y 1s given by : .
· 8y =±(_!!.... . _8u +!__ . 8v ) ·
y y u y v ... (3 '9)
The ab.ove equation shows that the resultant relative error is equal to the sum of the products'
fonl;led by' mu!tiplying the individual re~ative errors by the ratio of each term to the function.
(ii) Difference of two Quantities.
Let y=u-v
dy du dv
.--=---· ·,
y y y
-
38 ELECTRICAL MERSlJREMENTS.Al\D MEASURING INSTRUMENTS
1 ! '
Eqn. 3' lO is the same as Eqn. 3·9, It may, however, be mentioned that in this case when
u and u are almost equal ia magnitude i.e. when y= u-v<(u and also v~v then the relative error in y
would be very large.
(iii) Som or Difference of more tlum two Quantities. The sum Of difference of more than
two quantities may be treated in a similar way.
If we have y= ±u±v±w
Then the limiting error is given. oy :
BJ!=±(Jl-. tiu +.-E-. _av~---'!.~). . .. (3'11)
y y u y v Yw
(fr) Product of two Components.
Let y=uu
loge y =loge u+ loga v
Differentiating with respec;t to y
_l=l.~+~.~ or~=~+~·
y u ~ ~ ~ y u ~
Representing the errors in u and v a'.l ±Su and ±~v respectively, the error Sy in y is given by:
8y =
y .
±(?.u!!.. +~-) 0
... (3'12)
Thus from above we conr'~u<le that the' relative limiting error of product of terms is equal to
the sum of the relative errors of terms.
u
(v) Quotient Let y=--·
ll
Thus maximum possible err.pr occurs whe-n ~'j + ve and avv is. -ve or vice versa.
~ -
i
The above result is the sante '•US the ct>nespon-i;kPl bu.It for the 1roduct of two o.ua~tities.
i'
I.
ERROR3 IN MEASUREMENTS ANO 1HElR STATISTICAL ANALYSIS
or
_dy =-n qu +m dv
y u v
. hrmtmg
Relative . . . error of y is - = ±.- ( n Su
. 8y -U· +m.-Sv) ... (3'17)
y V
Rz=75± 1 ~0 x75=75±3'75 n
Ra=SO:-l:r~0 xso=50:l:2'5 n
The limiting value of resultant resistanc.e.
R=(J1+1s+so)±(l·ss+3·1s+2·S)=t62±s·1 ,l,
: . Magnitude of resistance= 162 0
Error in ohm =±8'1 0
. Percent limiting error =:I::~~~ X100=±5%.
· Example 3' 5. The resistance of a circuit is found by measuring current ff owing and the power
fed into the circuit. Find the limiting error in the measurement of resistance wllen the limiting errors ' 1
in the measurement of power aud current are respectively ±1'5% and ±l'Q%.
.4o .. .
. ELECTlUCAl ·MBASURBMENIS
.. . AND
. MBASURING IN 1TRUMliNTS
' ' . .• . . .
Deternih1e.the rnagnirude of the1 uilkni>wn reSfstance·and the.'Hmiti'ng error in. percent and in
ohm fort.he unknown ·resistilnee R~. . · · .- ·. ·
Solution : Unknowi:i
. ' .
resistance
' '
4639
Percentage Hmfring error %Er= ±co··oot 04XlOO) = ±0' 104% .
Limiting values of resistan~~ A.a.=4639(1±0'00104) .
=4639±5 0~4634 n 4644 n. ·
· ·Thus we conclude from ·th~ above examples that the guarantee vaJues are obtained by taking a
direct sum of the po~sible errors, adopting the algebraic signs that give. the worst possible case. In
fact setting of guarantee limits is necessarily a pessimistiC process. ·This is t~ue from manufacturer's
viewpoint as regards his promise to the buye.r, .and it· is also true of the·. user in setting accuracy
limits ill results of measurement. · ·
. 3·4 Known Errors. When the error of a quantity or an instrument is knowri\the effect or this
. err~r, .~hen combined with oq1er error~, can ?e computed in a manner si~i!ar -to -th\ combinations
.of hm1tmg errors. But the difference is that m case of koown errors the signs of relative errors are
given and must be preserved in the calculations. . \
Example 3·s. A resistance is fated at 3200 u and the current flowing thro.ugh thi~;Js 64 mA•.
• (q), Compute the power foss in the resistor. (b) It wa:s later found that the resistance· of the resistor
was 0'2 percent greater than the specified resistance.and the ammeter read 0·75 percent more than the
true current. Determine the known error in the computed power in part (a).
Soluti~n : (a} Power consumed P=I 2R=(64x l0_;3) 2 x3200=13'1 w. ·
(b). Relative error in power
Example 3·9, Current was measured during .a test .as 30'4 A, flowing· in a·iesistor of .
0·105 n. It was discovered later that the ammeter reading was low by 1'2 percent and the marked .
·resistance was high by 0'3.percent. Find the true power as a perc~ntage of the power that· was
originally calculated. · ·
Solution : True value of /=30'4(1-0'012)=30'035 A
True value of R=0' 105(1 t0'0003):::;:0'1053 0
True power ==/ 2R-=(30'035)2 X(0'10S3)=95 W
Originally measured power =(30'4) 2 x 0'105= 97'04 W
True power _2i_ ._ .
·~"
.. d x 100 = 97 ~
04 X 100-97 9 percent
Ongma11y measnre powei:
. '1•
We arrive at the same results by using the following wethod : ·
Power. P= 12R .
Totalrelativeerror= : . ~ + ~=2X(-0'012)+o·oo3=-0'021
8 2 1 3
. Soh~tkm :
(a) Total rn:iistance of resistors connected in parallel and neglecting their errors is :
t
nation of gross errors is probably impossible, one should try to anticipate and correct them. Some
gross errors arc.easily detected while others may be very difficult to, detect.
Gross errqrs may be of any amount and therefore their mathematical analysis_ is impossible.
However, they can be avoided by adopting two means. They are· :
1. Great care should be taken in reading and recording the data.
2. Two, tln·~i~ or even more readings should be taken for the quantity under measurcimen.~.
These readings should be taken preferably by .different experimenters and the teadings should be
taken at a diffe.ren~ reading point to avoid re-reading with the same error. It should he understood
mhat no re.Hance be placed on a single reading. It is always advisable to take a large number of read·
ings as a close agreement between readings assures that no gross error has been com.mitte d,
3'1 Systieum~k Enm·1~
Errors caused by loading effects of the meters can be avoided by using them intelligently.
For example when measuring a low resistance by ammeter-voltmeter method a high resistance volt-
meter should be used. ·
: In planning any measurement, the lo~ding. effect of instruments should be considered and
corrections for these effects should be made, 1f needed, or more suitable instruments should be used.
Preferably tpose methods should be used which result in. negligible or no loading effects.
3·7·2. Environmental Errors
These errors are due. to conditions external to the measuring device including conditions in the
area surrounding the instrument. These may be.effects of temperature pressure, humidity, dust, vibra-.
tfons or of external magnetic or electrostatic fields. The corrective measures employed to eliminate or
to reduce these undesirable effects are : :
1. Arrangements are made to keep the conditions as nearly as constant as possible. Por
example, temperature can be kept constant by keeping the equipment in ·a temperature controlled
enclosure. ·
, , 2. Using equipment which is immune to these effects. For example, variations in resistance
'with temperature can be minimized by using resistance materials which have a vety low resistance
temperature co-efficieµt.
3. Employing techniques which eliminate the effects of these disturbances. For example, the
effect of humidity dust etc. can be entirely eliminated by hermetically sealing the eq9ipment. ·
4. Applying computed corrections: Efforts are normally made to avoid the use:of'application
of computed corrections, but where these corrections are .needed and are necessary, they are incorpo·.
rated for the computations of the results.
3·7·3, Observational Errors
There are many sources of observational errors. As an example, the pointer of a voltmeter
rests slightly above the surface of the scale. Thus an error on account of PARALLAX will be in~ ·
ErtRORS IN MflASUidlM.BNTS AND THEIR STA"HSTICAL ANALYSIS 4S
curred unless the line of vision of the observer is exactly above the pointer. To minimize parallax
errors, highly accurate meters are provided with mirrored scales, as shown in Fig. 3'1.
2B0 2_~()
>~"J,i.L tJ"-~ I
J
G- -~"1
PointQf
rafledio~
I
I
/
Mirror I
I
(~
0 (i
''!I"~'('
£ya too for Left
~~,
toned ;;os·1tion
WRONG f.IO PARALLAX E.RROtk WRONG
to us. Thus mathematical laws or "Probability'' can be applied for the study of randollf errors. There
is no other way as the random errors are unknown and only statistical study c..rn lead· us to the best
approximation of the true value of the quantity under study.
3·10. Statistical Treatment of Data
The experimental data is obtained ic. two forms of tests :
(i) Multisample test) and (ii) Single-sample test.
Multisamp1e Test. In this test, repeated measurement of a given quantity are done using
different test conditions such as employing different instrumentsi different ways of measurement and
by employing different observers. Simply making measurements with the same equipment, procedure
technique and same observer do not provide multisample results. '
Single Sample Test. A single measurement (or succession of measurements) done under
identical conditions excepting for time is known as single-sample test.
In order to get the exact value of the quantity under measurement, tests should be done using
as many different procedures, techniques and experimenters as practicable. It should be borne in
mind that the statistical means which help us to arrive at correct res11lts are only valid for multi·
sample tests.
3' 10' t. Histogram.
When a number of multisample observations are taken experimentally there is a scatter of the
data about ~ome central value. One method presenting test results in the form of a Histogram. The
technique is ilJustrated in Pig. 3·3 representing th~ data given in Table 3' I. This table shows a set of
fifty re~dings of a length measurement. The most probable or central value of length is 100 cm and
the data are taken and recorded to the nearest o· 1 cm.
Table 3'1
-
I Length cm. Number of readings
i 99'1 1
I
I
. 99'8
99·9
100·0
4
12
19
100'1 10
100·2 3
100·3 1
~~
I I
almost symmetrically on either side. If smaller incremental steps, 1/ I
say 100 readings at o·os cm intervals are taken, the general form ~~
. I
•10
of the histogram will be almost the same but since the steps have 0t I
I
L-~.
cision) and regularity of the data. / ;/
A large dispersion indicates that some -~
. . ~ ·+1<1 1
factors involved in the measurement process are l<;i X1,
,, ;
... (3'21)
In practice, however, the number of observatton' is finite. When the number of observations
is greater than 20, S.D, is denoted by symbol cr while if it is less than 20, the symbol used is s. The
Standard Deviation of a finite number of data is given by ;
s=Jcl12+422~.:~i·-t_:_::"ll_J;1 2 -=~ J:~r- ... (3'22)
3'10'8. Variance. The variance is the mean square deviation, which is the same as S.D.,
except that square root is not extracted.
Variance V=(Standard Deviation) 2
=(S.D.)2:::a2= d12+d22+da2+ ... +dn2 .I ... (3'23)
n
'2d2 ... (3'24)
=--
n
But wLen the number of observations is Jess than 20
d2
Variance V s 2 =~ --- ... (3'25)
n--1
Example 3'13. A set of independent current 1ne:.isurements were taken by six observers and
were recorded as 12 8 A, 12'2 A, 12'5 A, 13 1 A. 12'9 A, and 12·4 A. Calculate (a) the arithmetic
mean, (b) the d_eviations from the me~n, (c) the avcni.ge devi.ation, (d) the· standard deviation, and
(e) variance.
Solution. ·(a) From Eqn. 3·17 the a;:':fhme\ic· mean. is
}l.= }:x =l~J-f:]2'2+12'5-!J}Jf:J.?'9+12'4 _ 12 .65 A.
n 6 .
l!tlROkS IN MEASUREMENIS ANr> THEIR STATISTICAL ANALYSis 49
(b) Fro~ Eqn. 3'18, the.deviations are:
di =x1-X=12·s-12·6s=+o·1s A
d2=x2-X=l2'2-12'65=-0'45 A
da=xa-X=12'5-12'65=-0'15 A
d4=X4-X=lll--12'65 . +0·45 A
d5=x5-X=t2'9·-12'6S=+o·2s A
d6=X6-X=l2'4-·12'65=--0'25 A.
(c) From Eqn. 3'20, average deviation:
D= ~-141 = 0·1s+0'4S+o·1sto·4s+o·2s+o·2s =0.283 A.
n .
(Note that for average deviation we have not to consider the signs).
(d) Since we have observations whose number is less than 20 therefor" we use Eqn. 3'25 for
determining the value of standard deviation
..,,,
I
s= Jn-1
Id =J
2
(0' 15) 2 +:F·~0.4S)2 -H-0'15)2+W·4s) 2 +(0'25)2+(-0'25)2
6-1
=0'399 A.
(e) Variance V=s2=(0'339)2~-0·11s A2.
3'U)'9. Normal or Gaussian Curve of Errors. The Normal or Gaussian law of errors is the
basis for the major part of study of random effects. This type of distribution is most frequently met
in practice. .
The law of probability states the normal occurrence of deviations from avc(age value of an
infinite number· of measurements or observations can be expressed by : ·
h -h2x2
Y=~ e .•• (3'26)
II Note : magnitude
The student here is cautioned. not to confuse x with I
of a quantity. Here x meuns Deviation. {
A A • ·~- t,2x2
1_in=nyc:,.x=nA e '• .f:.,,x.
If we ioltcg~~t~; the above expression for ~oo to +~ we shall have all the cases, or
00
·-00
Thus:
00
A!
re ~J;2x2
dx=l ••• (3'27)
"
-oo
Thus the i.u.tegrnJ of y from .-oo to +oo is equal to unity. The fracuon of the total number
of readings oc-Gvin:ing betw~©tn the values .X'l and x2 will be equal to the area under the curve between
these values o.f x.
x2
r --h'Ax9
h
r~1-3= vrr. J e dx m(3'28)
Xi
when: m.-2"'--=munber of readi11gs occurring between the values xi and x2•
If the ueb'. ooiwoon x1 and X?J is o· 5, then 50 per cent of the deviations fall between x1 and x 2•
fo general ·~iw p;mfr.lillbmty for iinding a deviation in an interval between xi and x:a becomes
x2 xs
1~{\.A..,1 <.•o'
'
11 .,~';("/
y ,-f.,,,""' --;::::::- f ,:P -h?.xJtd'11'
•'"1
h
M.4 ...., I '6; ,/~
••• (3·Jo)
A convenient measure of precision is the quantity r. It is called Probable Error or simply P.E.
The reason for this name is the fact mentioned above that half the observed values lie between the
limits ±r. If we determine r as the result of n measurements and then make an additional measure~
ment, the chances are 50-50 'percent that .the new value will lie between-rand tr. That is, the
cbances are even that any one reading will have an error no greater than ±r.
The location of point r can be found from Eqn. 3'28, by putting
r= ------
0'4764
This gives ... (3'33)
h
3·10·12. ·Average Deviation for the Normal Curve
The average deviation may be computed when more than one reading is present at a given
deviation by' multiplying the amount of the deviation by the number of points on the deviation. Then
this product is added to other similar products (without regard to sign) until all readings are taken
into account ; then divide by the number of readings.
In the case of normal curve,
tcx:i
Average deviation D = J I x I y dx
- ex:>
+oo
= -~'l_J
vrc e-h2x2x dx
= ~~j] ... (3 '34)
0
4764
From Eqn. 3'33, h= .9.' . Putting this value in Eqn. 3'35, we have, average deviation :
r
- r ... (3'3S)
D=0-8453
3'10'13. Standard Deviation for the Normal Curve
""'"'d2
The standard deviation is given by •)
CJ"=- (See Eqn. 3'1 l'i
n
S2
Follo'wing a method siroil~r to that foll.owed above1 we have :
+oo
2hr
cr= \7 n:-.: j e
-h~~x2 . 1
x2d.x .-~ --2 ~
0
or standard deviation for normal curve
1
a=-;--- ... (336)
V ~h
r
=0·614s •.. (3'37)
... (3'40)
for an infinite number of deviations forming the normal probability curve, where n is infinite. But
for a finite number of dc:viations, the probable error for one reading is :
3'10' 15. Studard Devimtioe ·~if t\1e:m. When we have a multiple sample data, it is evident
that the mean of various sets of data can be analvzed. by statistical means. This is done by taking
standard deviation of the mean.
(i
<ra= V2n !
... (3'46)
... (3'47)
E:iuimple 3'14. The following 10 observations were recorded when measuring a voltage:
41 '7, 42'0, 41'8, 42'0, 42'1, 41 '9 9 42'0, 41'9, 42'5 and 41 '8 volt. Find (I) the mean (ii) the standard
deviation (iii) the probable error of one reading (iv) the probable error of mean and (v) range.
Sobdi®n : For the sake of ease in caJcuJations, the observations are tabulated and manipulated
as under:
x a
--------- ------
41'7 -0·27 0'0729
42'0 +0·03 0'0009
41'8 -0'1'1 0'0289
42'0 +0'03 0·0009
42'1 +0'13 0'0169
41'9 -o 07 0'0049
42'0 +0·03 0'0049
I:xc:::419'7 I
~--------~~~~~-------·---~-~~
:Ed~=0'44! I
·
(i) Mean length X=
u
n-= 411o=4197
};x '97 .
volt.
·
. value of stan
(ii) The ' dard devtatton
. . 18 . cr=. -
. A
Jd2 y l'l
=
0'441
--=0'21
10
volt (See Eqn 3'21)
if the data is considered to be a set of infinite readings. However, the number of observations is
nnRy 1Oand therefore the standm:d deviation is :
~ /(12- . . J 0·441 .
s= 'V n~ 1 = 'V (lO-l) =0 22 volt (See Eqn, 3'25)
(iii) Probable error r1=0'6745 s=0'15 volt. (See Eqn. l '41)
(ill) Probable error of mean rm=. 1 r
'vn-1
1
·=:-?·-
v.9
~=::(/'(15 volt.
1
(See Eqn. 3'42)
(v) Range'--=42'5-41 '7 w.0'8 volt,
54 ~ELECTRICAL MEASUREMENTS AND MEASURING INSfRUME~TS
Ex2mple 3'15. In a test temperature is measured ~00 times with variations in apparatus
and procedures. After applying the corrections, the results are :
----------------~----,------------- ..
16 4 2 2
Calculate (a) arithmetic mean, (b) mean deviation, (c) standard deviation, (d) the probable
error of one reading, (e) the standard deviation and the probable error of the mean, (/) the standard
deviation of the standard deviation.
Sohltioun : The computations are done in a tabular form as under :
~------- -------- ----·- ----- - - ·------------- I
•
I
---- - ··---~·--
-·---------- - ----- , __ .
------'-- =102'8
I
191 '08
_______
I
---- -- -----
- -- -
4
(a) Mean temperature= i6r}~ =400'780°C
1
(b) Mean deviation 15=
1~~'?_= 1·02s·c I
~
. . .. I 191 ·os :' . -·- -o·c
(c) Standard dev1at10n cr = 'V - = 1 38 0.
100 ,
. (;:~) \:')\~
Standard deviation of the mean am=.\/ 100~=0'138°~
(f) Standard deviation of the standard deviation
0'111 0'138
ua=vI-=v 2 =0·0196°c.
3'11. S~ilylng ODDS
The probability of occurrence can be stated in terms of ODDS. Odds is the .number of
chanees that a particular reading will occur when the error limit is specified. For example~ if the errnr
limits are specified as ±0'6745 rJ• the chances are that 50% of the observations will lie between the
above limits or in other words we can say that odds are 1 to 1.
The odds can be calculated as under
odds+ 1 t yoroccurrence
odds =probab'l't ... (3'48)
1
The odds that th.e observation lies between ±cr limits are :
odds .
----=06828 or odds are 2'15: 1.
odds+l
Table 3'3 gives the deviationst the probability of occunenc~ and the odds.
TABLE 3·3
-
Deviation Probability Odds
±0'6145a 0'5000 i to J
±a 0'6828 2·1s to l
± 2a 0'9546 21 to 1
± 3a 0·~974 256to1
I'
3·12. Specifying Measuremest Data
I
I ',After doing the itatistfcal analysis of the muJtisampie dat:a, we muse specify the results The
resuHs are expressed as deviations about.a mean value. The deviations ru-e expressed as: •
(i) Standard deviation : The result is expres~ed as Jl ±('.J
The error limit in this case is the standard deviation. This means that 0'6828 (about 68%)
of the readings are within.the limits a=±l and the odds are 2'15 to 1. Thus there is approximately
a 2 to l possibility that a hew observation will 'fall beyond this Jimit..,
(ii) Probable error: The result is CXPJ.'OS~d as J:'±0'6745 a.
This means that SO% ofthe readings lie within this limit and the od4s are 1 to I. This means
that i&ere is ttn even possibility that ll now readins wiU lie within those limi~. ·
i'
56 ELECTRiCAL MEAS'9RBMENTS AND MEASURING INSTRUMBN'TS
(iii) :i2a limits: In case we went to increase our prnbability range we specify the
results as :
i±~.
of
Thus we assume tha11 0'9546 (or about 95%) readings fall within these limit;. These odds
1. I
... (3'51)
as in repeated meas~rements (8x1) 2 tends to become mean value of variance of xi, i.e.~ Vx1.
:. Eqn. 3'51 may be written as Vx= Vx1 Vx2+ .. , Vxn + + ... (3'52)
This shows the component variances are addative with weighing factors (8X/8x1) 2• The
weighted v~riance x1 can be written as ·
(-~!_)
3
Vx1 ::.c: V;J1 ... (3'53)
ox1 . ~.
. .. (3'54)
ERRORS iN lillASUREMENTS AND THEIR STA1W1CiCAL ANALYSIS 57
. -J--(-ai-){---:-·-(·
<1x= --
ax--)2·- ax )2 .
Jix:i.+ , - Vxz+ ... + -"-- Voo•i1
. ---- (
I
... (3 '55)
11111
It is clear from Eqn. 3·55 that }both component standard' deviation~; are addative with
weighing factors ( !X )2 ,etc. which express the relative influence of the various components on the
UX1 I
combined function.
Therefore we can write <>x = \/ a::x:12+cr::x:22+ ... +rJx 0 2 . .. (3'56)
It is important to note that the above expressions ·are valid only if component quantities
x1, x2, ... etc. are independent of each other and also that the increments are small so that the terms of
higher order than the first may be neglected. ActuaHy in engineering applications, the increments
are small as it is generally possible to keep the random effects under control,
3'13'1. Probable Error of Combination of Components
Suppose Xis a function of several component variables x1, .\·2, ... xu, each of which is an inde~
pendent variate, ~
Then we have the sf:andard devirition as
0X = I
1(' 2 y"
-i:- )2
CJ:~1 2 +
BX (-;::.,--- )2 u,ez2+·,,., +( -~~
I3 x )2 ~
arm'!'
N \ L'A) ox2 ' oXn J
But from E~n. 3'39, the probable: error r=0'6145 a,
or proabable error is .Proportional to standard deviation.
Hence we can write the probable error in X as
... (3'58)
where roo1, roo 2, etc. are the probable errosion x1, x2 etc. The contribution of probable error of x1 to
~· ~
the total error in Xis ( :x: ) 2
rx1
2
and this contribution may be written in another form as 2
rx1 .
+... +rxn2
I
.1
Therefore Eqn. 3'52 becomes r:v= y'rx12+rx22
where the weighted probable error of Xn becomes as rxn =
ax
( oxn )2 rx1
... (3'59)
... (3 '60)
Example 3'16. We have a parallel circuit having two branches. The current in one
branch is Ji::.-:::100±2 A an<l in the other is /z='.200:::1:5 A. Determine the value of the total current
l=h+I2,
(a) considering the errors in h and l2 as limiting errors,
I
~ and (b) considetfog the errors as st~vidurd cfoviations.
Comment upon the result.
Solutioo : (a) Now l = h +h
. , . ~I { 11 E!J lo o/2 \
.'. fractional error m J:=:.:-..;c-· =::!::! - :: · ---;.-+·~ ~ · )' 1
i
0
, 1. 11 1 ~s 1
58 ELECTRICAL MEASUREMENTS AND MEASURING INSTRUMENTS
= v{2)2~t--(5) 2 = 5'38 A
~ K[o.;66+1;0x 15 x (0'916)2x~ss}1'11 K.
Standard deviation of /is 0"1= ~f aO <=> ±(1 'll K) ( o· L< l ~O) rad
Suppose Xis a .WI1.Gtion of several var1i'tles, X-f(x1, x2, X3, ... , xn)
where xi, x2, xa, ... xn are independent variables with the same degree of odds. I
Let w:u be the resultant uncertainty and Will1, Wi1:2, wills, ... W:o11 be the uncertainties in the
independent variables xi, x2, xa, ... Xn respectively. The uncertainty in the result is given by : ji
... (3'62)
Example 3'19. A certain resistor has a voltage drop of 110'2 V and a current of 5·3 A.
The uncertainties in the measurements are ±0'2 V and ±0'06 A respectively. Calculate the power
dissipated in the resistor and the uncertainty in power.
Solution: Power P=voltage x current= VI= 110·2 x 5'3=584 W
Now .P= VI
~~=1=5'3 and~;= V=l10'2
Wv = ±0'2 and WI= ±0'06.
il
· Uncertainty in power
= y(~~ r
7
wv 2t {
8~ r wr2 =v'(5'3) 2 x (0'2)2+(110'2)2t(0'06):il
=±6'7 W=±i
584
x 100=±1'15%.
Example 3·20. Two resistors Ri and Ra are connected in series and then in parallel. The
values of resistances are ;
Ri = 100'0±0' l Q~ .R2=50±0'03 Q
Calculate the uncertainty in the combined resistance for both series and parallel arraqgements.
Soh'Rtion : When the two resistances are connected in series the resultant resistance is :
R=R1-f-R2
:.
()R
-=1 and·-- ==1
o.R
0R1 . R~
Hence) uncertainty in the total resistance is_
WR=±· V( 8R )2-
0R1
WR1 2 +(--
() R )2 wa2
8R2
=±v (0:111)2 x (0'1)2+(0~444) 2 x (J'03)2=0'01734 n.
The total resistance can be written as R=33'33±0'01734 n.
I
Ex1mple 3'21. A plot of land has measured dimensions of 50 by 150 m. The uncertainty
in the 50 m dimension is ±0'01 m. Calculate the uncertainty with which the 150 m dimension must
be measured to ensure that the total uncertainty in the area is not greater than 150 per cent of that
value it would have if 150 m dimension were exact.
Solution : Let
L=length of the plot= 150 m
B=width of the plot= 50 m
and A=area of the plot=LXB=l50x50=7500 m2.
Now A=LB
oA 8A .
--=Band -=L
ai ·. oB ·
Uncertainty in area
WA·=±
·-----
(
,J(it y~~~+- ~1 r- w~2 = ±yB2tt'~?+f2~~n-2
Case I. When there is no uncertainty in measurement of L.
WL=O
Uncertainty in measurement of area
WA==± VB 2 wr. 2 +L 2 wn~=±\IL 2 wn 2 =±Lwn= 150XO'Ol=1·5 m2.
Case II. When there is uncertainty in measurement of L.
The uncertainty in area is not to exceed l '5 x l '5 = ±2'25 m~.
WA=v iJ2w~2+£2;~2
or 2'25=y(50)2wx.2 +(150)2(o·o1·)2
Hence uncertainty in measurement .of Lis w1 = :10'0335 m.
,,I
Example 3'22. A resistor has a nominal value of 10 .Q ±0' l %. A voltage is applied across
the resistor and the power consumed in the resistor is calculated in two ways :
(i) from P=E~/R, .and (ii) from P=El.
Calculate the uncertainty in the power determination in each case when the measured values
of E and I are ; . ·
E=IOO V±1% and 1=10 A±1%
Comment upon the result.
o--L--·0-~ Solution : The circuit diagram is shown in Fig. 3'6.
. Ammeter l~
(i) £ 2-
P~o=-
R
Voltmeter
. oP _2E DP E2
Fig. 3'6 .. -oE-R and 8R =-R2
··61 · It£C1RJCAt MFASURF.ME:NtS AND MMSlJRINO lNS1'RUMEl.lltS
. , ..{
Hence uncertainty in power measurement.
w = JT87;--)z- WE-2.-+-:-(8-P )2-~R~ =~J(?:_l{)2wE2+f\-~=2 )2 w:;-
v >J \ oE 8E R R .
; ....
aP
oE=l and
o.P
·a1=.E.
Percentage uncertainty in power measurement
·~~~ 2 2
XlOO=r::J(oP..) win2+·(BP) w12X100= v'I 2wE2 tE2w11-x 100
P oEJ f)l
per cent of scale reading, What is the s!1H1c error if the instrumt·nt indh::'.tes 80 m8/s. fAuo. ±0'6 m8/s] !
I,
. . .. 4. A 0·10 A ammeter has a guanmteed ac\~1.uacy oft ·5 per ceut of rnh scale reading. The current measured
~0
by this 1nst.nm1rnt is 7'S A, Calculate the limitiJOg values or current and the percentage limiting error.
· [ADii, 2'S± 0·1sA; ±6%]
,.. 5. Aliquid flows through a pi!,Je h1tviog a diameter of 100 mm with a velocity of 1 ill/fl. Calculate the rate flow •
. ·. .••.. tr·~ dia~eltr i~ guaranteed within ±c,~ iiild the velodty is known to be within ±3% of measured value, find the
··:·~tt·Withm wh1~h rate of.flow can be sl>@Cified. [Ans. 7'S5x 10-• m•/s ±5%]
,·..· . 6. The resistance of an unknown resistor i:J d~termined by Wheatstone ruidge. The solution fo1 tbt un~
· ·k.1mwn ree.istance is stated as
· Calculate (a) the nominal value of the unkn<Jwn resistor, (h) the limiting erro1 of the unknown resnuor in
ohm• and (c) the limiting error in per cent of unknown resistor, [Ans. (a) 3075 Q 1 (b) ±76'88 Q and
1.
(c) ±2'S°/uJ.
Ek.ROOS IN MtlASURt.MEN1'6 AND_lHBIR STATISTICAL ANALYSIS 63
7. Three resistors have the followini rati~gs1:
wh~re Pis the pow~r In. watt, V the voltage in volt and I is the current in ampere. The relative errors in power 1 curr~nt
and voltnse are respectively ±5%, ±1% and ±1%. ·
Calculate t'he relative error ·in power factor. Also calculate the uncertainty in the power factor if the errors
were specified as uncertainties. [Ans. ±2.5%,· ± 1'5%)
9. A resistor is measured by the voltmeter-ammeter method. The voltmeter reading is ·123·4 Von the 250 V
scale and the ammeter reading is 283'5 mA .on the 500 mA scale. Both 111eters are guaranteed to be accurate within
±1 per cent of full sca!e reading. Calcl.late : (a) the·indicated value of resistance, lb) the limits withtn which the result
can be guaranteed. [Ans. 436 Q ±16'5 .Qj
10. T.lle fow lllmal of'a Hay's a.c. bridge are arranged as fullows :
AB is-a coil of unknown impedance.
n
BC is a non-inductance Ri"""' 1000 with an error of± I part in 10.000.
CD i!! a non-reactive resistor Ra=833±0'25 .Qin series with a no loss capacitor C=l '43 ±0'001 p.F.
DA is a non-reactive resistor Ra=l6800±1 part in 10,000.
The supply frequency is 50±0'1 Hz; The bridge is balanced. Determine Land R of the coil and the limits
of error. The balance conditions are :
L= CR1&_
l+c.>2C1Ra 1
R= R1RiR1C 1Cil 2 _
l+ro,C 2R82
{Ans.. L=21 ±G'l45 H; R=2480 ±29'5 UJ
11. 'Ibe stl'ess in a mild steel flat circular diaphrag·m is given by :
3 D1 p
s=·1&1-N/m1
where. D=Diameter of the diaphragn; mro,
t=thic~ness of diaphragm walls ; m,
and p=pressure applied ; N/m2•
3 2
A diaphragm has a diameter of 15 mm, thickness=0'2 men and the pressure applied is 300 x.10 N/m • Cal·
culate the stress. The kn0wn error in diameter is +1% and in thickness is 3%. Calculate6 the error 10 stress. 2
. (Ans. 316Xl0 N/m2, -4%1 -12'64 N/m ]
12. A set of independent ten measurements were made to determine the weight of a lead shot. The weights
m gramme were : ·
1·s10, 1'597, 1·591, 1'562, 1'577, r5so. t ·564, t ·5s6. 1'550~ rs1s.
Determine the (a) arithmetic mean, (b) average deviation (c) standard deviation (d) variance (e) probable error11
of one reading (/)problem error of the mean. [Ans. (a) 1'5752 g (b) 0'01068 (c) 0·01426 g (d) 2'0335x 10' 8
I ~ (e) 0 0024 g ( f) 0'00076 gJ
':) 13. The following values were obtained fro:n 1he m~asurements of! the value of a resistor : 147'2 .0,
I
1474 n, 147'9 Q, 148 1 Q, 147'1 Q, 147'5 .Q, 147'6 .0, 147'4 Q, 147'6 !land 147'5 0. Calculate (a) arithmetic mean
I ~b) average deviation 'c; standard deviation, treating the data as finite (d) .standard deviation treating the data as
population. [Ans. (a) 147'55 .0 (b) 0·22 Q (c) 0'295 Q (d) 0'28 Q]
ELECTRICAL MEASURBMENiS AN.b MBASORINO iNSfRtJMBNtS
Value 3 4 .5 6 7 8 9 IO 11
I I I
Frequency cf 6 2 1 ~'."",
2 3 7 6 4
occurrence
I
Calculate (o) arithmetic mean (b) mean deviation (c) standard deviation d) probable error of one reading
(e) the 11tanda~d deviation and the probable error of the mean ( f) standard deviation of standard deviation.
[Ans. (a) 7'062 (b) 1'45 (c; l '~65 (d) 1'258(e) 0'33, 0·226 ( J) 0'203]
15. A ca.pacitor C= 1'0± 0' 1 v.F is charged to a voltage 20J:1 V, where errors are probable errors. · Find
the charge on the capacitor and its probable error.t [Ans, 20>< 10-e coulomb, ±2'2X 10-e coulomb]
16. The resistance 'R of a copper wire is given by:
R=Roll +o:(t-20)]
where Ro""'4 0 ±0:2%
is the resistance of the wire at 20°C. Tbe value of resistance temperature co-efficient copper is 0·004 f!/0.-°C ±1%
und temperature t=25 ±1<'C, Find tbe resistance R and its probable error. [Ans. 4'08 Q ±0'3%] -~
I
the use of awkward numericaJ constants when we express ·a quantity of one kind which we have
derived from measurement of another quantity.
. In any given piece of theoretical work the Units of the various quantities must all belong t0i
a Consistent or coherent system.
In Science and Technology, two kinds of Units are used : 'ii\)
I
(I) Fundamental Units and (ii) Derived Units.
If we have A kinds of quantities to evaluate and B independent physical quantities expressing
relationships between them we can independently choose sizes of units of (A - B) quantities. The sizes
of Units of remaining B quantities can be derived with the help of B physical equations. A system
of Units so constructed is caJJed a Consistent System. (A-B) Units, which are independently chosen
are called Fmndame1tal Units. . The remaining B Units are called Derived -Units.
The .Fundamental Units in mechanics are measures of length. ~ass, and time. The sizes of
fundamental units, whether centimetre or metre or foot, gramme, or kilogramme or pound, ·second or
hour, are quite arbitrary and can be selected to fit a certain set of circumstances. Since length, mass
and time are fundamental to most other physical quantities besides those in mechanics, they are caJJed
the Primary Fundamental Units.
Measures of certain physi~u1 quantities in the thermal, eJectficaJ, 1uumi~ation fi.eJ~s ~re also
!epresented by fundamental units. These units are used only where these particular d1sc1pJmes are
mvolved and therefore they are called Auxiliary Fundamental Units.
. AH other units which can be expressed in terms of fundamental units with ~be help of physi.cal
eq~atl~ns are called Derived Units. Every derived unit originates from some ph) s1cal law or .equation
wht~h defines that unit. For example, the volume, V, of a room is equal to the product of its length
l, width b, and height h. Therefore
V=lbh
AI then the volume ofa room 6 m X 4 m x 5 mis 120 !11 3•
the metre is chosen as the unit of length,
Note that the number of measures (6X4X5=120) as wen the units (mxmxm=m 3) are m_ul~11,J1ed.
The derived unit for volume is thus ms. Also note that use of an awkard numerical mult1pher has
been avoided.
4·5, Dimensiom
Disregarding the problem of measurement and the concept of magnitude, it ~s evident_ t~t
every quantity has a qoamy which distinguishes it from all other quantities. This umque quality is
called Dimension. The dimension is written in a characteristics notation, as for example [L] for
length, [T] for time and so on.
I
A derived unit is always recognized by its Dimensions, which can be defined as the complete ·~
algebraic formula for the derived unit. Thus when quantity such as, area A of a rectangle is measured
in terms of other quantities i.e. length, /, and width, b, in this case, the relationship is expressed
mathematically as : ·
Area A=a constant></Xb ... (4'2)
Since 1 and b each have the dimensions of a length, written [L], the dimensions of area are :
[A] =[LJ[LJ=[L2] ... (4'3)
. .since the constant is a pure numerical ratio and is, therefore, dimensionless. As discussed
~arher i~depende?t units for derived quantities like volume, area et(< are unnecessary and even I
tnconvement, for 1f we take, the constant in Eqn. 4'2 as unity, and if metre (m) is unit of length, metre
square (m2) can be used as unit for area.
. . As stated earlier, in· mechanics the three fundamental units are length, mass and time. Their
dt01ens1onal symbols are : ·
UNITS, SYSTEMS; DiMBNSiONS AND STANDAll.DS 67
The square brackets .incticate dimensional notation only. H should be understood that
equality is in terms of dimensions only, and it should not be mixed up with numerical values.
4.6. Dimensions or Mecbamc1d Quantities
All mechanical quantities can be expressed in terms of the three fundamental quantities-
length, mass and time. A few examples are being given below :
.
1. Ve1oc1ty
length
= -..--t.
1me
- [v]= i~ =[LT-1]
... rrr .
1J [ . , ..
2. Acceleration= v:~oc~!Y_
.J
i;AffiC
[a =~·['.f]-- ~1 -., 1.
3. Force =mass Xacceleration
. I
[FJ::~(M][LT-i]=[MLr- 2]
...
'.
4. Work =force x distance [W] =l!iliLT- 2][L}= [ML21:r-2]
•. /Ill'.
5. Power ==
work .f,P],~; Vr.!r~r-2J=[ML2r-a1
time
6. Energy -power x time [ Wj "" [ML2T-3][T]= [ML2f-2]
7. Momentu~ =mass x velocity =[M][LT-1]=[MLT-1]
As a second principle, they decided that all other units should be derived from the. aforemen-
tioned three fundamental units of length, mass and time. The third principle they proposed was that
all multiples and submultiples of basic units be in the decimal system, and thev devised the svstem of
. prefixes in use ·today. ·
The proposals of the French Academy were approved and introduced as the metric syste1n
of units in France in 179 :· The metric system aroused considerable interest elsewhere and finally,
in 1875~ seventeen countries signed the so-called Mette Convention, making the metric system of units
legal system. Since 1875, the international autori.ty concerning metric units has been the Conference,
Generale des Poids et Mesures (CGPM). The CGPM maintains the Bureau International des Poids
et Mesures (BIPM) at Sevres near Paris. Britain and the United States, although signatories of the
convention, recognized its legality only in international transactions but did not accept the metric
system for their own domestic use.
Britain, in the meantime, had been working on a system of electrical units. The.British
Association for the advancement of Science appointed a Committee on Electrical Standards m 1861
under the Chairmanship of Clerk Maxwell. They decided on the centimetre and the gramme as the
fundamental units of length and. mass. From this developed the centimetre-gram-second or CGS
absolute system of units.
4'8, CGS System or Units
Till now the most commonly used units In electrical work were CGS units. These units
involve the use of unit of a fourth quantity in addition to units of mass, length, and time .. Experirne~Q
ta1Jy observed facts of either electrostatics or electromagnetism are used for defining CGS umts
for electrical work. Therefore, there are two systems of CGS units.
4'8'1. Electromagnetic Units (e.m. units). Units based on electromagnetic effects are known
as electromasnetic units or e.m. units and the system is. known as electromagnetic system of units or
e.m.u. system. This system involves the units of four quantities ; permeahility (µ) of the medium and
the units of length mass and time. The value of permeability of free space (vacuum) is taken as unity
in this system.
4'8'2. Electrostatic Units (e.s. unit:;). Units based on electrostatic effects are known as electro-
.. static units or e.s. units and the system is known as electrostatic system of units or e.s. u. system. This
system involves the units of four quantities; permittivity (E) of tbe medium and the unit:> of length,
mass and time. The value of permittivity of free space is taken as unity in this system.
The e.m.u. system is more convenient to use from t_he point of view of most electrical
measurements, and is, much frequently used than the electrostatic system. If a quantity is expressed
_in C.6.S. system without any additional designation "electromagnetic" m ''electrostatic'' it may be
taken that the electromagnetic system has been used.
4'9. Practical. Units. Practical units are derived either from the absolutf:' units or by refor·
ence \o arbitrary standards. These units should be e~sy to handle and manage.
In working with either of the absolute systems mentioned above, some most commonly used
units are found to be inconveniently large or small. The practical units employed are decimal illulti~
ptes, or submulHples of the absolute electromagnetic units.
The practical units. based on e.m. units, are made larger or smaller than the corresponding
e.m. units by an appropriate power of 10, to bring those which are mostly used in experimental work to
convenient and manageable magnitudes.
The British Association Committee on electrical units decided that the practical unit of resis~
tance (ohm) should be 109 C.G.S. electromagnetic units of resistance. It was also decided to make
the.practical unit of potential difference (volt) equal to 108 C.G.S. electromagnetic units of poten· •
tial difference'. These two magnitudes being fixed, we can fix the practical 'Jnits of other quantities.
For example, the practical unit of current (ampere) would be
_ _-yo!~ - _lOB_e_~Ei~ :_ l -1 .
ampere-:-. ohm - 109 e.m. - . uni'ts - O e.m. umt.
t.rNns, SYSTEMS, DlMENSiONS ANb STANDARDS 69
Thus the practical unit of current is l/10 of a C.G.S; electromagnetic unit. Table 4'1 shows the
symbols and magnitudes of practical units in terms of corresponding C.G.S. electromagnetic units of
different quantities.
TABLE 4.1
iliii ~-------------.-----.------------------·-- -----
Practical Number of C.G.S. em. Defining
No. Quantity unit Symbol units in one practical uni! eqrntlon
The sizes of practical units of resistance and potential difference were arbitrarily chosen to be
of convenient magnitude and the other units are then derived from these units.
The practical unit of power is watt or joule per second and is equal to 101 ergs per second.
The practical unit of energy is joule and is equal to 107 ergs, Actually in commercial measure men~
energy is expressed in kilowatt hour and one kilowatt hours is equal to 3'6x101s ergs.
4·10. Dimensional Equations. All mechanical quantities can be represented in terms of the
dimensions of length~ mass and time, but electrical quantities need at least one other fundamental
dimension.
4·11. Dimensions in Electrostatic System. In this system the dimension of permittivity E is
taken as the fundamental dimension. We use Coulomb's law as our starting point.
1. Charge. According to Coulomb's Jaw, the force exerted between two charges Q1 and Q2 is
·- F=Q1Q2
€d2
where dis the distance between charges Qi and Q2· Putting the dimensions of force and distance.
weh~e: ·
,_ [Q2]
[ML1 21 = [E][L2]
:. Dimensions of charge [Q]=[E1!21 M112 £8121-1]
2. Current. Curren.t is charge per unit time.
Let (/) represent the dimensions of current :
[I]= [Q] _ [~~ _Nflf _f_ ' T~:_l] -[E112 Ji112Ls12r2]
2 3 2
[T]- [T] ·
3. Potent111l Difference or Emf. Potential difference is work done per unit charge.
70 f!LEClRiCM. MEAStiREMENtS AND Ml:!ASUliNd INS1Rt1MENTS
2 2
[E]=[WJ ___ [¥L r- ] __ ::;=1E-112Af1'2L1/2y-1]
[Q] -[El'2M1f2La12r-1]
where [E] and [ W] represent the dimensions of emf and work respectively.
4. Capadhmce. Capacitance C is equal to-~- · l
I
. , . [Q]- [E112 Ml/2 La12r-1]
Therefore, d1mens10ns of capacitance are [C]=rer= [E-112 M 112Lit2r-I] [fl]
6. Im1uctaoce, Inductance
emf e
L -r-at-e-of-c-hange of cunent =d!/dt
· Dimensions of inductance are :
[E] [E] [T] [E-112M1'2L112r-t]{T)
[L]= [l]/[T] =-(/]- = [El/2Mll2£312T-2) [E-1L-1T2]
4·120 Dimensions in Electroo.rngnetic Sy~tem. The fourth dimension in this system is the
dimension of permeability µ. The starting point is the law of force between poles.
m1m2
1. Pole strength. Force F= µd 2
[MLT-2] = (µ][L2)
[m2]
3. Current.· We know that the magoetfaing force at the centre of a loop of radius r is :
H = 2n I 'fl'is .m e.m. umts.
-~-- 1
.
r
1
--o---=~H \IC Iocity.
"
y(.l('
Equating the: dirncm;ionai equndons of othef quantities, Wf;, always get ·Hie same result i.e.
some powe.c of 1/ µ.·f is always involved. H sbodd be: understood that H is no~ possible to obtain
separate express\ons fqr IJ. and E,
fo any syskm of units the permeability of free space r.io and oermi111vil.y of free space itn am
related by relation :
ir:
where vis the wlodtv of light in the svstem of units eonsidcn:,d, !~~or ~;xarnple~ CG.S. units the
value of o is nearly ec!ual to3 x mm which ls the ve!odty light and oi elcff.'.trornagneuo:: v;.i3_ves in free
space. (The velocity of light fa 3 x l 010 centimetre per second).
Frmn the above relationship the dimensions of any electrical qm~ittity ca11 be 10onver!ed from
those of electrostatic r;ygtem to those of the dectromagnetic syt:tern, and vice v~r:-m.
~·
For exarnp lei t ne 1.iJJuens10u&
1 •
(hf pot{mtw. "ft't::rence m
• l e11 . e .s. n. system :JJ"i;
rm '*[f-1,~iM112vNr-11
and in e.m.u. §ystem are [E]==fr,,11 2.Mlt2.L31?.y-·21
Tl ELECTRICAL MEAt Ull.EMEl'>ltS A.t\l> MFAStJRJNG JNS1RtJMENtS
""
9
xi 020 e.m. units of capacitance= 9 x iozo x10u=9 x~ 011 farad
or 1 farad=9X10 11 e.s. units of capacitance.
Since the farad is too large a unit for many practical cases, the microfar.:d (µF) or pico· farad
(pl are used as more convenient units.
1)
Similarly milli-henry or micro-henry are often used as more convenient units of inductance..
Table 4'2 shows the conversion factors for converting different e.m. units.
TABLE 4'2
Conversions of e.m u. to e.s.u. and vice versa
---··--.. -----
' Dime11sio11s I
Number of e.m .
No. Quantity Symbol
.--- e.m.:----i----,,-e.s=~-·---1 units in one e.s.
unit
1
1
=10-i'. .1-) 2
,,
M.I\.~:-.~.?.Lt of
. . ..,("'/!:!
c. .1.~. ~JIL
QOV~er
·~· f
-··=(
umi o power
~QOX 100
1
J.
1
)!:!('
•
J_ \-
,
3
= ~o 7
·1
~ l.l J• •
.
M.K.S. imili pow1;.r:..-::: 107>< C.G.3. e.rn. m1its of pov11er
,;)f
Thus we see that the units of all purely electrical quantities in i'ri.K.S. system, w~th units
of power and energy, are identic~l with prac~icai sy,stcm of units comprising, the ampere, .volt~ ohm~
coulomb, farad} henry 9 watt, .1oule etc. fheretore~ we conclude that 1f M.K.S, umta are used
throughout, the fundamental defo1itions of electrical quantities give the practical units directly, without
any necessity of conversion fact.om.
M K.S. ~nit of flux is 1 weber= 1os lines or max we Us.
M.K.S. unit of flux density is 1 weber per square mctre=104 gmJ.ss.
Tb..:. ..:dvau.tages of M.K.S. system of units are :
( i) This system connects the practical units directly with tht': fmufamental faws of electricity
1
and magnetism..
(ii) This system gives specified formulse for expressions of electromar~netism involving only
prnctkal units. ·
(iii) Tkcre are ccrt,dn forrnu!ac in C.G.S. system wherein p, is freqmmtly omitted owing to the I
permeability of air being unity. This 11.bsence of v from the formulae ·wr011g!y i!lldkates that the ~
permeability of· medium has no effect on the quantity being conskiered, .fo M.KS. sy.st~m the
presence of p. in the expression. emphasises the fract that the dfoct depemfo upon tb.t~ 1nedium. For
cxampli;: the force p©r unit length between two strnight cond.1~<,;tors is ·~::.;. in the M.K..S. system.
'I'his means that tbe formula makes it chmr tha~; the force is. deptndru.lt upon the pern1~:ability
of medium .
.f Jt5. R~tfom~Hsi;el M.K.S.A. Sy»te~ll. The M.K.S. ~ystem in its rationalised form, utilizes four f!:
fundamental. units.. fbese are metre, kilogrc;.m~ second and ampere. The use of ampere as a fourth ':""'~
fundamental unit has been recommended by International Electro-technical Commission in July 1950. I
The dimensional equations for various quantities i:o terms of L, M, T and I are in Table '4'3.
There are .many expressions in the M K.S. system which contain the term 4n. This makes the
formulae a little awkward to work with. Eliminatioo. of 41t from all the formulae is impossible but
tTNITS, S'\'.STEMS, DiMBNSION8 AND STANbARDS 75
the aim is to transfer the term 4n to formulae which are not very important or where it may be
reasonably expected to occur as in volume of sphere. Thus rationalization means the elimination of
term 4rc from frequently used expressions by redefinition of certain units. This rationalization is
carried out by assuming the value of permeability of free space as 4rc x 10-7 ia place l 0- 7 as in the
unrationalised system.
The basis of rationalization is the conception that unit flux issues from the surface of a body
containing a unit charge. This flux is 411; in the case of unrationalized system. This fundamental
change redefines all the relationships.
TABLE 4·3
Rationalised M.K.S. A. System
l. Current I [I]
wol'k done
E = --·-----,----
3. Emf E charge
4. Resistance R £,,=/R
B
'l=__!lux
I
/.• 6. Flux Dtnsity area
H= mmf.
8. Magnetizing f ort'e JI length
1••
SI. Reluctance R
10. Inductance L
D=~ectric flux
12. El~ctdc flux density D area
14. Capacitance c
---···~·- .. - - - - - - - - - - - - - - - - · - - - - - - - ·---···--- ----~- ------------ -· ---~--
It should be emphasized again that rationalizati~n does not l}lean elimination of 4rc from all
formulae but only from important formulae which are frequently used. The term 471 disappears in
some re)ationshipc only to appear in another place in others. The advantage of rationalization is to
be judged ~Y its effects on.the relationships which are considered very important.
ELECTRICAL MEASUREMENTS AND MEAStriUNG lNSIRliMENTS
tf 16. SI Units. An international organization of which most of the advanced and .developing·
countries, including India, are members, called the General Conference of Weights and Measures
TABLE 4'4
Unit
Base Units
Symbol
l
!
Length metre m
I
Mass kilogramme kg
I
Time second s
Intensity of electric ampere A
current
Thermodynamic temM kelvin K
perature
Luminous intensity candela cd
*Amount of substance mole .mol
TABLE 4·5
Recommended Prefixes and Symols
DERIVED UNITS
TABLE 4·7
Recommended Multiples and SubmultipJes of Basic Units
I Mg tonne
Second (s) - ks ms µs ns - Use of min~
hour and day
is also · allowed
. A ~electlon'ofd~rived qits is given in Table 4·s. Some of the units are given special names.
The recommended· multipte-s.and subm:ultiple~ are also given.
TABLE 4'8
Force. kN mN v.N.
Work (and energy) ·01 MJ kJ mJ
Power ow MW kW mW µW nW pW
Pressure GN/m2 MN/m~ kN/m2 -
Velocity km/s
Electrical potential MV kV mV !AV
Electrical resistance MO kO
Nominal unit 109 100 ms 10-3 to~a 10-9 10-12
Multiplier
80 ELECTRiCAL MEASUREMENTS ANO MEASURING INSTRUMENTS
. - There are many othepinits which are not part of SI but are recognised by International
Committee for Weights and Measures'. They are listed in Table 4'9..
TABLE 4·9
. ·Example 4·1. If we .adopt new units of voltage and current such that one new unit voltage
= 100 volt and one·new unft of current= 100 ampere, how would the units of resistance, capacitance,
inductance be affected ? . .
Solution : Let the old units of resistance;. inductance, capacitance, voltage and current be
R, L, C, E and I respectively and the new units of resistance, inductance, capacitance, voltage and
current be R', V, C', E', and 1' respectively.
· (i) We have resistance :
. , E' . ·100 E E .
·and R =-·-= -------- =-=R
· I' 100 I I .
Thus the unit of resistance is ·equal to the old unit of resistance and, therefore, there is no
change.·
- E . , ·E' 100 E . E
(ii) Inductance~= -dl/dt- and . L =-dl'/dt dl/dt -L.
100 d!/dt
Thus the new unit of inductance is equal to the old unit of inductance ·and, therefore, there is
no change.
charge · It ·
(iii) Capacitance C and
emf· = E .
. · · . ·. . ·
.
Example 4·2. If we adopt new units of length, mass and time such that
One new unit of length. 10 !Petre. One new unit of mass=== l/IO kilogram.
·. UNlIS, sYs~EM.', DJMEN~IONS AND STANDARDS 81
One new unit of time= 100 second.
How would the units of force, power, energy, and current be affected 1
The unit of permittivity remains the same.
Solution : (i) The dimensions of force are [MLT-2],
:. New unitofcurrent=(+·r
12
' ( l/fD_)1'2( J[-)1 12 ( J~O )-l
· =10-2 old units ofcurrent=10-2 A.
Exa~pte 4·3, D.~rive the dime~sion_a~ equations for pole strength cu:~r.ent,. e.m.f., magnetic
flux, magnetic flux density, m m.f., magnehzmg force, reluctance, permea.nce', m the e.m. syuem of
units.
. '
Solution :. (i) The force exerted between two poles is F=. !-'-d 2
' . ~~
(viii.) Rll
c uctance=~-
m.m.f.
.uux
[:t] [µ-112 Ml-12 £11! r-1) . _ _
1 1
[R]= (q)f = .[µ172 _M112 £3/9 r-1]-[µ L ]
1
(lx) Perm eanee At:.:2 t
re1uc ance
[A]= [RJ-1 =[µL]
Exampfo 41'4. Derive the dimensions or (I) e.m.f., (ii) magnetic flux density. (ill) electric
flux density, (iv) current density~ (v) permeability; (vi) permittivity, (vii) resistivity and (viii) con-
ductivity in L, M~ T, I system of dimensions.
.
Soluhon: (f.) Emf= --
work-done--
charge
... [B]
[emf] [time]_ __ [r 1ML 2T- 3][TJ -[J-lMT-2]
[area] - [L2]
. . . . electric flux .charge
(111) Electric flux density=----· = - -
area · ·ar~
,. .'
. current
(iv) Current dens1ty=---
area
... [~]= lJJr=[IL·-~]
(v). . Flux -ffux density x area .·. (<f)] =[I"'lM r-2][L 2] =[rlML 2r-21
Mmf- turns X current [1]=[1]
mmf -[J2 M-1 L-2 f2]
Reluctance [.,] = [/]
= fiui II\ [r1Mi2r-21-
Jeng th
We have, Reluctance - bT
permea 11ty x area ... permea.b'l'
1 tty==
length
re1uc tance x area