Sergey N.

Makarov
Reinhold Ludwig
Stephen J. Bitar

Practical Electrical
Engineering
Second Edition
Practical Electrical Engineering
Sergey N. Makarov • Reinhold Ludwig •
Stephen J. Bitar

Practical Electrical
Engineering
Second Edition
Sergey N. Makarov Reinhold Ludwig
ECE Department ECE Department
Worcester Polytechnic Institute Worcester Polytechnic Institute
Worcester, WA, USA Worcester, MA, USA

Stephen J. Bitar
Worcester Polytechnic Institute
Worcester, MA, USA

ISBN 978-3-319-96691-5 ISBN 978-3-319-96692-2 (eBook)

https://doi.org/10.1007/978-3-319-96692-2

Library of Congress Control Number: 2018952468

© Springer Nature Switzerland AG 2016, 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or
part of the material is concerned, specifically the rights of translation, reprinting, reuse of
illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way,
and transmission or information storage and retrieval, electronic adaptation, computer software, or
by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt
from the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, express or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with
regard to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
To Antonina, Margot, and Juliette
Contents

1 From Physics to Electric Circuits . . . . . . . . . . . . . . . . . . . . . . 1

1.1 Electrostatics of Conductors . . . . . . . . . . . . . . . . . . . . . . 3
1.1.1 Charges, Coulomb Force, and Electric Field . . . . 3
1.1.2 Electric Potential and Electric Voltage . . . . . . . . . 4
1.1.3 Electric Voltage Versus Ground . . . . . . . . . . . . . 5
1.1.4 Equipotential Conductors . . . . . . . . . . . . . . . . . . 7
1.1.5 Use of Coulomb’s Law to Solve Electrostatic
Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2 Steady-State Current Flow and Magnetostatics . . . . . . . . . 11
1.2.1 Electric Current . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2.2 Difference Between Current Flow Model and
Electrostatics . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.2.3 Physical Model of an Electric Circuit . . . . . . . . . 13
1.2.4 Magnetostatics and Ampere’s Law . . . . . . . . . . . 14
1.2.5 Origin of Electric Power Transfer . . . . . . . . . . . . 16
1.3 Hydraulic and Fluid Mechanics Analogies . . . . . . . . . . . . 18
1.3.1 Hydraulic Analogies in the DC Steady State . . . . 18
1.3.2 Analogies for Alternating-Current (AC)
Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.3.3 Analogies for Semiconductor Circuit
Components . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Part I DC Circuits: General Circuit Theory—Operational

Amplifier
2 Major Circuit Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
2.1 Resistance: Linear Passive Circuit Element . . . . . . . . . . . 31
2.1.1 Circuit Elements Versus Circuit Components . . . 31
2.1.2 Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
2.1.3 υ-i Characteristic of the Resistance:
Open and Short Circuits . . . . . . . . . . . . . . . . . . 34
2.1.4 Power Delivered to the Resistance . . . . . . . . . . . 35
2.1.5 Finding Resistance of Ohmic Conductors . . . . . . 36

vii
Contents

2.1.6 Application Example: Power Loss

in Transmission Wires and Cables . . . . . . . . . . . 39
2.1.7 Physical Component: Resistor . . . . . . . . . . . . . . 41
2.1.8 Application Example: Resistive Sensors . . . . . . . 42
2.2 Nonlinear Passive Circuit Elements . . . . . . . . . . . . . . . . . 46
2.2.1 Resistance as a Model for the Load . . . . . . . . . . 46
2.2.2 Nonlinear Passive Circuit Elements . . . . . . . . . . 47
2.2.3 Static Resistance of a Nonlinear Element . . . . . . 48
2.2.4 Dynamic (Small-Signal) Resistance
of a Nonlinear Element . . . . . . . . . . . . . . . . . . . 49
2.2.5 Electronic Switch . . . . . . . . . . . . . . . . . . . . . . . 50
2.3 Independent Sources . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
2.3.1 Independent Ideal Voltage Source . . . . . . . . . . . . 52
2.3.2 Circuit Model of a Practical Voltage Source . . . . 54
2.3.3 Independent Ideal Current Source . . . . . . . . . . . . 55
2.3.4 Circuit Model of a Practical Current Source . . . . 57
2.3.5 Operation of the Voltage Source . . . . . . . . . . . . . 58
2.3.6 Application Example: DC Voltage Generator
with Permanent Magnets . . . . . . . . . . . . . . . . . . 59
2.3.7 Application Example: Chemical Battery . . . . . . . 61
2.4 Dependent Sources and Time-Varying Sources . . . . . . . . . 64
2.4.1 Dependent Versus Independent Sources . . . . . . . 64
2.4.2 Definition of Dependent Sources . . . . . . . . . . . . 64
2.4.3 Transfer Characteristics . . . . . . . . . . . . . . . . . . . 66
2.4.4 Time-Varying Sources . . . . . . . . . . . . . . . . . . . . 67
2.5 Ideal Voltmeter and Ammeter: Circuit Ground . . . . . . . . . 69
2.5.1 Ideal Voltmeter and Ammeter . . . . . . . . . . . . . . . 69
2.5.2 Circuit Ground: Fluid Mechanics Analogy . . . . . 70
2.5.3 Types of Electric Ground . . . . . . . . . . . . . . . . . . 71
2.5.4 Ground and Return Current . . . . . . . . . . . . . . . . 71
2.5.5 Absolute Voltage and Voltage Drop
Across a Circuit Element . . . . . . . . . . . . . . . . . . 72
3 Circuit Laws and Networking Theorems . . . . . . . . . . . . . . . . 89
3.1 Circuit Laws: Networking Theorems . . . . . . . . . . . . . . . . 91
3.1.1 Electric Network and Its Topology . . . . . . . . . . . 91
3.1.2 Kirchhoff’s Current Law (KCL) . . . . . . . . . . . . . 93
3.1.3 Kirchhoff’s Voltage Law (KVL) . . . . . . . . . . . . . 95
3.1.4 Power-Related Networking Theorems . . . . . . . . . 98
3.1.5 Port of a Network: Network Equivalence . . . . . . 99
3.2 Series and Parallel Network/Circuit Blocks . . . . . . . . . . . 100
3.2.1 Sources in Series and in Parallel . . . . . . . . . . . . . 100
3.2.2 Resistances in Series and in Parallel . . . . . . . . . . 102
3.2.3 Reduction of Resistive Networks . . . . . . . . . . . . 104
3.2.4 Voltage Divider Circuit . . . . . . . . . . . . . . . . . . . 105
3.2.5 Application Example: Voltage Divider
as a Sensor Circuit . . . . . . . . . . . . . . . . . . . . . . 107

viii
Contents

3.2.6 Application Example: Voltage Divider

as an Actuator Circuit . . . . . . . . . . . . . . . . . . . . 109
3.2.7 Current Limiter . . . . . . . . . . . . . . . . . . . . . . . . . 111
3.2.8 Current Divider Circuit . . . . . . . . . . . . . . . . . . . 111
3.2.9 Wheatstone Bridge . . . . . . . . . . . . . . . . . . . . . . 113
3.3 Superposition Theorem and Its Use . . . . . . . . . . . . . . . . . 115
3.3.1 Linear and Nonlinear Circuits . . . . . . . . . . . . . . 115
3.3.2 Superposition Theorem or Superposition
Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
3.3.3 Y (Wye) and Δ (Delta) Networks:
Use of Superposition . . . . . . . . . . . . . . . . . . . . . 120
3.3.4 T and Π Networks: Two-Port Networks . . . . . . . 122
3.3.5 General Character of Superposition
Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
4 Circuit Analysis and Power Transfer . . . . . . . . . . . . . . . . . . . 143
4.1 Nodal/Mesh Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
4.1.1 Importance of Circuit Simulators . . . . . . . . . . . . 145
4.1.2 Nodal Analysis for Linear Circuits . . . . . . . . . . . 145
4.1.3 Supernode . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149
4.1.4 Mesh Analysis for Linear Circuits . . . . . . . . . . . 150
4.1.5 Supermesh . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
4.2 Generator Theorems and Their Use . . . . . . . . . . . . . . . . . 154
4.2.1 Equivalence of Active One-Port Networks:
Method of Short/Open Circuit . . . . . . . . . . . . . . 154
4.2.2 Application Example: Reading and Using
Data for Solar Panels . . . . . . . . . . . . . . . . . . . . . 155
4.2.3 Source Transformation Theorem . . . . . . . . . . . . 156
4.2.4 Thévenin’s and Norton’s Theorems: Proof Without
Dependent Sources . . . . . . . . . . . . . . . . . . . . . . 159
4.2.5 Finding Thévenin and Norton Equivalents
and Using Them for Circuit Solution . . . . . . . . . 161
4.2.6 Application Example: Generating Negative
Equivalent Resistance . . . . . . . . . . . . . . . . . . . . 165
4.2.7 Short Summary of Circuit Analysis Methods . . . . 167
4.3 Power Transfer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
4.3.1 Maximum Power Transfer . . . . . . . . . . . . . . . . . 168
4.3.2 Maximum Power Efficiency . . . . . . . . . . . . . . . . 170
4.3.3 Application Example: Power Radiated
by a Transmitting Antenna . . . . . . . . . . . . . . . . . 171
4.3.4 Application Example: Maximum Power
Extraction from Solar Panel . . . . . . . . . . . . . . . . 172
4.4 Analysis of Nonlinear Circuits: Load Line
Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176
4.4.1 Iterative Method for Nonlinear Circuits . . . . . . . . 178

ix
Contents

5 Operational Amplifier and Amplifier Models . . . . . . . . . . . . . 201

5.1 Amplifier Operation and Circuit Models . . . . . . . . . . . . . 203
5.1.1 Amplifier Operation . . . . . . . . . . . . . . . . . . . . . 203
5.1.2 Application Example: Operational Amplifier
Comparator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
5.1.3 Amplifier Circuit Model . . . . . . . . . . . . . . . . . . 207
5.1.4 Ideal-Amplifier Model and First
Summing-Point Constraint . . . . . . . . . . . . . . . . . 209
5.2 Negative Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
5.2.1 Idea of the Negative Feedback . . . . . . . . . . . . . . 211
5.2.2 Amplifier Feedback Loop: Second
Summing-Point Constraint . . . . . . . . . . . . . . . . . 211
5.2.3 Amplifier Circuit Analysis Using Two
Summing-Point Constraints . . . . . . . . . . . . . . . . 213
5.2.4 Mathematics Behind the Second
Summing-Point Constraint . . . . . . . . . . . . . . . . . 217
5.2.5 Current Flow in the Amplifier Circuit . . . . . . . . . 218
5.2.6 Multiple-Input Amplifier Circuit:
Summing Amplifier . . . . . . . . . . . . . . . . . . . . . . 219
5.3 Amplifier Circuit Design . . . . . . . . . . . . . . . . . . . . . . . . 221
5.3.1 Choosing Proper Resistance Values . . . . . . . . . . 221
5.3.2 Model of a Whole Voltage Amplifier
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
5.3.3 Voltage Amplifier Versus Matched
Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
5.3.4 Cascading Amplifier Stages . . . . . . . . . . . . . . . . 227
5.3.5 Amplifier DC Imperfections and Their
Cancellation . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
5.3.6 DC-Coupled Single-Supply Amplifier:
Virtual-Ground Circuit . . . . . . . . . . . . . . . . . . . . 232
5.4 Difference and Instrumentation Amplifiers . . . . . . . . . . . . 234
5.4.1 Differential Input Signal to an Amplifier . . . . . . . 234
5.4.2 Difference Amplifier: Differential Gain
and Common-Mode Gain . . . . . . . . . . . . . . . . . 235
5.4.3 Application Example: Instrumentation
Amplifier . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
5.4.4 Instrumentation Amplifier in Laboratory . . . . . . . 240
5.5 General Feedback Systems . . . . . . . . . . . . . . . . . . . . . . . 242
5.5.1 Signal-Flow Diagram of a Feedback
System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
5.5.2 Closed-Loop Gain and Error Signal . . . . . . . . . . 242
5.5.3 Application of General Theory to Voltage
Amplifiers with Negative Feedback . . . . . . . . . . 244
5.5.4 Voltage, Current, Transresistance, and
Transconductance Amplifiers with the Negative
Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245

x
Contents

Part II Transient Circuits

6 Dynamic Circuit Elements . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
6.1 Static Capacitance and Inductance . . . . . . . . . . . . . . . . . . 273
6.1.1 Capacitance, Self-Capacitance, and Capacitance
to Ground . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
6.1.2 Application Example: ESD . . . . . . . . . . . . . . . . 275
6.1.3 Parallel-Plate Capacitor . . . . . . . . . . . . . . . . . . . 276
6.1.4 Circuit Symbol: Capacitances in Parallel
and in Series . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
6.1.5 Application Example: How to Design
a 1-μF Capacitor? . . . . . . . . . . . . . . . . . . . . . . . 279
6.1.6 Application Example: Capacitive
Touchscreens . . . . . . . . . . . . . . . . . . . . . . . . . . 282
6.1.7 Self-Inductance (Inductance) and Mutual
Inductance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
6.1.8 Inductance of a Solenoid With and Without
Magnetic Core . . . . . . . . . . . . . . . . . . . . . . . . . 285
6.1.9 Circuit Symbol: Inductances in Series
and in Parallel . . . . . . . . . . . . . . . . . . . . . . . . . . 287
6.1.10 Application Example: How to Design
a 1-mH Inductor? . . . . . . . . . . . . . . . . . . . . . . . 288
6.2 Dynamic Behavior of Capacitance and Inductance . . . . . . 290
6.2.1 Set of Passive Linear Circuit Elements . . . . . . . . 290
6.2.2 Dynamic Behavior of Capacitance . . . . . . . . . . . 290
6.2.3 Dynamic Behavior of Inductance . . . . . . . . . . . . 293
6.2.4 Instantaneous Energy and Power
of Dynamic Circuit Elements . . . . . . . . . . . . . . . 295
6.2.5 DC Steady State . . . . . . . . . . . . . . . . . . . . . . . . 296
6.2.6 Behavior at Very High Frequencies . . . . . . . . . . 297
6.3 Application Circuits Highlighting Dynamic
Behavior . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
6.3.1 Bypass Capacitor . . . . . . . . . . . . . . . . . . . . . . . 299
6.3.2 Blocking Capacitor . . . . . . . . . . . . . . . . . . . . . . 301
6.3.3 Decoupling Inductor . . . . . . . . . . . . . . . . . . . . . 301
6.3.4 Amplifier Circuits With Dynamic Elements:
Miller Integrator . . . . . . . . . . . . . . . . . . . . . . . . 302
6.3.5 Compensated Miller Integrator . . . . . . . . . . . . . . 303
6.3.6 Differentiator and Other Circuits . . . . . . . . . . . . 304
7 Transient Circuit Fundamentals . . . . . . . . . . . . . . . . . . . . . . . 319
7.1 RC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
7.1.1 Energy-Release Capacitor Circuit . . . . . . . . . . . . 322
7.1.2 Time Constant of the RC Circuit
and Its Meaning . . . . . . . . . . . . . . . . . . . . . . . . 324
7.1.3 Continuity of the Capacitor Voltage . . . . . . . . . . 325

xi
Contents

7.1.4 Application Example: Electromagnetic

Railgun . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
7.1.5 Application Example: Electromagnetic
Material Processing . . . . . . . . . . . . . . . . . . . . . . 328
7.1.6 Application Example: Digital Memory Cell . . . . 329
7.1.7 Energy-Accumulating Capacitor Circuit . . . . . . . 330
7.2 RL Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
7.2.1 Energy-Release Inductor Circuit . . . . . . . . . . . . . 333
7.2.2 Continuity of the Inductor Current . . . . . . . . . . . 336
7.2.3 Energy-Accumulating Inductor Circuit . . . . . . . . 337
7.2.4 Energy-Release RL Circuit with the Voltage
Supply . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
7.2.5 Application Example: Laboratory Ignition
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
7.3 Switching RC Oscillator . . . . . . . . . . . . . . . . . . . . . . . . . 342
7.3.1 About Electronic Oscillators . . . . . . . . . . . . . . . 342
7.3.2 Bistable Amplifier Circuit with the Positive
Feedback . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 342
7.3.3 Triggering . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
7.3.4 Switching RC Oscillator . . . . . . . . . . . . . . . . . . 345
7.3.5 Oscillation Frequency . . . . . . . . . . . . . . . . . . . . 346
7.3.6 Circuit Implementation: 555 Timer . . . . . . . . . . . 347
7.4 Single-Time-Constant (STC) Transient Circuits . . . . . . . . 349
7.4.1 Circuits with Resistances and Capacitances . . . . . 349
7.4.2 Circuits with Resistances and Inductances . . . . . . 351
7.4.3 Example of a Non-STC Transient Circuit . . . . . . 353
7.4.4 Example of an STC Transient Circuit . . . . . . . . . 354
7.4.5 Method of Thévenin Equivalent
and Application Example: Circuit
with a Bypass Capacitor . . . . . . . . . . . . . . . . . . 355
7.5 Description of the Second-Order Transient Circuits . . . . . 358
7.5.1 Types of Second-Order Transient Circuits . . . . . . 358
7.5.2 Series-Connected Second-Order RLC
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
7.5.3 Initial Conditions in Terms of Circuit
Current and Capacitor Voltage . . . . . . . . . . . . . . 361
7.5.4 Step Response and Choice of the Independent
Function . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362
7.5.5 Parallel Connected Second-Order RLC
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
7.6 Step Response of the Series RLC Circuit . . . . . . . . . . . . . 366
7.6.1 General Solution of the Second-order ODE . . . . . 366
7.6.2 Derivation of the Complementary Solution:
Method of Characteristic Equation . . . . . . . . . . . 366
7.6.3 Finding Integration Constants . . . . . . . . . . . . . . 368

xii
Contents

7.6.4 Solution Behavior for Different Damping

Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 368
7.6.5 Overshoot and Rise Time . . . . . . . . . . . . . . . . . . 369
7.6.6 Application Example: Nonideal Digital
Waveform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370

Part III AC Circuits

8 Steady-State AC Circuit Fundamentals . . . . . . . . . . . . . . . . . 401
8.1 Harmonic Voltage and Current: Phasor . . . . . . . . . . . . . . 403
8.1.1 Harmonic Voltages and Currents . . . . . . . . . . . . 403
8.1.2 Phase: Leading and Lagging . . . . . . . . . . . . . . . 405
8.1.3 Application Example: Measurements
of Amplitude, Frequency, and Phase . . . . . . . . . . 408
8.1.4 Definition of a Phasor . . . . . . . . . . . . . . . . . . . . 408
8.1.5 From Real Signals to Phasors . . . . . . . . . . . . . . . 410
8.1.6 From Phasors to Real Signals . . . . . . . . . . . . . . . 411
8.1.7 Polar and Rectangular Forms: Phasor
Magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
8.1.8 Operations with Phasors and Phasor
Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413
8.1.9 Shorthand Notation for the Complex
Exponent . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 416
8.2 Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417
8.2.1 The Concept of Impedance . . . . . . . . . . . . . . . . 417
8.2.2 Physical Meaning of Impedance . . . . . . . . . . . . . 419
8.2.3 Magnitude and Phase of Complex
Impedance . . . . . . . . . . . . . . . . . . . . . . . . . . . . 420
8.2.4 Application Example: Impedance
of a Human Body . . . . . . . . . . . . . . . . . . . . . . . 422
8.3 Principles of AC Circuit Analysis . . . . . . . . . . . . . . . . . . 423
8.3.1 AC Circuit Analysis: KVL, KCL,
and Equivalent Impedances . . . . . . . . . . . . . . . . 423
8.3.2 Complete Solution for an AC Circuit: KVL
and KCL on Phasor Diagram . . . . . . . . . . . . . . . 424
8.3.3 Source Transformation . . . . . . . . . . . . . . . . . . . 425
8.3.4 Thévenin and Norton Equivalent Circuits . . . . . . 427
8.3.5 Summary of AC Circuit Analysis at a Single
Frequency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 429
8.3.6 Multifrequency AC Circuit Analysis:
Superposition Theorem . . . . . . . . . . . . . . . . . . . 429
9 Filter Circuits: Frequency Response, Bode Plots,
and Fourier Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445
9.1 First-Order Filter Circuits and Their Combinations . . . . . . 447
9.1.1 RC Voltage Divider as an Analog Filter . . . . . . . 447

xiii
Contents

9.1.2 Half-Power Frequency and Amplitude

Transfer Function . . . . . . . . . . . . . . . . . . . . . . . 452
9.1.3 Bode Plot, Decibel, and Roll-Off . . . . . . . . . . . . 453
9.1.4 Phase Transfer Function and Its Bode Plot . . . . . 456
9.1.5 Complex Transfer Function: Cascading Filter
Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
9.1.6 RL Filter Circuits . . . . . . . . . . . . . . . . . . . . . . . 460
9.2 Bandwidth of an Operational Amplifier . . . . . . . . . . . . . . 463
9.2.1 Bode Plot of the Open-Loop Amplifier Gain . . . . 463
9.2.2 Unity-Gain Bandwidth Versus Gain-Bandwidth
Product . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 464
9.2.3 Model of the Open-Loop AC Gain . . . . . . . . . . . 465
9.2.4 Model of the Closed-Loop AC Gain . . . . . . . . . . 466
9.2.5 Application Example: Finding Bandwidth
of an Amplifier Circuit . . . . . . . . . . . . . . . . . . . 467
9.2.6 Application Example: Selection of an Amplifier IC
for Proper Frequency Bandwidth . . . . . . . . . . . . 468
9.3 Introduction to Continuous and Discrete Fourier
Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470
9.3.1 Meaning and Definition of Fourier
Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 470
9.3.2 Mathematical Properties of Fourier
Transform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 472
9.3.3 Discrete Fourier Transform and Its
Implementation . . . . . . . . . . . . . . . . . . . . . . . . . 473
9.3.4 Sampling Theorem . . . . . . . . . . . . . . . . . . . . . . 475
9.3.5 Applications of Discrete Fourier Transform . . . . . 476
9.3.6 Application Example: Numerical
Differentiation via the FFT . . . . . . . . . . . . . . . . 476
9.3.7 Application Example: Filter Operation
for an Input Pulse Signal . . . . . . . . . . . . . . . . . . 478
9.3.8 Application Example: Converting Computational
Electromagnetic Solution from Frequency Domain
to Time Domain . . . . . . . . . . . . . . . . . . . . . . . . 479
10 Second-Order RLC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . 493
10.1 Theory of Second-Order Resonant RLC Circuits . . . . . . . 495
10.1.1 Self-Oscillating Ideal LC Circuit . . . . . . . . . . . . 495
10.1.2 Series Resonant Ideal LC Circuit . . . . . . . . . . . . 497
10.1.3 Series Resonant RLC Circuit: Resonance
Condition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 498
10.1.4 Quality Factor Q of the Series Resonant RLC
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
10.1.5 Bandwidth of the Series Resonant RLC
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 502
10.1.6 Parallel Resonant RLC Circuit: Duality . . . . . . . 505

xiv
Contents

10.2 Construction of Second-Order RLC Filters . . . . . . . . . . . 508

10.2.1 Second-Order Band-Pass RLC Filter . . . . . . . . . 508
10.2.2 Second-Order Low-Pass RLC Filter . . . . . . . . . . 511
10.2.3 Second-Order High-Pass RLC Filter . . . . . . . . . . 512
10.2.4 Second-Order Band-Reject RLC Filter . . . . . . . . 514
10.2.5 Second-Order RLC Filters Derived
from the Parallel RLC Circuit . . . . . . . . . . . . . . 516
10.3 RLC Circuits for Near-Field Communications
and Proximity Sensors . . . . . . . . . . . . . . . . . . . . . . . . . . 518
10.3.1 Near-Field Wireless Link . . . . . . . . . . . . . . . . . . 518
10.3.2 Transmitter Circuit . . . . . . . . . . . . . . . . . . . . . . 519
10.3.3 Receiver Circuit . . . . . . . . . . . . . . . . . . . . . . . . 520
10.3.4 Application Example: Near-Field Wireless
Link in Laboratory . . . . . . . . . . . . . . . . . . . . . . 522
10.3.5 Application Example: Proximity Sensors . . . . . . 523
11 AC Power and Power Distribution . . . . . . . . . . . . . . . . . . . . . 535
11.1 AC Power Types and Their Meaning . . . . . . . . . . . . . . . . 537
11.1.1 Instantaneous AC Power . . . . . . . . . . . . . . . . . . 537
11.1.2 Time-averaged AC Power . . . . . . . . . . . . . . . . . 538
11.1.3 Application Example: rms Voltages and AC
Frequencies Around the World . . . . . . . . . . . . . . 540
11.1.4 rms Voltages for Arbitrary Periodic AC
Signals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 541
11.1.5 Average AC Power in Terms of Phasors:
Power Angle . . . . . . . . . . . . . . . . . . . . . . . . . . . 543
11.1.6 Average Power for Resistor, Capacitor,
and Inductor . . . . . . . . . . . . . . . . . . . . . . . . . . . 544
11.1.7 Average Power, Reactive Power, and Apparent
Power . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 545
11.1.8 Power Triangle . . . . . . . . . . . . . . . . . . . . . . . . . 547
11.1.9 Application Example: Wattmeter . . . . . . . . . . . . 550
11.2 Power Factor Correction: Maximum Power Efficiency
and Maximum Power Transfer . . . . . . . . . . . . . . . . . . . . 551
11.2.1 Power Factor Correction . . . . . . . . . . . . . . . . . . 551
11.2.2 Application Example: Automatic Power Factor
Correction System . . . . . . . . . . . . . . . . . . . . . . . 554
11.2.3 Principle of Maximum Power Efficiency
for AC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . 554
11.2.4 Principle of Maximum Power Transfer
for AC Circuits . . . . . . . . . . . . . . . . . . . . . . . . . 556
11.3 AC Power Distribution: Balanced Three-Phase Power
Distribution System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 558
11.3.1 AC Power Distribution Systems . . . . . . . . . . . . . 558
11.3.2 Phase Voltages: Phase Sequence . . . . . . . . . . . . . 559

xv
Contents

11.3.3
Wye (Y) Source and Load Configurations
for Three-Phase Circuits . . . . . . . . . . . . . . . . . . . 561
11.3.4 Application: Examples of Three-Phase Source
and the Load . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
11.3.5 Solution for the Balanced Three-Phase Wye-Wye
Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
11.3.6 Removing the Neutral Wire in Long-Distance
Power Transmission . . . . . . . . . . . . . . . . . . . . . . 565
11.4 Power in Balanced Three-Phase Systems: Delta-connected
Three-Phase Circuits . . . . . . . . . . . . . . . . . . . . . . . . . . . 568
11.4.1 Instantaneous Power . . . . . . . . . . . . . . . . . . . . . 568
11.4.2 Average Power, Reactive Power,
and Apparent Power . . . . . . . . . . . . . . . . . . . . . 570
11.4.3 Application Example: Material Consumption
in Three-Phase Systems . . . . . . . . . . . . . . . . . . . 570
11.4.4 Balanced Delta-Connected Load . . . . . . . . . . . . . 572
11.4.5 Balanced Delta-Connected Source . . . . . . . . . . . 572
12 Electric Transformer and Coupled Inductors . . . . . . . . . . . . . 589
12.1 Ideal Transformer as a Linear Passive Circuit
Element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
12.1.1 Electric Transformer . . . . . . . . . . . . . . . . . . . . . 591
12.1.2 Ideal Open-Circuited Transformer:
Faraday’s Law of Induction . . . . . . . . . . . . . . . . 591
12.1.3 Appearance of Transformer Currents . . . . . . . . . 595
12.1.4 Ampere’s Law . . . . . . . . . . . . . . . . . . . . . . . . . 595
12.1.5 Ideal Loaded Transformer . . . . . . . . . . . . . . . . . 596
12.1.6 Ideal Transformer Versus Real Transformer:
Transformer Terminology . . . . . . . . . . . . . . . . . 598
12.1.7 Mechanical Analogies of a Transformer . . . . . . . 601
12.2 Analysis of Ideal Transformer Circuits . . . . . . . . . . . . . . 602
12.2.1 Circuit with a Transformer in the Phasor
Form . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602
12.2.2 Referred (Or Reflected) Source Network
in the Secondary Side . . . . . . . . . . . . . . . . . . . . 602
12.2.3 Referred (Or Reflected) Load Impedance
to the Primary Side . . . . . . . . . . . . . . . . . . . . . . 604
12.2.4 Transformer as a Matching Circuit . . . . . . . . . . . 605
12.2.5 Application Example: Electric Power Transfer
via Transformers . . . . . . . . . . . . . . . . . . . . . . . . 607
12.3 Some Useful Transformers . . . . . . . . . . . . . . . . . . . . . . . 610
12.3.1 Autotransformer . . . . . . . . . . . . . . . . . . . . . . . . 610
12.3.2 Multiwinding Transformer . . . . . . . . . . . . . . . . . 611
12.3.3 Center-Tapped Transformer: Single-Ended
to Differential Transformation . . . . . . . . . . . . . . 612
12.3.4 Current Transformer . . . . . . . . . . . . . . . . . . . . . 614

xvi
Contents

12.4 Real-Transformer Model . . . . . . . . . . . . . . . . . . . . . . . . 616

12.4.1 Model of a Nonideal Low-Frequency
Transformer . . . . . . . . . . . . . . . . . . . . . . . . . . . 616
12.4.2 Model Parameters and Their Extraction . . . . . . . 616
12.4.3 Analysis of Nonideal Transformer Model . . . . . . 618
12.4.4 Voltage Regulation and Transformer
Efficiency . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 621
12.4.5 About High-Frequency Transformer
Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 622
12.5 Model of Coupled Inductors . . . . . . . . . . . . . . . . . . . . . . 624
12.5.1 Model of Two Coupled Inductors . . . . . . . . . . . . 624
12.5.2 Analysis of Circuits with Coupled
Inductors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 625
12.5.3 Coupling Coefficient . . . . . . . . . . . . . . . . . . . . . 628
12.5.4 Application Example: Wireless Inductive
Power Transfer . . . . . . . . . . . . . . . . . . . . . . . . . 630
12.5.5 Application Example: Coupling of Nearby
Magnetic Radiators . . . . . . . . . . . . . . . . . . . . . . 634

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 651

xvii
Chapter 1

Chapter 1: From Physics to Electric Circuits

Overview
Prerequisites:
- Knowledge of university physics: electricity and magnetism
Objectives of Section 1.1:
- Show that the electric voltage and the electric potential may be treated as two
equivalent quantities
- Define the electric voltage—work per unit charge—in the form of a line integral and
show its independence on the integration path for conservative fields
- Relate voltage to the potential energy of the electric field
- Introduce three-dimensional potential distributions and realize the guiding function
of metal wires
- Formulate and understand major conditions of electrostatics of conductors
- Visualize surface charge distributions in the electrostatic case
Objectives of Section 1.2:
- Introduce electric current density as a function of the applied electric field
- Visualize steady-state current flow in a single conductor along with the associated
electric potential/voltage distribution
- Visualize electric and magnetic-field distributions for a two-wire DC transmission
line
- Obtain initial exposure to the Poynting vector
- Realize that electric power is transferred via Poynting vector even in DC circuits
- Indicate a path toward circuit problems where the field effects become important
Objectives of Section 1.3:
- Review basic hydraulic (fluid mechanics) analogies for DC circuit elements
- Present major hydraulic analogies for dynamic circuit elements in AC circuits
- Briefly discuss hydraulic analogies for semiconductor components
Application Examples:
- Human body subject to applied voltage
- Human body in an external electric field

© Springer Nature Switzerland AG 2019 I-1

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_1
Chapter 1 From Physics to Electric Circuits

Keywords:
Electricity, Electric field intensity, Electric field, Electric field magnitude, Lines of force, Electric
potential, Electric voltage, Line integral, Contour integral, Conservative field, Potential energy of the
electric field, Voltage drop, Voltage difference, Ground reference, Neutral conductor, Common
conductor, Voltage versus ground, Equipotential lines, Volumetric charge density, Surface charge
density, Gauss’ theorem, Equipotential surface, Self-capacitance, Electrostatic discharge, Effect of
electrostatic discharge on integrated circuits, Boundary element method, Electric current density,
Material conductivity, Transmission line, Direct current (DC), Electric load, Ideal wire, Kirchhoff’s
voltage law (KVL), Magnetic field, Magnetic-field intensity, Ampere’s law, Cross (vector) product,
Poynting vector, Poynting theorem, Wireless communications, Wireless power transfer, Fluid
mechanics analogy of an electric circuit, Hydraulic analogy of an electric circuit, Voltage source
(hydraulic analogy), Resistance (hydraulic analogy), Current source (hydraulic analogy), Capacitance
(hydraulic analogy), Inductance (hydraulic analogy), Electric transformer (hydraulic analogy), NMOS
transistor (hydraulic analogy), Bipolar junction transistor (hydraulic analogy)

I-2
Chapter 1 Section 1.1: Electrostatics of Conductors

Section 1.1 Electrostatics of Conductors

This introductory chapter is optional in the sense that the reader does not need its content
as a prerequisite for the subsequent chapters. The aim of this chapter is to illustrate why
electric circuits trace their origin to electromagnetic fields. The chapter highlights several
field concepts which form the theoretical foundation of electric circuits. At the same time
it makes clear why, for the majority of electric circuits, the electric and magnetic fields are
often ignored without affecting the final results. When this is the case, the electric circuits
and components follow useful and simple hydraulic analogies discussed below.

1.1.1 Charges, Coulomb Force, and Electric Field

Electric Charges
The smallest electric charge is known as the elemental charge of an electron,
q ¼1:60218  1019 C (coulombs). In electrical engineering, we deal with much
larger charges, which, for this reason, are assumed to be infinitely divisible. There are no
positive movable charges in metal conductors. Therefore, when we talk about positive
charges, it is implied that we have a lack of electrons at a certain location, e.g., at the surface.
Oppositely, the negative charge is the excess of electrons at a certain location.

Definition of the Electric Field

Electrostatics plays a key role in explaining the operations of electric capacitors and all
semiconductor devices. The word electricity is derived from the Greek word for amber.
Probably Thales of Miletus was the first who discovered, about 600 B.C., that amber,
when rubbed, attracts light objects. An electrostatic force acting on a charge q is known as
the Coulomb force. This Coulomb force is a vector; it is measured in newtons (or N)

~
F ¼ q~
E ½N ð1:1Þ

Equation (1.1) is the definition of the electric field intensity vector, ~ E, often called the
electric field. This electric field is created by other (remote or nearby) charges. In the
general case, the electric field exists both in free space and within material objects,
whether conductors or dielectrics. The electric field ~ E is measured in volts (V) per
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  
meter (V/m). The field magnitude, E ¼ E þ E þ E ¼  ~
2
x
2
y z
2
E, has the same units. From
Eq. (1.1),

1 N  1m 1 J
1V ¼ ¼ ð1:2Þ
1 C 1 C

I-3
Chapter 1 From Physics to Electric Circuits

lines of force

+Q
+ + +
plus (+) terminal A
+ + + +
E

+
voltage power
supply 0 VAA’
- x

- - A’
- -
minus (-) terminal
- - -
-Q

Fig. 1.1. Electric field emanating from a voltage supply with open-circuit output terminals.

Electric Field of a Laboratory Power Source

As an example, we consider an electric field generated by a laboratory voltage source
turned on. Figure 1.1 shows the realistic electric field distribution modeled numerically.
The power supply has two output terminals shaped as two metal cylinders. The plastic
cover is ignored. Nothing is connected to the terminals yet. Still, the power supply
already performs a “charge separation”: the plus (+) terminal is charged positively
(total charge is +Q), whereas the minus () terminal is charged negatively (total charge
is Q). Those charges are schematically shown in Fig. 1.1. As a result, an electric field is
created. The electric field in Fig. 1.1 is directed along particular lines, which we call lines
of force. This electric field surrounds the power supply terminals. Every line of force
starts at the positive terminal and ends at the negative terminal. The strength of the
electric field everywhere in space is linearly proportional to the supply voltage as studied
next. However, the field shape always remains exactly the same.

1.1.2 Electric Potential and Electric Voltage

The electric potential φ measured in volts (V) and electric voltage V measured in volts
(V) are two identical quantities once they refer to the same observation point A and to the
same reference point A0 . Both terms may be used; the electric potential is frequently
denoted by V. The potential is more common in physics. Work is done against the electric
forces when a charge is moved in an electric field. The electric potential or electric
voltage V AA0 between points A (#1) and A0 (#2) is work in joules per coulomb (per unit
charge) to bring a positive charge from reference point A0 (#2) to observation point A (#1),
i.e., against the electric field—see Fig. 1.1. The work per unit charge over a short straight
vector distance d~ l is

I-4
Chapter 1 Section 1.1: Electrostatics of Conductors

1
 ~F  d~ E  d~
l ¼ ~ l ¼ Edl cos θ ð1:3Þ
q
   
 
l, E ¼  ~
E and d~
where θ is the angle between ~ E  , and dl ¼ d~
l . The total work or V AA0 is
the sum of all such small contributions conventionally written in the form of an integral
0
ðA ð
A

E  d~
V AA0 ¼  ~ l¼ E  d~
~ l ð1:4Þ
0 A
A

The integral in Eq. (1.4) is a line integral, also called a contour integral. In the general
case, it is evaluated along a curve connecting points A and A0 . In the particular case of
Fig. 1.1, this curve is just a straight line.

Exercise 1.1: Assume for simplicity that the electric field along the line of force from A to
A0 in Fig. 1.1 is strictly uniform and has the magnitude of 50 V/m. The line length is
0.02 m. Find voltage (or potential) V AA0 .
ðA ð 0:02m
~ ~ V
Answer: V AA0 ¼  0 E  d l ¼  50  cos π  dl ¼ 50  0:02 m ¼ 1 V.
A 0 m

The electrostatic field (and any slowly varying electric field) is called a conservative
field. There are two equivalent definitions of a conservative field:
1. Electric voltage or electric potential V AA0 is path independent; it only depends on the
position of A and A0 , but not on the shape of the curve between A and A0 .
2. The line integral in Eq. (1.4) over any closed contour is zero.
The equivalence of these definitions is proved by treating two different integration
contours between A and A0 as two parts of one closed contour. The independence of the
integration path suggests that the voltage is equal to the potential energy of a unit charge
in the electric field. Strictly speaking, it is the change in the potential energy.

1.1.3 Electric Voltage Versus Ground

The voltage between two arbitrary points V AA0 (e.g., between two terminals of a resistor) is
a convenient measure when analyzing electric circuits with discrete circuit components.
In this case, it is called the voltage drop (or voltage difference) across a circuit element. At
the same time, it is equally convenient to define the “global” or absolute voltage between
an arbitrary point in space A and some fixed point A0 , which is assigned the voltage value
zero. This fixed point (or the set of points) is called the ground reference. In general, the
ground reference may be chosen arbitrarily. In physics of localized conductors, it is

I-5
Chapter 1 From Physics to Electric Circuits

customary to choose the ground at infinity. In electrical engineering, the ground reference is
either the physical (earth) ground or some neutral (common) conductor assigned to zero
volts. Thus, the absolute voltage versus ground denoted by V ð~ rÞ is still defined by Eq. (1.4)
where point A is now characterized by the position vector ~ r. By definition, it becomes zero
when ~ r approaches ground, i.e., A0 . The equivalent representation of Eq. (1.4) for
conservative fields may be shown to be

~
E ð~
rÞ ¼ gradV ð~
rÞ ¼ ∇V ð~
rÞ ð1:5Þ

Thus, the electrostatic field is expressed as the gradient of the electric potential or of
the (absolute) electric voltage everywhere in space. In other words, it means the
electric field does not have closed loops, but starts and ends at the charges. Equation
(1.5) is of significant value since it replaces a complicated vector ~
E by the single scalar
voltage V.

Exercise 1.2: Electric voltage/potential with respect to ground is given in Cartesian

coordinates by V ð~
rÞ ¼ y ½V. Determine the electric field everywhere in space.
Answer: E y ¼ 1V=m, E x ¼ E z ¼ 0.

As an example, we choose the x-axis in Fig. 1.1 as the ground reference. The positive
supply terminal is chosen to have a voltage equal to +0.5 V versus ground, and the
negative terminal is assigned a voltage equal to 0.5 V versus ground. Those values will
uniquely determine charges Q in Fig. 1.1. The function V ð~ rÞ is now plotted using the
lines of equal potential, or equipotential lines. The result is shown in Fig. 1.2. It will be
proved next that the surface of any metal (or other) conductor in electrostatics is an
equipotential surface. All points on this surface have the same value of the electric
potential: +0.5 V for the plus terminal and 0.5 V for the minus terminal in Fig. 1.2.
Using Eq. (1.5) it can be verified that the equipotential lines and the lines of force are
always perpendicular to each other; you can see this in Fig. 1.2.

I-6
Chapter 1 Section 1.1: Electrostatics of Conductors

y 0.1 V

lines of force
0.2 V
0.3 V
+Q 0.4 V
+ + +
plus (+) terminal A B
+ + + +
E
voltage power 0V C
supply 0
x

- - A’
- -
minus (-) terminal B
B’ equipotential lines

-Q -0.4 V
-0.3 V
-0.2 V

-0.1 V

Fig. 1.2. Electric field and electric potential (electric voltage) of a 1-V voltage supply. Equipo-
tential lines are thin solid curves, while the lines of force are the thicker curves.

Exercise 1.3: In Fig. 1.2, determine voltage differences: V BB0 , V A0 A , V AB , V AC , V B0 C .

Answer: V BB0 ¼ 1 V, V A0 A ¼ 1 V, V AB ¼ 0 V, V AC ¼ 0:5 V, V B0 C ¼ 0:5 V.

1.1.4 Equipotential Conductors

Consider a conductor, which is characterized by a sufficient number of free charges. The
conductor is subject to an applied electric voltage, or to an applied electric field, or to
those effects combined. The charge density in the conductor is the difference between the
concentration of ions (positive charges) and electrons (negative charges) multiplied by
the charge of an electron. The charge density can be either volumetric, measured in C/m3,
or surface, measured in C/m2. The following is true when dealing with electrostatics:
1. The electric field everywhere within the conductor is zero, ~ E¼~ 0. Otherwise, the
Coulomb force given by Eq. (1.1) will act on free charges and cause permanent
charge motion (current flow), which is impossible.
2. The volumetric charge density everywhere within the conductor is zero. If this were
not true, then according to Gauss’ theorem, there would be a nonzero electric field
within the conductor, which is impossible based on statement #1.
3. The surface charge density is not zero. In fact, the surface charge is distributed so as
to assure that statement #1 is satisfied.
4. For the electric field given by Eq. (1.5), the tangential component of the electric
field ~Et is continuous across an interface. Since ~ Et ¼ ~0 inside the conductor, ~ Et

I-7
Chapter 1 From Physics to Electric Circuits

must be zero over the entire surface of the conductor too. This is seen in Figs. 1.1
and 1.2 where the lines of force are perpendicular to the conductor surface.
5. Since ~Et ¼ ~0 on the conductor surface, any line integral between two points on this
surface is zero. Consequently the potential or voltage remains the same for any point
on this surface. The conductor surface is thus an equipotential surface.
These statements have an immediate practical application. Consider two conductors
(wires) attached to the power supply terminals as seen in Fig. 1.3.

+Q’>Q
0.5 V
plus (+) terminal 0.5 V 0.5 V
0.5 V
+

+
voltage power
supply 1V 1V 1V
- - -
-0.5 V
minus (-) terminal -0.5 V -0.5 V
-0.5 V
-Q’<-Q

Fig. 1.3. Voltage source from Fig. 1.2 with two wires connected. There is still no current flow.

Everywhere along the upper wire, the voltage is +0.5 V with respect to ground.
Furthermore, along the lower wire, the voltage is 0.5 V with respect to ground.
Everywhere in space the voltage difference between the two wires is therefore 1 V. The
0
wires may be extremely long. The charge Q required to maintain the corresponding
voltage difference of 1 V increases when the combined area of the metal conductors
increases. The conducting wires thus “guide” the electric field to a remote point. Without
the attached wires, the field would be spread out in space as seen in Figs. 1.1 and 1.2.

Exercise 1.4: In Fig. 1.3, two wires happen to be very close to each other at a certain
location; they are separated by 1 mm. What is approximately the electric field strength at
this location?
Answer: On the order of 1000 V/m, the wire isolation has little influence.

Exercise 1.5: In Fig. 1.3, a human body is in contact with the upper wire. What is the
body’s voltage versus ground set at 0 V?
Answer: Since the human body is still a conductor, its voltage is +0.5 V.

I-8
Chapter 1 Section 1.1: Electrostatics of Conductors

1.1.5 Use of Coulomb’s Law to Solve Electrostatic Problems

The theory of electrostatics solely relies upon the distribution of surface charges, since no
other charges exist. An important case is a conductor subject to voltage V applied from
one terminal of the voltage source; the connection position does not matter. The other
terminal is usually located very far away, ideally at infinity; it is customarily assigned a
value of 0 V. As a result, the conductor acquires an extra positive charge Q if V is positive.
The quantities of interest are the value of Q itself and the resulting surface charge
distribution. The ratio Q/V is the self-capacitance of the conductor. Consider a compli-
cated conductor—a human body. The idea of the solution is simple and elegant.
The entire body surface is divided into many small elements with a constant
charge density each. We assign an unknown charge qi to every such element with number
i (i ¼ 1, . . . , N ) and center position ~
ri . Charge qi is very similar to the point charge.
It follows from Coulomb’s law that it generates the electric potential in space given by

qi
V ð~
rÞ ¼ ð1:6Þ
4πε0 j~
r ~
ri j

Here, ε0 ¼ 8:85419  1012 F=m is the electric permittivity of air. The net voltage of the
jth element is the sum of all such voltage contributions, i.e.,

  X N
q
V ~
rj ¼  i  ¼ 1 V, j ¼ 1, ::, N ð1:7Þ
i¼1 4πε0 ~ ri 
rj ~

Equations (1.7) forms a system of N equations for N unknown charges. The diagonal
terms need a special treatment. By solving this algebraic system of equations using linear
algebra, we obtain the unknown charges; their sum is the net charge Q. Figure 1.4 shows
the surface charge distribution found this way for two human subjects. Emphasize that the
charge (and the strongest electric field) concentrates at the sharpest parts of the body:
elbows, hands, feet, and the head. The total excess body charge Q in both cases is
approximately 50  1012 C. This is a very small charge; the same charge is stored in
a 50 pF capacitor at 1 V. The method of this example is widely used in electrostatic
simulations including modeling electrostatic discharge and its effect on integrated
circuits.

I-9
Chapter 1 From Physics to Electric Circuits

a) surface charge density b)

-11 2
x10 C/m

Fig. 1.4. Surface charge distribution over the human body based on an applied voltage of 1 V. Both
subjects are ECE graduate students.

A human beneath the power line is subject to an electric field. A similar solution applies
given that the body surface is still the equipotential surface. Figure 1.5 shows the surface
charge distribution for two human subjects in a vertical electric field of 1 V/m. The negative
charges concentrate close to the head, whereas the positive charges concentrate in the lower
body. Figure 1.5 also shows the electric potential distribution around the body. Dense
equipotential lines mean a high local electric field. The local field may exceed the external
field by a factor of 10 or more. All results are linearly scaled with the applied electric field.

a) surface charge b)
density
-11 2 1.6 V
1.6 V x10 C/m
1.5 V
1.5 V
8 1.4 V
1.4 V 1.3 V
1.3 V 1.2 V
6
1.2 V 1.1 V
4 1.0 V
1.1 V
0.9 V
1.0 V 2 0.8 V
0.9 V 0 0.7 V

0.8 V 0.6 V
-2
0.7 V 0.5 V
-4
0.6 V 0.4 V
-6
0.5 V 0.3 V
0.4 V -8 0.2 V
0.3 V 0.1 V
-10
0.2 V

Fig. 1.5. Human body subject to an applied electric field: surface charge and potential distribution.

I-10
Chapter 1 Section 1.2: Steady-State Current Flow and Magnetostatics

Section 1.2 Steady-State Current Flow and Magnetostatics

1.2.1 Electric Current
An electric current in a material is the directed motion of free positive charges. Conse-
quently, the electrons, which are often the only free charges in the conductor, move in the
opposite direction. The electric current density ~j everywhere in the material is a vector, and
it is measured in amperes per meter square, i.e., A/m2. The total current I through a
conductor with the uniform current density is the product of the current density magnitude
and the conductor cross section A; I is measured in amperes. Circuits with lumped
components deal with the total currents only. The current density (and the total current)
is directly proportional to the applied electric field,

rÞ ¼ σ ~
~jð~ E ð~
rÞ ð1:8aÞ
at any point of interest ~
r. Here, σ is the material conductivity with units of siemens/m, i.e.,
S/m. Note that 1 S ¼ 1/Ω. If there is no electric field, there is no electric current in the
material and vice versa. In metals, the conductivity σ is very high. Therefore, even a
vanishingly small electric field inside a metal conductor creates a large electric current.

Exercise 1.6: An AWG #00 (American Wire Gauge) aluminum wire has the conductivity
of 4.0  107 S/m and the diameter of 9.266 mm. Determine the total current in the wire
when the electric field inside the wire is 0.01 V/m. This value is more than 10,000 times
less than the field between the terminals of a 5-V laboratory supply separated by 2 cm.
Answer: 27 A.

For steady-state current flow, an electrostatic potential exists; it is given by Eq. (1.5).
Therefore, the current density can be expressed as the gradient of the potential,

~jð~
rÞ ¼ σ∇V ð~
rÞ ð1:8bÞ

1.2.2 Difference Between Current Flow Model and Electrostatics

In electrostatics, there are no charges and there is no electric field within conductors; only the
surface charges exist. The steady-state current problem is quite different. The volumetric
uncompensated charge density still does not exist within the conductor, but the electric field
does. Along with this, the surface charge density is always present, similar to electrostatics.
Physically, the difference arises from the different boundary conditions. The conductor
surface is divided into two distinct parts: the surface of electrodes, Se, where the voltage
(or current) is applied, and the rest of conductor surface, Sc, which is in contact with air. The
following is true for the steady-state current flow:

I-11
Chapter 1 From Physics to Electric Circuits

1. Everywhere on the surface Sc, the current density component perpendicular to the
surface is zero, i.e., ~j  ~
n ¼ 0, where ~
n is the unit normal vector to the surface, and
dot denotes the scalar product of two vectors. In other words, no current can flow
from the conductor into air.
2. On the surface of electrodes Se, the voltage is given: for example, +0.5 V on the left
electrode and 0.5 V on the right electrode. Alternatively, the inflowing current,
~j  ~
n, may be given.

How Does the Conductor “Guide” the Electric Field?

We consider the current flow in a conducting cylinder with two circular electrodes shown
in Fig. 1.6b. The electrostatic counterpart of the problem is given by the same coaxial
electrode pair in air, see Fig. 1.6a. The electrodes have the radius a; they are separated by
25a. The electrode voltages are 0.5 V. The exact value of the cylinder conductivity does
not matter; the same results will be obtained. The electrostatic problem and the steady-
state current problem are both solved as described in the previous section. Results of both
solutions are given in Fig. 1.6 where the equipotential lines and the electric field vectors
are plotted. Some general observations from this figure are worthy of note:
1. The current-carrying conductor “guides” the electric field as shown in Fig. 1.6b
which, otherwise, would be spread out in space; see Fig. 1.6a.
2. In the long conductor, the electric field and the electric current are both directed
along the conductor axis; they are uniform across any conductor cross section,
which is simultaneously an equipotential surface. In other words, current flow in the
long conductor is one dimensional, like water flow in a pipe. This is also true if the
conductor is bent or has a noncircular cross section
3. The voltage decreases linearly along the conductor from the most positive to the
most negative value. The voltage drop per unit length is constant; it is only a
function of the applied voltage.
It is seen from Fig. 1.6b that within in a current-carrying conductor of length l:

V
E¼ ð1:9Þ
l

where E is the magnitude of the electric field (its direction is along the conductor axis), and
V is the voltage across the conductor (1 V in the present case). Equation (1.9) is a simplified
version of Eq. (1.4) for uniform fields. In many textbooks, it is used to derive Ohm’s law.
Note that the electric field in Fig. 1.6b is not continuous across the conductor-air
interface. A component of the electric field perpendicular to the conductor boundary
suddenly appears. This component is due to the surface charges on the conductor–air
interface (not shown in the figure).

I-12
Chapter 1 Section 1.2: Steady-State Current Flow and Magnetostatics

a) 5
0.5 V -0.5 V
0.1 V -0.1 V

E E
y/a

0.2 V -0.2 V

-5
b) 5
0.5 V -0.5 V

E E
y/a

0 0.4 V 0.3 V 0.2 V 0.1 V 0V -0.1 V -0.2 V -0.3 V -0.4 V

Se Sc Se
-5
-15 -10 -5 0 5 10 15
x/a

Fig. 1.6. (a) Two electrodes subject to 0.5 V in free space; (b) the same electrodes but with a
conducting cylinder between them. Equipotential lines and electric field vectors are shown.

Exercise 1.7: A common AWG #22 copper wire is used in the laboratory to form a coil
with the radius of 0.1 m and 100 turns. This coil is subject to an applied voltage of 0.5 V.
Determine the total current in the wire if its diameter is 0.64516 mm; the copper conduc-
tivity is 5.8  107 S/m. Hint: find the electric field in the wire first.
Answer: 0.1509 A.

1.2.3 Physical Model of an Electric Circuit

Thus far, we have considered the current flow in only one wire. However, a simple
electric circuit uses two wires and it includes at least three elements:
1. A voltage power supply, which in steady-state (direct current or DC) case generates
a constant voltage between its terminals.
2. An electric load that consumes electric power. The load may be modeled as a
resistant material of a much smaller conductivity.
3. Two wires, which extend from the source to the load. Those wires form a transmis-
sion line. In laboratory settings, both wires may be arbitrarily bent.

I-13
Chapter 1 From Physics to Electric Circuits

Figure 1.7 shows a physical model of the simple circuit. We study its electric part first.
In Fig. 1.7a, the highest electric field magnitude is observed exactly between the two
conducting wires, at the line connecting its centers. At the same time, the electric field in
the wires is usually very small. Still, sufficient current flows there due to the high material
conductivity; see Eq. (1.8a). However, the electric field in the load cylinder is not small.
This is indicated by denser equipotential surfaces. In Fig. 1.7a, the equipotential surfaces
are all separated by 0.05 V. There is a net voltage drop of 0.3 V along each wire and the
voltage drop of 0.4 V across the load, resulting in a total potential drop of 1 V. This
equality is KVL (Kirchhoff’s voltage law). Ideally, when the conductivity of the wires is
infinitely high, the entire source voltage appears across the load cylinder. The field in the
wires becomes vanishingly small, but enough current still flows. The wire of infinite
conductivity, or the ideal wire, is a useful abstraction.

Exercise 1.8: If the voltage drop along each wire in Fig. 1.7a were 0.01 V, what would be
the value of the voltage across the load?
Answer: 0.98 V.

Exercise 1.9: In contrast to Fig. 1.7b, the field within the wire in Fig. 1.6b is not small,
regardless of its conductivity. Why is it so?
Answer: There is no load in Fig. 1.6b; the entire voltage drop is purposely forced to occur
across the wire.

1.2.4 Magnetostatics and Ampere’s Law

We next study the magnetic part of the circuit in Fig. 1.7b. The same electric current
I flows in the entire circuit. The current in one of the conductors in Fig. 1.7b creates the
magnetic field (magnetic-field intensity) H ~ with units of A/m around the conductor.
If the conductors are considered to be sufficiently long, the magnitude (absolute value)
~ ð~
of the field H rÞ anywhere in space, except within the conductor, is given by

I
H ð~
rÞ ¼ ð1:10Þ
2π j~
rj
Equation (1.10) is a particular form of Ampere’s law for an infinite straight wire of the
~ forms concentric circles around the wire; its direction follows
total current I. The vector H
the right-hand rule. When two wires are present, as in Fig. 1.7b, the resulting combined
magnetic field is the vector sum of two solutions given by Eq. (1.10) for two conductors

I-14
Chapter 1 Section 1.2: Steady-State Current Flow and Magnetostatics

a) Electric field and equipotential surfaces

for direct current flow
0.05 V
lines of force (E)

1V +
-

S’ equipotential
surfaces

load
0.4 V

+
electric field (E) -
equipotential
surfaces
b) Electric/magnetic fields and
direct current flow
lines of force (E)
Poynting vector
1V +
-

S’ electric
current I

load
lines of
magnetic
field (H)

lines of magnetic field (H)

Fig. 1.7. Accurate physical model of an electric circuit.

I-15
Chapter 1 From Physics to Electric Circuits

having the opposite current directions, respectively. The lines of the combined magnetic
field are shown in Fig. 1.7b. They are always perpendicular to the lines of force for the
electric field. The magnetic-field magnitude in Fig. 1.7b also has its maximum exactly
between the two conducting wires, at the line connecting its centers.

Exercise 1.10: Determine the magnetic-field magnitude in the middle between two
parallel long wires carrying current of 1 A each and separated by 2 cm.
Answer: H ð~
rÞ ¼ 31.8 A/m.

Another useful form of Ampere’s law is the magnetic field of an infinite “current
sheet,” i.e., when current flows in a thin conducting sheet in one direction. The sheet may
be thought of as an infinite number of parallel thin wires carrying the same current. If
j [A/m] is the current density per unit of sheet width, then the resulting constant field is

1
H¼ j ¼ const ð1:11Þ
2

1.2.5 Origin of Electric Power Transfer

We know from physics classes that electric power P delivered to the load is given by the
product P ¼ V I where V is the voltage across the load and I is the current through it. How
exactly is this power transferred to the load? To answer this question, we assume a
transmission line in the form of two parallel sheets of width W and spacing l shown in
Fig. 1.8. Each sheet carries current density with the magnitude j per unit of sheet length.

electric field (E)

Poynting vector (S)
+

W j
V
-

ad
to lo

magnetic field (H)

Fig. 1.8. Transmission line in the form of two current sheets.

When the material conductivity is infinite, the voltage between the two sheets is the
load voltage V. For l=W << 1, Eq. (1.9) yields the electric field within the transmission
line, E ¼ V =l. The magnetic field is found using Eq. (1.11). The result is H ¼ j since both
sheets contribute to the field within the line. Next, we define a vector

I-16
Chapter 1 Section 1.2: Steady-State Current Flow and Magnetostatics

~
S¼~ ~
E H ð1:12Þ
where the symbol  stands for the cross product or vector product. The vector ~ S is called
the Poynting vector; it is shown in Fig. 1.8. The units for the Poynting vector are given by
power per unit area, i.e., 1V=m  1 A=m ¼ 1 W=m2 . Thus, the Poynting vector char-
acterizes the directional energy flux density in space. Its magnitude in Fig. 1.8 is given by
S ¼ EH ¼ V j=l ¼ V I=ðlW Þ where I ¼ jW is the net current in every conductor.
Multiplying the Poynting vector by the area A ¼ lW where the fields are concentrated,
we obtain the remarkable result,

AS ¼ V I ¼ P ð1:13Þ

In other words, the power is transferred by the fields. This result is perhaps less important
for wired circuits where the fields are directly linked to charges and currents. However, it
is important for the transition from wired to wireless circuits. At a sufficiently high
frequency, significant electric and magnetic fields will be radiated by an antenna into
empty space. These fields will carry power flux density given by Eq. (1.12).

Exercise 1.11: Two conductors extending from the source to the load are parallel
1-cm-wide perfectly conducting sheets separated by 1 mm. The (vertical) electric field
between the plates is 100 V/m; the (horizontal) magnetic field is 100 A/m. Determine
electric power delivered to the load. Assume no field fringing.
Answer: P ¼ 0:1W.

I-17
Chapter 1 From Physics to Electric Circuits

Section 1.3 Hydraulic and Fluid Mechanics Analogies

1.3.1 Hydraulic Analogies in the DC Steady State
In DC and low-frequency alternating circuits, the electric and magnetic fields outside
conductors may be ignored without affecting the final results. When this is the case, the
electric circuits follow rather precisely useful fluid mechanics (or hydraulic) analogies.
Major hydraulic analogies are depicted in Fig 1.9. Let us consider a water pump
connected to a filter in Fig. 1.9a—left. The water pump creates a constant pressure
difference p between its terminals, which forces water to move through the filter.
Although the constant pressure (torque) pump is less common than a water pump of
constant flux, we can use it as an analogy since it exactly corresponds to the voltage
source, which maintains constant voltage difference V in Fig. 1.9a—right. The filter can
be thought of as an electric resistance: it opposes the water flow, and a certain pressure
difference, or voltage, is required to overcome this resistance. Electric current corre-
sponds to fluid velocity times the tube cross section which constitutes the total water flux.
For the entirely closed pumping system in Fig. 1.9a, the water pressure inside the system
can have an arbitrary reference level p0. This level may be quite different from the
ambient atmospheric pressure. Similarly, an isolated electric circuit may have an arbitrary
voltage V0 versus ground, due to static charge accumulation.
The condition V 0 ¼ 0 will be achieved by grounding the circuit. Figure 1.9b–d also
specifies hydraulic analogies for the separate circuit elements. We consider steady-state
flow of incompressible fluid. Emphasize that the DC voltage source in Fig. 1.9c is
analogous to a constant-torque water pump, which creates a constant pressure difference

a) p+p0 V+V0
Water pump velocity
(pressure drop) filter I
resistance
p + V +
- - (load)
velocity I
p0
V0
b) resistance sand filter

c) constant torque pump

voltage source

-
+

d) constant speed pump

current source

Fig. 1.9. Hydraulic analogies in the DC steady state.

I-18
Chapter 1 Section 1.3: Hydraulic and Fluid Mechanics Analogies

between its terminals, whereas a DC electric current source in Fig. 1.9d is similar to a
constant-speed water pump, which creates constant fluid flux.

1.3.2 Analogies for Alternating-Current (AC) Circuits

This case illustrated in Fig. 1.10 corresponds to alternating flow of incompressible fluid in a
piping system. The AC voltage source corresponds to a harmonically (sinusoidally)
oscillating piston in Fig. 1.10a, with a constant-torque amplitude. A capacitance in
Fig. 1.10b is represented by a flexible membrane. The capacitance value, C, corresponds
to the inverse stiffness, 1/k, of the membrane, also called the compliance. When k ! 1 and
C ! 0 (a rigid membrane), the capacitance value tends to zero. The membrane becomes a
solid wall, which blocks the alternating fluid flow entirely. In another limiting case (k ! 0
or C ! 1), the membrane has no effect on the fluid flow. Intermediate cases correspond to
a partial blocking. A massive wheel with a rotational inertia in Fig. 1.10c represents an
inductance. The inductance value, L, corresponds to the mechanical mass m of the wheel.
When m ! 1 or L ! 1, the wheel does not respond to fluid oscillations and
blocks the alternating fluid flow entirely. In the opposite case (m ! 0 or L ! 0), the
wheel has no effect on the fluid flow. Intermediate cases correspond to a partial blocking.

a) oscillating piston

AC voltage source

b) flexible membrane
capacitance

c) massive wheel
of mass m
inductance

d) electric transformer

primary secondary D1 D2

e) electric transformer

N N
pivot

Fig. 1.10. Hydraulic analogies for alternating-current (AC) circuit elements.

I-19
Chapter 1 From Physics to Electric Circuits

An electric transformer shown in Fig. 1.10d operates with alternating currents.

One mechanical analogy is a gear transmission or gearbox. In terms of angular speed
ω [rad/s] and developed torque T ½N  m, one has T 2 ¼ ðD2 =D1 ÞT 1 , and
ω2 ¼ ðD1 =D2 Þω1 , where D1,2 are pitch diameters of gear wheels. Here, torque is the
voltage and speed is the current. D1,2 are similar to the number of turns, N1,2, of
the primary and secondary coils of the transformer, respectively. This analogy ignores
the field effect—magnetic coupling between the coils. Therefore, it will fail in the DC
case. A more realistic transformer analogy is shown in Fig. 1.10e. The model with four
pistons transforms power from one circuit to another in the AC case only. It is drawn for a
1:1 transformer. When a transformer with a turn ratio of 2:1 is required, the area of output
pistons is doubled. This doubles the output current, but the output voltage (the force) will
be halved.

1.3.3 Analogies for Semiconductor Circuit Components

Semiconductor circuit components are similar to fluid valves, which are either externally
controlled or are controlled by the fluid-flow pressure itself. Figure 1.11a shows a hydraulic
analogy for a diode. This picture highlights its major function: a one-way valve. Fig. 1.11b
illustrates the operation of an n-channel metal-oxide-semiconductor field-effect transistor,
or NMOS transistor. This transistor is a valve controlled by a third voltage terminal. For
another bipolar junction transistor or BJT in Fig. 1.11c, not only the control voltage but
also the control current is important. In other words, to keep the valve open, we must also
supply a small amount of current (fluid) at the control terminal.
a) higher pressure lower pressure
semiconductor diode

+ -
b) control pressure
NMOS transistor
flexible membrane
control voltage

c) control pressure/current
junction transistor
control voltage/current

Fig. 1.11. Analogies for semiconductor circuit components.

I-20
Chapter 1 Summary

Summary
Electrostatics
Electric voltage/electric potential V AA0 ¼ 1 V )Work of 1 J is necessary to bring the 1 C of
charge from point A0 to point A against the field;
ðA
V AA0 ¼  ~ E  d~
l for any contour 1, 2, or 3;
0
A
~
E ð~
rÞ ¼ gradV ð~ rÞ ¼ ∇V ð~ rÞ for potential V ð~
rÞ
everywhere in space including materials;
V ¼ lE in uniform fields (most important).
Coulomb force on charge q ~
F ¼ q~ E [N]
The force is directed along the field for positive charges and
against the field for negative charges

Gauss law Total flux of the electric field through closed surface S times
the permittivityð is the total charge enclosed by S.
Q
¼ ~ n dS (ε0 =8.8541012 F/m)
E~
ε0
S

Equipotential conductors Within conductor(s) with applied voltage:

– Electric field is exactly zero;
– Volumetric charge density (C/m3) is exactly zero.
On surface(s) of conductor(s) with applied voltage:
– Every point has the same voltage (conductor surface is
equipotential surface);
– Surface charge density (C/m2) exists;
– Emanating electric field is perpendicular to the surface
(tangential field is zero: ~
E t ¼ 0)
Outside conductor(s):
– Equipotential lines and lines of force are perpendicular
to each other
Voltage between two wires – Metal wires “guide” the electric field/voltage to a remote
point
– Given the same voltage, charges on wires increase when
their length increases.
Electrical induction in electrostatics – Conductors 1 and 2 are subject to applied voltages V1, V2;
– Conductor 3 has zero net charge;
– Conductor 3 acquires certain voltage V3;
– Surface charges in conductor 3 are separated as shown in
the figure.

(continued)
I-21
Chapter 1 From Physics to Electric Circuits

Steady-state current flow

Associated electric field – Current density (A/m2) and total current I exist if and
only if there is an electric field within the material;
rÞ ¼ σ ~
~jð~ E ð~
rÞ;
– There is little field in the conductor wires of a high
conductivity;
– The field is high in air gap between two wires;
– The field is equally high within the load;
– Both wires carry total surface charges Q defined by
supply voltage, wire separation, and wire length
Simple formulas for wire con- I ¼ AEσ
ductors
I ¼ π r2 Eσ
E ¼ ðV 1  V 2 Þ=l

Non-uniform current flow Compare:

In the steady-state current-flow model:
– Electric field within the conductor is not zero;
– The conductor surface is not the equipotential surface
In the electrostatic model:
– Electric field within the conductor is zero;
– The conductor surface is the equipotential surface.

ð
Current conservation law ~j  ~
n dS ¼ 0
S

There are no sources and sinks of electric current within a

conductor except the surface electrodes

Magnetostatics
Ampere’s law in a general form Line integral of the magnetic field over a closed contour is
the total current enclosed by this contour
ð ð
~ ~
H  d l ¼ ~j  ~
n dS ¼ I enc , j is measured in A/m2
S S

Particular forms of Ampere’s Wire: H ð~ r ¼ j~

rÞ ¼ 2πI r , rj (particular form)
law j
Sheet of current: H ¼ ¼ const, j is in A/m
2

I-22
Chapter 1 Problems

Problem 1.5. The figure below shows the

Problems electric potential distribution across the semi-
1.1 Electrostatics of conductor pn-junction of a Si diode. What
Conductors kinetic energy should the positive charge
1.1.2 Electric Potential and Electric (a hole) have in order to climb the potential
hill from anode to cathode given that the hill
Voltage
“height” (or the built-in voltage of the
1.1.3 Electric Voltage Versus Ground
pn-junction) is V bi ¼ 0:7 V? The hole charge
1.1.4 Equipotential Conductors is the opposite of the electron charge. Express
Problem 1.1. Determine voltage (or potential)
your result in joules.
VAB and VBA (show units) given that the electric
field between points A and B is uniform and has
the value of 5 V/m. Point A has coordinates
anode cathode
(0, 0); point B has coordinates (1, 1).
B Vbi

E
A
Problem 1.6. Is the electric field shown in the
figure that follows conservative? Justify your
Problem 1.2. Determine voltages VAB, VBD,
answer.
and VBC given that the electric field shown in
the figure that follows is uniform and has the A E
2 cm
value of (A) 10 V/m, (B) 50 V/m, and
(C) 500 V/m.
1 cm
A E C
10 cm
C B
0
D
5 cm
B
Problem 1.7. List all conditions for voltage and
0 cm
D electric field used in electrostatic problems.

Problem 1.3. Assume that the electric field Problem 1.8. The figure below shows a 345 kV
along a line of force AA0 has the value 1l V/ power tower used in MA, USA—front view. It
m where 0  l  1 m is the distance along the also depicts electric potential/voltage and elec-
line. Find voltage (or potential) V AA0 . tric field distributions in space:
A. Determine which figure corresponds to
Problem 1.4. The electric potential versus the electric potential and which to the
ground is given in Cartesian coordinates by V magnitude of the electric field.
ð~
rÞ ¼ x þ y  z [V]. Determine the B. Provide a detailed justification of your
corresponding electric field everywhere in space. answer.

I-23
Chapter 1 From Physics to Electric Circuits

steel bar a) Problem 1.10. In your circuit, two wires

connected to a 10-V voltage supply happen to
symmetry plane
be very close to each other at a certain location;
grounded pole
they are separated by 2 mm.
2.84 m
A. What is the voltage between the wires at
this location?
B. What is approximately the electric field
3.15 m
strength at this location?
281.7 kV conductors
Problem 1.11. Is the figure that follows cor-
rect? Black curves indicate metal conductors.
b) Vpeak=281.7 kV b) Why yes or why not?

85

2V

+
1V
1V -
35

54
85
132
206
1.2 Steady-state Current
85
Flow and Magnetostatics
1.2.1 Electric Current
1.2.2 Difference Between Current Flow
c) Vpeak=281.7 kV
c) Model and Electrostatics
1.2.3 Physical Model of an Electric
7%
14%
Circuit
21%
Problem 1.12. List the conditions for voltage
28%
and electric field used in the steady-state elec-
36% tric current problems.
43%
50%
57%
64%
Problem 1.13. Figure that follows shows the
99%
92%
85%
78%
lines of force and equipotential lines for a DC
71%
current flow in a conductor due to two elec-
trodes. List all mistakes of this drawing.
Problem 1.9. Figure that follows shows some lines of force
1V
isolated conductors. Determine the following 2V

voltage differences: VAB, VAC, VCB.

equipotential
C
3V lines
A
0.5 V

0.5 V
B Problem 1.14. An AWG #10 (American Wire
Gauge) aluminum wire has the conductivity of
4.0107 S/m and the diameter of 2.58826 mm.

I-24
Chapter 1 Problems

Determine the total current in the wire when the Problem 1.18. Two conductors running from the
electric field inside the wire is (A) 0.001 V/m; source to the load are two parallel 0.5-cm-wide
(B) 0.005 V/m. thin plates of infinite conductivity. The (vertical)
electric field between the plates is 100 V/m; the
Problem 1.15. A copper wire (AWG #24) in (horizontal) magnetic field between the plates is
the form of a coil with the radius of 0.1 m and 100 A/m. The load power is 0.1 W. Assuming no
1000 turns is subject to applied voltage of 5 V. field fringing, determine (A) plate separation,
Determine the total current in the wire if its (B) load voltage, and (C) load current.
diameter is 0.51054 mm; the copper conductiv-
ity is 5.8  107 S/m. Problem 1.19. Repeat the previous problem
when the electric field between the plate elec-
Problem 1.16. The figure below shows a trodes increases by a factor of two, but the
conducting cylinder of radius R ¼ 1 cm, length magnetic field decreases by a factor of two.
L ¼ 5 cm, and conductivity σ 1 ¼ 1:0 S=m in air.
Two electrodes are attached on both cylinder
sides; the electrode radius is exactly the cylinder 1.3 Hydraulic and fluid
radius. Electrode voltages are exactly 1 V:
mechanics analogies
V1=1V z V2=-1V Problem 1.20. For the hydraulic setup shown
electrode #1 R electrode #2 in the figure, draw its electrical counterpart
(an electric circuit) using the circuit symbols.
y

constant
speed pump
L filter

A. Determine and sketch to scale the elec-

tric potential everywhere inside the cyl-
inder and on its surface.
B. Determine and sketch to scale the elec- Problem 1.21. For the hydraulic setup shown
tric field everywhere inside the cylinder. in the figure, present its electrical counterpart
C. Attempt to sketch the electric potential (an electric circuit). Note a connection to a large
distribution outside the cylinder. reservoir with atmospheric pressure.
D. Repeat tasks A and B when the voltage
electrodes are replaced by current elec- constant
torque pump
trodes with the applied electric current filter
density of 1 A/m2.
1.2.4 Magnetostatics/Ampere’s Law
1.2.5 Origin of Electric Power Transfer 0
Problem 1.17. A DC magnetic field of reservoir at atmospheric pressure

1000 A/m is measured between two parallel

wires of an electric circuit separated by 0.5 m.
What is the circuit current?

I-25
 Part I
DC Circuits: General Circuit
Theory—Operational Amplifier
Chapter 2

Chapter 2: Major Circuit Elements

Overview
Prerequisites:
- Knowledge of university physics: electricity and magnetism (optional)
- Knowledge of vector calculus (optional)

Objectives of Section 2.1:

- Realize the difference between circuit elements and circuit components
- Review (derive) Ohm’s law
- Become familiar with the
-i characteristic of the resistance including limiting cases
- Realize the importance of ohmic losses in long cables
- Become familiar with discrete fixed resistors and with resistive sensing elements

Objectives of Section 2.2:

- Realize the meaning of a passive nonlinear circuit element and its
-i characteristic
- Define two resistance types (static and dynamic) for a nonlinear passive circuit
element
- Present two examples of nonlinear elements: ideal diode and a threshold switch

Objectives of Section 2.3:

- Introduce the concept of independent voltage and current sources and become
familiar with their
-i characteristics
- Introduce the concept of practical voltage/current sources including their

-i characteristics
- Obtain initial exposure to the operation principles of voltage sources including
specific examples

Objectives of Section 2.4:

- Become familiar with the concept of a dependent source
- Become familiar with four major types of dependent sources
- Obtain initial exposure to transfer characteristics of dependent sources
- Become familiar with ideal time-varying and AC sources

© Springer Nature Switzerland AG 2019 II-29

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_2
Chapter 2 Major Circuit Elements

Objectives of Section 2.5:

- Formalize the meaning of voltmeter and ammeter from the viewpoint of open and
short circuits
- Obtain a clear understanding of the circuit ground and its role in the circuit
- Review different ground types

Application Examples:
- Power loss in transmission lines and cables
- Resistive sensing elements
- DC voltage generator with permanent magnets
- Chemical battery

Keywords:
Circuit elements, Circuit components, υ-i characteristic, Resistance, Polarity, Voltage difference,
Voltage drop, Voltage polarity, Passive reference configuration, Ohm’s law, Linear passive circuit
element, Conductance, Siemens, mho, Short circuit, Open circuit, Ohmic conductor, Mobility of
charge carriers, Material conductivity, Material resistivity, Electric circuit, American Wire Gauge
(AWG), Resistor, Fixed resistors, Surface Mound Devices (SMD) (resistance value, geometry
size), Variable resistor, Potentiometer, Resistive sensors, Photoresistor, Photocell, Negative
temperature coefficient (NTC), Thermistor equation, Thermistor constant, Thermocouple,
Peltier-Seebeck effect, Strain gauge, Strain sensitivity, Gauge factor, Strain gauge equation,
Potentiometric position sensor, Nonlinear passive circuit elements, Non-ohmic circuit elements,
Radiation resistance, Ideal diode, Shockley equation, Static resistance, Dynamic resistance,
Small-signal resistance, Differential resistance, Incremental resistance, (DC) Operating point,
Quiescent point, Electronic switch, Solid-state switch, Switch threshold voltage, Two-terminal
switch, Three-terminal switch, Unidirectional switch, Bidirectional switch, Independent ideal
voltage source, Active reference configuration, Nonlinear passive circuit elements, Non-ohmic
circuit elements, Radiation resistance, Static resistance, Ideal diode, Shockley equation, Dynamic
resistance, Small-signal resistance, Differential resistance, Incremental resistance, (DC) Operating
point, Quiescent point, Electronic switch, Solid-state switch, Switch threshold voltage, Two-
terminal switch, Three-terminal switch, Unidirectional switch, Bidirectional switch, Independent
ideal voltage source, Active reference configuration, Practical voltage source, Maximum available
source current, Maximum available source power, Open-circuit source voltage, Short-circuit
circuit current, Internal source resistance, Independent ideal current source, Practical current
source, Charge separation principle, Faraday’s law of induction, Lorentz force, Instantaneous
generator voltage, Average generator voltage, Battery voltage, Battery capacity, Battery energy
storage, Dependent sources, Voltage-controlled voltage source, Current-controlled voltage source,
Voltage-controlled current source, Current-controlled current source, Open-circuit voltage gain,
Transresistance, Transconductance, Short-circuit current gain, Voltage amplifier, Current
amplifier, Transresistance amplifier, Transconductance amplifier, Transfer characteristic, AC
voltage source, Ideal voltmeter, Ideal ammeter, Earth ground, Chassis ground, Common
(neutral) terminal (ground), Forward current, Return current, Absolute voltages in a circuit

II-30
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

Section 2.1 Resistance: Linear Passive Circuit Element

2.1.1 Circuit Elements Versus Circuit Components
Circuit Elements
Similar to mechanical mass, spring, and damper used in analytical dynamics, circuit
elements are simple hypothetic ideal models. Every circuit element is characterized by its
unique voltage/current dependence called the υ-i characteristic. Most of the υ-i charac-
teristics reflect general physical laws. A list of the circuit elements includes:
1. Resistance
2. Capacitance
3. Inductance
4. Ideal electric transformer
5. Voltage source (independent and dependent)
6. Current source (independent and dependent)
7. Ideal switch
8. Ideal (Shockley) diode
9. Logic gates (NOT, AND, OR).
Circuit elements may be linear (resistance) or nonlinear (ideal diode), passive (resis-
tance) or active (voltage source), static (resistance) or dynamic (capacitance/inductance),
or both. Although all circuit elements studied here are static ones, the extension to the
case of time-varying voltage and current is often trivial.

Circuit Components
Circuit components are numerous hardware counterparts of the circuit elements. Exam-
ples of the circuit components include resistor, capacitor, inductor, battery, etc. The circuit
components may be modeled as combinations of the ideal circuit elements with one
dominant desired element (e.g., resistance) and several parasitic ones (e.g., parasitic
inductance and capacitance of a physical resistor). Another example is a battery, which
is modeled as an ideal voltage source in series with a (small) resistance. In practice, we
attempt to model any existing or newly discovered circuit component as a combination of
the well-known circuit elements. The same is valid for more complicated structures
targeted by electrical, mechanical, and biomedical engineers. An example is a human
body, the response of which is modeled as a combination of resistance and capacitance.

2.1.2 Resistance
Symbols and Terminals
Figure 2.1 shows the circuit symbol for resistance with current direction and voltage
polarity: positive voltage applied to the left terminal and a negative voltage applied to the
right terminal cause a current to flow from left to right, as depicted in Fig. 2.1b. As a
circuit element, the resistance is fully symmetric: terminals 1 and 2 in Fig. 2.1 may be
interchanged without affecting its operation. Thus, the resistance does not have a polarity.

II-31
Chapter 2 Major Circuit Elements

a) b)

+
-

+
V
1 2
-
I

Fig. 2.1. Resistance symbol along with the voltage and current.

Voltage Across the Resistance

The voltage difference (or voltage drop or simply voltage) across the resistance, V, in
Fig. 2.1 is a signed quantity. The voltage is measured in volts (V), named in honor of
Italian physicist Alessandro Volta (1745–1827), who invented the first battery. Plus and
minus signs across the resistance indicate the voltage polarity. Specifically, a plus sign
denotes a (presumably) higher absolute voltage level versus ground than the minus sign;
see Fig. 2.1b. For example, let us assume that the value V in Fig. 2.1a is positive and equal
to 1 V. This means that the electric field spends positive work equal to
1V
1C ¼ 1J ð2:1Þ
when moving the charge of 1 C through the resistance from left to right in Fig. 2.1a.
Similarly, the positive work of one joule is to be spent by an external force to move one
coulomb of charge across a potential difference of one volt against the electric field. In
power electronics, quantities of 1 kV (1000 V), even 1 MV (106 V) for voltage, are
customary. In sensors and cellular phone circuits, for example, voltages are usually much
lower. Values of 1 mV (103 V) or even 1 μV (106V) may be recorded. Voltage applied
to the resistance causes electric current flow.

Current Through Resistance: Passive Reference Configuration

The net current I flowing through the resistance is shown in Fig. 2.1a by an arrow.
The current is measured in amperes (A), named in honor of French physicist and mathe-
matician André-Marie Ampère (1775–1836). For example, the value of I in Fig. 2.1a is
1 A. This means that one coulomb of charges passes through the resistance in one second:

1A
1s ¼ 1C ð2:2Þ

The electric current flow through the resistance (and any other circuit element) is a
directed quantity; the arrow shows its direction. A useful fluid mechanics analogy for
the resistance is water (electric current) that flows down the “voltage” hill in Fig. 2.1b.
The relation between voltage polarity and current direction depicted in Fig. 2.1b is
known as the passive reference configuration. It is commonly used for all passive circuit
elements such as resistances, diodes, capacitances, and inductances. Physically, the passive
reference configuration means that the resistance consumes electric power, but does not
create it. In power electronics, currents of several A, even kA (1000 A), are customary.

II-32
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

In digital and communication circuits, however, currents are usually low; therefore, units
of 1 μA (106A) or 1 mA (103A) are commonly used.

Ohm’s Law: Resistance and Conductance

According to Ohm’s law, the voltage V across the resistance and the current I through the
resistance are related by a simple linear expression

V ¼ RI ð2:3Þ

with the proportionality constant R known as the resistance. This expression was first
established by German mathematician and physicist Georg Simon Ohm (1789–1854) in
1827 but was coldly received by the scientific community at that time. It took nearly
14 years before the Royal Society of London finally recognized his work and his
discovery is now known as Ohm’s law. The unit of resistance R carries his name ohm
and the Greek symbol Ω. The unit follows from Eq. (2.3) as volt over ampere:
1 V
1 Ω¼ ð2:4Þ
1 A
The resistance is the linear passive circuit element. Resistance values vary typically
between 1 Ω and 100 MΩ. The reciprocal of the resistance is the conductance, G:
1
G¼ ð2:5Þ
R
The unit of conductance, Ω1, is called siemens (S) in honor of Ernst Werner von Siemens
(1816–1892), a German inventor and the founder of what is today Europe’s largest
electrical engineering company (Siemens AG). An older American equivalent of that unit
Ω
is mho ( ) or ohm spelled backwards! Conductance is useful in the circuit analysis.

Exercise 2.1: A voltage of 20 V is applied to a 1-MΩ resistance. Determine the current

through the resistance.
Answer: 20 μA.

Exercise 2.2: A voltage of 20 V is applied to a 1-mS conductance. Determine current

through the conductance.
Answer: 20 mA.

II-33
Chapter 2 Major Circuit Elements

2.1.3 υ-i Characteristic of the Resistance: Open and Short Circuits

Figure 2.2 plots the linear dependence given by Eq. (2.3) for two distinct resistances. The
corresponding plot is known as the υ-i characteristic (or the υ-i dependence). We use
small letters υ-i to maintain consistency with the following study of time-varying circuits.
The υ-i characteristic is the “business card” of the circuit element—every circuit element
has its own υ-i characteristic. Once the υ-i characteristic is known, the circuit element is
characterized completely. The following is true with reference to Fig. 2.2a:
1. The slope of the υ-i dependence for the resistance is equal to 1/R or G.
2. Smaller resistance leads to a stepper υ-i dependence (large currents).
3. Larger resistance leads to a flatter υ-i dependence (small currents).
4. The negative part of the υ-i dependence simply means simultaneous switching
voltage polarity and current direction, respectively, in Fig. 2.1.

a) smaller b) short
I resistance I circuit

larger
resistance open
0 0 circuit

V V

Fig. 2.2. υ-i Characteristics for resistances and for the short and open circuits, respectively.

Open and Short Circuits

Two limiting cases of the resistance υ-i characteristics are the short circuit and the open
circuit, as seen in Fig. 2.3.
R 0
short circuit

R open circuit

Fig. 2.3. Transformation of resistance to a short and an open circuit, respectively.

When R ! 0, the resistance becomes a short circuit, or an ideal wire. There is no

voltage drop V across the wire, but any current I can flow through it. Therefore, the υ-i
characteristic of the short circuit is the straight vertical line in Fig. 2.2b. When R ! 1,
the resistance becomes an open circuit, or an ideal vacuum gap. There is no current
I through the gap at any value of the applied voltage V. Therefore, the υ-i characteristic of
the open circuit is the straight horizontal line in Fig. 2.2b.

II-34
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

Exercise 2.3: Every vertical division in Fig. 2.2a is 0.1 A; every horizontal division is 1 V.
Find resistances for two υ-i dependencies in the figure.
Answer: R ¼ 4 Ω and R ¼ 25 Ω, respectively.

Exercise 2.4: An ideal switch is open when V < 0 and is closed when V  0. Plot the υ-i
characteristic given that only a positive current I > 0 can flow.
Answer: Horizontal line I ¼ 0 when V < 0 and vertical line V ¼ 0 when I > 0.

2.1.4 Power Delivered to the Resistance

Voltage V across the resistance is work in joules necessary to pass 1 C of charge through
the resistance. Since there are exactly I coulombs passing through the resistance in one
second, the power P delivered to the resistance must be the product of work per unit
charge and the number of charges passing through the element in one second: P ¼ V I.
The power P has indeed the units of watts (1 V
1 A ¼ 1 J=1 s ¼ 1 W). When the υ-i
characteristic of the resistance is examined, the power is equal to the area of the shaded
rectangles in Fig. 2.4. Using Ohm’s law, Eq. (2.3) gives us three equivalent definitions of
the absorbed power by a resistance:

P ¼ VI Basic definition, valid f or any passive circuit element ð2:6Þ

2
V
P¼ Power f or resistance in terms of voltage ð2:7Þ
R
P ¼ RI 2 Power f or resistance in terms of current ð2:8Þ

smaller
I resistance

P1
larger
resistance
P2

0 V

Fig. 2.4. υ-i Characteristics for the resistances and power rectangles. P1,2 are absorbed powers.

II-35
Chapter 2 Major Circuit Elements

Despite their obvious nature, all three equations are useful in practice. In particular,
Eq. (2.7) indicates that a small resistance absorbs more power than the large resistance at
the same applied voltage; this is seen in Fig. 2.4. Imagine for a moment that we know the
voltage across the resistance, but do not know the current. This happens if a number of
circuit elements are connected in parallel to a known voltage source. Then Eq. (2.7) is
used to find the power. However, if the current is known, but the voltage is not (a number
of elements connected in series to a current source), then Eq. (2.8) is employed.

Example 2.1: A voltage of 10 V is applied to a 2.5-Ω resistance. Determine the absorbed

electric power in three possible ways as stated by Eqs. (2.6) through (2.8).
Solution: To apply two of the three equations, current through the resistance is needed.
From Ohm’s law, I ¼ V =R ¼ 4 A. The electric power delivered to the resistance can
thus be determined in three ways:
P ¼ V I ¼ 40 W Basic power definition, passive reference configuration
V2
P¼ ¼ 40 W Power f or resistance in terms of voltage
R
P ¼ RI 2 ¼ 40 W Power f or resistance in terms of current

2.1.5 Finding Resistance of Ohmic Conductors

An ohmic conductor satisfies Ohm’s law given by Eq. (2.3). Finding its resistance is
equivalent to the derivation of Ohm’s law under certain assumptions. Let us consider a
conducting circular cylinder subject to an applied voltage V in Fig. 2.5. The cylinder has
length l and a cross-sectional area A.

equipotential
surfaces
electric field +
E V/4 -
+
I A + I
+
+
+
x

+
V -

Fig. 2.5. Finding the resistance of a conducting cylinder.

Finding Total Current

The net electric current in a metal or other conductor is defined as the net flux of positive
charge carriers directed along the conductor axis x:

II-36
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

I ¼ Aqnυ ð2:9Þ

Here, qn is the volumetric charge density of free charges q with concentration n in

coulombs per cubic meter, C/m3, and υ is the magnitude of the average charge velocity in
m/s. In the one-dimensional model of the current flow, the average velocity vector is
directed along the x-axis seen in Fig. 2.5. Since the electrons have been historically
assigned a negative charge, the electric current direction is opposite to the direction of
electron motion in a conductor. The electron carries an elemental charge of q ¼
1.60218
1019 C. Because A and qn are constants for a given conductor, the electric
current is simply associated with the charge’s mean velocity υ.

Finding Average Carrier Velocity

In order to find υ, we use the following method. The total voltage drop V applied to a
sufficiently long, conducting cylinder is uniformly distributed along its length following
the equally spaced equipotential surfaces; this is schematically shown in Fig. 2.5. This
fact has been proved in Chapter 1. Such a voltage distribution corresponds to a constant
uniform electric field within the cylinder, which is also directed along the cylinder axis.
The magnitude of the field, E, with the units of V/m, is given by

E ¼ V =l ð2:10Þ

The electric field creates a Coulomb force acting on an individual positive charge q. The
Coulomb force is directed along the field; its magnitude F is given by

F ¼ qE ð2:11Þ

The key is a linear relation between the charge velocity υ and force F or, which is the same,
a linear relation between the charge velocity υ and the applied electric field E, i.e.,

υ ¼ μE ð2:12Þ

where μ is the so-called mobility of charge carriers, with the units of m2/(Vs).
Carrier mobility plays an important role in semiconductor physics. With the help
of Eqs. (2.10) and (2.12), the expression for the total current Eq. (2.9) is transformed to



l l l
V ¼ I¼ I ¼ RI, σ ¼ qnμ, R ¼ ð2:13Þ
Aqnμ Aσ Aσ

This is the expression for the resistance of a cylindrical conductor. Material conductivity
σ is measured in S/m. Its reciprocal is the material resistivity ρ ¼ 1=σ measured in Ωm.

II-37
Chapter 2 Major Circuit Elements

Example 2.2: Estimate resistance R of a small doped Si disk with the length l of 5 μm, cross
section of A ¼ 104 cm2, uniform electron, concentration (carrier concentration) of
n ¼ 1017 cm3 , and carrier mobility of μn ¼ 1450 cm2/(Vs).
Solution: Resistance calculations are usually simple when the one-dimensional model of a
conducting cylinder or a disk is used. However, one must be careful with the units. Units of
cm are customary in semiconductor physics. Therefore, one should first convert all different
units of length to meters (or to centimeters). After that, we use the definition of the resistance
given by Eq. (2.13) and obtain (units of meters are used):

l 5
106
R¼ ¼ 8 ¼ 0:215 Ω ð2:14Þ
Aqnμn 10
1:602
1019
1023
0:145

Table 2.1 lists conductivities of common materials. What is the major factor that
determines the conductivity of a particular conducting material? According to
Eq. (2.13), there are two such parameters: charge concentration and charge mobility.

Table 2.1. DC conductivities of conductors, semiconductors, and insulators (25  C, multiple

sources).
Material Class σ (S/m) Material Class σ (S/m)
Silver Conductor 6.1
107 Seawater Semiconductor 4
Copper Conductor 5.8
107 Human/ani- Semiconductor 0.1–2.0
mal tissues
Gold Conductor 4.1
107 Germanium Semiconductor 2
Aluminum Conductor 4.0
107 Fresh water Semiconductor 0.01
Brass Conductor 2.6
107 Wet soil Semiconductor 0.01–0.001
Tungsten Conductor 1.8
107 Dry soil Semiconductor 0.001–0.0001
Zinc Conductor 1.7
107 Intrinsic sili- Semiconductor 4.4
104
con (Si)
Nickel Conductor 1.5
107 Gallium arse- Semiconductor 106
nide (GaAs)
Iron Conductor 1.0
107 Glass Insulator 1012
Tin Conductor 0.9
107 Porcelain Insulator 1014
Lead Conductor 0.5
107 Hard rubber Insulator 1015
Graphite Conductor 0.003
107 Fused quartz Insulator 1017
Carbon Conductor 0.003
107 Teflon Insulator 1023
Magnetite Conductor 0.002
107

II-38
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

It is mostly the different concentration of free charge carriers n that makes the resistance
of two materials quite different. For example, n ¼ 8:46
1028 m3 in copper (a good
conductor), whereas it may be n ¼ 1016 m3 in a moderately doped silicon crystal (doped
semiconductor). However, it is also the difference in mobility μ that represents the
“friction” experienced by the “gas” of free charges with density n that is moving through
the solid or liquid conductor under the applied voltage (electric field).

Exercise 2.5: Using Table 2.1, determine the total resistance of an aluminum wire having
a length of 100 m and a cross-sectional area of 1 mm2.
Answer: 2.5 Ω.

2.1.6 Application Example: Power Loss in Transmission Wires and Cables

All metal wires and cables are ohmic conductors. Electric power absorbed by an ohmic
conductor is transformed into heat. This is known as electric power loss. We can apply
Eqs. (2.6)–(2.8) and Eq. (2.13) in order to determine the loss of electric power in
transmission lines and/or cables. This question has significant practical importance.
Figure 2.6 outlines the corresponding electric circuit. The electric circuit is a closed
path for electric current. Resistance RL characterizes the load. Only Ohm’s law is used
to analyze this circuit, along with the current continuity. No other circuit laws are
necessary. We also consider a voltage source in Fig. 2.6. The voltage sources will be
studied next.

source transmission line (TL) a load

I I
+
+ I RL V
- I I
-
b
20 km

Fig. 2.6. A long transmission power line carrying a steady-state current I to the load resistance.

According to Eq. (2.13), the wire resistance is inversely proportional to its diameter.
In the USA, the American Wire Gauge (AWG) system was developed to classify the wire
diameters of conductors. You probably have heard an electrician refer to a gauge
12 household wiring. This implies a wire diameter of about 2 mm, or 0.0800 . Table 2.2
reports common AWG numbers and maximum current strengths.

II-39
Chapter 2 Major Circuit Elements

Table 2.2. American Wire Gauge (AWG) wire parameters. The maximum current is given for
solid copper (Source: Handbook of Electronic Tables and Formulas for American Wire Gauge).
Resistance per Maximum current
Diameter Diameter 1000 ft in (A)
AWG # (inches) (mm) or 304.8 m (Ω) for power transmission
24 0.0201 0.51054 25.67 0.577
22 0.0254 0.64516 16.14 0.92
20 0.0320 0.81280 10.15 1.50
18 0.0403 1.02362 6.385 2.30
16 0.0508 1.29032 4.016 3.70
14 0.0640 1.62814 2.525 5.90
12 0.0808 2.05232 1.588 9.30
10 0.1019 2.58826 0.999 15.0
Gauges 10 through 1 are not shown
0 (1 aught) 0.3249 8.252 0.09827 150
00 (2 aught) 0.3648 9.266 0.07793 190
000 (3 aught) 0.4096 10.404 0.06180 239
0000 (4 aught) 0.4600 11.684 0.04901 302

Example 2.3: An AWG 0 aluminum transmission grid cable schematically shown in

Fig. 2.6 has a wire diameter of 8.25 mm and a cross-sectional area of 53.5 mm2.
The conductivity of aluminum is 4.0
107 S/m. The total cable length (two cables must
run to a load) is 40 km. The system delivers 1 MW of DC power to the load. Determine the
power loss in the cable when load voltage V and load current I are given by:
1. V ¼ 40 kV and I ¼ 25 A
2. V ¼ 20 kV and I ¼ 50 A
3. V ¼ 10 kV and I ¼ 100 A
Why is high-voltage power transmission important in power electronics?
Solution: We find the total cable resistance from Eq. (2.13):

L L 40
103
R¼ρ ¼ ¼ ¼ 18:7 Ω ð2:15Þ
A σA 4:0
107
53:5
106
The same load current I flows through the load modeled by a resistor RL and through the
cables in Fig. 2.6. Therefore, power loss in the cables may be found using Eq. (2.8).
Knowing the load voltage (or the voltage across the cable) is not necessary. The power
loss in the cables is thus given by P ¼ RI 2. For the three different cases corresponding to
the same load power, we obtain

II-40
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

Example 2.3 (cont.):

1. P ¼ RI 2 ¼ 11:7 kW or 1.17 % of the load power

2. P ¼ RI 2 ¼ 46:8 kW or 4.68 % of the load power (2.16)
3. P ¼ RI 2 ¼ 187 kW or 18.7 % of the load power
Clearly, the high-voltage power transmission allows us to reduce the power loss in long
cables very significantly while transmitting the same power to the load. Therefore, the
high-voltage transmission lines passing through the country have typical voltages between
100 kV and 800 kV.

In circuit analysis in the laboratory, we usually consider ideal or perfectly conducting

wires whose resistance is zero. This is justified since the wire lengths for most practical
circuit applications are so short that the voltage drop is negligibly small.

2.1.7 Physical Component: Resistor

Fixed Resistors
Resistance is constructed intentionally, as a separate circuit component. This component
is called the resistor. A common axial leaded carbon film 0.25-W resistor deployed on a
solderless protoboard is seen in Fig. 2.7a. Those carbon film resistors are typically
manufactured by coating a homogeneous layer of pure carbon on high-grade ceramic
rods. After a helical groove is cut into the resistive layer, tinned connecting leads of
electrolytic copper are welded to the end-caps. The resistors are then coated with layers of
tan-colored lacquer. The common Surface Mount Device (SMD) thin-film resistor is
shown in Fig. 2.7b. Manufacturing process variations result in deviations from the normal
resistor values; they are known as tolerances and reported to the end user through an extra
color ring (for leaded axial resistors) or an extra digit (for SMD resistors). Typical power
ratings for the axial resistors are 1/6 W, ¼ W, ½ W, 1 W, 2 W, and 3 W. When the power
delivered to the resistor considerably exceeds the particular rating, the resistor may burn
out, releasing a prominent “carbon” smell. The axial resistors have color codes shown in
Fig. 2.7c. To find the value of the resistor depicted in Fig. 2.7a, we first encounter the
tolerance code, which will typically be gold, implying a 5 % tolerance value. Starting
from the opposite end, we identify the first band, and write down the number associated
with that color; in Fig. 2.7a it is 9 (white). Then, we read the next band (brown) and
record that number; it is 1. After this we read the multiplier black, which is 0. The resistor
value in Fig. 2.7a is consequently R ¼ 91
100 ¼ 91 Ω.
The surface mount resistors, also known as SMD resistors, do not have color codes.
The SMD resistors are labeled numerically as 102 ¼ 10
102 ¼ 1 kΩ, 271 ¼ 27

101 ¼ 270 Ω, etc. Along with this, the SMD resistors, similar to other SMD compo-
nents, do have codes corresponding to their geometry size. Each size is described as a
four-digit number. The first two digits indicate length, and the last two digits indicate

II-41
Chapter 2 Major Circuit Elements

a) axias leaded resistor b) SMD thin-film resistor

c) Color codes for axial resistors

Black Brown Red Orange Yellow Green Blue Violet Gray White
0 1 2 3 4 5 6 7 8 9

Fig. 2.7. (a) A leaded axial resistor, (b) a thin-film resistor, and (c) color codes for leaded resistors.

width (in 0.0100 , or 10 mils units). Some popular SMD resistor sizes are 0603
(0.0600
0.0300 , or 60
30 mils, or 1.6
0.8 mm), 0805 (0.0800
0.0500 ), and 1206
(0.1200
0.0600 ).

Variable Resistors (Potentiometers)

The simplest variable resistor is a potentiometer. A picture is shown in Fig. 2.8a along
with its equivalent electric schematic in Fig. 2.8b. The potentiometer is used either as a
voltage divider, discussed later in the text, or as a variable resistor. By rotating the
potentiometer shaft, it is possible to obtain any resistance value up to the maximum
potentiometer value. The adjustable resistance is obtained between terminals 1 and 2 or
2 and 3 of the potentiometer, respectively. You should remember that the potentiometer
is a nonpolar device. This means that it can be placed into the circuit in any orientation.
With this knowledge the joking engineer telling you that “all resistors in your circuit are
backwards” should not cause any fear.

a) b) 1

25 kΩ

2
1 2 3
3

Fig. 2.8. A rotary 25-kΩ potentiometer rated at 0.25 W and its equivalent circuit schematic.

2.1.8 Application Example: Resistive Sensors

There are a variety of sensor types—resistive sensors—which use electric resistance
variation to measure a mechanical or a thermal quantity. Some of them are shown in
Fig. 2.9. As a first example, we consider a temperature sensor based on a thermistor
(a resistor), with a resistance that varies when ambient temperature changes; see Fig. 2.9a.

II-42
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

The word thermistor is a contraction of the words “thermal” and “resistor.” As a second
example of a resistance subject to ambient conditions, we will consider a photoresistor
(photocell) shown in Fig. 2.9b. The final example is a strain gauge shown in Fig. 2.9c.

Fig. 2.9. Sensing elements which change their resistances when ambient conditions change.

Thermistor
The thermistor changes its resistance as temperature increases or decreases. General-
purpose thermistors are made out of metal oxides or other semiconductors. Successful
semiconductor thermistors were developed almost simultaneously with the first transis-
tors (1950s). For a metal-oxide thermistor, its resistance decreases with increasing
temperature. Increasing the temperature increases the number of free carriers (electrons)
and thus increases the sample conductivity (decreases its resistance). Shown in Fig. 2.9a
is a very inexpensive NTC—negative temperature coefficient—leaded thermistor.
According to the manufacturer’s datasheet, it reduces its resistance from approximately
50 kΩ at room temperature (about 25  C) by 4.7 % for every degree Celsius (or Kelvin)
and reaches about 30 kΩ at body temperature according to the thermistor equation:


1 1
R1 ¼ R2 exp B  ð2:17Þ
T1 T2

where T1, T2 are two absolute temperatures always given in degrees K. Temperature T2
corresponds to a room temperature of 25  C so that R2¼R25 C, temperature T1 is the
observation temperature, and B is the thermistor constant, which is equal to 4200 K in the
present case. Equation (2.17) is a nontrivial result of the solid-state physics theory. We
emphasize that Eq. (2.17) is more accurate than the temperature coefficient of the
thermistor—the above referenced value of 4.7 %. Typical applications include temper-
ature measurement, control, compensation, power supply fan control, and printed circuit
board (PCB) temperature monitoring. Inexpensive thermistors operate from 30  C to
approximately +130  C. At higher temperatures, thermocouples should be used.

Thermocouple
Figure 2.9 does not show one more important temperature sensor—the thermocouple—
which is used to measure large temperatures and large temperature differences. It operates

II-43
Chapter 2 Major Circuit Elements

based on a completely different principle. The thermocouple does not significantly

change its resistance when temperature changes. Instead, it generates an electric current
and the associated voltage, when the junction of the two metals is heated or cooled,
known as the Peltier-Seebeck effect; this voltage can be correlated to temperature.
Therefore, the thermocouple, strictly speaking, is not a resistive sensor.

Photoresistor (Photocell)
An idea similar to the thermistor design applies. Quanta of light incident upon the
photocell body create new free charge carriers—new electron-hole pairs in a semicon-
ductor. If the concentration of free charges increases, the resistance of the sample
decreases according to Eq. (2.13). The resistance is inversely proportional to the concen-
tration. The photocell in Fig. 2.9b is characterized by very large nonlinear variations of
the resistance in response to ambient light.

Strain Gauge
The strain gauge measures mechanical strain. The operation is based on Eq. (2.13), which
defines the resistance through material conductivity σ, the length of the resistor l, and its
cross section A. When the resistor, which may be a trace on the base of a metal alloy, is
stretched, its length l increases and its cross section A decreases. Hence, the resistance
R increases due to both of these effects simultaneously; changes in the resistance may be
made visible for small strains. Shown in Fig. 2.9c is an inexpensive uniaxial strain gauge
with a nominal resistance of 350 Ω; typical resistances are 120, 350, 600, 700, and 1000 Ω.
The gauge changes its resistance R in proportion to the strain sensitivity SG of the wire’s
resistance, also called the gauge factor (GF). For a strain gauge, the relative resistance
variation, ΔR/R, is estimated based on known values of the strain sensitivity, SG, and
strain, ε. The strain gauge equation has the form

ΔR=R ¼ S G ε ð2:18Þ

The dimensionless strain sensitivity SG varies around 2. The strain (a relative elongation)
is a dimensionless quantity. It is measured in micro-strains, με, where one με is 106.
Typical strain values under study are on the order of 1000 με. Using Eq. (2.18) this yields
a relative resistance variation as small as 0.2 %. Because of this, the circuits for the strain
measurements should be designed and built with great care. Temperature compensation
efforts are also required. Since the relative resistance changes are very small, the strain
gauge is a linear device: the strain is directly proportional to resistance variations.

Potentiometric Position Sensor

Another general resistive sensor is a potentiometric (or potentiometer) position sensor. Its
operation becomes apparent when we rotate the potentiometer dial in Fig. 2.8a. A change
in the resistance is directly proportional to the rotation angle. The resistance variation can

II-44
Chapter 2 Section 2.1: Resistance: Linear Passive Circuit Element

be converted to voltage variation and then measured. Similar potentiometer sensors for
measuring linear motion also exist.

Sensitivity of Resistive Sensors

One major difference between different resistive sensing elements is a very different
degree of the relative resistance variations. For the photocell, the resistance variation is up
to 100 times. For the thermistor, the resistance variation can be as much as 50 %. For the
strain gauge, the resistance variations do not exceed 0.5 %.

Circuit Symbols
There are several similar but not identical standards for circuit symbols related to
resistance: International standard IEC 60617, American ANSI standard Y32 (IEEE Std
315), etc. Figure 2.10 shows popular circuit symbols for variable resistances.

a) generic variable b) potentiometer c) thermistor d) photoresistor e) strain gauge

resistor

Fig. 2.10. Circuit symbols for variable resistances: (a) generic variable resistance,
(b) potentiometer, (c) thermistor, (d) photoresistor, and (e) strain gauge.

II-45
Chapter 2 Major Circuit Elements

Section 2.2 Nonlinear Passive Circuit Elements

2.2.1 Resistance as a Model for the Load
It might appear at first glance that the resistance causes mainly power loss and heating.
However, the concept of heating elements in household appliances or power losses in
long cables covers only a small subset of applications. Important resistance applications
are related to the resistive sensors studied previously. Resistances are widely used in the
circuits to provide different voltage values, i.e., bias circuit components such as diodes
and transistors. Last but not least, resistances model an arbitrary power-absorbing device,
the load, which can be mechanical, acoustical, microwave, optical, etc. From a circuit
point of view it does not matter how the electric power supplied to a load is eventually
transformed. The circuit delivers certain power to the load, but cares little about whether
the power is converted into heat to warm a heating pan, light to illuminate our house,
mechanical power to drive a motor, or electromagnetic radiation generated by a cell
phone. Circuit analysis is concerned with the power delivered to a power-absorbing
device, the load, leaving its utilization and conversion to other engineering disciplines.
Therefore, many practical loads can be replaced by a simple load resistance RL. Such a
resistance is often called the equivalent resistance, RL ¼ Req . It is also valid for AC
circuits. For AC circuits it is convenient to use rms (root mean square) voltages, which are
equivalent to DC voltages and thus provide us with the same power value delivered to the
load. Figure 2.11 shows two examples of the load replacement with the equivalent
resistance: a light source radiating visible light and an antenna radiating microwaves.

a) monopole
antenna a
ground plane
RL
coaxial cable
a
b
b
b)
a a

RL

b b

Fig. 2.11. (a) Radiating monopole antenna is modeled as a resistance. (b) Light source is
approximately modeled as a resistance.

Example 2.4: A small commercial monopole antenna shown in Fig. 2.11a is rated at Req
¼ 50 Ω in the ISM band of 902–928 MHz. When an rms voltage of 10 V is applied to the
antenna, what is the total amount of power radiated by the antenna?

II-46
Chapter 2 Section 2.2: Nonlinear Passive Circuit Elements

Example 2.4 (cont.):

Solution: We use Eq. (2.7) and obtain P ¼ V 2rms =Req ¼ 2 W. All of this power is radiated
in the form of an outgoing electromagnetic wave. There is no heat loss. The resistance Req
is called the radiation resistance in such a case. The above analysis is valid only in a
certain frequency range.

A load that exactly follows Ohm’s law Eq. (2.3) is called the linear load. While the
transmitting antenna in Fig. 2.11a is a linear load, an incandescent light bulb in Fig. 2.11
is not. Most of the loads deviate from the linear Ohm’s law.

2.2.2 Nonlinear Passive Circuit Elements

Nonlinear passive circuit elements do not satisfy Ohm’s law with the constant resistance
R over a wide range of voltages. Therefore, they are also called non-ohmic circuit elements.
The non-ohmic elements may be described by a similar expression:
V
V ¼ RðV ÞI , RðV Þ  ð2:19Þ
I ðV Þ
but with a variable resistance R(V). For passive elements, RðV Þ > 0. The resistance R(V)
is known as the static or DC resistance. Figure 2.12 depicts the υ-i characteristics for
three distinct passive circuit elements.

ohmic conductor incandescent light bulb ideal (Shockley) diode

a) I b) I c) I

0 0 0
V V V

Fig. 2.12. Three υ-i characteristics: (a) linear—resistance; (b) nonlinear—incandescent light bulb;
and (c) nonlinear—ideal or Shockley diode.

The first element is an ohmic element (ohmic conductor) with a constant resistance R.
The corresponding υ-i characteristic is a straight line—the circuit element is linear. The
second element corresponds to an incandescent light bulb. Its resistance R increases when
the applied voltage V increases (the conductivity of the radiating filament of wire decreases
with increasing absorbed power and wire temperature). Hence, the υ-i characteristic bends
down and deviates from the straight line—see Fig. 2.12b. This element only approximately

II-47
Chapter 2 Major Circuit Elements

follows Ohm’s law. It is therefore the nonlinear circuit element. The third element in
Fig. 2.12 corresponds to an ideal (Shockley) diode. The diode does not conduct at negative
applied voltages. At positive voltages, its υ-i characteristic is very sharp (exponential).
The diode is also the nonlinear circuit element. Strictly speaking, the υ-i characteristic of the
incandescent light bulb does not belong to the list of circuit elements due to its limited
applicability. However, the ideal diode is an important nonlinear circuit element. The
nonlinear elements are generally polar (non-symmetric) as Fig. 2.12c shows.

2.2.3 Static Resistance of a Nonlinear Element

Once the υ-i characteristic is known, we can find the static resistance R(V) of the nonlinear
circuit element at any given voltage V0. For example, the υ-i characteristic of the ideal
diode shown in Fig. 2.12c is described by the exponential Shockley equation:

 
V
I ¼ I S exp 1 ð2:20Þ
VT
In Eq. (2.20), the constant IS [A] is the diode saturation current. The saturation current is
very small. The constant VT [V] is called the thermal voltage.

Example 2.5: Give a general expression for the diode resistance R(V) using Eq. (2.20)
and find its terminal values at V ! 0 and V ! 1, respectively. Then, calculate static diode
resistance R0 and diode current I0 when the voltage across the diode is V0 ¼ 0.55 V.
Assume that I S ¼ 1
1012 A and V T ¼ 25:7 mV.
Solution: Using Eq. (2.20) we obtain
V
RðV Þ ¼ 
 
V ð2:21Þ
IS exp 1
VT
When V ! 0, we can use a Taylor series expansion for the exponent. Keeping only the first
nontrivial term, one has expðV =V T Þ  1 þ V =V T . Therefore,
VT
RðV Þ ! when V ! 0 ðor V =V T << 1Þ ð2:22Þ
IS
This value is very large, in excess of 1 GΩ. The diode is thus the open circuit with a good
degree of accuracy.
On the other hand, at large V, the exponential factor in Eq. (2.21) greatly increases.
Therefore,
RðV Þ ! 0 when V ! 1 ð2:23Þ

II-48
Chapter 2 Section 2.2: Nonlinear Passive Circuit Elements

Example 2.5 (cont.):

The diode becomes virtually a short circuit as indicated by the almost vertical slope in
Fig. 2.12c. Finally, we obtain the particular values for V 0 ¼ 0:55 V:

R0 ¼ 275 Ω, I 0 ¼ 2:00 mA ð2:24Þ

2.2.4 Dynamic (Small-Signal) Resistance of a Nonlinear Element

Equally, and perhaps even more important, is the concept of a dynamic (or small-signal)
resistance, r, of the nonlinear circuit element. Other equivalent names include differential
resistance or incremental resistance. Quite often the voltage across the element and the
current through it are given by
V ¼ V 0 þ υ, I ¼ I0 þ i ð2:25Þ
where V0 and I0 are the DC (constant-value) voltage and current related to each other
through the static resistance, V 0 ¼ R0 I 0 . These values are set with the help of an external
DC circuit. On the other hand, quite small components υ and i describe a very weak time-
varying (AC or pulse) signal. Though weak, this signal contains the major information to
be processed. A receiver circuit in your cell phone is an example. Now, how are υ and
i related to each other? The answer is still given by Ohm’s law but written in terms of the
dynamic resistance, i.e.,

dV 
υ ¼ r i, r  ð2:26Þ
dI V ¼V 0 , I¼I 0
This derivative is to be evaluated at the operating point V0, I0 (also called the quiescent
point or Q-point). Figure 2.13 provides a graphical proof of Eq. (2.26) using the example
of an ideal diode. The zoomed-in version of Fig. 2.12c has been used. The dynamic
(small-signal) resistance is thus the inverse slope of the υ-i characteristic at the operating
point. The exact mathematical proof is performed using a Taylor series expansion.

Exercise 2.6: Determine the small-signal resistance r for the ohmic circuit element with
V ¼ RI, R ¼ const.
Answer: r ¼ R for any operating point.

The dynamic diode resistance plays a decisive role in the design of amplifiers based
on junction transistors. The ideal-diode circuit element becomes a part of the transistor
circuit model. The dynamic resistance is also critical for amplifiers which use other
transistor types. From the mathematical point of view, finding the static and dynamic
resistance is simply finding the function and its first derivative at the operating point.

II-49
Chapter 2 Major Circuit Elements

I, mA
5

Operating point tangent at Q

or Q-point

2.5
2.0 I0
i

1/r
0
0.5 v V, volts 0.6
V0

Fig. 2.13. Finding dynamic resistance for an ideal diode.

Example 2.6: Give a general expression for the dynamic diode resistance r using
Eq. (2.20) at an arbitrary operating point with current I0. Then, calculate the dynamic
diode resistance r when the voltage across the diode is V0 ¼0.55 V. Assume that I S ¼ 1

1012 A and V T ¼ 25:7 mV.
Solution: Using Eq. (2.20) we obtain
 
I VT VT
V ¼ V T ln 1 þ ) r ðI Þ ¼ )r ð2:27Þ
IS I þ IS I0
since the saturation current IS may be neglected. At V0 ¼ 0.55 V we obtain I0 ¼ 2.00 mA—
see Eq. (2.24). Therefore,

r ¼ 12:8 Ω when V 0 ¼ 0:55 V: ð2:28Þ

2.2.5 Electronic Switch

In an electronic switch (or a solid-state switch), an electric quantity (voltage or current)
acts as a stimulus—it opens and closes the switch. The electronic switch is inherently a
nonlinear device. Figure 2.14 shows a two-terminal switch and its (idealized) υ-i charac-
teristic. The voltage across the switch V is generated by the main circuit. When this voltage
reaches a certain switch threshold voltage VTh, the switch becomes closed, it can conduct
any current. Further, the voltage across the switch does not change. Switches of this type
usually involve pn-junction diodes. For example, the diode υ-i characteristic from
Fig. 2.12c may approximate the step function in Fig. 2.14b. The unidirectional switch
shown in Fig. 2.14 can pass the current only in one direction. Bidirectional switches can
pass current in both directions.

II-50
Chapter 2 Section 2.2: Nonlinear Passive Circuit Elements

a) Rest of the circuit b) I

a
I Open when V<V Th

+
V 0
VTh V
-

I Closed when V=V Th

b
Rest of the circuit

Fig. 2.14. (a) Two-terminal unidirectional threshold switch and (b) its ideal υ-i characteristic.

Figure 2.15 shows another, three-terminal electronic switch. The voltage V controlling
the switch operation is now generated by a separate (control) circuit. It is still referenced
to the common circuit ground. When the control voltage reaches a certain switch
threshold voltage VTh or exceeds it, the switch closes. The switches of this type involve
transistors, either junction or field effect. A distinct feature of the switch in Fig. 2.15 is
that the control voltage may have arbitrary values, including V > V Th . Therefore, its υ-i
characteristic involves all states to the right of the vertical line in Fig. 2.14.

Rest of the circuit

a
I Open when V<VTh

I
Closed when V>VTh
0V

Fig. 2.15. Three-terminal unidirectional threshold switch (pull-down switch).

The switch shown in Fig. 2.15 and its pull-up counterpart are the “heart” of any digital
circuit, which is essentially a nonlinear switching circuit. An introduction to digital
switching circuits is provided elsewhere.

Exercise 2.7: Based on conditions of example 2.5, determine when a diode switch closes.
This condition approximately corresponds to the diode current of 10 mA.
Answer: The diode voltage should be equal to 0.594 V or approximately 0.6 V.

II-51
Chapter 2 Major Circuit Elements

Section 2.3 Independent Sources

2.3.1 Independent Ideal Voltage Source
An independent ideal voltage source is an important circuit element. Figure 2.16a shows
the corresponding circuit symbol for the DC (steady-state) source. As a circuit element,
the voltage source is not symmetric: terminals 1 and 2 (commonly labeled as plus and
minus or red and black) may not be interchanged without affecting its operation. In other
words, the voltage source is a polar device. The voltage source generates a positive
voltage difference (or voltage drop or simply voltage) across its terminals, V S > 0, the
polarity of which is shown in Fig. 2.16. The term independent means that voltage VS does
not vary because of different parameters of an electric circuit (not shown in the figure),
which is implied to be connected to the voltage source.

a) b)
VS +
1 - 2
+

-
I

Fig. 2.16. Symbol for an ideal voltage source along with the voltage and current behavior.

Current Through the Voltage Source: Active Reference Configuration

The current I flowing through the voltage source is shown in Fig. 2.16 by an arrow. The
relation between voltage polarity and current direction depicted in Fig. 2.16 is known as
the active reference configuration. It is commonly used for all active circuit elements such
as voltage and current sources, either dependent or independent. A useful fluid mechanics
analogy for the voltage source is water (electric current) that is pushed up the “voltage”
hill in Fig. 2.16b by external means. Alternatively, one may think of a water pump that is
characterized by a constant pressure drop. The active reference configuration means that
the voltage source supplies electric power to the circuit. This configuration is the opposite
of the passive reference configuration for the resistance.

υ-i Characteristic of the Voltage Source

Figure 2.17a plots the υ-i characteristic of the ideal voltage source. The term ideal
literally means that the υ-i characteristic is a straight vertical line: the ideal voltage source
is capable of supplying any current to any circuit while keeping the same voltage VS
across its terminals. In reality, it is not the case since the high currents mean high powers.
Therefore, a laboratory power supply—the physical counterpart of the ideal voltage
source—will be eventually overloaded as shown in Fig. 2.17b. Figure 2.17c shows a
common way of drawing the υ-i characteristic for the voltage source with the axes
interchanged. For the purposes of consistency, the x-axis will always be used as the
voltage axis throughout the text.

II-52
Chapter 2 Section 2.3: Independent Sources

a) I b) I overload c) V

VS

overload
0 0 0
VS V VS V I

Fig. 2.17. υ-i Characteristics for (a) ideal voltage source used in the circuit analysis and (b) its
physical counterpart—a regulated laboratory power supply. (c) Typical way of drawing the υ-i
characteristic for the voltage source with the axes interchanged.

Symbols for Independent Voltage Source

Multiple symbols may be used in a circuit diagram to designate the independent ideal
voltage source—see Fig. 2.18. All these symbols are equivalent from the circuit point of
view, as long as we imply the ideal source. However, their physical counterparts are quite
different. The general symbol in Fig. 2.18a implies either an AC to DC converter (the
laboratory power source) or a battery. The symbol in Fig. 2.18b relates to a battery and
Fig. 2.18c–d depicts battery banks.

a) 5V b) 5V - c) 5V - d) 5V -
+

+
+

-
= = =
+

Fig. 2.18. (a) Generic DC voltage source, (b) single battery, and (c) and (d) battery banks. All
symbols in the circuit diagram are equivalent.

Example 2.7: Solve an electric circuit shown in Fig. 2.19—determine circuit current I and
voltage across the resistance V.
Solution: We use the graphical solution—plot the υ-i characteristic of the 2 kΩ resistance
and the υ-i characteristic of the voltage source on the same graph to scale—see Fig. 2.19b.
The intersection point is the desired solution: V ¼ 3 V, I ¼ 1:5 mA. Indeed, this simple
solution implicitly uses circuit laws (KVL and KCL) studied next.

II-53
Chapter 2 Major Circuit Elements

a) b) I, mA

2
I
1
3 V = VS

+
+ R=2 kW V 0
- - VS V, volts
-1

I -2

-4 -2 0 2 4

Fig. 2.19. Electric circuit solution in graphical form.

2.3.2 Circuit Model of a Practical Voltage Source

Any practical voltage source is modeled as a combination of an ideal voltage source VS
and an ideal resistance R in series—see Fig. 2.20a. The resistance R reflects the non-ideality
of the practical source: it limits the maximum available source current and the maximum
available source power by (similar to the current-limiting resistor):
I max ¼ V S =R, Pmax ¼ V S I max ¼ V 2S =R ð2:29Þ
Voltage VS is called the open-circuit voltage of the source for an obvious reason.
Similarly, current Imax is called the short-circuit current of the source. Once both
quantities are measured in laboratory, resistance R (called the internal source resistance)
may be found using Eq. (2.29).

a) b) I
practical voltage source
I
+

R
VS + V
- VS V
-

Fig. 2.20. Circuit model of a practical voltage source and its υ-i characteristic.

Exercise 2.8: The open-circuit voltage of a voltage source is 9 V; the short-circuit current
is 2 A. Determine the internal source resistance.
Answer: 4.5 Ω.

II-54
Chapter 2 Section 2.3: Independent Sources

The υ-i characteristic of the practical voltage source is the plot of source current I versus
voltage V available from the source in Fig. 2.20a. This voltage is generally less than VS
since any nonzero current I causes a voltage drop of RI across resistance R. One has
VS  V
V ¼ V S  RI ) I ¼ ð2:30Þ
R
This υ-i characteristic is plotted in Fig. 2.20b by a solid line. The deviation from the
straight vertical line characterizes the degree of non-ideality. Emphasize that any labora-
tory power supply is indeed a practical voltage source. However, using a special circuit,
its input is regulated so that the output voltage does depend on the output current, at least
over a reasonable range of currents. Therefore, instead of Fig. 2.20b we arrive at a more
reliable voltage source from Fig. 2.17b.

Exercise 2.9: Determine internal source resistance for the source illustrated in Fig. 2.20b
given that every horizontal division is 3 V and every vertical division is 1 A.
Answer: 0.6 Ω.

2.3.3 Independent Ideal Current Source

An independent ideal current source is dual of the ideal voltage source. Figure 2.21a shows
the corresponding circuit symbol for the DC (constant-current) source. As a circuit
element, the constant-current source is directional: terminals 1 and 2 indicate the direction
of the current flow. The current source generates a positive constant current, I S > 0, which
flows from terminal 2 to terminal 1 in Fig. 2.21. The term independent means that current IS
does not vary because of different parameters of an electric circuit (not shown in the figure),
which is implied to be connected to the current source.

a) - b)
+

V
+

1 2
IS -

Fig. 2.21. Symbol for the ideal current source along with voltage and current designations.

Voltage Across the Current Source: Active Reference Configuration

Once the current source is connected to a circuit, a voltage V will be created across it. The
voltage polarity is indicated in Fig. 2.21a. The relation between voltage polarity and
current direction shown in Fig. 2.21 is again the active reference configuration, similar to
the voltage source. A useful fluid mechanics analogy for the electric current source is a

II-55
Chapter 2 Major Circuit Elements

water pump that creates a constant water supply (e.g., 0.5 ft3/s). Indeed, this water pump
will be characterized by a certain pressure difference across its terminals, which is the
analogy of voltage V in Fig. 2.21.

υ-i Characteristic of the Current Source

Figure 2.22a plots the υ-i characteristic of an ideal current source. Compared to the ideal
voltage source, the graph is rotated by 90 degrees. The term ideal again means that the υ-i
characteristic is the straight horizontal line: the ideal current source is capable of creating
any voltage across its terminals while keeping the same current IS flowing into the circuit.
In reality, it is not the case since high voltages mean high powers. Therefore, a laboratory
current power supply—the physical counterpart of the ideal current source—will even-
tually be overloaded as shown in Fig. 2.17b. The current laboratory supplies are rarely
used (one common use relates to transistor testing); they are less common than the
voltages supplies. However, the current sources are widely used in transistor circuits,
both integrated and discrete. There, the current sources are created using dedicated
transistors. Furthermore, photovoltaic sources are essentially current sources.

a) I b) I

IS IS
overload

0 0
V V

Fig. 2.22. υ-i Characteristics of (a) an ideal current source and (b) its physical counterpart.

Symbols for Independent Current Source

A few equivalent symbols may be used in a circuit diagram to designate the independent
ideal current source; see Fig. 2.23. All these symbols are equivalent as long as we imply the
ideal source. The symbol in Fig. 2.23a is used in North America, the symbol in Fig. 2.23b is
European, and the symbol in Fig. 2.23c may be also found in older texts.

a) b) c)
= =

1 mA 1 mA 1 mA

Fig. 2.23. Equivalent symbols of the current source in the circuit diagram.

II-56
Chapter 2 Section 2.3: Independent Sources

Example 2.8: Solve an electric circuit shown in Fig. 2.24—determine voltage V across the
source and across the resistance.
Solution: Similar to the voltage source, we use a graphical solution—plot the υ-i charac-
teristic of the 2-kΩ resistance and the υ-i characteristic of the current source on the same
graph to scale; see Fig. 2.24b. The intersection point gives us the desired solution:
V ¼ 3 V. Note that the solutions for this example and the solution for Example 2.7
coincide. This means that, under certain conditions, we can interchange both sources
without affecting the circuit performance. Indeed, the graphical solution implicitly uses
the circuit laws (KVL and KCL) studied in detail next.

a) b) I, mA

IS 2
IS
1.5 mA = IS R=2 kW 1
+

V V 0
- - V, volts
-1
IS
-2

-4 -2 0 2 4

Fig. 2.24. Electric circuit solution in the graphical form.

2.3.4 Circuit Model of a Practical Current Source

Any practical current source is modeled as a combination of the ideal current source IS
and the ideal resistance R in parallel—see Fig. 2.25a. The resistance R reflects the
non-ideality of the practical source: it limits the maximum available source voltage and
the maximum available source power by

V max ¼ RI S , Pmax ¼ V max I S ¼ RI 2S ð2:31Þ

Voltage Vmax is again called the open-circuit voltage of the source. Similarly, current IS is
called the short-circuit current of the source. Once both the quantities are measured,
resistance R (called the internal source resistance) may be found using Eq. (2.31).

Exercise 2.10: The open-circuit voltage of a current source is 9 V; the short-circuit current
is 2 A. Determine the internal source resistance.
Answer: 4.5 Ω.

II-57
Chapter 2 Major Circuit Elements

a) b) I
practical current source
I IS

+
IS R V
- V

Fig. 2.25. Circuit model of a practical current source and its υ-i characteristic.

The υ-i characteristic of the practical current source is the plot of current
I available from the source versus voltage V across the source in Fig. 2.25b. This current
is generally less than IS since a portion of IS flows through the internal resistance R, i.e.,
V
I ¼ IS  ð2:32Þ
R
This υ-i characteristic is plotted in Fig. 2.25b by a solid line. The deviation from the
straight horizontal line characterizes the degree of non-ideality.

Exercise 2.11: Determine the internal source resistance for the source illustrated in
Fig. 2.25b given that every horizontal division is 3 V and every vertical division is 1 A.
Answer: 15 Ω.

2.3.5 Operation of the Voltage Source

Operation of a voltage power supply of any kind (an electric generator, a chemical
battery, a photovoltaic cell, etc.) might be illustrated based on the charge separation
principle very schematically depicted in Fig. 2.26. We need to deliver electric power to a
load modeled by an equivalent resistance RL. First, we consider in Fig. 2.26 a charged
capacitor with a charge Q connected to a load resistor RL at an initial time moment. The
capacitor voltage V is related to charge by V ¼ Q=C where C is the (constant) capaci-
tance. The capacitor starts to discharge and generates a certain load current IL. At small
observation times, the change in Q is small, so is the change in V. Therefore, the capacitor
initially operates as a voltage power supply with voltage V. However, when time pro-
gresses, the capacitor discharges and the voltage V eventually decreases.

II-58
Chapter 2 Section 2.3: Independent Sources

Charge separation mechanism

-

+ +++++++
+Q -
IL
- IL
-
-
-
-Q -
-
V

IL RL

Fig. 2.26. Power source schematically represented as a capacitor continuously charged by a charge
separation mechanism—the charge pump.

How could we keep V constant, i.e., continuously charge the capacitor? A charge
separation mechanism should be introduced between the hypothetic capacitor plates to
continuously compensate for the charge leakage. That mechanism may have the forms:
1. For an electromechanical generator, this is the Lorentz force that acts on individual
electrons in a conductor and pushes them to one conductor terminal while creating
the opposite charge density on the opposite conductor terminal. The macroscopic
effect of the Lorentz force is the Faraday’s law of induction.
2. For a battery, these are chemical reactions at the electrodes which cause a charge
separation, i.e., positive metal ions dissolve in the electrolyte and leave excess
electrons in the metal electrode on the left in Fig. 2.26.
3. For the photovoltaic cell, this is a built-in potential of the semiconductor
pn-junction that separates light-generated negative carriers (electrons) and positive
carriers (holes) as shown in Fig. 2.26.
Indeed, the capacitor analogy in Fig. 2.26 is only an illustrative approach, especially for
electromechanical power generation. Below, we will consider a few specific examples.

2.3.6 Application Example: DC Voltage Generator with Permanent

Magnets
A realistic electromechanical voltage source—a basic DC generator with permanent mag-
nets—is shown in Fig. 2.27. This generator setup makes use of the Lorentz force:
 
~
f q ~ υ
~ B ð2:33Þ

The Lorentz force acts on charge q moving with a velocity ~ υ in an external magnetic field
~
with the vector flux B measured in tesla (T). The force itself is measured in newtons. The
cross symbol in Eq. (2.33) denotes the vector product of two vectors evaluated according
to the right-hand rule. Shown in Fig. 2.27a are two permanent magnets (stator of the
generator) responsible for creating the magnetic flux ~ B emanating from the north pole

II-59
Chapter 2 Major Circuit Elements

(N) and terminating at the south pole (S). The armature (rotor) rotates clockwise in
Fig. 2.27b with the armature velocity ~
υ.

a) b)
stator B v A= d
B v d B
f
rotor +
+
N S
f v
- +
brushes brushes - - ++

- -
+

+
V V

Fig. 2.27. Charge separation in a DC electromechanical generator.

When the flux density ~ B is applied, every positive charge +q in the armature segment
l will experience a Lorentz force with the magnitude f ¼ þqυB which will move this
charge toward the right terminal of the armature in Fig. 2.27b. Similarly, every negative
charge –q in the armature would experience the equal but oppositely directed Lorentz
force f ¼ qυB which will move this charge toward the left terminal. Hence, a charge
separation occurs along the armature which will give rise to an induced voltage V. Total
work W of the Lorentz force on a charge q along the entire armature path in Fig. 2.27b is
given by W ¼ 2lf . This work divided by the amount of charge determines the equivalent
voltage that will be developed on the generator terminals, i.e., the instantaneous gener-
ator voltage V ¼ W =q ¼ 2lυB. If the armature rotates at an angular speed ω (rad/s), the
charge velocity perpendicular to the field is given by υ ¼ d=2ω cos θ (m/s). Plugging in
this expression and averaging over angles θ from 0 to π/2, we obtain the average
generator voltage in the form

V ¼ ldBhcos θi ¼ ð2=π ÞABω ½V ð2:34Þ

where A is the armature area. If the rotor has N turns, the result is multiplied by N. The
same expression for the voltage is obtained using the Faraday’s law of induction. A
regulator circuit is necessary to obtain a flat DC voltage without ripples. Any brushed DC
motor operates as a generator when its shaft is rotated with a certain speed. The generated
open-circuit voltage may be observed in laboratory with the oscilloscope.

Exercise 2.12: Determine average open-circuit generator voltage in Fig. 2.27 given
A ¼ 0:1 m2 , B ¼ 0:2 T, ω ¼ 20 rad/s (191 rpm), and the armature with 20 turns.
Answer: 5.1 V.

II-60
Chapter 2 Section 2.3: Independent Sources

2.3.7 Application Example: Chemical Battery

A chemical reaction in a battery induces a continuous charge separation. The “charge
pump” so constructed, once connected to a load, is able to create a continuous electric
current into a load and a voltage difference across it. You are probably aware of the quest
to improve the venerable battery. Extensive coverage in the media and in technical
journals frequently reports on new chemical compounds and control circuits. They target
smaller, more powerful rechargeable batteries for such diverse devices as portable
computers, cell phones, sensors, and automobiles. Specifically, it is the automotive sector
which implements hybrid vehicle technology where powerful electro motors in conjunc-
tion with high-performance batteries are supplementing, even completely replacing,
conventional combustion engines. For a standard chemical battery, the two important
parameters are battery voltage and battery capacity. The capacity, Q, is a new quantity
that is needed because of a battery’s inability to provide constant current and power for an
infinite time duration. How a battery behaves over time is illustrated in Fig. 2.28 where
we monitor the power and current as a function of time. A key time constant is the
so-called discharge time, which is critically dependent on the attached load.

a) b)
PB, W IB, A

#1 #1

#2 #2

t t
T1 T2 T1 T2

Fig. 2.28. Generic plots of delivered power, PB, and electric current, IB, for two different loads
labeled #1 and #2. The discharge times T1 and T2 correspond to the loads #1 and #2, respectively.

When a load of resistance RL is connected to the battery, and the battery’s internal
resistance R is negligibly small compared to that resistance, the circuit current, IB, and the
power delivered by the battery, PB, are determined based on Ohm’s law:
VB
IB ¼ , PB ¼ V B I B ð2:35Þ
R
The total energy, EB, stored in the battery and then delivered to the circuit is a fixed
constant. Its value depends on the battery type and size. The total energy in joules is given
by the time integral of delivered power over time, i.e.,
ð
1

EB ¼ PB ðt 0 Þdt 0 ð2:36Þ
0
We can assume that the total energy is a finite constant; it follows from Eq. (2.36) that the
delivered power must drop to zero at a finite time T. This is schematically shown in

II-61
Chapter 2 Major Circuit Elements

Fig. 2.28 for two different load resistances, #1 and #2, which require two different circuit
currents. Even though the two power curves in Fig. 2.28a are different, the area under
those curves, denoting the total energy stored in the battery, remains the same to a
sufficient degree of accuracy. The battery’s terminal voltage VB also remains approxi-
mately constant over the entire operation cycle and even afterwards. It is the battery’s
current IB that finally sharply decreases with time and causes a drop in power, as seen in
Fig. 2.28b. Let us consider the simplest case where the current is a constant for t < T and
at t ¼ T drops to zero and stays zero for t > T. From Eq. (2.36), it follows that
ð
1 ðT
EB ¼ PB ðt Þdt ¼ V B I B dt 0 ¼ ½T I B V B
0 0
ð2:37Þ
0 0

The expression in the square brackets is the definition of the battery capacity, Q:
EB
Q  TIB ¼ ð2:38Þ
VB
Since the battery terminal voltage is always known, its capacity determines the total
energy stored in the battery. The capacity is measured in A·h (Ah) or for small batteries in
mA·h (mAh). The capacity rating that manufacturers print on a battery is based on the
product of 20 h multiplied by the maximum constant current that a fresh battery can
supply for 20 h at 20 C while keeping the required terminal voltage. The physical size
of batteries in the USA is regulated by the American National Standards Institute
(ANSI) and the International Electrotechnical Commission (IEC). Table 2.3 lists the
corresponding parameters of some common batteries.

Example 2.9: A 12-V battery rated at a capacity of Q ¼ 100 A·h may deliver 5 A over a
20-h period, 2.5 A over a 40-h period, or 10 A over a 10-h period. Find the total energy
delivered by the battery provided that its internal resistance is negligibly small.
Solution: The total energy delivered by the battery is equal to
EB ¼ V B Q ¼ 12  100 V  A  h ¼ 1200 W  h ¼ 4:32 MJ ð2:39Þ
It remains constant for each case. This example shows that the electric energy can be
measured either in joules or in Wh or more often in kWh. Clearly, 1 Wh ¼ 3600 J.

Circuit Model of a Battery

As a practical voltage source, a battery always has a small, but finite internal resistance,
R. Battery’s equivalent circuit therefore includes the ideal voltage source and the internal
resistance in series—see Fig. 2.29. Even though the values of R are small, the internal
resistance has critical implications affecting both the battery’s efficiency and its ability to
provide a high instantaneous power output. In general, it is difficult to directly measure

II-62
Chapter 2 Section 2.3: Independent Sources

R
- + +
9V
= 9V
-

Fig. 2.29. Circuit model of a battery: the ideal voltage source in series with an internal resistor.

the internal resistance of batteries, you would need a calibrated load resistor and sophis-
ticated measurement equipment to precisely measure voltages and currents.

Exercise 2.13: A 12-V battery has an internal resistance of 10 Ω. What are the maximum
current and the maximum power that the battery can output?
Answer: I max ¼ 1:2 A, Pmax ¼ 14:4 W

Many battery types have been developed for a wide range of applications. They differ
both in battery energy storage per kg of weight, or unit volume, and in power delivery per
kg of weight, or per unit volume. In particular, modern heavy-duty, deep-cycle batteries
may sport the following properties:
Energy storage : 150 W  h=l; ð2:40Þ
Power density : 2 kW=l: ð2:41Þ
You can compare Eqs. (2.40) and (2.41) with the last row of Table 2.3 and establish the
approximate density of the battery device.

Table 2.3. Characteristics of batteries (from multiple datasheets).

Voltage
Battery size/type Rechargeable (cell) Capacity (Ah) Resistance (R)
AAA No 1.5 1.3 100–300mΩ for
AA No 1.5 2.9 alkaline battery
C No 1.5 8.4 per cell
D No 1.5 20.5
9V No 9.0 0.6 ~400 mΩ
Lithium batteries Yes 3.6–3.7 0.7–1.5 ~300 mΩ
Lead acid starter bat- Yes 12.6 ~600A for 30 s at 32  F <100 mΩ
tery (automotive, deep before voltage drops to
cycle) 7.20 V
Deep-cycle marine, Yes Variable: ~ 30 Wh per kg of weight, ~200 mΩ
electric vehicles or ~ 108 kJ per kg of weight

II-63
Chapter 2 Major Circuit Elements

Section 2.4 Dependent Sources and Time-Varying Sources

2.4.1 Dependent Versus Independent Sources
If the strength of the source (voltage or current) does not vary because of variation of the
circuit parameters, the source is an independent source. The voltage and current sources
considered previously are the independent sources. However, if the strength of the source
is controlled by some dedicated circuit parameters, the sources are called dependent
sources. Figure 2.30 shows circuit symbols (diamonds) for the dependent voltage and
current sources. The ideal dependent sources are the important circuit elements, along
with the independent sources. We explain the notations in Fig. 2.30 as follows.
1. The ideal independent and dependent sources may generate not only the steady-
state (DC) voltages and currents, but also arbitrary time-varying voltages and
currents. To underscore this fact, we will use the lowercase notations for voltages
and currents, respectively.
2. The dependent sources generate (or output) voltage or current in response to some
input voltage or current—the stimulus. To underscore this fact, we will use the
subscript OUT for the generated voltage or current strengths. This is in contrast to the
subscript S, which always denotes the independent sources.
3. The stimulus voltage and the stimulus current (not yet shown in Fig. 2.30) will be
denoted by υin and iin, respectively.

voltage sources current sources

vS + vout +
-
iS iout
-

Fig. 2.30. Circuit symbols for ideal independent and dependent sources, respectively. Lowercase
notations for voltages and currents are used to underscore possible time variations.

2.4.2 Definition of Dependent Sources

The stimulus voltage υin is the voltage across a certain resistance. Likewise, the stimulus
current iin is the current through a certain resistance. Figure 2.31 shows four major types
of dependent sources where the stimulus and the response are combined into one block—
the shaded rectangle. Such a combination reflects the physical reality since this block
usually corresponds to a single circuit component—a transistor or an amplifier. Empha-
size that the resistances in Fig. 2.31 may be reduced to an open circuit for dependent
voltage sources or to a short circuit for dependent current sources, respectively, if
required. Also note that another circuit element may be present in place of the resistance,

II-64
Chapter 2 Section 2.4: Dependent Sources and Time-Varying Sources

for example, the ideal diode. The bottom (ground) nodes in every circuit in Fig. 2.31 may
be interconnected to emphasize the same voltage reference.

a) voltage-controlled voltage source b) voltage-controlled current source

+
vin + vout=Avin vin iout=Gvin
- - -

c) current-controlled voltage source d) current-controlled current source

iin iin
+ vout=Riin iout=Aiin
-

Fig. 2.31. Four major types of dependent sources.

Voltage-Controlled Voltage Source

This dependent source is shown in Fig. 2.31a. The source voltage or the output voltage
follows the input voltage according to a linear law
υout ¼ Aυin ð2:42Þ
where the dimensionless constant A is called the open-circuit voltage gain of the
dependent source. However, units of V/V or V/mV are often used. For example, the
expressions A ¼ 5 V=mV and A ¼ 5000 are equivalent. Equation (2.42) is valid
irrespective of the circuits connected to the dependent source to the right and to the left
in Fig. 2.31a. In this sense, the voltage-controlled voltage source is the ideal circuit
element. Such a source is a voltage amplifier.

Voltage-Controlled Current Source

This dependent source is shown in Fig. 2.31b. The source current or the output current
follows the input voltage according to a linear law:

iout ¼ Gυin ð2:43Þ

where the constant G with units of A=V ¼ Ω1 ¼ S is called the transconductance of
the dependent source, similar to the name conductance. For example, the expressions

II-65
Chapter 2 Major Circuit Elements

G ¼ 0:5 A=mV and G ¼ 500 S are equivalent. Emphasize that the transconductance
has nothing in common with the conductance (inverse resistance) of a passive resistor.
Equation (2.43) is also valid irrespective of the circuits connected to the dependent source
to the right and to the left in Fig. 2.31b. In this sense, the voltage-controlled current source
is again the ideal circuit element. Such a source is a transconductance amplifier.

Current-Controlled Voltage Source

This dependent source is shown in Fig. 2.31c. The source voltage or the output voltage
follows the input current through the resistance in Fig. 2.31c according to a linear law:

υout ¼ Riin ð2:44Þ

where the constant R with units of V=A ¼ Ω is called the transresistance of the dependent
source, similar to the name resistance. For example, the expressions R ¼ 5V=mA and
R ¼ 5000 Ω are equivalent. Emphasize that the transresistance has nothing in common
with the resistance of a passive resistor. Equation (2.44) is again valid irrespective of the
circuits connected to the dependent source to the right and to the left in Fig. 2.31c. In this
sense, the current-controlled voltage source is also an ideal circuit element. Such a source
is a transresistance amplifier.

Current-Controlled Current Source

The last dependent source is shown in Fig. 2.31d. The source current or the output current
follows the input current according to a linear law:

iout ¼ A iin ð2:45Þ

where the dimensionless constant A is called the short-circuit current gain of the
dependent source. However, units of A/A or A/mA are often used. For example, the
expressions A ¼ 0:5 A=mA and A ¼ 500 are equivalent. We repeat that Eq. (2.45) is
valid irrespective of the circuits connected to the dependent source to the right and to the
left in Fig. 2.31d—the voltage-controlled voltage source is the ideal circuit element. Such
a source is a current amplifier.

2.4.3 Transfer Characteristics

The dependent sources do not possess the υ-i characteristic. Instead, a transfer charac-
teristic of the source is used, which relates the output voltage or current to the input
voltage or current. For example, the transfer characteristic of the voltage-controlled
voltage source follows Eq. (2.42). It is a straight line in the υin, υout plane (the xy-
plane), with the slope equal to A. Other linear transfer characteristics are obtained
similarly.

II-66
Chapter 2 Section 2.4: Dependent Sources and Time-Varying Sources

Example 2.10: Solve a circuit shown in Fig. 2.32a—determine current i through the 1-kΩ
resistance. The independent voltage source is given by υS ¼ 0:5 þ 2 cos 2t ½V ; the open-
circuit voltage gain of the dependent voltage source is 5 V/V.
Solution: The input voltage is simply the independent-source voltage, υin ¼ υS . The output
voltage is υout ¼ 5υin ¼ 5υS . The output current follows Ohm’s law:
υOUT
i¼ ¼ 2:5 þ 10 cos 2t ½mA ð2:46Þ
1 kΩ
Note that all circuit parameters now become time dependent. However, this does not
change the solution compared to the steady-state case.

a) b)
+

+
vS + vin + vout 1 kW i iS vin + vout 1 kW i
- - - - -

Fig. 2.32. Two circuits with the dependent voltage-controlled voltage source.

Exercise 2.14: Solve an electric circuit shown in Fig. 2.32b—determine current i through
the 1-kΩ resistance. The independent current source is given by iS ¼ 0:5 þ 2 cos 2t ½mA ;
the open-circuit voltage gain of the dependent source is 5 V/V. The leftmost resistance in
Fig. 2.32b (often called the input resistance) is 1 kΩ.
Answer: i ¼ 2:5 þ 10 cos 2t ½mA (the same answer as in Example 2.10).

2.4.4 Time-Varying Sources

Figure 2.33 shows a number of commonly used symbols for the voltage source—the ideal
circuit element—which differentiate its time-related behavior. Figure 2.33a shows the
steady-state ideal DC voltage source. Figure 2.33b indicates an arbitrary (either steady-
state or variable) ideal voltage source. Figure 2.33c–d indicates a time-harmonic ideal AC
(alternating current) voltage source described by a cosine function in the form

υS ðt Þ ¼ V m cos ðωt þ φÞ ½V ð2:47Þ

where Vm is the AC source amplitude with the units of volts, ω is the AC source angular
frequency, and φ is the phase in degrees or radians. The AC current sources do not have
special symbols—the symbols from Fig. 2.30 are used. The same is valid for the
dependent AC sources, both voltage and current.

II-67
Chapter 2 Major Circuit Elements

a) b) c) d)

VS + vS + vS + vS
- - -

Fig. 2.33. Symbols for independent voltage source which imply (a) the DC source (capital VS), (b)
an arbitrary source (lowercase υS), and (c) and (d) AC time-harmonic sources (lowercase υS).

AC Source Polarity
Since the voltage in Eq. (2.47) is alternating, the polarity of the AC voltage source is also
variable. This circumstance is reflected in Fig. 2.33d where the source polarity is not
shown at all. However, for reference purposes, and when the multiple sources of the same
frequency are present in the circuit, it is always useful to designate the source polarity.
Reversing the AC source polarity means changing the phase in Eq. (2.47) by
180 .

II-68
Chapter 2 Section 2.5: Ideal Voltmeter and Ammeter: Circuit Ground

Section 2.5 Ideal Voltmeter and Ammeter: Circuit Ground

2.5.1 Ideal Voltmeter and Ammeter
The ubiquitous voltmeter and ammeter are devices designed to measure voltages and
currents. Both devices are usually assembled in one unit known as a digital multimeter
(DMM). From the circuit point of view, the ideal voltmeter is an open circuit which
conducts zero current as shown in Fig. 2.34. An ideal ammeter is a short circuit which
conducts any current with zero resistance—see the same figure. In reality, the voltmeter
will conduct a small leakage current, and the ammeter will exhibit a small resistance.

a b a open circuit b
+V -

a b a short circuit b
+A -

Fig. 2.34. Circuit equivalencies for ideal voltmeter and ammeter.

These features guarantee that the connection of the measurement device will not
change the circuit operation. Figure 2.35 shows the proper connection of the voltmeter and
ammeter to measure current through circuit element A and voltage across this element.

I I
+A - X
-
+

V
+V -

Fig. 2.35. Correct connection of voltmeter and ammeter for voltage and current measurements.

The ammeter is always connected in series with element X. In other words, to connect
the ammeter we must break the circuit either before or after element X. Since the ammeter
has no resistance, it acts just like an ideal wire and thus does not perturb the electric
circuit. On the other hand, the voltmeter is always connected in parallel with element X.
The circuit current I in Fig. 2.35 cannot flow through the voltmeter, which acts as an open
circuit. As required, it will flow through element X. We conclude that an ideal voltmeter
does not perturb the circuit either. Generally, voltage measurements are simpler to
perform than current measurements.

Wrong Connections of Ammeter and Voltmeter

The ammeter connected in parallel will short out the element A: the current will flow
through the ammeter. If the element A were a load, there would no longer be a load
resistance in the circuit. And with no attached load, the power supply will deliver the

II-69
Chapter 2 Major Circuit Elements

largest possible current, which will likely burn out the ammeter fuses or destroy other
circuit elements. The voltmeter is an open circuit. Connecting it in series is equivalent to
physically breaking the circuit. The circuit will no longer properly function.

2.5.2 Circuit Ground: Fluid Mechanics Analogy

Consider first a fluid mechanics analogy of an ungrounded electric circuit shown in
Fig. 2.36a.

a) p+p0 V+V0
Water pump
(pressure drop) velocity I
filter resistance
p + + (load)
V
- -
I

p0 V0
b) p V
Water pump
(pressure drop) velocity I
filter resistance
p + + (load)
V
- -
I

0
0V

Fig. 2.36. A large reservoir at atmospheric pressure attached to a pumping system serves as an
analogy to the ground connection in an electric circuit.

A water pump creates a constant pressure difference p between its terminals, which
forces water to move through the filter. The pressure water pump is less common than a
water pump of a constant flux; however, it exactly corresponds to the voltage power
supply of the electric circuit. For entirely closed (isolated) pumping systems, such as
those shown in Fig. 2.36a, the water pressure inside the system can in principle have an
arbitrary pressure deviation p0 from the ambient atmospheric pressure. A large p0 is in
practice undesirable since if the system breaks, then a large pressure difference with
regard to atmospheric pressure will cause high-speed water leakage. Similarly, an isolated
electric circuit may have an arbitrary voltage V0 versus ground voltage, due to static
charge accumulation. We could make the reference level equal to atmospheric pressure
(make p0 equal to zero) if we connect tubing to a large water reservoir at atmospheric
pressure as shown in Fig. 2.36b. There is indeed no water flow through such a connection;
but the pressure level is normalized. A similar situation takes place for the electric ground
shown in Fig. 2.36b. By connecting a point in the circuit to a ground, we normalize the
circuit voltage to the earth’s voltage level, which we define to be 0 V, and eliminate any
static charges. There is no current flow through the ground connection, except, maybe, for
the first time moment. Therefore, this connection is only a voltage reference point. A
similar analogy holds for a current source (pump of a constant flux).
II-70
Chapter 2 Section 2.5: Ideal Voltmeter and Ammeter: Circuit Ground

2.5.3 Types of Electric Ground

Figure 2.37 shows three different types of electric ground connections. The first one is the
earth ground. A true earth ground, as defined by the National Electrical Code (USA),
physically consists of a conductive pipe or rod driven into the earth to a minimum depth
of 8 feet. Obviously, it is not always possible to physically connect the circuit directly to
the earth. Some examples include a cell phone, an automobile, or an airplane.

Earth ground Chassis ground Common (neutral)

ground

0V 0V 0V

Fig. 2.37. Different ground types: earth ground, chassis ground, and common (neutral) ground.

The second ground type is the chassis ground. It is the physical metal frame or structure
of an automobile, an airplane, a desktop computer, a cell phone, or other electrical devices;
the term case is very similar in meaning. The chassis ground primarily involves a connec-
tion to the metal case. It is implied that the case should eventually discharge due to contact
with other objects or with earth. The term ground plane for planar printed circuits, which is
usually the copper bottom of a printed circuit board, is equivalent to chassis ground. The
third ground type in Fig. 2.37 is the common terminal or common ground. The word
common is typical for many circuits including the amplifier circuits considered next, when a
dual-polarity power supply is used. Here two identical batteries are connected in series, plus
to minus. The common terminal of the dual power supply so designed serves as the
reference ground; even a metal case is not necessarily required. The AC analog of the
common ground is the neutral terminal of your wall plug. Frequently, different ground
types may be interconnected. For example, the neutral terminal of the wall plug should be
connected to earth ground at a certain location. The chassis ground of a large truck may be
connected to the physical ground by a little flexible strip nearly touching the asphalt.

2.5.4 Ground and Return Current

We have already seen that electric current in a circuit always flows in closed loops. This
is a simple and yet a very critical property of an electric circuit. The steady-state current
that flows to a load is sometimes called forward current, whereas the current that returns
to the power supply is the return current. Can the chassis ground itself be used as a part
of this loop for the return current? The answer is yes, and Fig. 2.38a depicts this
situation as an example. Here a 9 V battery is powering an incandescent light bulb.
For the chassis ground in Fig. 2.38a, the circuit is correctly drawn, but putting two wires
into the soil, as shown in Fig. 2.38b, will fail owing to the high resistance of the earth.
The use of the ground to establish a path for the return current is quite common for the
chassis ground (automotive electronics) and also for the common ground. However, it
should not be attempted for the earth ground connection. Emphasize that, in many

II-71
Chapter 2 Major Circuit Elements

circuit diagrams, the difference between the chassis ground, the common ground, and
the true earth ground is often ignored. Namely, the symbol of the earth ground used in
the circuit often implies either the chassis ground or the common ground, i.e., the
(physically grounded or not) current return path.

a)

- + - +
9V = 9V

metal chassis short circuit

b)

I I

- + - +
9V = 9V
0V 0V
physical ground (soil) open circuit

Fig. 2.38. (a) The return current path for the chassis ground is metal; it can be replaced by a wire.
(b) There is no current return path, since soil (dry or wet) is a very poor conductor. The circuit is
therefore open and not functioning.

2.5.5 Absolute Voltage and Voltage Drop Across a Circuit Element

The electric ground serves as a voltage reference point in a circuit. It allows us to use two
types of voltages in the circuit:
1. The absolute voltage at a certain circuit node
2. The voltage drop or simply the voltage across a circuit element
Figure 2.39 shows the concept. Voltages Va,b,c,d are absolute voltages measured versus
ground at nodes a, b, c, d in Fig. 2.39. Voltages VA,B,C give the voltage drop across the
circuit elements A, B, and C. Indeed, the ideal wires remain the equipotential surfaces
(have the same absolute voltage). Taking into account the polarity of the voltages VA,B,C
shown in Fig. 2.39, one has for the node voltages
V a ¼ 0 V, V b ¼ V a þ V A ¼ 10 V, V c ¼ V b  V B ¼ 5 V, V d ¼ V c  V C ¼ 0 V ð2:48Þ

Note that both voltage types—absolute voltage and voltage across a circuit element—are
often denoted by the same letter V (in the DC case) and may be easily misplaced. Both of
them are widely used in electric circuit analyses. The hint is that the voltage across a circuit
element always has the polarity labeled with
sign, whereas the absolute voltage often
has not.
II-72
Chapter 2 Section 2.5: Ideal Voltmeter and Ammeter: Circuit Ground

Vb=10 V Vc=5 V
VB=5 V
-

+
b B c

+
VA=10 V A C VC=5 V
- -
a d

0V
Va=0 V Vd=0 V

Fig. 2.39. Absolute voltages measured versus ground in a grounded electric circuit and voltages
across individual circuit elements. Note that there is no voltage drop across ideal wires.

Exercise 2.14: Determine the absolute voltages at nodes 1 through 6 in the circuit shown
in Fig. 2.40.
Answer: Clearly, V 1 ¼ 0V since node 1 is directly connected to ground. Then,

V 2 ¼ V 1 þ V A ¼ 6 V, V 3 ¼ V 2 þ V B ¼ 12 V, V 4 ¼ V 3  V C ¼ 9 V,
V 5 ¼ V 4  V D ¼ 3 V, V 6 ¼ V 5  V E ¼ 0 V

VC=3 V
-
+

3 C 4
+

VB=6 V B D VD=6 V
- -
2 5
+

VA=6 V A E VE=3 V
- -

1 6
0V

Fig. 2.40. Voltages across circuit elements in a grounded electric circuit.

II-73
Chapter 2 Major Circuit Elements

Summary
Passive circuit elements
Name and symbol υ–i Characteristic Physical counterpart (component)
Resistance Resistor

V ¼ RI, σ ¼ qnμ, R ¼ Aσl

P ¼ V I ¼ RI 2 ¼ V 2 =R

Short circuit: Short wire of almost zero

resistance
Ammeter

Open circuit: Air gap of almost infinite

resistance
Voltmeter

Ideal diode Electronic diode

Transistor junctions
Solar cell
h
i
I ¼ I S exp VVT  1
Static resistance: R0  V 0 =I ðV 0 Þ
Dynamic resist.: r  dV =dIjV 0 , I 0
Threshold switch Diode
Transistor
Open circuit when V < V Th
Short circuit when V ¼ V Th

(continued)

II-74
Chapter 2 Summary

Active circuit elements

Name and symbol υ–i Characteristic Physical counterpart (component)
Independent voltage Practical voltage source
source

Independent current Practical current source

source

Voltage-controlled Transistor
voltage source Amplifier

υout ¼ Aυin
A—open-circuit voltage gain
[V/V, V/mV] (dimensionless)

Current-controlled Transistor
voltage source Amplifier
υout ¼ Riin
R—transresistance [V/A, V/mA]
(units of resistance, Ω)

Voltage-controlled Transistor
current source Amplifier
iout ¼ G υin
G—transconductance [A/V]
(units of conductance, Ω1)

Current-controlled Transistor
current source Amplifier
iout ¼ Aiin
A—short-circuit current gain
[A/A, A/mA] (dimensionless)

II-75
Chapter 2 Major Circuit Elements

R=2.5 W
Problems short circuit

I, A
2.1 Resistance: Linear 5

Passive Circuit Element

2.1.2 Resistance
2.1.3 υ-i Characteristic of the Resistance: 0
V, volts
Open and Short Circuits
2.1.4 Power Delivered to the Resistance
2.1.5 Finding Resistance of Ohmic -5
-5 0 5
Conductors
Problem 2.1. Plot υ-i characteristics of the
Problem 2.4. Given υ-i characteristics of a
following resistances: (A) 8 kΩ, (B) 2 kΩ,
resistance determine the corresponding con-
(C) 1 kΩ, and (D) 500 Ω. Clearly label each
ductance. Show units.
characteristic.
a) I, mA
I, mA
6
2
3
1
0
0 V, volts
V, volts -3
-1
-6
-2
-4 -2 0 2 4
-4 -2 0 2 4 b) I, mA

6
Problem 2.2. Plot υ-i characteristics of the
following resistances: (A) 1.667 Ω. (B) Open 3
circuit on the same graph. 0
V, volts
R=1.667 W open circuit -3

I, A -6
5
-4 -2 0 2 4

Problem 2.5. An incandescent energy-saving

0
light bulb (“soft white”) from General Electric is
V, volts rated to have the wattage of 57 W when the
applied AC voltage is 120 V rms (root mean
square). This means that the corresponding DC
-5 voltage providing the same power to the load is
-5 0 5 exactly 120 V. When the bulb is modeled as a
resistance, what is the equivalent resistance value?
Problem 2.3. Plot υ-i characteristics of the
following resistances: (A) 2.5 Ω. (B) Short cir-
cuit on the same graph.
120 V +
-

II-76
Chapter 2 Problems

Problem 2.6. The power absorbed by a resistor B. Solve task A when the wire cross section
from the ECE laboratory kit is 0.2 W. Plot the is increased to 2.5 mm2.
υ-i characteristics of the corresponding resis-
tance to scale given that the DC voltage across Problem 2.11. Determine the total resistance of
the resistor was 10 V. the following conductors:
A. A cylindrical silver rod of radius
I, mA 0.1 mm, length 100 mm, and conductiv-
8
ity 6.1
107 S/m.
B. A square graphite bar with the side of
4 1 mm, length 100 mm, and conductivity
0 3.0
104 S/m.
V, volts C. A semiconductor doped Si wafer with
-4 the thickness of 525 μm. Carrier mobility
-8 is μ ¼ 0:15 m2/(Vs). Carrier concentra-
-4 -2 0 2 4
tion is n ¼ 1023 m3 . Carrier charge is
1.6
1019 C. The resistance is mea-
Problem 2.7. The number of free electrons in sured between two circular electrodes
copper per unit volume is n ¼ 8:46
1028 m13 . with the radius of 1 mm each, which
The charge of the electron is 1.60218
1019 are attached on the opposite sides of the
C. A copper wire of cross section 0.25 mm2 is wafer. Assume uniform current flow
used to conduct 1A of electric current. between the electrodes.
A. Sketch the wire, the current direction, a)
and the direction of electron motion.
B. How many coulombs per one second is b)
transported through the conductor?
C. How fast do the electrons really move?
In other words, what is the average elec- c)

tron velocity?

Problem 2.8. Repeat the above problem when

the conductor’s cross section is increased to
5 mm2.

Problem 2.9. A copper wire having a length of Problem 2.12. A setup prepared for a basic
1000 ft and a diameter of 2.58826 mm is used wireless power-transfer experiment utilizes a
to conduct an electric current of 5 A. square multi-turn loop schematically shown in
A. What is wire’s total resistance? Compare the figure, but with 40 full turns. A #22 gauge
your answer to the corresponding result copper wire with the diameter of 0.645 mm is
of Table 2.2. used. Total loop resistance, R, is needed. Please
B. What is the power loss in the wire? assist in finding the loop resistance (show units).

Problem 2.10
A. A copper wire having a length of 100 m
and a cross section of 0.5 mm2 is used to
conduct an electric current of 5 A. What
is the power loss in the wire? Into what is
this power loss transformed? Problem 2.13. Estimate resistance, Rn (show
units), of the n-side of a Si pn-junction diode in

II-77
Chapter 2 Major Circuit Elements

the figure that follows. We model the n-side by The transistor has three terminals (metal con-
a Si bar having the following parameters: tacts): drain (with voltage V DS > 0 vs. source),
1. Length of L ¼ 0.0005 cm ¼ 5 μm. gate (with voltage V GS > 0 vs. source), and
2. Cross section of A ¼ 0.01 cm
source itself (grounded). The source is also
0.01 cm ¼ 1
104 cm2. connected to a metal conductor on the other
3. Uniform electron concentration (carrier side of the semiconductor body. Accordingly,
concentration) of n ¼ 1017 cm3 . This there are two types of the electric field within
value is typical for a Si diode the semiconductor body: the horizontal field
pn-junction. Carrier mobility is created by VGS, and the vertical field created
μn ¼ 1450 cm2/(Vs). by VDS. The horizontal field fills a conducting
V
- channel between the drain and the source with
+

V
-
+

anode cathode charge carriers, but has no effect on the vertical

= p n charge motion. The resulting carrier concen-
tration in the channel is given by n ¼ N ðV GS 
V Th Þ > 0, N ¼ const, V Th ¼ const: The indi-
Problem 2.14. Estimate resistance, Rp (show vidual carrier charge is q. Given the channel
units), of the p-side of a Si pn-junction diode in cross section A, the carrier mobility μ, and
the figure that follows. We model the p-side by the channel length L, determine transistor
a Si bar having the following parameters: current ID and transistor resistance (drain-to-
1. Length of L ¼ 0.0005 cm ¼ 5 μm. source resistance) RDS. Express both results in
2. Cross section of A ¼ 0.01 cm
terms of quantities listed above including VGS
0.01 cm ¼ 1
104 cm2. and VTh.
3. Uniform hole concentration (carrier con-
centration) of n ¼ 1017 cm3 . This value
is typical for a Si diode pn-junction. Car- 2.1.6 Application Example: Power Loss
rier mobility is μn ¼ 500 cm2/(Vs). in Transmission Wires and Cables
V
- Problem 2.16. An AWG 0000 aluminum trans-
+

V
-
+

anode cathode mission grid cable has the wire diameter of

= p n 11.68 mm and the area of 107 mm2. The con-
ductivity of aluminum is 4.0
107 S/m. The
total cable length (two cables must run to a
Problem 2.15. A cross section of the most
load) is 100 km. The system delivers 10 MW
popular NMOS transistor is shown in the
of DC power to a load. Determine the power
following figure.
loss in the cable (show units) when load voltage
Drain and current are given by:
VDS 1. V ¼ 200 kV and I ¼ 50 A
x
VGS
- 2. V ¼ 100 kV and I ¼ 100 A
+

L
3. V ¼ 50 kV and I ¼ 200 A
+ + + + + +

Gate ID Why do you think the high-voltage power trans-

VGS mission is important in power electronics?
source transmission line (TL) load
0
a
Source Channel with I I
electron carriers + RL
0V - I I
b

50 km

II-78
Chapter 2 Problems

Problem 2.17. Solve the previous problem length of 1000 m. Each line uses AWG#10
when the total cable length (two cables must aluminum wire with the diameter of 2.59 mm.
run to a load) is increased to 200 km. The conductivity of aluminum is 4.0
107 S/
m. The house load is an electric range with the
Problem 2.18. An AWG 00 aluminum trans- resistance of 20 Ω. Determine total power
mission grid cable has the wire diameter of delivered by the generator, Ptotal, total power
9.266 mm. The conductivity of aluminum is loss in the transmission lines, Ploss, and total
4.0
107 S/m. A power transmission system useful power, Puseful (show units).
that uses this cable is shown in the figure that
follows. The load power is 1 MW. Determine 2.1.7 Physical Component: Resistor
the minimum necessary load voltage V that Problem 2.20. A leaded resistor has color
guarantees us a 1 % relative power loss in the bands in the following sequence: brown,
cables. black, red, gold. What is the resistor value?

source transmission line (TL) load Problem 2.21. Potentiometer operation may be
schematically explained as moving sliding con-
+
+
I I tact #2 in the following figure along a uniform
V?
- I I conducting rod with the total resistance of
-
20 kΩ. Determine resistance between terminals
100 km 1 and 2 as well as between terminals 2 and 3 of
the potentiometer, when the sliding contact is at
Problem 2.19. An AC-direct micro- one fifth of the rod length.
hydropower system is illustrated in the figure
that follows. a) 1 b) 1

20 kW 20 kW 4/5

1/5
2 2
3 3

2.2 Nonlinear Passive Circuit

Elements
source transmission line (TL) to each house load
2.2.2NonlinearPassiveCircuitElements
2.2.3 Static Resistance
+
I I
RL
2.2.4Dynamic(Small-Signal)Resistance
- I I
2.2.5 Electronic Switch
Problem 2.22. A nonlinear passive circuit ele-
1 km ment—the ideal diode—is characterized by
Reprinted from Micro-Hydropower Systems the hυ-i characteristic

i in the form
Canada 2004, ISBN 0-662-35880-5. I ¼ I S exp V
VT  1 with I S ¼ 1
1013 A
The system uses a single phase induction gen-
erator with the rms voltage (equivalent DC and V T ¼ 25:7 mV. Find the static diode resis-
voltage) of 240 V. The system serves four tance R0 and the diode current I0 when
small houses, each connected to the generator (A) V0 ¼ 0.40 V, (B) V0 ¼ 0.50 V,
via a separate transmission line with the same (C) V0 ¼ 0.55 V, and (D) V0 ¼ 0.60 V.

II-79
Chapter 2 Major Circuit Elements

Problem 2.23. Find the dynamic (small-signal) Justify your answer.

resistance r of a nonlinear passive circuit ele-
I
ment—the ideal diode—when the operating

+
DC point V0, I0 is given by the solutions to
V V
the previous problem. Consider all four cases.
- -
Problem 2.24. A nonlinear passive circuit ele- I
ment is characterized by the υ-i characteristic in
V =V S
the form I ¼ I S pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi with I S ¼ 1 A and
2
1þðV =V S Þ
V S ¼ 1 V. Plot the υ-i characteristic to scale. 0
Next, find the static element resistance R0, the V
element current I0, and the corresponding
dynamic element resistance r when
(A) V0 ¼ 0.1 V, (B) V0 ¼ 1.0 V, and
(C) V0 ¼ 5.0 V.

1
I, A 2.3 Independent Sources
2.3.1 Independent Ideal Voltage Source
2.3.2 Circuit Model of a Practical Volt-
age Source
0
V, volts Problem 2.27. In the following figure, deter-
mine if the element is a resistance or a voltage
source. Find the power delivered to element
-1 A or taken from element A in every case.
-5 0 5
A) 1 mA
Problem 2.25. Repeat the previous problem A
when I S ¼ 0:5 A and V S ¼ 0:5 V. All other VA=20 V
-
+

parameters remain the same. Consider the follow-

ing DC operating points: (A) V0 ¼ 0.05 V, B) -1 mA
(B) V0 ¼ 0.50 V, and (C) V0 ¼ 2.50 V. A
- VA=20 V
+

Problem 2.26. A DC circuit shown in the

following figure includes two interconnected C) -1 mA
passive elements: an ideal diode and a resis- A
tance. One possible circuit solution is given by VA=-20 V
-
+

an intersection of two υ-i characteristics marked

by a circle in the same figure. This solution pre- Problem 2.28. Based on voltage and current
dicts a non-zero circuit current and a positive measurements, determine if the circuit element
voltage across both circuit elements. is a resistance or a voltage source. Readings of
A. Is this solution an artifact (a mistake has the ammeter and voltmeter are shown in the
been made somewhere)? following figure.
B. Is this solution true (the circuit so
constructed might function)?

II-80
Chapter 2 Problems

A) a) I
+A -

+
+
3 V = VS + V
+1 A - -
+V -
I
-5 V b)
B) 1 I, A
+A -

-1 A
+V -

-5 V 0
C) V, volts
+A -

-1 A
+V - -1
-5 0 5
+5 V

Problem 2.29. The figure that follows shows a Problem 2.31. Plot to scale the υ-i characteris-
circuit with a passive nonlinear circuit element tic of the practical voltage source shown in the
shown by a rectangle. following figure.

a) a) practical voltage source

I I
+
+

1W

+
3 V = VS + V
- - 3V + V
I -
-
b)
1 I, A
b)
1 I, A

0
V, volts
0
V, volts

-1
-5 0 5
-1
-5 0 5
The polarity (direction of current inflow for
passive reference configuration) of the Problem 2.32. The following figure shows a
element is labeled by a sign plus. The υ-i circuit with a passive nonlinear circuit element
characteristic of the element is also shown in labeled by a rectangle. Element’s polarity
the figure. Determine current I and voltage V. (direction of current inflow for passive refer-
ence configuration) of the element is indicated
Problem 2.30. Repeat the previous problem for
by a sign plus. The υ-i characteristic of the
the circuit shown in the figure that follows.

II-81
Chapter 2 Major Circuit Elements

element is also shown in the figure. Determine to the circuit element or taken from it in
circuit current I. every case.
a) R=2 W
Problem 2.34. The following figure shows a
I
circuit with a passive nonlinear circuit element

+
+
4 V = VS + V shown by a rectangle. Element’s polarity
- -
(direction of current inflow for passive refer-
I ence configuration) of the element is labeled by
b)
1
I, A a sign plus. The υ-i characteristic of the element
is also shown in the figure. Determine current
I and voltage V.
a)
0 I

+
V, volts

+
600 mA = IS V
-
I
-1 b)
-5 0 5 1 I, A

2.3.3 Independent Ideal Current Source

2.3.4 Circuit Model of a Practical 0
V, volts
Current Source
Problem 2.33. Readings of the ammeter and
voltmeter are shown in the following figure.
-1
-5 0 5
A)
- A+
Problem 2.35. Repeat the previous problem for
+2 mA the circuit shown below.
+V -

-5 V a)
B)
+
+

+A -
600 mA = IS V
-
-0.5 A
- V+
b)
+10 V I, A
C) 1
- A+

+1 A
- V+

+10 V 0
V, volts

Based on voltage and current measurements,

determine if the element is a resistance or a
current source. Then, find the power delivered -1
-5 0 5

II-82
Chapter 2 Problems

Problem 2.36. Plot to scale the υ-i characteris-

800 mA= IS
tic of the practical current source shown in the
following figure. + 3 V= VS R=2 W
-
a) practical current source
I
1 I, A

+
8.33 W
0.8 A V
-
0
b) V, volts
1 I, A

-1
-5 0 5
0
V, volts

2.3.7 Application Example: Chemical

-1
Battery
-5 0 5 Problem 2.39. The electronics aboard a certain
sailboat consume 96 W when operated from a
24 V source.
Review Problems A. If a certain fully charged deep-cycle
Problem 2.37. For every circuit element shown marine battery is rated for 24 V and
in the following figure, plot its υ-i characteristic 100 A h, for how many hours can the
on the same graph. electronics be operated from the battery
without recharging? (The ampere-hour
rating of the battery is the battery capac-
600 mA= IS
ity—the operating time to discharge the
+ 4 V= VS R=5 W
- battery multiplied by the current).
B. How much energy in kilowatt hours is
I, A
initially stored in the battery?
1

Problem 2.40. A motor of a small, unmanned

electric vehicle consumes 120 W and operates
from a 24-V battery source. The source is rated
0
V, volts for 200 Ah.
A. For how many hours can the motor be
operated from the source (a battery bank)
-1 without recharging?
-5 0 5 B. How much energy in kilowatt hours is
initially stored in the battery source?
Problem 2.38. Repeat the previous problem:
for every circuit element shown in the figure
below, plot its υ-i characteristic on the same Problem 2.41. A certain sensing device
graph. operates from a 6-V source and consumes
0.375 W of power over a 20-h time period.
The source is a combination of four fully
charged AAA batteries, 1.5 V each, assembled

II-83
Chapter 2 Major Circuit Elements

in series. The batteries discharge by the end of

the 20-h period. iin 40 W 2 kW
A. What is the expected capacity of a typ- vS + +
- vout i
-
ical AAA battery used, in mAh?
B. How much energy in Joules was stored
in each AAA battery?
Problem 2.46. Solve an electric circuit
Problem 2.42. How many Joules are in 1 kWh shown in the following figure—determine volt-
and how many Nm does this correspond to? age υ through the 30-Ω resistance. The inde-
pendent current source is given by
2.4 Dependent Sources iS ¼ 0:05  0:2 cos 2t ½mA ; the transcon-
ductance of the dependent source is 10 A/V.
and Time-Varying Sources
2.4.1 Dependent Versus Independent 0.1 kW 30 W

+
Sources iS vin iout v
2.4.2 Definition of Dependent sources - -
2.4.3 Transfer Characteristics
2.4.4 Time-Varying Sources
Problem 2.43. Draw circuit diagrams for four Problem 2.47. Solve an electric circuit
major types of dependent sources, label stimu- shown in the following figure—determine volt-
lus voltage/current and output voltage/current. age υ through the 1-kΩ resistance. The inde-
Describe operation of each dependent source. pendent voltage source is given by
υS ¼ 0:05 þ 0:1 cos 4t ½V ; the short-circuit
Problem 2.44. Solve an electric circuit shown current gain of the dependent source is 10 A/A.
in the following figure—determine current
i through the 2-kΩ resistance. The independent 50 W 1 kW
iin
current source is given by iS ¼ 0:2  1:5 cos 5

+
vS + iout
t ½mA ; the open-circuit voltage gain of the - v
-
dependent source is 12 V/V.

3 kW 1 kW Problem 2.48. Solve an electric circuit

+

iS vin +
- vout i shown in the following figure—determine cur-
- rent i through a nonlinear passive circuit ele-
ment shown by a rectangle. Element’s polarity
(direction of current inflow for passive
Problem 2.45. Solve an electric circuit shown reference configuration) is labeled by a sign
in the following figure—determine current plus. The υ-i characteristic of the element is
i through the 2-kΩ resistance. The independent also shown in the figure. The independent
voltage source is given by υS ¼ 0:3 þ 0:7 cos current source is given by iS ¼ 0:1 A ; the
6t ½V ; the transresistance of the dependent open-circuit voltage gain of the dependent
source is 250 V/A. source is 5 V/V.

II-84
Chapter 2 Problems

a)
2.5 Ideal Voltmeter
6 kW

+
iS vin + vout i
and Ammeter: Circuit
-
-
Ground
2.5.1 Ideal Voltmeter and Ammeter
b)
1 I, mA 2.5.3 Types of Electric Ground
2.5.4 Ground and Return Current
Problem 2.50. You attempt to measure electric
current through a resistance as part of a circuit.
0 Is the following figure appropriate? What is the
V, volts current across the 51-Ω resistor?
I
-1
51 W
-5 0 5
+A-
Problem 2.49. Solve the electric circuit shown
in the following figure—determine voltage υ Problem 2.51. You attempt to measure voltage
across the 1-kΩ resistance. across a resistance in the circuit. Is the follow-
a)
ing figure correct? What is the voltmeter’s read-
1 kW
ing, assuming an ideal instrument?
+

iin
+

vS + iout v I=20 mA
- -
51 W

+V -
b) I, mA
0.1
Problem 2.52. Two circuits with an incandes-
cent light bulb are shown in the following
figure. Will they function? Explain.
0
V, volts a)
5V
-
+

-0.1
-1.5 0 1.5
b)
5V
A nonlinear passive circuit element is shown by -
+

a rectangle. Element’s polarity (direction of cur-

rent inflow for passive reference configuration)
of the element is labeled by a sign plus. The υ-i
characteristic of the element is also shown in the Problem 2.53. A 9-V battery is connected to a
figure. The independent voltage source is given 7.5-kΩ resistor shown in the following figure.
by υS ¼ 0:6 V; the short-circuit current gain of
the dependent source is 100 A/A.

II-85
Chapter 2 Major Circuit Elements

Find current I in every case. (a) The negative 2.5.5 Absolute Voltage and Voltage Drop
terminal is left disconnected; (b) the negative Across a Circuit Element
terminal is connected to the positive terminal Problem 2.55. Determine if the circuit element
through the resistor; (c) both terminals are shown in the following figure is a resistance, a
connected to chassis ground. voltage source, or a wire (short circuit). Absolute
voltages at points a and b are measured versus
a) 7.5 kW
ground.
I
- + I
Va Vb
9V
a b
0V
0V
b)
7.5 kW
1. V a ¼ 3 V, V b ¼ 3V, I ¼ 1 A
2. V a ¼ 3 V, V b ¼ 1V, I ¼ 1 A
- + I

9V
3. V a ¼ 2 V, V b ¼ 5 V, I ¼ 2 A:

c) 7.5 kW Problem 2.56. Determine if the circuit element

shown in the following figure is a resistance, a
voltage source, or a wire (short circuit). Absolute
- + I
voltages at points a and b are measured versus
9V ground.
I
Va Vb
Problem 2.54. What is the voltmeters’ (amme-
ter’s) reading in the figure below? a b

10 W 0V
a)

+V - 1. V a ¼ 6 V, V b ¼ 3 V, I ¼ 1 A
+ +
5V
- #1
V
-
#2 2. V a ¼ 1 V, V b ¼ 1 V, I ¼ 1 A
3. V a ¼ 7V, V b ¼ 5V, I ¼ 2 A:
0V

b) 10 W

+V -
+
5V + A #2
- #1 -

0V

II-86
Chapter 2 Problems

Problem 2.57. Determine absolute voltages at Problem 2.58. Determine voltages across cir-
nodes 1 through 6 in the circuit shown in the cuit elements A, B, C, D, and E in the circuit
following figure. shown in the following figure.

VC=4 V
- VC
-
+

+
3 C 4 7V C 12 V

+
+

+
VB=10 V B D VD=4 V VB B D VD
- - - -
2 5 5V 6V

+
+

VA=5 V A E VE=7 V VA A E VE
- - - -

1 6 0V 0V
0V 0V

II-87
Chapter 3

Chapter 3: Circuit Laws and Networking

Theorems

Overview
Prerequisites:
- Knowledge of circuit elements, their
-i characteristics, and Ohm’s law (Chap. 2)

Objectives of Sect. 3.1:

- Understand the meaning of an electric network and its topology (nodes, branches,
loops, meshes)
- Review the Kirchhoff’s current law, its use, and value
- Review the Kirchhoff’s voltage law, its use, and value
- Become familiar with the Tellegen’s theorem and Maxwell’s minimum heat theorem

Objectives of Sect. 3.2:

- Be able to combine sources and resistances in series and parallel
- Practice in the reduction of resistive networks using series/parallel equivalents
- Realize the function and applications of the voltage divider circuit
- Realize the function of the current divider circuit
- Understand the function and applications of the Wheatstone bridge

Objectives of Sect. 3.3:

- Understand the role and place of linear circuit analysis
- Learn the superposition theorem
- Understand the decisive value of superposition theorem for linear circuit analysis
- Learn about immediate applications of the superposition theorem
- Obtain the initial exposure to Y and
networks and to T and  networks

Application examples:
- Voltage divider as a sensor circuit
- Voltage divider as an actuator circuit
- Superposition theorem for a cellphone

© Springer Nature Switzerland AG 2019 III-89

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_3
Chapter 3 Circuit Laws and Networking Theorems

Keywords:
Electric network, Branches of electric network, Nodes of electric network, Loops of electric network,
Meshes of electric network, Essential mesh, Branch currents, Branch voltages, Series connection,
Parallel connection, Shunt connection, Kirchhoff’s current law, Kirchhoff’s voltage law, Maxwell’s
minimum heat theorem, Tellegen’s theorem, Power conservation law for electric networks, One-port
network, Equivalent electric networks, Equivalent electric circuits, Series battery bank, Battery pack,
Dual-polarity power supply, Common ground of the dual-polarity power supply, Virtual ground of the
dual-polarity power supply, Parallel battery bank, Series and parallel combinations (of resistances, of
conductances),, Equivalent resistance, Equivalent circuit element, Reduction of resistive networks,
Voltage divider circuit, Voltage division rule, Sensor circuit sensitivity, Maximum sensitivity of the
voltage divider circuit, Current limiter, Current-limiting resistor, Current divider circuit, Current
division rule, Wheatstone bridge (definition of, difference signal, difference voltage, balanced),
Linear circuit (definition of, homogeneity, additivity, superposition), Nonlinear circuit (definition of,
linearization, dynamic or small-signal resistance), Superposition theorem, Superposition principle, Y
network, Δ network, Two-terminal networks, Three-terminal networks, Conversion between Y and Δ
networks, Replacing a node by a loop, Δ to Y transformation, Y to Δ transformation, Balanced Y
network, Balanced Δ network, Star to delta transformation, T network, T pad, Π network, Π pad, Two-
terminal network (definition of, input port, output port)

III-90
Chapter 3 Section 3.1: Circuit Laws: Networking Theorems

Section 3.1 Circuit Laws: Networking Theorems

Electric components have to be interconnected to obtain functional circuits which perform
specific tasks like driving a motor or monitoring a power plant. Interconnected circuit components
form an electric network. In turn, any electric network can be evaluated using two simple yet
very general laws: Kirchhoff’s current law (KCL) and Kirchhoff’s voltage law (KVL). Series,
parallel, and other combinations of circuit elements, whether linear or not, can be explained
and solved using these laws.1 They were established by Gustav Kirchhoff (1824–1887), a
German physicist and mathematician, in 1845, while Kirchhoff was a 21-year-old student at the
University of Königsberg in East Prussia. The analyses of all electric circuits are based on KCL
and KVL.

3.1.1 Electric Network and Its Topology

An electric network can be studied from a general mathematical point of view. If the specific
electrical properties are abstracted, there remains a geometrical circuit, characterized by sets of
nodes, branches, and loops. These three items form the topology of an electric network, and the
interconnection of its elements can be represented as a graph. The study of electric network
topology makes it possible to:
A. Identify identical circuit blocks in electric circuits which can be drawn in a variety of ways.
Examples include series/parallel connections, as well as wye (Y or T) and delta (Δ or Π) blocks
as considered in this chapter.
B. Analyze general properties of very large electric circuits such as electric power grids. For
example, there are important relationships between the power grid topology and risk identifi-
cation and mitigation.
C. Relate electric circuits to other disciplines. For example, electric networks have a remarkable ability
to model the dynamical behavior of complicated biological systems.

Nodes, Branches, Loops, and Meshes

We consider an electric network with four arbitrary circuit elements A to D as shown in Fig. 3.1a.
A branch is a two-terminal circuit element. All four two-terminal elements in Fig. 3.1a are
therefore branches.

1
The KCL and KVL concepts are so powerful that they even find applications in equivalent form in magnet systems such
as transformers and motors. For example, the magnetic flux in a yoke with air gaps can be model according to KCL
and KVL.

III-91
Chapter 3 Circuit Laws and Networking Theorems

Fig. 3.1. An electric network and assigning branch voltages/currents.

A node is a point of interconnection of two or more branches. All small six circles in Fig. 3.1a
are formally identified as nodes. Every node i can be assigned a certain voltage Vi with respect to
circuit ground. If a short circuit (a connecting wire) connects two or more nodes, these nodes
constitute a single node since they have the same voltage. The circuit may be redrawn to reduce
the number of nodes and keep only the meaningful nodes (single nodes) with the distinct voltages
as shown in Fig. 3.1b. The circuits in Fig. 3.1a and in Fig. 3.1b are identical. A loop is any closed

III-92
Chapter 3 Section 3.1: Circuit Laws: Networking Theorems

path between two or more branches. There are three loops in Fig. 3.1b. A mesh (or essential mesh)
is a loop that does not contain any other loops within it. There are two meshes in Fig. 3.1b. A
planar (or two-dimensional) electric network with b branches, n nodes, and m meshes satisfies,
after keeping only nodes with distinct voltages (single nodes), the equality
b¼nþm1 ð3:1Þ
which is sometimes called the fundamental theorem of network topology. It is proved by
considering the electric network as a polygonal graph in two dimensions, where each edge is a
branch and each single node is a vertex.

Branch Currents and Voltages

The branch currents and their directions may be assigned arbitrarily, see Fig. 3.1c. The physical
currents either coincide with them or are directed in opposite directions. This can easily be found
by checking the sign of the current value once the analysis is complete. If the branch voltage
polarities have to be assigned afterward, they should satisfy the same reference configuration
for all branches. Let us say the passive reference configuration with regard to the previously
assigned current directions is seen in Fig. 3.1d. The branch voltages (voltage drops) in Fig 3.1d are
expressed through the node voltages according to:
V A ¼ V 1  V 3, V B ¼ V 1  V 2, VC ¼ VD ¼ V2  V3 ð3:2Þ
Conversely, the branch voltages may be assigned arbitrarily at first. If the directions of the branch
currents have to be assigned afterward, they should again satisfy the same reference configuration.

Exercise 3.1: Establish whether or not the networks in Fig. 3.1a and in Fig. 3.1b satisfy
Eq. (3.1).
Answer: The answer is yes for both figures if we consider single nodes. However, if we
consider every small circle in Fig. 3.1a as a node, Eq. (3.1) will not be satisfied.

Series and Parallel Connections

Two or more branches (circuit element) are in series if they exclusively share a common node.
Elements A and B in Fig. 3.1d are in series. Elements B and C are not since node 2 is also shared
by element D. Two or more branches (circuit elements) are in parallel if they are connected to the
same two nodes. Elements C and D in Fig. 3.1d are in parallel. The parallel connection is also called
the shunt connection; the parallel element may be called shunt (or shunting) elements.

3.1.2 Kirchhoff’s Current Law (KCL)

Let us begin our investigation with KCL. It specifically applies to the nodes in an electric network.
Kirchhoff’s current law simply states that the net current entering the node is zero. In other words,
the sum of inflowing currents is equal to the sum of outflowing currents. This statement is also

III-93
Chapter 3 Circuit Laws and Networking Theorems

known as the current conservation law, which is the electrical counterpart of the mass conservation
law in fluid mechanics. For N currents entering a node, KCL can be cast in the form

X
N
Ii ¼ 0 ð3:3Þ
i¼1

where N denotes the total number of nodal currents. The current directions may be assigned
arbitrarily; the same results will eventually be obtained. The nodal current is taken with a plus
sign if it is entering the node, i.e., the current arrow is directed toward the node. It carries a minus
sign if it is leaving the node, i.e., the current arrow points in the opposite direction. The current
value itself (positive or negative) is substituted afterward. Interestingly, if this law did not hold, an
uncompensated charge could accumulate in a node over time. This uncompensated charge and its
associated Coulomb force would eventually destroy the operation of the underlying electric
circuit. To illustrate the use of KCL, we consider Fig. 3.2 with four different node types for a
collection of branches A, B, C, D which could represent arbitrary circuit elements.

Fig. 3.2. Different types of the nodes in an electric network subject to KCL.

While the node in Fig. 3.2a is simple, more complicated node configurations may be observed
in a circuit (see the following figures). You should note that in a node we may move individual
joints to one common joint without affecting the circuit’s operation. A node transformation to a
single joint is a convenient tool used when working with more complicated nodes.

III-94
Chapter 3 Section 3.1: Circuit Laws: Networking Theorems

Exercise 3.2: Write KLC for the nodes shown in Fig. 3.2.
Answer: In the case of Fig. 3.2a, IA + IC ¼ IB for any values of IA, IB, IC. In the case of Fig. 3.2b,
IA + IC ¼ IB + ID for any values of IA, IB, IC, ID. In the case of Fig. 3.2c, IA + IB ¼ IC + ID for any
values of IA, IB, IC, ID. In the case of Fig. 3.2d, 2A + IC ¼ 5A + IE for any values of IC, IE.

Example 3.1: In some cases, the use of KCL may be sufficient to determine all currents in an
electric network. Solve for the unknown currents IA, IC, ID in a network shown in Fig. 3.3.
Solution: First, we note that the wire connection on the left states that the current along the wire
is preserved. This implies IA ¼  1 A. KCL for node 1 gives
1A þ I C ¼ 5 A ) I C ¼ 4 A ð3:4aÞ

Next, KCL applied to node 2 yields

5A ¼ I D þ 2A ) I D ¼ 3 A ð3:4bÞ

Thus, the circuit is solved. Node 3 has not been used; it can be employed to check the
correctness of the solution: IA + ID + 2A ¼ IC or 4 A ¼ 4 A.

Fig. 3.3. Solving in a network using KCL.

3.1.3 Kirchhoff’s Voltage Law (KVL)

This law specifically applies to the loops in an electric network. The Kirchhoff’s voltage law states
that the sum of voltages for any closed loop is zero. In other words, the total amount of work
needed to move a unit electric charge one loop turn is zero.2 If this law were not applicable, a

2
The physical counterpart of KVL is Faraday’s law of induction in the static case. When there is a variable magnetic field
penetrating a wire loop, KVL is no longer valid.

III-95
Chapter 3 Circuit Laws and Networking Theorems

charge moving in a closed loop would constantly accelerate and eventually escape the circuit or
constantly decelerate and eventually stop moving. In order to formulate KVL for a closed loop, we
need to identify the loop direction. It is usually chosen to be clockwise (see the dotted arrows in
Figs. 3.4 and 3.5). KVL in its general form states

X
N
Vi ¼ 0 ð3:5Þ
i¼1

where N is the total number of circuit elements in a loop. The voltage Vi is taken with a plus sign if
the loop arrow is entering the positive voltage polarity and with a minus sign otherwise. The
polarities of voltages Vi can be assigned arbitrarily; the same result is obtained. To demonstrate
this fact, in Fig. 3.4 we first consider a simpler network with different voltage polarities. One
network branch is purposely designated as a voltage source, while the rest are arbitrary circuit
elements.

Fig. 3.4. KVL applied to a closed loop with one voltage source and three passive circuit elements. The
dotted arrow indicates current flow in clockwise direction.

Example 3.2: Write KVL for the circuit shown in Fig. 3.4 that includes an ideal voltage source
and three other circuit elements.
Solution: In the case of Fig. 3.4a, we start with the source and strictly apply our convention of
positive and negative polarity based on the prescribed loop arrow direction:
5V þ V A þ V B  V C ¼ 0 ð3:6aÞ

In the case of Fig. 3.4b, we have to change signs of VA, VB, VC and thus obtain
5V  V A  V B þ V C ¼ 0 ð3:6bÞ

Note that these two cases only differ by voltage polarities.

III-96
Chapter 3 Section 3.1: Circuit Laws: Networking Theorems

Example 3.2 (cont.):

Since reversing voltage polarities in Fig. 3.4b was taken into account by changing the sign of
the voltage, both figures yield the identical result after substituting values:

5 V þ 3 V þ 3 V  1 V ¼ 0 ð3:6cÞ
This observation highlights the fact that the voltage polarities for circuit elements may initially
be assigned arbitrarily: applying KVL will ultimately lead to the correct signs of the voltage
values.

Figure 3.5 shows another electric network with all branches now consistently labeled; the
voltages in red denote the actual values with respect to the initially assigned directions.

Fig. 3.5. KVL applied to a network with three loops, two meshes, and five circuit elements.

Exercise 3.3: Determine the number of branches, nodes, loops, and meshes for the network
shown in Fig. 3.5.
Answer: There are five branches, four single nodes, three loops, and two meshes.

Example 3.3: In some cases, the use of KVL may be sufficient to determine all voltages in an
electric network. Solve for the unknown voltages in the network shown in Fig. 3.5.
Solution: The application of KVL for loop 1 (mesh 1) has the form

5V  7V þ V C ¼ 0 ) V C ¼ 12V ð3:6dÞ

Moreover, KVL for loop 2 (mesh 2) results in (VC is already known)

12V þ 6V þ V E ¼ 0 ) V E ¼ 6V ð3:6eÞ

III-97
Chapter 3 Circuit Laws and Networking Theorems

Example 3.3 (cont.):

Finally, applying KVL for loop 3 is the solution check:
5 V  7 V þ 6 V þ 6 V ¼ 0 ð3:6f Þ
If at least one of the voltages VA, VB, VD in Fig. 3.5 is unknown, then a unique solution does not
exist. Through the use of KCL more information must be acquired.

3.1.4 Power-Related Networking Theorems

This section would be incomplete without an introductory discussion of two power-related
networking theorems. The first one is Maxwell’s minimum heat theorem formulated by James
Clerk Maxwell in 1891. It states that, for a linear electric network of resistive circuit elements and
voltage/current sources, the currents distribute themselves in such a way that the total dissipated
power (generated heat) in the resistances is a minimum. If there are NR resistive circuit elements,
we can state
X
NR X
NR X
NR
V2
V iI i ¼ Ri I 2i ¼ i
¼ min ð3:7aÞ
i¼1 i¼1 i¼1
Ri

where voltages Vi and currents Ii must all satisfy the passive reference configuration. Indeed,
Eq. (3.7a) is equivalent to the condition where the net power generated by all sources is also
minimized. Maxwell’s minimum heat theorem can be proved using KCL, KVL, and Ohm’s law.
The second theorem is Tellegen’s theorem formulated by Bernard D.H. Tellegen in 1952. It
postulates that, for an arbitrary electric network with a total of N circuit elements of arbitrary
(linear or nonlinear, passive or active) nature, the equality

X
N
V iI i ¼ 0 ð3:7bÞ
i¼1

must hold if all voltages Vi and currents Ii satisfy one (let’s say the passive) reference configuration.
Tellegen’s theorem serves as a power (or energy) conservation law for all electric networks. It can be
proved based on KCL and KVL only. Tellegen’s theorem has other important implications and
generalizations. To illustrate both theorems, we consider two simple examples.

Example 3.4: Use Maxwell’s minimum heat theorem and determine the unknown current x
through resistance R1 in Fig. 3.6a.
Solution: Equation (3.7a) yields

R1 x2 þ R2 ðx  I S Þ2 ¼ min ð3:8aÞ

This function is minimized when its derivative with respect to x is zero. Therefore,

III-98
Chapter 3 Section 3.1: Circuit Laws: Networking Theorems

Example 3.4 (cont.):

R2
x¼ IS ð3:8bÞ
R1 þ R2
which is the current division principle studied in the next section. The voltage division
principle may be established similarly.

Fig. 3.6. Examples for Maxwell’s minimum heat theorem and Tellegen’s theorem.

Example 3.5: Prove Tellegen’s theorem for the network shown in Fig. 3.6b.

Solution: Both elements follow the passive reference configuration. By KCL we conclude
IB ¼  IA. By KVL, VB ¼ VA. Therefore, VAIA + VBIB ¼ 0, which is the simple poof. In practice,
element A may be a voltage source, and element B may be a resistance. We can also use the active
reference configuration for the source but need to define its power to be negative in such a case.

3.1.5 Port of a Network: Network Equivalence

So far we have studied only closed electric networks. A network can have a port, through which it
is interconnected to another network as shown in Fig. 3.7. The network may be active or passive.
This specific network has two (input or output) terminals a and b, which form a single port. Thus,
the network in Fig. 3.7 is known as a one-port network. All series/parallel combinations of sources
and resistances are typically one-port networks.

Fig. 3.7. Generic system for establishing a network equivalence.

Two arbitrary one-port networks in the form of Fig. 3.7 are said to be equivalent networks
(or equivalent electric circuits) when their υ-i characteristics at terminals a and b coincide. This
means that for any given voltage, there is the same current entering the network or leaving it and
vice versa. Therefore, these networks are indistinguishable.

III-99
Chapter 3 Circuit Laws and Networking Theorems

Section 3.2 Series and Parallel Network/Circuit Blocks

3.2.1 Sources in Series and in Parallel
Series-Connected Battery Bank
The simultaneous use of KCL and KVL allows us to analyze the behavior of combinations of
active circuit elements and establish their equivalence. The physical counterparts are various
battery banks, which are interconnections of the identical batteries. Figure 3.8 shows a series
battery bank, also called a battery pack, with two or more batteries connected in series. The battery
symbol implies an ideal voltage source.

Fig. 3.8. Series combinations of battery cells and their equivalent representations.

We intend to find the resulting voltage and current of this combination. To determine the
equivalent voltage, we close the circuit loop shown in Fig. 3.8a by introducing a virtual

III-100
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

circuit element with terminals a and b, with an unknown voltage V between these terminals. This
element simulates the rest of the circuit, which closes the current path. Applying KVL for the loop
shown in Fig. 3.8a results in
V þ 9 V þ 9 V ¼ 0 ) V ¼ 18 V ð3:9Þ
Moreover, using KCL, we obtain the same current flows throughout the left-handed circuit of
Fig. 3.8a; this current is exactly equal to IB.3 Therefore, the series combination of two batteries in
Fig. 3.8a is equivalent to one battery bank that provides double the voltage, or 18 V, compared to
the unit cell. However, it delivers a current of a single cell. The same method can now be applied
to multiple battery cells connected in series; one such battery bank is shown in Fig. 3.8b.

Dual-Polarity Voltage Power Supply

Two batteries, or other voltage sources, connected in series can be used as a dual-polarity power
supply as shown in Fig. 3.9. The middle terminal provides a virtual ground for the circuit or the
common ground. Both negative and positive voltages with respect to the common port can now be
created in the circuit. Such a source is of particular importance for operational amplifier circuits
and for transistor circuits. Every multichannel laboratory power supply may operate as a dual
power supply. The common terminal may (but does not have to) additionally be connected to the
earth ground.

Fig. 3.9. A dual-polarity power supply constructed with two battery cells.

Parallel-Connected Battery Bank

As an alternative to the series connected battery bank, we can investigate the parallel battery bank
shown in Fig. 3.10. To determine the equivalent voltage of the combination, we again close the
circuit loop by introducing a virtual circuit element with terminals a and b and with an unknown
voltage V between these terminals. The use of KVL gives
V ¼9V ð3:10Þ
Thus, the voltage of the battery bank in Fig. 3.10 is still equal to the unit cell voltage. However,
applying KCL to both rightmost nodes in Fig. 3.10 indicates that the current doubles. Therefore, the

3
If realistic battery cells are capable of delivering different currents when short circuited, then the lowest cell current will
flow under short circuit condition.
III-101
Chapter 3 Circuit Laws and Networking Theorems

parallel combination of two batteries is equivalent to one battery bank that provides the same voltage
of 9 V as one unit cell but at twice the current strength.

Fig. 3.10. A parallel combination of battery cells and its equivalent single battery representation.

Series Versus Parallel Connection

What is the difference between series and parallel combinations of two 9 V batteries? First, let us
find the power delivered to the circuit. We assume IB ¼ 1 A in both cases, even though the specific
value of the current is not important. For the series combination in Fig. 3.8a, the delivered power
is 18 V  1 A ¼ 18 W. For the parallel combination in Fig. 3.10, the delivered power is again
9 V  2 A ¼ 18 W. Thus, as far as the power rating is concerned, there is no difference.
You should, however, remember that we always deliver power to a load. For the series combina-
tion, the implied load resistance becomes 18 V/1 A ¼ 18 Ω. For the parallel combination, the
anticipated load resistance will be 9 V/2 A ¼ 4.5 Ω. Thus, it is the load resistance that determines
which combination should be used. This question is of great practical importance.

Combinations of Current Sources

Combinations of current sources are studied similarly. They are important for photovoltaic and
thermoelectric semiconductor devices.

3.2.2 Resistances in Series and in Parallel

Series Connection
After the sources have been analyzed, we turn our attention to series and parallel combinations of
resistances (or conductances). Their physical counterparts are various circuit loads, for example,
individual households connected to the same power grid or individual motors driven by the same
source. Figure 3.11 depicts the series combination of two resistances R1 and R2.

Fig. 3.11. Two resistances in series and their equivalent single resistance representation.

III-102
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

Note that the current direction for both resistances corresponds to a passive reference config-
uration: current flows “down the voltage hill.” Again, we close the circuit loop by introducing a
virtual circuit element with terminals a and b and with an unknown voltage V between those
terminals. This virtual circuit element, which simulates the rest of the circuit, allows us to close the
current path. KVL for the loop shown in Fig. 3.11 results in

V ¼ V 1 þ V 2 ¼ IR1 þ IR2 ¼ I ðR1 þ R2 Þ ¼ IReq ) Req ¼ R1 þ R2 ð3:11aÞ

Thus, two resistances can be replaced by one equivalent resistance, which is the sum of the
individual resistances. It follows from Eq. (3.11a) that the equivalent resistance gives us the same
circuit current (and the same power into the load) as the original resistance combination does, for any
applied voltage. This is the formal description of the equivalent resistance to be generalized later.
Equation (3.11a) can easily be extended to any arbitrary number of resistances connected in series.
Equation (3.11a) may be also formulated in terms of conductances, the reciprocals of resistances,
1 1 1
¼ þ ð3:11bÞ
Geq G1 G2

Parallel Connection
Next, we consider the parallel combination of resistances shown in Fig. 3.12. KVL applied to the loop
shown in the figure and for another loop between two resistances indicates that the voltages across
every resistance are equal to V. KCL applied to either rightmost node results in

Fig. 3.12. Two resistances connected in parallel and the equivalent resistance.

V V V 1 1 1
I ¼ I1 þ I2 ¼ þ ¼ ) ¼ þ ) Geq ¼ G1 þ G2 ð3:12Þ
R1 R2 Req Req R1 R2
Therefore, the parallel combination of two resistances is equivalent to one resistance, which has a
value equal to the reciprocal of the sum of the reciprocal values of both. The equivalent resistance
again gives us the same circuit current as the original resistance combination does, for any applied
voltage. We emphasize that the equivalent resistance is always smaller in value than each of the
resistances to be combined in parallel. Note that the conductances simply add up for the parallel
combinations. Equation (3.12) can again easily be extended to any arbitrary number of resistances
connected in parallel.

Meaning of Equivalent Circuit Element

In summary, the study of series/parallel active and passive circuit elements leads us to the
following simple definition of an equivalent circuit element, either passive or active. The

III-103
Chapter 3 Circuit Laws and Networking Theorems

equivalent circuit element possesses the same υ-i characteristic as the υ-i characteristic of the
original circuit, for which voltage and current are acquired at its terminals a and b.

3.2.3 Reduction of Resistive Networks

The reduction of a network of many resistances to a single equivalent resistance is a topic of
practical importance. There are many examples of distributed resistive networks. A contemporary
example is a rear window defroster, which is a distributed resistive heater. The corresponding
solution is usually based on:
(i) Step-by-step use of series/parallel equivalents
(ii) Moving, splitting, or reducing modes
(iii) Reliance on fluid mechanics analogies, which may be helpful for resistive networks
Although a unique solution always exists, its practical realization may be quite difficult. The
following examples outline the procedure for the reduction of resistive networks.

Example 3.6: Find the equivalent resistance between terminals a and b for the resistive
network shown in Fig. 3.13a.
Solution: We should not start with terminals a and b but with the opposite side of the circuit.
First, the three resistances furthest to the right are combined in series in Fig. 3.13a. The next step
is the parallel combination of the resulting resistance and the 1.5 kΩ resistance in Fig. 3.13b. The
final step is another series combination; this results in the final equivalent resistance value of
1875 Ω. We need to point out again that it is impossible to:
(i) Combine in series two resistances separated by a node.
(ii) Move the resistance through a node or move the node through a resistance.
These restrictions hold for other circuit elements as well, beside resistors.

Fig. 3.13. Step-by-step circuit reduction via series and parallel resistance combinations.
III-104
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

Example 3.7: Find the equivalent resistance between terminals a and b for the resistive
network shown in Fig. 3.14a.

Solution: First, we can split and move a node along the wire, leading to a circuit shown in
Fig. 3.14b. Next, we combine two pairs of resistances in parallel. The next step, in Fig. 3.14c, is
the series combination of three resistances. The last step is the solution of a parallel circuit,
leading to the equivalent resistance of 66.67 Ω.

Fig. 3.14. Step-by-step circuit reduction to a single equivalent resistance.

Exercise 3.4: Using a fluid mechanics analogy of identical water flow in two symmetric
channels, find the equivalent resistance of the network shown in Fig. 3.15.
Answer: 2.5 kΩ

Fig. 3.15. The resistive network discussed in Exercise 3.4.

3.2.4 Voltage Divider Circuit

The purpose of the voltage divider circuit is to provide a voltage different from the supply voltage.
The voltage divider circuit is associated with resistances in series. It is perhaps the most important

III-105
Chapter 3 Circuit Laws and Networking Theorems

basic electric circuit or block within another circuit. The voltage division principle is used in
sensor circuits, actuator circuits, and bias circuits. Furthermore, any input and output port of a
(transistor) amplifier is essentially a voltage divider. We consider a particular voltage divider form,
shown in Fig. 3.16, connected to a DC voltage supply. This circuit, as with any other electric
circuit, can be analyzed by using KCL and KVL simultaneously. We prefer using this method
although the combination of two resistances in series will provide an equivalent solution.

Fig. 3.16. A conventional voltage divider consists of two resistances and an ideal voltage source.

KCL states that the current I through both resistance and the voltage supply is the same.
Applying KVL to the circuit loop allows us to find the circuit current:
VS
V S þ V 1 þ V 2 ¼ 0 ) V S ¼ V 1 þ V 2 ¼ I ðR1 þ R2 Þ ) I ¼ ð3:13Þ
R1 þ R2
Once the circuit current I is known, then Ohm’s law can be used. This yields the voltage divider rule
R1 R2
V1 ¼ V S, V2 ¼ VS ð3:14Þ
R1 þ R2 R1 þ R2
Equation (3.14) states that the major function of the voltage divider is to divide the voltage of the
power source between two resistances in direct proportion so that:
(i) The larger resistance always acquires a higher voltage, and the smaller resistance acquires a
smaller voltage.
(ii) The individual voltages always add up to the supply voltage, i.e., V1 + V2 ¼ VS.

Exercise 3.5: A voltage divider circuit uses a 10 V DC source and two resistances: R1 ¼ 5 Ω
and R2 ¼ 100 Ω. What are the voltages V1, V2 across the resistances?
Answer: V 1 ¼ 0:48 V, V2 ¼ 9:52 V, V 1 þ V 2 ¼ 10 V:

The voltage divider with multiple resistances R1, R2, . . ., RN is solved in the form

VS Ri
V S ¼ V 1 þ V 2 þ .. . þ V N ) I ¼ and V i ¼ VS ð3:15Þ
R1 þ R2 þ . .. þ RN R1 þ R2 þ .. . þ RN

III-106
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

3.2.5 Application Example: Voltage Divider as a Sensor Circuit

Consider a resistive sensing element (thermistor, strain gauge, photoresistor, etc.) as denoted by
R2(x) in Fig. 3.17. The element changes its resistance R2(x) when an external parameter x changes.
Parameter x could be temperature, pressure, humidity, solar radiation, or any other physical
parameter that undergoes an environmental change. A simple sensor configuration is a direct
connection to a voltage source and to the DMM for voltage measurements (see Fig. 3.17a).
However, no matter how the sensor resistance changes, the sensor will always output the source
voltage. A solution to the this problem is a voltage divider circuit shown in Fig. 3.17b. The extra
resistance R1 is fixed. According to Eq. (3.14), voltage V2 in Fig. 3.17b varies depending on the
influence of R2(x):
R2 ðxÞ
V 2 ¼ V 2 ðxÞ ¼ VS ð3:16Þ
R1 þ R2 ðxÞ

Fig. 3.17. (a) Incorrect sensor circuit; (b) a sensor circuit on the basis of a resistive voltage divider
where R2(x) changes its resistance depending on the process parameter x.

The variable voltage V2(x) is measured by the voltmeter. The dependence of V2 on R2 is clearly
nonlinear in Eq. (3.16). Although Eq. (3.16) can be linearized by choosing a sufficiently large R1
to make the denominator nearly constant, we will show later that such an operation greatly
decreases device sensitivity. Let us assume that the external parameter x in Eq. (3.16) changes
from a lower limit x1 to an upper limit x2, i.e., x1  x  x2. As a result, the sensing resistance
changes monotonically, but not necessarily linearly, from R0 ¼ R2(x1) to R00 ¼ R2(x2). We also
assume that if x1  x  x2, then R0 > R00 . The sensor circuit’s sensitivity, S, is given by
 
V 2 ðx1 Þ  V 2 ðx2 Þ V
S¼ ð3:17Þ
x2  x1 units of x

The sensitivity is expressed in terms of voltage variation per one unit of x. A higher sensitivity
implies a larger voltage variation and thus provides a better sensor resolution and improved
robustness against noise.

Design of the Sensor Circuit for Maximum Sensitivity

Let us pose the following question: what value should the fixed resistance R1 assume in order to
achieve the highest sensitivity of the voltage divider sensor? It is clear that R1 cannot be very small

III-107
Chapter 3 Circuit Laws and Networking Theorems

(otherwise the voltage reading will always be VS and the sensitivity will be zero) and that R1
cannot be very large (otherwise the voltage reading will be always 0 V and the sensitivity will be
zero). The sensitivity is thus a positive function that is zero at R1 ¼ 0 and at R1 ¼ 1. According to
the extreme value theorem, a global maximum should exist between these two values. We denote
the unknown resistance R1 with variable t, substitute V2 from Eq. (3.14), and rewrite Eq. (3.17) in
the form
 0 
VS R R00
S¼  ð3:18aÞ
x2  x1 t þ R0 t þ R00

It is convenient to transform this result into a simpler expression S ¼ VSS0f(t), where a constant S0
is called the intrinsic sensitivity of the resistive sensing element and f(t) is the sole function of the
first resistance, i.e.,
 0 
R  R00 t
S0 ¼ , f ðt Þ ¼ 0 ð3:18bÞ
x2  x1 ðR þ tÞðR00 þ tÞ

This function f(t) is to be maximized. At the function’s maximum, the derivative of f(t) versus
t should be zero. Using the quotient rule for the differentiation of a fraction, it follows from
Eq. (3.18b) that
R0 R00  t 2
f 0 ðt Þ ¼  0 2 ð3:18cÞ
R þ t ðR00 þ tÞ
2

The final result following from the condition f 0 (t) ¼ 0 is surprisingly simple
pffiffiffiffiffiffiffiffiffi
t ¼ R1 ¼ R0 R00 ð3:18dÞ

In other words, the fixed resistance of the voltage divider circuit should be equal to the geometric
mean of two extreme resistances of the sensing element itself.


Example 3.8: For the NTC-503 thermistor sensing element, x1 ¼ 25 C (room temperature),

x2 ¼ 37 C C (approximate temperature of a human body), R0 ¼ 50 kΩ, and R00 ¼ 30 kΩ.
What is the sensitivity of the voltage divider sensor if VS ¼ 9 V and (A) R1 ¼ 5 kΩ,
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(B) R1 ¼ 500 kΩ, and (C) R1 ¼ 30  50  39 kΩ?
Solution: We substitute the numbers in Eq. (3.18a) and find the sensitivity. The corresponding
sensitivity plot as a function of R1 is given in Fig. 3.18a. The particular sensitivity values are
(A) S ¼ 39 mV/ C; (B) S ¼ 26 mV/ C; and (C) S ¼ 95 mV/ C.

III-108
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

Example 3.9: For the SGT-1/350-TY11 strain gauge, the nominal resistance is 350 Ω. The
resistance variation of 0.1% for the tensile strain ε is observed, and the intrinsic device
sensitivity is S0 ¼ 700 Ω/ε. What is the sensitivity of the voltage divider sensor if VS ¼ 4.5 V
and (A) R1 ¼ 50 Ω, (B) R1 ¼ 5 kΩ, and (C) R1 ¼ 350 Ω? For positive sensitivity numbers,
interchange x1,2 in Eq. (3.18a).
Solution: In this example, R0 ¼ 350.35 Ω and R00 ¼ 349.65 Ω. We use Eq. (3.18b) and plot
the sensitivity as a function of R1; this is seen in Fig. 3.18b. The particular values are (με
are micro-strain units) (A) S ¼ 0.24 mV/1000με; (B) S ¼ 0.55 mV/1000με; and
(C) S ¼ 2.25 mV/1000με.

Fig. 3.18. Sensitivity curves for the divider sensor circuits with a thermistor and a strain gage.

3.2.6 Application Example: Voltage Divider as an Actuator Circuit

The circuit shown in Fig. 3.19 contains a voltage divider circuit block with one resistive sensing
element, i.e., resistance R1. The block is connected to a three-terminal electronic switch
(a transistor). A variable voltage V controlling the switch operation is created by the voltage
divider. When the control voltage V reaches a certain threshold voltage VTh or exceeds it, the
switch closes. A DC motor is now connected to the source, and the rotor starts to spin. Switches of
this type involve field-effect transistors. We emphasize that there is no current into the control
switch terminal, only the control voltage counts. If R1 is a thermistor and the DC motor is a fan
motor, the entire circuit may operate as a basic temperature controller in an enclosure or in a room.
The control voltage versus circuit ground, the reference point, is given by (cf. Eq. (3.16))
R2
V ðxÞ V 2 ðxÞ ¼ VS ð3:19Þ
R1 ðxÞ þ R2

where variable x corresponds to the ambient temperature.

III-109
Chapter 3 Circuit Laws and Networking Theorems

Fig. 3.19. An actuator circuit based on the voltage divider principle.

Example 3.10: The circuit in Fig. 3.19 uses an NTC-503 thermistor sensing element with
R1 ¼ 50 kΩ at 25  C (room temperature) and R1 ¼ 30 kΩ at 37  C. The fixed resistance
(assumed temperature independent) is R2 ¼ 12 kΩ. The threshold voltage of the switch
is VTh ¼ 2.3 V. The supply voltage is 9 V. Determine the circuit behavior at 25 C and at
37 C, respectively.
Solution: According to Eq. (3.19), the control voltage at 25 C is equal to 1.74 V. This value is
below the threshold voltage. Consequently, the switch is open and the motor is not connected to
the source. However, the control voltage at 37 C is 2.57 V. This value is above the threshold
voltage and the switch is now closed: the motor is connected to the source and is spinning.
Figure 3.20 shows the corresponding laboratory setup. Note that in reality the threshold voltage
of the transistor switch is not exactly a constant; it depends on the temperature.

Fig. 3.20. Laboratory realization of the circuit shown in Fig. 3.19.

III-110
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

3.2.7 Current Limiter

The current limiter (or simply the current-limiting resistor) is a particular case of the voltage
divider circuit and is of a significant practical importance. Its concept is shown in Fig. 3.21. A load
resistance RL is connected to the ideal voltage source in series with another (smaller) resistance R,
which physically acts as a current-limiting resistor.

Fig. 3.21. Voltage divider circuit in the current-limiting configuration. The 10 Ω resistance is used to
limit the circuit current.

If the load resistance is fixed at a rather high value, the circuit in Fig. 3.21 does not pose any
problem, and the current-limiting resistance of 10 Ω becomes insignificant. For example, in
Fig. 3.21, the circuit current is 100 mA in the absence of the first resistance R. The power
delivered to the load resistance is P ¼ RLI2 ¼ 1 W. However, the load resistance may be variable,
and it may attain really small values. When this happens, the circuit current increases. In Fig. 3.21,
it becomes equal to 10 A when the load resistance decreases to 1 Ω and the current-limiting resistor
is missing. The power delivered to the load resistance also increases to P ¼ 100 W. This large
power may overheat and eventually destroy the small-scale load (a thermistor is one example).
The role of resistance R is to limit the total current when the load resistance is either variable or
constant but small. For example, in the circuit of Fig. 3.21, the maximum possible circuit current is
VS VS
I¼ < ¼ 1A ð3:20Þ
R þ RL R
irrespective of the value of the load resistance. Therefore, the power delivered to a 1 Ω-load
becomes always less than 1 W instead of the initial value of 100 W.

3.2.8 Current Divider Circuit

The current divider circuit shown in Fig. 3.22 is associated with resistances connected in parallel.
It is dual to the voltage divider circuit in the sense that the roles of voltage and current are
interchanged. Figure 3.22 shows the concept.

Fig. 3.22. A current divider circuit with an ideal current source.

III-111
Chapter 3 Circuit Laws and Networking Theorems

To solve the circuit, we again prefer to use KCL and KVL simultaneously. KCL gives
IS ¼ I1 þ I2 ð3:21Þ

Based on KVL for both loop 1 and loop 2, the voltage across the current source (voltage between
its terminals a and b) is equal to the voltage across either resistor and is equal to V. Application of
Ohm’s law gives the expression for this voltage,
V V IS R1 R2
IS ¼ þ )V ¼ ¼ IS ð3:22Þ
R1 R2 1 1 R1 þ R2
þ
R1 R2

Therefore, we obtain the current division rule in the form

V R2 V R1
I1 ¼ ¼ IS, I2 ¼ ¼ IS ð3:23Þ
R1 R1 þ R2 R2 R1 þ R2

Equation (3.23) teaches us that the major function of the current divider is to divide the current of
the power source between two resistances in an inverse proportion so that:
(i) The larger resistance always acquires the smaller current, and the smaller resistance
acquires the larger current.
(ii) The individual currents add up to the source current, i.e., I1 + I2 ¼ IS.
In other words, the electric current always chooses a path of least resistance. If one resistance is
replaced by a wire the entire source current will flow through the wire; the second resistance will
be shorted out by the wire.

Exercise 3.6: A current divider uses a 3 mA current source and two resistances: R1 ¼ 200 Ω
and R2 ¼ 600 Ω. What are the currents I1, I2 through the resistances?

Answer: I1 ¼ 2.25 mA, I2 ¼ 0.75 mA. The smaller resistance acquires the larger current.

Example 3.11: The current divider circuit can be assembled with a voltage source as shown in
Fig. 3.23. Find currents I1, I2 and the total circuit current I.
Solution: The circuit in Fig. 3.23 can be analyzed in a number of ways. Perhaps the simplest
way is to recognize that, according to KVL, the voltages across all three elements in Fig. 3.23
are equal to each other and equal to 10 V. Therefore,
10V 10V 10V R1 R2
I1 ¼ , I2 ¼ , I ¼ I1 þ I2 ¼ , Req ¼ ð3:24Þ
R1 R2 Req R1 þ R2
Another way to solve the same circuit is to combine resistances in parallel, find the circuit
current I, and then apply the current division principle.

III-112
Chapter 3 Section 3.2: Series and Parallel Network/Circuit Blocks

Fig. 3.23. A current divider circuit with a voltage source.

3.2.9 Wheatstone Bridge

The Wheatstone bridge was invented by the British scientist and mathematician, Samuel Christie
(1784–1865), and first used for resistance measurements by Sir Charles Wheatstone in 1843. It is
shown in Fig. 3.24. From Fig. 3.24, we can recognize that the Wheatstone bridge is, in fact, a
combination of two independent voltage divider blocks: one with two fixed resistances R1, R2
and another with one fixed resistance R3 and some other resistance R4 denoted here by R4 ¼ R(x).
When the resistance measurements are implied, R4 is the fixed unknown resistance. However,
common modern applications use the Wheatstone bridge as a part of the sensor circuit. In this
case, R(x) is a variable resistance (a sensing element), where x is a physical quantity to be
measured. The second voltage divider is the voltage divider sensor circuit; the first voltage divider
is fixed. Circuit ground (absolute voltage reference) may be introduced as shown in Fig. 3.24.

Fig. 3.24. The Wheatstone bridge is a combination of two independent voltage dividers connected to
the same voltage source. The second voltage divider is a sensor circuit.

It has been shown that the voltage divider circuit is a basic sensor circuit. Now, why do we need
two voltage dividers? The answer to this question will be based on the fact that, with the help of
the fixed divider, we can eliminate a DC voltage offset in the sensor voltage reading of the
“master” voltage divider and thus enable the use of a difference signal (and a difference amplifier)

III-113
Chapter 3 Circuit Laws and Networking Theorems

to amplify the likely very weak sensor voltages. The key point is that the output voltage of the
Wheatstone bridge is not Va or Vb but the differential voltage between terminals a and b in
Fig. 3.24, Vab ¼ Va  Vb.

Example 3.12:
A. A simple voltage divider circuit with R3 ¼ 350 Ω and R(x) ¼ 350 Ω  0.1% (the strain
gauge) in Fig. 3.24b is used for mechanical strain measurements with a 4.5 V voltage
power supply. What is the output voltage Vb of the sensor circuit?
B. The same voltage divider circuit is augmented with another fixed voltage divider having
R1 ¼ R2 ¼ 350 Ω to form the Wheatstone bridge shown in Fig. 3.24c. What is the output
voltage Vab of the sensor circuit now?

Solution: In case A, we use the voltage division rule and obtain

V b ¼ 2:25 V  1:125 mV ð3:25aÞ

It is difficult to process such voltages since we cannot really amplify them. Amplification of
2.251125 V by a factor of 100 gives 225.1125 V; such a large voltage simply cannot be
obtained with the common amplifier circuits. In case B, however, the value of Vb should be
subtracted from Va ¼ 2.25 V, which yields the sensor voltage in the form
Vab ¼
1:125 mV ð3:25bÞ

If we now amplify 1.125 mV by a factor of 100, a conventional value of 0.1125 V would be

obtained. Along with this fact, the differential sensor voltage has another significant advantage,
which is its immunity against circuit noise.

General Model of Wheatstone Bridge

Using the voltage division rule twice, the differential voltage Vab of the Wheatstone bridge in
Fig. 3.24c becomes (R4 ¼ R(x))
 
R2 R4
Vab ¼  VS ð3:26Þ
R1 þ R2 R3 þ R4

The Wheatstone bridge is balanced when Vab ¼ 0. From Eq. (3.26) one obtains the necessary and
sufficient condition for the balanced Wheatstone bridge,
R1 R3
¼ ð3:27Þ
R2 R4

III-114
Chapter 3 Section 3.3: Superposition Theorem and Its Use

Section 3.3 Superposition Theorem and Its Use

3.3.1 Linear and Nonlinear Circuits
Linear Circuit
The superposition theorem studied in this section is only valid for linear circuits. A linear circuit is
a circuit that includes only the linear circuit elements (elements with a linear or straight υ-i
characteristic):
1. Resistance
2. Capacitance (υ-i relationship is time-dependent but still linear)
3. Inductance (υ-i relationship is time-dependent but still linear)
4. Voltage source (independent and linearly dependent)
5. Current source (independent and linearly dependent)
Every linear circuit satisfies both the homogeneity and additivity properties. To explain those
properties, we consider a linear circuit with an input parameter x (input voltage or current) and an
output parameter f(x) (output voltage or current). A function f is the characteristic of the circuit
itself; it must be a linear function. Namely, when an input parameter is a linear superposition
ax1 + bx2 of two individual stimuli, x1, x2, the output is also a linear superposition of two
individual responses, i.e.,
f ðax1 þ bx2 Þ ¼ af ðx1 Þ þ bf ðx2 Þ ð3:28Þ

For example, if we double all source strengths in a linear circuit, voltages across every passive
circuit element and currents through every circuit element will also double.

Nonlinear Circuit and Circuit Linearization

A nonlinear circuit will include nonlinear circuit elements, elements with a nonlinear υ-i charac-
teristic. Any circuit with semiconductor components (such as diodes, transistors, solar cells) is a
nonlinear circuit. Since the vast majority of electronic circuits include semiconductor components,
a legitimate question to ask is what value do the linear circuits have in this case? One answer is
given by a linearization procedure, which makes it possible to reduce the nonlinear circuit to a
linear one, in a certain domain of operating parameters. Mathematically, circuit linearization
means that a nonlinear relationship V(I ) is expanded into a Taylor series,

dV
V ðI Þ ¼ V 0 þ ðI  I 0 Þ þ . . . ð3:29Þ
dI V ¼V 0 , I¼I 0

III-115
Chapter 3 Circuit Laws and Networking Theorems

about a certain operating point V0, I0, where only the constant and the linear terms are retained.
The derivative in Eq. (3.29), the so-called dynamic or small-signal resistance r can now be used in
place of the familiar resistance R for the linear ohmic circuit elements.

3.3.2 Superposition Theorem or Superposition Principle

The superposition theorem, often called the superposition principle, applies to circuits with more
than one voltage and/or source. It states that the complete circuit solution is obtained as a linear
superposition of particular solutions, for every power source separately. In other words, we are
zeroing (or turning off) all the power sources except for one, find the solution, and then add up all
such solutions. The following rules apply:
1. To turn off a voltage source, we replace it by a short circuit or an ideal wire (see Fig. 3.25a).
The voltage across the ideal wire is exactly 0 V.
2. To turn off a current source, we replace it by an open circuit or an air gap (see Fig. 3.25b).
The current through the gap is exactly 0 A.
3. The dependent sources do not need to be zeroed. They remain the same for every particular
solution and may affect every particular solution.

Fig. 3.25. Superposition theorem for two independent sources (either voltage or current).

III-116
Chapter 3 Section 3.3: Superposition Theorem and Its Use

Fig. 3.26. Application of the superposition principle to a circuit with voltage sources.

Example 3.13: Find current I1 in the circuit shown in Fig. 3.26.

Solution: The first step is shown in the figure; we apply the superposition theorem and obtain
two simpler circuits. Each of those circuits is solved using series equivalents. For the circuit
with the 12 V power source,

i1 ¼ 12 V=5 kΩ ¼ 2:4 mA ð3:30Þ

The minus sign is due to the fact that the actual circuit current flows in the opposite direction.
For the circuit with the 18 V source,

i2 ¼ 18 V=5 kΩ ¼ 3:6 mA ð3:31Þ

The total circuit current I1 is therefore given by I1 ¼ i1 + i2 ¼ 1.2 mA. Circuit voltage across
each resistance can be found as well.

III-117
Chapter 3 Circuit Laws and Networking Theorems

Fig. 3.27. Application of the superposition principle to voltage sources.

Example 3.14: Find current I1 in the circuit shown in Fig. 3.27.

Solution: The first step is again shown in the figure; we apply the superposition theorem and
obtain two simpler circuits. Each of those circuits is solved using series/parallel equivalents.
For the circuit with the 15 V power source,

Req ¼ 1 kΩ þ 3 kΩ þ 0:75 kΩ ¼ 4:75 kΩ ð3:32Þ

The circuit current is 3.1579 mA; i1 is 75% of this value (from current division). For the circuit
with the 10 V source, Req ¼ 3.8 kΩ and i2 is 2.1053 mA. Thus, current I1 is given by
I 1 ¼ i1 þ i2 ¼ 4:4737 mA ð3:33Þ

III-118
Chapter 3 Section 3.3: Superposition Theorem and Its Use

Fig. 3.28. Application of the superposition principle to current sources.

Example 3.15: Find current I1 in the circuit shown in Fig. 3.28.

Solution: The first step is shown in the figure; we apply the superposition theorem and obtain
two simpler circuits by disconnecting current sources. The first circuit predicts i1 ¼ 15 mA,
while the second circuit predicts i2 ¼ 10 mA. Therefore,
I 1 ¼ i1 þ i2 ¼ 25 mA ð3:34Þ
In other words, each of the currents i1, i2 flows in its own loop. The solution does not depend on
any particular resistor value.

The superposition theorem is the direct consequence of circuit linearity. Interestingly, the
superposition theorem is applicable not only to DC circuits but also to AC and transient circuits.
The superposition theorem does not hold for electric power though, since power is the product of
voltage and current. The circuit power and the power delivered to individual elements may be
correctly obtained only from the final solution.

Application Example: Superposition Theorem for a Cellphone

Why do we ultimately need to solve circuits with multiple sources? Look at your cellphone. There
is a circuit inside, which receives a very weak radio-frequency input voltage signal from the
antenna. This is the first source. The signal is processed and amplified by transistors powered by
the cellphone battery. This is the second source (or sources). The signal is then demodulated
through interaction with an internal high-frequency signal generator. This is the third source.

III-119
Chapter 3 Circuit Laws and Networking Theorems

Moreover, every component in that circuit is in fact “noisy”; it creates a small thermal noise
voltage, which is modeled by its own tiny voltage source. A careful solution of the entire circuit
with multiple voltage and current sources including noise sources is possible by superposition.
This solution allows electrical engineers to extract the weak input signal from otherwise over-
whelming noise and properly design the cellphone circuitry.

3.3.3 Y (Wye) and Δ (Delta) Networks: Use of Superposition

The series and parallel resistance configurations are not the only meaningful network blocks.
Situations often arise when the resistances are neither in parallel nor in series. In Fig. 3.29a, b, we
introduce two new networking blocks, which are in importance second to the series/parallel
equivalents.

Fig. 3.29. (a) Y (wye) and (b) Δ (delta) networks; (c, d) applying the superposition theorem to
establish network equivalence.

The first block is known as the Y (wye) network. It represents a nodal connection of three
arbitrary resistances. The second block is known as the Δ (delta) network. It represents a loop
connection of three arbitrary resistances. Both Y and Δ networks have three terminals 1, 2, and 3;
they are therefore known as three-terminal networks. This is in contrast to series/parallel resis-
tance circuits, which are usually two-terminal networks.
The Y and Δ networks occur either independently or as part of a larger network. Important
applications include three-phase power electronics circuits, filter circuits, and impedance-
matching networks in high-frequency circuits. The theory that follows holds for AC circuits
too, when the resistances become general impedances.

III-120
Chapter 3 Section 3.3: Superposition Theorem and Its Use

Conversion Between Y and Δ Networks

A problem of significant practical importance is the conversion between Y and Δ networks in
Fig. 3.29a, b. This conversion is equivalent to replacing a node by a loop (strictly speaking, by a
mesh) and vice versa in a more complicated circuit or network. Such a replacement may
significantly simplify the overall circuit analysis. The conversion is established based on the
superposition theorem. Two arbitrary networks are equivalent if their υ-i characteristics are the
same. In other words, by connecting three arbitrary sources to terminals 1, 2, and 3 of the Y
network, we must obtain terminal voltages and currents identical to those of the Δ network with
the same sources. We select three current sources I1, I2, I3 in Fig. 3.29c, d. The solution with three
sources is obtained as a superposition of three partial solutions, with two sources open-circuited at
a time. Let us keep the source I1 and replace I2, I3 by open circuits first. Voltages V12 for both
networks will be the same when the equivalent resistances R12 between terminals 1 and 2 are the
same. A similar treatment holds for terminals 1 and 3 (source I3 is kept), and terminals 2 and
3 (source I2 is kept), respectively. Therefore, with reference to Fig. 3.29c, d, we have
Rb ðRa þ Rc Þ
R12 ¼ R1 þ R3 ¼ Rb kðRa þ Rc Þ ¼ ð3:35aÞ
Ra þ Rb þ Rc
Rc ðRa þ Rb Þ
R13 ¼ R1 þ R2 ¼ Rc kðRa þ Rb Þ ¼ ð3:35bÞ
Ra þ Rb þ Rc
Ra ðRb þ Rc Þ
R23 ¼ R2 þ R3 ¼ Ra kðRb þ Rc Þ ¼ ð3:35cÞ
Ra þ Rb þ Rc

Next, we add Eq. (3.35a) and Eq. (3.35b) and subtract from this result Eq. (3.35c). This gives us an
expression for R1. To obtain R2, we add Eq. (3.35b) and Eq. (3.35c) and subtract Eq. (3.35a). R3 is
obtained by adding Eq. (3.35a) and Eq. (3.35c) and subtracting Eq. (3.35b). The result has the
form of a Δ to Y transformation:
Rb Rc Ra Rc Ra Rb
R1 ¼ , R2 ¼ , R3 ¼ ð3:36Þ
Ra þ Rb þ Rc Ra þ Rb þ Rc Ra þ Rb þ Rc

The inverse transformation, a Y to Δ transformation, follows

R1 R2 þ R1 R3 þ R2 R3 R1 R2 þ R1 R3 þ R2 R3 R1 R2 þ R1 R3 þ R2 R3
Ra ¼ , Rb ¼ , Rc ¼ ð3:37Þ
R1 R2 R3

Balanced Y and Δ Networks

When all resistances of the Y network are equal to RY , the Y network is said to be balanced. When
all resistances of the Δ network are equal to RΔ, the Δ network is also balanced. For balanced
networks, two previous equations reduce to

III-121
Chapter 3 Circuit Laws and Networking Theorems

RΔ
RY ¼ , RΔ ¼ 3RY ð3:38Þ
3

Example 3.16: Find equivalent resistance of a network between terminals a and b in

Fig. 3.30a.
Solution: To simplify the network, we use the Δ to Y transformation and obtain the network
shown in Fig. 3.30b. Since all resistances of the bottom Δ network are equal, we can use the
simplified Eq. (3.38) to find the new resistance values. The remaining circuit is solved using
series/parallel combinations, which gives us Req ¼ 2.2 kΩ.

Fig. 3.30. Network simplification using Δ to Y transformations.

The conversions between Y and Δ networks date back to Arthur E. Kennelly (1861–1939),
an Indian American engineer who established them in 1899. Note that the transformation between
Y and Δ networks is also called the star to delta transformation.

3.3.4 T and Π Networks: Two-Port Networks

The Y and Δ networks are equivalent to T and Π networks, respectively, which are shown in
Fig. 3.31. The T and Π networks are predominantly used as two-port networks or two-port
networking blocks. Every port should have two terminals. To create the two ports with two
terminals each, we simply split terminal 2 in Fig. 3.31a, c into two terminals: 2 and 4 in Fig. 3.31b,
d, respectively. Port #1 is the input port of the network. Port #2 is the output port of the network.
Multiple two-port networks may be connected in chains. Equations (3.36), (3.37), and (3.38)
allow us to establish the connections between the T and Π networks, which are the same as the
connections between the Y and Δ networks. Hence, any two-port T network may be replaced by a
two-port Π network and vice versa. The T network is sometimes called the T pad and the Π
network the Π pad. Both networks are used as attenuators, filters, and antenna tuners. In the last
two cases, capacitances and inductances occur in place of resistances.

III-122
Chapter 3 Section 3.3: Superposition Theorem and Its Use

Exercise 3.7: A two-port T network in Fig. 3.31b is characterized by R1 ¼ 3 kΩ and

R2 ¼ R3 ¼ 5 kΩ. Establish resistance values for the equivalent Π network in Fig. 3.31d.

Answer: Ra ¼ 18.33 kΩ, and Rb ¼ Rc ¼ 11 kΩ.

Fig. 3.31. Conversion of Y and Δ three-terminal networks to equivalent T and Π two-port, four-
terminal networks. (a) Y network. (b) Two-port T network. (c) Δ network. (d) Two-port Π network.

3.3.5 General Character of Superposition Theorem

The superposition theorem makes it possible to analyze not only the two-port electric networks but
also various networks with multiple ports such as sensor arrays, antenna arrays, multiple-input and
multiple-output (MIMO) communication systems, arrays of magnetic resonance imaging (MRI)
coils, etc. It is so widely used that, quite often, we do not even mention its name and consider the
corresponding result as “obvious.”

III-123
Chapter 3 Circuit Laws and Networking Theorems

Summary

(continued)

III-124
Chapter 3 Summary

(continued)

III-125
Chapter 3 Circuit Laws and Networking Theorems

III-126
Chapter 3 Problems

Problems
3.1 Circuit Laws: Networking
Theorems
3.1.1 Electric Network and Its Topology
Problem 3.1. In the network graph shown in the
figure below:
A. Find the number of branches, single nodes 3.1.2 Kirchhoff’s Current Law (KCL)
(nodes with distinct voltages), and meshes. Problem 3.4. Find current IB for the node shown
B. Prove the equality b ¼ n + m  1 for the in the figure that follows.
number of branches b, meshes m, and single
nodes n.

Problem 3.2. Repeat problem 3.1 for the circuits Problem 3.5.
shown in the following figures. A. Find current ID for the node shown in the
figure below.
B. Redraw this node (and the circuit between
terminals a and b) in an equivalent form
eliminating the horizontal wire.

Problem 3.6. Determine current iC for the circuit

Problem 3.3. Repeat problem 3.1 for the circuit shown in the figure that follows.
shown in the figure that follows. Each straight
segment is now a branch.
III-127
Chapter 3 Circuit Laws and Networking Theorems

Problem 3.10. Determine currents iC, iF, iH in the

figure that follows.
Problem 3.7. Find a relation between currents iC,
iE for the circuit shown in the figure that follows.
Does the problem have a unique solution?

Problem 3.11. Find all unknown currents for the

circuit shown in the figure below.

Problem 3.8. Find currents iA, iB, iD, iH for the

circuit shown below.

3.1.3 Kirchhoff’s Voltage Law (KVL)

Problem 3.12.
A. Find voltage VC for the circuit shown in the
figure that follows;
Problem 3.9. Determine currents iC, iF, iH for the B. How would the solution change if VD were
circuit shown in the figure that follows. equal to 0 V?

III-128
Chapter 3 Problems

C. Could the value VE ¼ 0 V be used in this Problem 3.15. Determine voltage VE for the
problem? circuit shown below.

Problem 3.13. Determine voltage VF in the cir-

cuit shown in the figure below.

Problem 3.16. Equipotential lines for a human

body subject to a vertical electric field with the
strength of 1 V/m are shown in the figure that follows.
A. Determine voltages VAB, VBC, VCA.
B. Establish the KVL loop and formulate
KVL for these three voltages.

Problem 3.14. Find the unknown voltages VA, VB,

VG, VH for the circuit shown in the figure below.

3.1.4 Power-Related Theorems

Problem 3.17. For the circuit shown in the figure
that follows:
III-129
Chapter 3 Circuit Laws and Networking Theorems

(i) Determine which circuit elements are Problem 3.19. For the circuit shown in the figure:
resistances and which are sources. 1. Use KVL to solve for the unknown
(ii) Find the power (delivered to the circuit or voltages.
taken from the circuit) for every circuit 2. Use KCL to solve for the unknown
element. currents.
(iii) Assuming that the powers of the sources 3. For each of six circuit elements, determine
are negative, find the sum of all powers if the element is a resistance or a source.
in the circuit. 4. Assuming that the source powers are nega-
tive, find the algebraic sum of all powers in
the circuit.

Problem 3.18. For the circuit shown in the

figure:
1. Use KVL and KCL to solve for unknown
currents and voltages.
2. Find the power (delivered to the circuit or
taken from the circuit) for every circuit
element.
3. Assuming that the powers of the sources are
negative, find the algebraic sum of all pow-
ers in the circuit.
3.2 Series and Parallel
Network/Circuit Blocks
3.2.1 Sources in Series and in Parallel
Problem 3.20. The electronic circuits onboard
an 18-foot long Parker motor boat consume 96 W
when operated from a 24-V source. The source is
a combination of two fully charged deep-cycle
batteries, each of which is rated for 12 V and
100 ampere hours.
1. Should the batteries be connected in series
or in parallel?
2. For how many hours can the electronics be
operated from the battery bank without
recharging?
3. How much energy in kilowatt hours is ini-
tially stored in each battery?

III-130
Chapter 3 Problems

Problem 3.21. A certain sensing device con- below.All resistancesare equal: R1 ¼ . . .¼ R15 ¼10
sumes 0.375 W of power over a 20 h period and Ω. Determine the heat power (power delivered to
operates from a 6 V source. The source is a the defroster).
combination of four fully charged AAA batteries,
1.5 V each. The batteries discharge by the end of
the 20 h period.
1. Should the batteries be connected in series
or in parallel?
2. What is a typical capacity of the AAA
battery?
3. How much energy in watt hours is initially
stored in each battery?

Problem 3.22. A load has the resistance of 1.5 Ω

and requires the applied voltage of 3 V. A number
of battery cells are given, each of which is rated
for 1.5 V. Each cell may deliver no more than 1 A
of current.
1. Construct and draw a battery bank that
could be used to drive the load.
2. Is the solution to the problem unique?
Problem 3.26. Find the equivalent resistance
Problem 3.23. Repeat the previous problem between terminals a and b.
when the required load voltage is changed to 6 V.
3.2.2 Resistances in Series and in Parallel
3.2.3 Reduction of Resistive Networks
Problem 3.24. Determine the equivalent resis-
tance between terminals a and b.

Problem 3.27. Find the equivalent resistance

between terminals a and b.

Problem 3.28. Find the equivalent resistance

Problem 3.25. The equivalent electric circuit for between terminals a and b.
a car rear window defroster is shown in the figure

III-131
Chapter 3 Circuit Laws and Networking Theorems

Problem 3.33. Find the equivalent resistance

between terminals a and b.

Problem 3.29. Determine the equivalent resis-

tance between terminals a and b.

Problem 3.34. Determine the equivalent resis-

tance of the network (resistance between termi-
nals a and b) shown in the figure below.
Problem 3.30. Find the equivalent resistance
between terminals a and b.

Problem 3.31. Determine the equivalent resis-

tance (resistance between ports a and b) of the Problem 3.35. Determine the equivalent resis-
network shown in the figure that follows. tance between terminals a and b (show units).

Problem 3.32. Determine the equivalent resis-

tance between terminals a and b.

Problem 3.36. A network shown in the figure

below is known as a ladder. The ladder network
includes one particular section with a series
(R1 ¼ 1 Ω) and a shunt (R2 ¼ 1 Ω) resistance,
known as the L-section. This section is then
repeated an infinite number of times to the right.
Determine the equivalent resistance between ter-
minals a and b of the ladder network.

III-132
Chapter 3 Problems

A. Determine the equivalent resistance

between terminals a and b of the ladder
network when R1 ¼ 2R3 ¼ 10 Ω and
R2 ¼ 10 Ω.
B. How different is your result from
the equivalent resistance of the finite ladder
network with only four sections?

Hint: For the semi-infinite ladder network in the

previous figure, the equivalent resistance, Req, will
not change after adding a new section up front as
shown in the figure that follows.

3.2.4 Voltage Divider Circuit

Problem 3.39. Redraw the circuit shown in the
figure that follows and plot the distribution of the
circuit voltage versus ground reference point to
scale between two circuit points a and b as a
function of distance x from point a.

Problem 3.37.
A. Repeat the previous problem for the semi-
infinite ladder network circuit shown in the
same figure when R1 ¼ R3 ¼ 10 Ω and
R2 ¼ 25 Ω.
B. How different is your result from
the equivalent resistance of the finite ladder
network with only four sections?

Problem 3.38. Another important ladder net-

work type (with the T-section) is shown in the
figure.

III-133
Chapter 3 Circuit Laws and Networking Theorems

Problem 3.40. Repeat the task of the previous

problem for the disconnected circuit shown in the
figure below.

Problem 3.41. For the circuit shown in the figure

that follows, use the voltage division principle to
calculate V1, V2, V3.

Problem 3.42. For the circuit shown in the fol-

lowing figure:
A. Calculate V1, V2, V3.
B. Find the voltage across the current source. 3.2.5 Application Example: Voltage Divider
as a Sensor Circuit
3.2.6 Application Example: Voltage Divider
as an Actuator Circuit
Problem 3.44. An NTC thermistor-based tem-
perature sensor should operate between 25 C and
65 C from a 6 V DC power supply. The therm-
istor’s resistance changes from R0 ¼ 50 kΩ to
R00 ¼ 20 kΩ in this temperature range.
A. Present a circuit diagram for the simple
temperature sensor.
Problem 3.43. For the circuits shown in the B. Determine the value of the unknown
figure that follows, determine voltages V1, V2, resistance for the maximum circuit
absolute voltage Va at node a versus ground sensitivity.
(show units), and circuit current I.

III-134
Chapter 3 Problems

C. Determine the maximum circuit B. Can you derive an analytical formula that
sensitivity. gives the voltage variation across
D. What is the circuit sensitivity when the R2 ¼ R1  Δ as a linear function of an
unknown resistance is set to 1 kΩ? arbitrary (but very small) resistance var-
iation Δ? [Hint: use your calculus back-
Problem 3.45. In the previous problem, using ground—the Maclaurin series versus a
software of your choice (MATLAB is small parameter.]
recommended), plot the circuit sensitivity to 3.2.7 Current Limiter
scale as a function of the value of the unknown
Problem 3.49. A thermistor is connected to an
resistance in the range from 1 kΩ to 100 kΩ.
ideal voltage power source of 9 V. Determine the
value of the current-limiting resistor R based on
Problem 3.46. A strain gauge with nominal the requirement that the power delivered to the
resistance of 120 Ω is used in conjunction thermistor should be always less than 0.1 W. The
with a 2.5 V DC voltage source. Its nominal lowest possible value of R should be chosen.
resistance changes by 0.2% when the gauge Consider two cases:
operates in the permissible strain range, which 1. Thermistor resistance is exactly 100 Ω.
is 1000με. 2. Thermistor resistance changes from
A. Present a circuit diagram for the simple 200 Ω to 100 Ω.
strain gauge sensor.
B. Determine the value of the unknown
resistance for the maximum circuit
sensitivity.
C. Determine the maximum circuit
sensitivity.
D. What is the circuit sensitivity when the
unknown resistance is set to1 kΩ?
3.2.8 Current Divider Circuit
Problem 3.47. In the previous problem, Problem 3.50. For the circuit shown in the figure
using software of your choice (MATLAB is that follows:
recommended), plot the circuit sensitivity A. Calculate voltage between terminals
to scale as a function of the value of the a and b. Show its polarity on the figure.
unknown resistance in the range from 10 Ω to B. Use the current division principle to cal-
1000 Ω. culate branch currents i1, i2.

Problem 3.48. A voltage divider circuit with

R1 ¼ 700 Ω (the fixed resistance) and R2 ¼
700 Ω  0.1% (the strain gauge) is used.
The voltage power supply is rated at 4.5 V.
A. Show that voltage across the strain
gauge varies in the range
2.25 V  1.125 mV.

III-135
Chapter 3 Circuit Laws and Networking Theorems

Problem 3.51. Find branch currents i1, i2 for the Problem 3.56. Find the voltage between termi-
circuit depicted in the figure below. nals a and b for the circuit shown in the figure that
follows.

Problem 3.52. Find branch currents i1, i2 for the

circuit shown in the figure that follows.
Problem 3.57. Find the voltage between termi-
nals a and b for the circuits shown in the figure
that follows.

Problem 3.53. Find the voltage between terminals

a and b (voltage across the current power source)
for the circuit shown in the figure that follows.

Problem 3.54. The voltage source in the circuit is

delivering 0.2 W of electric power. Find R.

Problem 3.58. You are given:

(i) A photoresistor that changes its resis-
tance from 20 kΩ for brightness to
500 kΩ for darkness
(ii) A 9 V battery
(iii) A voltmeter (DMM)
(iv) Any other necessary precise resistors
Construct (present the circuit diagram) a Wheat-
stone bridge sensor circuit that:
3.2.9 Wheatstone Bridge (i) Has zero voltage reading for brightness
Problem 3.55. Describe in your own words the
function and the scope of the Wheatstone bridge.

III-136
Chapter 3 Problems

(ii) Has maximum possible voltage reading 2. Label one (or two) strain gauges used, and
for darkness (has maximum sensitivity) specify the values of all deployed
resistances.
Problem 3.59. You are given: 3. Show power supply and DMM
1. A SGT-1/350-TY11 uniaxial strain gauge connections.
with the nominal resistance of 350 Ω Combined Voltage and Current Dividers
(no strain) Problem 3.61. For the circuit shown in the
2. A 4.5 V voltage source figure:
3. Any number of precise resistors, of any A. Find currents i, i1, i2 (show units).
value B. Find power P delivered by the voltage
When tensile strain is applied, a resistance varia- source to the circuit.
tion up to +0.1% is observed. Present the circuit C. Find voltages V1, V2.
diagram for the Wheatstone bridge sensor circuit
that:
(i) Has zero voltage reading at no strain
(ii) Has maximum possible voltage
response when strain is present (has
maximum sensitivity)
(iii) Outputs positive voltages when the
resistance of the strain gauge increases

Problem 3.60. Resistance of a strain gauge Problem 3.62. Find voltage V across the 20 kΩ
increases when its length increases (a bended resistance and current i for the circuit shown in
surface under test becomes convex) and the figure that follows.
decreases when its length decreases (a bended
surface under test becomes concave). The
corresponding strains are the tensile strain and
the compressive strain. You are given two strain
gauges (#1 and #2), which are attached to oppo-
site sides of a thin bended surface under test, at
the same position. The gauge resistance at nor-
mal conditions (no bending) is R ¼ 100 Ω. You
are also given any number of fixed resistors, of
any value.
1. Suggest and sketch a sensor circuit which
will convert changes in the resistance into Problem 3.63. For the circuit that follows
measurable voltage changes. This circuit determine:
should possibly have: A. Current i2 through the 0.25 kΩ resistance
(a) Zero sensor output voltage at normal B. Power P absorbed by the 600 Ω
sensor conditions (no bending) resistance
(b) Maximum voltage sensitivity to
changes in resistance (sensitivity to
the strain)

III-137
Chapter 3 Circuit Laws and Networking Theorems

3.3 Superposition Theorem

and Its Use
3.3.2 Superposition Theorem or Superposi-
tion Principle
Problem 3.67. Solve the circuits shown in the
figure that follows (find current I and show units)
using superposition theorem.
Problem 3.64. For the circuit shown in the figure
determine:
A. Current i2 through the 0.25 kΩ
resistance,
B. Power P absorbed by the 0.3 kΩ
resistance.

Problem 3.68. Find current I (show units) using

the superposition theorem.
Problem 3.65. Determine current i (show units)
in the circuit that follows.

Problem 3.69. Find current I (show units) using

the superposition theorem.
Problem 3.66. Determine current i and voltage υ
for the circuit shown in the figure (show units).

III-138
Chapter 3 Problems

Problem 3.71. For the circuit shown in the fig-

ure, determine the absolute voltage (versus chas-
sis ground) and the electric current at the circuit
point a.

Problem 3.72. Determine the absolute voltage

(versus chassis ground) and the current i at the
circuit point a.

Problem 3.70.
(a) Find current I (make sure to show units)
using the superposition theorem.
(b) Find voltage V (make sure to show
units) using the superposition theorem.

Problem 3.73. For the circuit shown in the figure

that follows, find the current across the 10 Ω
resistance. Show its direction in the figure.

III-139
Chapter 3 Circuit Laws and Networking Theorems

Problem 3.77. Convert the network shown in the

figure that follows from Δ to Y:
Problem 3.74. Determine the unknown current
A. Draw the corresponding Y network.
I in the circuit (solve the problem by
B. Label its terminals.
superposition).
C. Determine and label the corresponding
resistance values.

Problem 3.75. Determine the unknown

current I in the circuit (solve the problem by
superposition).
Problem 3.78. Convert the two-port network
shown in the figure that follows from T to Π:
1. Draw the corresponding Π network.
2. Label its terminals and ports.
3. Determine and label the corresponding
resistance values.

3.3.3 Y (Wye) and Δ (Delta) Networks: Use

of Superposition
3.3.4 T and Π Networks: Two-Port Passive
Networks
Problem 3.76. Convert the network shown in the
figure that follows from Y to Δ:
A. Draw the corresponding Δ network.
B. Label its terminals.
C. Determine and label the corresponding Problem 3.79. Convert the two-port network
resistance values in the figure. shown in the figure that follows from Π to T:

III-140
Chapter 3 Problems

A. Draw the corresponding T network.

B. Label its terminals and ports.
C. Determine and label the corresponding
resistance values.

Problem 3.80. For the bridge network, deter-

mine the equivalent resistance between terminals
a and b.

III-141
Chapter 4

Chapter 4: Circuit Analysis and Power

Transfer

Overview

Prerequisites:
• Knowledge of major circuit elements, their
-i characteristics, and Ohm’s law (Chap. 2)
• Knowledge of basic networking theorems (Chap. 3)

Objectives of Sect. 4.1:

• Become familiar with the nodal analysis and be able to apply it to solve in arbitrary
linear circuits
• Become familiar with the mesh analysis and be able to apply it to solve in arbitrary
linear circuits

Objectives of Sect. 4.2:

• Become familiar with the method of short/open circuit
• Establish and prove the source transformation theorem
• Establish and prove Thévenin and Norton’s theorems
• Be able to apply Thévenin and Norton’s equivalents for circuit solution

Objectives of Sect. 4.3:

• Establish the maximum power theorem and become familiar with the power
efficiency concept
• Be able to apply the concepts of Thévenin and Norton’s equivalents and maximum
power theorem in practice

Objectives of Sect. 4.4:

• Obtain an initial exposure to nonlinear circuit analysis
• Be able to solve in a simple nonlinear circuit

© Springer Nature Switzerland AG 2019 IV-143

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_4
Chapter 4 Circuit Analysis and Power Transfer

Application Examples:
Reading and using data for solar panels
Power radiated by a transmitting antenna
Maximum power extraction from solar panel
Solving the circuit for a generic solar cell

Keywords:
Nodal analysis, Mesh analysis (mesh current analysis), Supernode, Supermesh, Method of short/open
circuit (definition of open-circuit network voltage, short-circuit network current), Source
transformation theorem, Circuit equivalent (see equivalent circuit), Thévenin’s theorem
(formulation, proof, special cases), Thévenin equivalent, Norton’s theorem, Norton equivalent, R-2R
ladder network, negative equivalent (Thévenin) resistance, Maximum power theorem (principle of
maximum power transfer), Power efficiency, Analysis of nonlinear circuits, Load line (definition,
method of), Iterative method for nonlinear circuits (definition of explicit iterative scheme, implicit
iterative scheme), Solar cell (c-Si, open-circuit voltage, short-circuit photocurrent density, fill factor,
characteristic equation of), Solar panel (series cell connection, open-circuit voltage, short-circuit
photocurrent, fill factor, maximum power load voltage, maximum power load current)

IV-144
Chapter 4 Section 4.1: Nodal/Mesh Analysis

Section 4.1 Nodal/Mesh Analysis

4.1.1 Importance of Circuit Simulators
The series and parallel equivalent circuits along with Y and Δ transformations provide a practical
tool for solving simple circuits involving typically only a few elements. However, for more
elaborate circuits, circuit simulators such as SPICE (Simulation Program with Integrated Circuit
Emphasis) and their various modifications become indispensable tools for the professional
engineer. SPICE was developed by the Electronics Research Laboratory at the University of
California, Berkeley, and first presented in 1973. These circuit simulators are quite general and
allow us to model circuits with passive and active elements including semiconductor devices such
as diodes, transistors, and solar cells. Since those elements typically exhibit nonlinear current-
voltage behaviors, elaborate solution strategies are needed. The circuit simulators use quite
interesting algorithms: they often operate in the time domain, even for DC circuits. For example,
a solution for a DC circuit is obtained as the steady-state limit of a transient solution, for voltage
and/or current sources turned on at a certain time instance. The key of the time-domain approach is
its inherent ability to solve nonlinear problems, with passive and active circuit elements. In this
section, we will not discuss the specifics of the numerical circuit simulation. Instead, we will
provide the foundation of the nodal analysis (or node analysis) and the mesh analysis (or the mesh
current analysis), which are two important features of any professional circuit simulator. The
nodal and mesh analyses in their pure form do not involve time-domain methods. They are
primarily applicable only to linear circuits, also referred to as linear networks.

4.1.2 Nodal Analysis for Linear Circuits

The nodal analysis is a general method of solving linear networks of arbitrary complexity, which
is based on KCL and Ohm’s law. Let us consider a circuit shown in Fig. 4.1a, which is a resistive
bridge circuit with a bridging resistance. This circuit may be solved using Δ to Y conversion; see,
for instance, example 3.16 of Chap. 3. Here, we prefer to use the nodal analysis directly. The nodal
analysis operates with the absolute values of the node voltages in the circuit with respect to ground
reference. It may be divided into a number of distinct steps:
1. A ground reference needs to be assigned first; this is a node where the voltage is set to 0 V. To this
end, we ground the negative terminal of the voltage power supply.
2. Next, we select nontrivial (also called non-reference) nodes for which we do not know the
voltages. These are nodes 1 and 2 in Fig. 4.1b. The two additional nodes are eliminated from the
analysis since the voltages there are already known.

IV-145
Chapter 4 Circuit Analysis and Power Transfer

3. We label absolute node voltages versus ground reference as V1, V2—see Fig. 4.1c.
4. Then currents for every nontrivial node are labeled, assuming that all currents are outflowing—
see Fig. 4.1c. This condition can be replaced by all inflowing currents.
5. Next, KCL is applied to every nontrivial node. We express the currents as the difference of two
absolute voltages: the voltage at the beginning of the current arrow (voltage at the master node)
minus the voltage at the end of the current arrow (voltage at any other node) and then divide this
difference by the appropriate resistance. Hence, we arrive at a system of linear equations for the
nodal voltages. Currents are no longer involved.
6. After the resulting system of linear equations is solved, all circuit parameters can be determined
as necessary.

a)
1 kW 1 kW
1 kW
5V +
-
3 kW 1 kW

1. Select ground reference

2. Select and label nontrivial nodes
b) 5V
1 kW 1 kW
1 kW
5V + 1 2
-
3 kW 1 kW
0V 3. Label unknown node voltages
0V 4. Label outflowing currents
c) 5V for each node

1 kW 1 kW
1 kW
5V + V1 V2
-
5. Write KCL for each node
3 kW 1 kW in terms of node voltages
0V 6. Solve the resulting system
0V of equations

Fig. 4.1. Major steps of the nodal analysis applied to a bridge circuit.

Note that the resulting system of linear (or nodal) equations may have one, two, three, or more
equations depending on the complexity of the circuit as well as on your own “smart” choice of the
nontrivial nodes. The following two examples will apply the nodal analysis to a circuit with a
voltage source.

IV-146
Chapter 4 Section 4.1: Nodal/Mesh Analysis

Example 4.1: Solve the circuit shown in Fig. 4.1a using nodal analysis and find the supply
current.

Solution: Steps 1 to 4 are indicated in Fig. 4.1b, c. Applying KCL to node 1 and then to node
2 (the order is not important), one has.
V1
5 V V1
0 V V1
V2
þ þ ¼0 ð4:1aÞ
1 kΩ 3 kΩ 1 kΩ

V2
5 V V2
0 V V2
V1
þ þ ¼0 ð4:1bÞ
1 kΩ 1 kΩ 1 kΩ

This is a system of linear equations for two unknown voltages. It simplifies to

7=3V 1
V 2 ¼ 5 V ð4:2aÞ
3V 2
V 1 ¼ 5 V ð4:2bÞ

Equations (4.2) are solved, for example, via Gaussian elimination. This yields
V 1 ¼ 3:33 V, V 2 ¼ 2:78 V ð4:3Þ

The circuit current (current of the voltage source) is therefore

ð5 V
V 1 Þ=1 kΩ þ ð5 V
V 2 Þ=1 kΩ ¼ 3:89 mA:

a) 1 kW b) 1 kW

1 kW 1 kW 1 kW V1 1 kW
10 V V2
1 2
+ 10 V 1 kW 1 kW + 10 V 1 kW 1 kW
- -

0V 0V 0V

Fig. 4.2. Major steps of the nodal analysis applied to a circuit with a voltage source.

Example 4.2: Solve the circuit shown in Fig. 4.2a employing the nodal analysis.
Solution: Steps 1 to 4 are indicated in Fig. 4.2b. Applying KCL to node 1 and then to node
2, one obtains a system of equations with two unknown voltages:

IV-147
Chapter 4 Circuit Analysis and Power Transfer

Example 4.2 (cont.):

V 1
10 V V 1
0 V V 1
V 2
þ þ ¼0 ð4:4aÞ
1 kΩ 1 kΩ 1 kΩ
V 2
10 V V 2
0 V V 2
V 1
þ þ ¼0 ð4:4bÞ
1 kΩ 1 kΩ 1 kΩ

In setting up these equations, it does not matter which sequence of nodes are selected.
Simplifying Eqs. (4.4) gives
3V 1
V 2 ¼ 10 V ð4:5aÞ
3V 2
V 1 ¼ 10 V ð4:5bÞ

The solution is obtained by symmetry, i.e., V1 ¼ V2 ¼ 5 V. The circuit current provided by the
power supply is 10 mA. All other branch currents can then be found using Ohm’s law. An
interesting feature of the circuit shown in Fig. 4.2a is that the marked 1 kΩ resistor can be
considered as “dead,” since there is no current flowing through it (the voltage difference across
this resistor is exactly zero). This resistor can be removed from the circuit without affecting the
behavior of the circuit in terms of voltages and currents. It might appear at first sight that the
circuits shown in Figs. 4.1 and 4.2 have a different network topology. In fact, they do not. To
prove this, you should attempt to redraw the circuit in Fig. 4.2a; your redrawn topology should
be identical with the circuit in Fig. 4.1a.

Circuits with a Current Source

When a current source is present in a circuit, the solution becomes even simpler: one makes use of
the existing current and substitutes its value into the KCL equation written for a particular node.
For example, KLC applied to node 1 in Fig. 4.3 includes the outflowing current of
1 mA; the
current sign must be taken into account. The same idea is applied to circuits with multiple current
power supplies. When only the current sources are present, the ground may be connected to the
incoming terminal of a current source.

1 kW

1 1 kW 1 kW 2

1 mA 10 V + 1 kW
-

Fig. 4.3. A circuit with a current source solved via nodal analysis.

IV-148
Chapter 4 Section 4.1: Nodal/Mesh Analysis

Exercise 4.1: Determine the voltage across the current source in Fig. 4.3 using the nodal
analysis.

Answer: 8.6 V.

4.1.3 Supernode
The nodal analysis requires a “good” eye to see possible simplifications when labeling the nodes.
Let us examine a particular case and point out a few useful subtleties.1 Figure 4.4a depicts a
network with two voltage sources. The property of the 5 V source is such that it is not fixed to a
particular ground connection—we therefore call it a floating source. Setting up the nodal analysis
becomes tricky, since we do not know the current through this source. However, a supernode may
be formed as shown in Fig. 4.4b.

1 kW
a)

1 kW 5V
-
+

10 V + 1 kW 1 kW
-

supernode

b) 1 kW

5V
1 kW V1
10 V - V2
+

10 V + 1 kW 1 kW
-

0V 0V 0V

Fig. 4.4. A network with a floating voltage source between nodes 1 and 2.

When KCL is applied to any closed contour around the supernode, the net current must add to
zero in such a case. With reference to Fig. 4.4b this yields
V 1
10 V V 1
0 V V 2
0 V V 2
10 V
þ þ þ ¼0 ð4:6aÞ
1 kΩ 1 kΩ 1 kΩ 1 kΩ

1
Subtleties are often euphemism for “playing” around with the circuit, like redrawing the wire connections and
rearranging the circuit elements. This is done to find simpler solution approaches.

IV-149
Chapter 4 Circuit Analysis and Power Transfer

What is the second equation for the two unknowns V1 and V2? It simply turns out to be the relation
between the supernode voltages themselves. Since V2
V1 is the voltage of the power source, we
obtain
V2 ¼ V1 þ 5 V ð4:6bÞ

Equations (4.6) can now be solved even without the calculator, eliminating one of the unknowns,
which yields
V 1 ¼ 2:5 V, V 2 ¼ 7:5 V ð4:7Þ

The circuit is solved. All currents can be found using the nodal voltages and Ohm’s law.

Example 4.3: Solve the circuit shown in Fig. 4.4a by using the standard nodal analysis,
without the supernode concept.
Solution: We have to specify an unknown current Ix through the 5 V source, which flows, say,
from left to right in Fig. 4.4a. This convention results in the following two equations for the two
nodes.
V 1
10 V V 1
0 V
þ þ Ix ¼ 0 ð4:8aÞ
1 kΩ 1 kΩ
V 2
0 V V 2
10 V
þ
Ix ¼ 0 ð4:8bÞ
1 kΩ 1 kΩ

Next, we can add both equations and thereby eliminate Ix. The result is exactly Eq. (4.6a) for
the supernode. We must add one more condition to solve this equation. Eq. (4.6b) is the only
choice, i.e.,
V2 ¼ V1 þ 5 V ð4:8cÞ

Interestingly, we arrive at the supernode concept again but in a more complicated way. This is
the reason why the supernode approach is such useful tool.

4.1.4 Mesh Analysis for Linear Circuits

The mesh analysis (or the mesh current analysis) is using loops instead of nodes. Only loops that
do not contain any other loops—the meshes—are employed. The meshes as elements of the
networking topology were defined in Sect. 3.1.1 of Chap. 3. Accordingly, instead of KCL the
mesh analysis makes use of KVL. Hence, we need to choose mesh currents for every mesh.
Figure 4.5 depicts the concept for a circuit with three meshes. Note that this circuit is identical to
the circuit from Fig. 4.1. A ground connection does not have to be introduced for the mesh

IV-150
Chapter 4 Section 4.1: Nodal/Mesh Analysis

method. Let us denote the mesh current for mesh 1 in Fig. 4.5 by I1, the mesh current for mesh 2 by
I2, and the mesh current for mesh 3 by I3.

mesh #3 mesh #1

1 kW 1 kW
I1
1 kW
5V + I3
-
I2
3 kW 1 kW
mesh #2

Fig. 4.5. Circuit analysis based on the mesh analysis; circuits in Fig. 4.5 and in Fig. 4.1 coincide.

The KVL equations for the three meshes are based on Ohm’s law for the proper reference
configuration. We do not need the fourth (large) loop encompassing the entire circuit. For resistors
that are shared by two adjacent meshes, we combine either the difference or the sum of the two
adjacent mesh currents. The mesh equations become
Mesh1 : 1 kΩ  ðI 1
I 3 Þ þ 1 kΩ  I 1 þ 1 kΩ  ðI 1
I 2 Þ ¼ 0 ð4:9aÞ
Mesh2 : 3 kΩ  ðI 2
I 3 Þ þ 1 kΩ  ðI 2
I 1 Þ þ 1 kΩ  I 2 ¼ 0 ð4:9bÞ
Mesh3 :
5V þ 1 kΩ  ðI 3
I 1 Þ þ 3 kΩ  ðI 3
I 2 Þ ¼ 0 ð4:9cÞ

We have arrived at a system of three equations for the three unknown mesh currents I1, I2, and I3.
After division by 1 kΩ and combining similar terms it simplifies to
Mesh1 : þ 3I 1
I 2
I 3 ¼ 0 ð4:10aÞ
Mesh2 :
I 1 þ 5I 2
3I 3 ¼ 0 ð4:10bÞ
Mesh3 :
I 1
3I 2 þ 4I 3 ¼ 5 mA ð4:10cÞ

In contrast, the nodal analysis applied to the same circuit requires only two equations for
two unknown node voltages, see Example 4.1. The final solution is indeed the same. Thus, the
nodal analysis is more beneficial for small networks when a voltage source or sources are present.
If, however, instead of a voltage source, a current source were present in Fig. 4.5, the nodal
analysis would require three equations. By contrast, the mesh analysis would require only two
equations, because I3 is defined by the current source. Reasoning like this gives us clues which
method is most suitable. When mixed power supplies like voltage and current sources are
involved, there is usually no real difference between the two methods. The choice often becomes
a matter of taste.

IV-151
Chapter 4 Circuit Analysis and Power Transfer

Exercise 4.2: Determine the mesh currents for the circuit in Fig. 4.5.

Answer: I1 ¼ +2.22 mA, I2 ¼ +2.78 mA, I3 ¼ +3.89 mA.

4.1.5 Supermesh
Consider the circuit shown in Fig. 4.6. The straightforward mesh analysis should use KVL written
for meshes 1 and 2. However, KVL cannot be formulated directly since we do not know the
voltage across the current source. A solution is to combine meshes 1 and 2 into a supermesh and
write KVL around its periphery. The mesh equations become
Supermesh : 1 kΩ  ðI 1
I 3 Þ þ 1 kΩ  I 1 þ 1 kΩ  I 2 þ 3 kΩ  ðI 2
I 3 Þ ¼ 0 ð4:11aÞ

Mesh3 :
5 V þ 1 kΩ  ðI 3
I 1 Þ þ 3 kΩ  ðI 3
I 2 Þ ¼ 0 ð4:11bÞ

KCL for the central branch of the source : I 1
I 2 ¼ 1 mA ð4:11cÞ

mesh #3 mesh #1

1 kW 1 kW
I1
1 mA
5V + I3
-
I2
3 kW 1 kW
mesh #2

Fig. 4.6. Circuit solved with the supermesh method.

After division by 1 kΩ and combining similar terms, the system of Eq. (4.11) is simplified to
þ2I 1 þ 4I 2
4I 3 ¼ 0

I 1
3I 2 þ 4I 3 ¼ 5 mA ð4:12Þ
þI 1
I 2 ¼ 1 mA

Exercise 4.3: Determine the mesh currents for the circuit in Fig. 4.6.

Answer: I1 ¼ +3 mA, I2 ¼ +2 mA, I3 ¼ +3.5 mA.

IV-152
Chapter 4 Section 4.1: Nodal/Mesh Analysis

Example 4.4: Outline the solution approach for the circuit shown in Fig. 4.6 using the
standard mesh analysis, without the supermesh concept.

Solution: The voltage across the current source is introduced as an extra unknown, Vx. Then,
we write three KVL equations for three meshes in Fig. 4.6, which will contain four unknowns:
I1, I2, I3, and Vx. An extra equation is needed, which is KCL applied to the central branch:
I1
I2 ¼ 1 mA. As a result, we need to solve a system of four simultaneous equations. This is
considerably more work than in the previous case. This is why the supermesh approach is such
a useful tool for the mesh analysis.

IV-153
Chapter 4 Circuit Analysis and Power Transfer

Section 4.2 Generator Theorems and Their Use

4.2.1 Equivalence of Active One-Port Networks: Method of Short/Open
Circuit
In Chap. 3, we considered passive linear networks involving only resistors, which we have
transformed into equivalent circuits. Active linear networks, which include sources and resistors
simultaneously, can be subjected to similar transformations. We know that two electric single-port
networks are equivalent when their terminal υ-i characteristics are identical. For passive resistive
networks studied in Chap. 3, we connected arbitrary source(s) across the network terminals and
checked the resulting υ-i characteristics. For active networks with sources and resistances, we can
use the same method. Alternatively, we can connect arbitrary resistance(s) across the network
terminals and check either the resulting voltage or current. A test resistance to be connected will be
denoted here by R. If for two networks the voltages across the resistance R (or currents through it)
are identical for all values of R, the networks are equivalents.

Method of Short/Open Circuit

However, testing all possible values of resistance R connected to terminals a and b of a network in
Fig. 4.7 is not necessary. Note that an active linear network may ultimately have only two
elements: a source and a resistance. To uniquely determine the two elements (their values), only
two equations are necessary. It is therefore customary to check only two (limiting) values of the
test resistance:
R ! 1 and R ¼ 0 ð4:13Þ

test of open-circuit voltage test of short-circuit current

a a
+

arbitrary electric VOC arbitrary electric ISC

network - network
b b

Fig. 4.7. Method of short/open circuit for an active, one-port network.

IV-154
Chapter 4 Section 4.2: Generator Theorems and Their Use

Conveniently, this corresponds to open- and short-circuit conditions. In the first case, the voltage
between terminals a and b is the open-circuit network voltage VOC. In the second case, the current
flowing from terminal a to terminal b is the short-circuit network current ISC. The pair VOC, ISC is the
key for the method of short/open circuit. This method states that two active linear circuits are
equivalent when their VOC and ISC coincide. Network equivalency relates not only to the linear active
networks with two components, but, as will be shown soon, it is valid for all active linear networks.

4.2.2 Application Example: Reading and Using Data for Solar Panels
The method of short/open circuit is also very useful for active nonlinear networks, including
nonlinear sources. An example is a solar cell or a combination thereof, a solar panel. Every solar
panel has the measured data for VOC and ISC listed on its backside. The short-circuit current is
simultaneously the photocurrent of the solar cell. Table 4.1 collects this data for common
crystalline silicon (or c-Si) solar panels. It is organized in such way that VOC is given per one
cell in the panel and ISC is given in terms of photocurrent density, JP, per unit cell area. The cells in
the panel are connected in series.

Table 4.1. Manufacturers’ specified parameters for different c-Si solar panels from five different
manufacturers (1–230 W output power range, circa 2012–2015). The cell area is either measured
directly or extracted from the datasheet.
Cell area JP ¼ ISC/A
Solar panel Cells, N VOC/N, V A, cm2 A/cm2
1 W BSPI-12 power up c-Si panel 36 0.59 2.36 0.030
10 W BSP-1012 power up c-Si panel 36 0.59 ~22.0 0.030
65 W BSP-1012 power up c-Si panel 36 0.61 121.7 0.032
230 W sharp ND-U230C1 c-Si panel 60 0.62 241.0 0.034
175 W BP solar SX3175 c-Si panel 72 0.61 156.25 0.033
6 W global solar GSE-6 c-Si panel 44 0.52 16.6 0.027
200 W GE Energy GEPVp-200 54 0.61 249.3 0.032
c-Si panel
Average 0.593 0.0311

Table 4.1 demonstrates that c-Si solar cells have approximately the same open-circuit voltage
of 0.6 V per cell. The open-circuit voltage does not depend on the area of the cell. The short-circuit
photocurrent density is also approximately the same for c-Si solar cells from different manufac-
turers. On average, it is given by JP ¼ 0.03 A/cm2. These values corresponds to an incident light

intensity of 1000 W/m2 at T ¼ 25 C. The photocurrent density does not depend on the area of the
cell. Obviously, the total photocurrent does.

Exercise 4.4: A c-Si solar panel (or solar module) has the open-circuit voltage of 23.4 V? How
many individual solar cells does it have?

Answer: Approximately 39.

IV-155
Chapter 4 Circuit Analysis and Power Transfer

Exercise 4.5: A c-Si solar panel is needed with the open-circuit voltage of 12 V and a short-
circuit current of 3 A. Design the panel: give the number of cells to be connected in series and
the required unit cell area.

Answer: Twenty cells with the area of 100 cm2 (10  10 cm) each.

4.2.3 Source Transformation Theorem

The most fundamental transformation of active linear networks is the source transformation
theorem. The source transformation theorem involves the substitution of an independent voltage
source VT in series with resistance RT for an independent current source IN with resistance RN and
vice versa, as seen in Fig. 4.8a, b. The meaning of indexes N and T will become apparent soon. The
identical theorem applies to the dependent sources shown in Fig. 4.8c, d.

a) b)
RT
a a

VT + IN RN
-
b b

c) d)
RT
a a

vT + iN RN
-

b b

Fig. 4.8. Transformation of dependent and independent sources.

Let us prove this theorem by establishing the circuit equivalence. The pair VOC, ISC is to be
found for every network. For the two networks in Fig. 4.8a, b, we have

VT
V OC ¼ V T , I SC ¼ ð4:14aÞ
RT
V OC ¼ RN I N , I SC ¼ IN ð4:14bÞ

Equations (4.14) have a unique solution in the form of the source transformation theorem
VT
RN ¼ RT , I N ¼ ð4:15Þ
RT

IV-156
Chapter 4 Section 4.2: Generator Theorems and Their Use

If Eqs. (4.15) are satisfied, both networks in Fig. 4.8a, b have equal VOC and ISC. This ensures
that their entire υ-i characteristics are also the same. To confirm this fact, an arbitrary resistance
R could be connected across the port. The resulting voltages is found directly by solving
the voltage divider and the current divider circuits, respectively. Both voltages are equal to
VTR/(R+RT). Thus, the source transformation theorem is proved.

Exercise 4.6: A network has a 10 V voltage source in series with a 20 Ω resistance. It is

replaced by a current source IN in shunt with resistance RN. Find IN and RN.
Answer: IN ¼ 0.5 A, RN ¼ 20 Ω.

Exercise 4.7: A linear active circuit records an open-circuit voltage of 5 V and a short-circuit
current of 1 mA. Determine its equivalents in the form of a voltage source in series with a
resistance and in the form of a current source in shunt with a resistance.
Answer: VT ¼ 5 V, RT ¼ 5 kΩ and IN ¼ 1 mA, RN ¼ 5 kΩ.

Often, the source transformation theorem allows us to simplify the circuit analysis through
simple network manipulations.

Example 4.5: Find current I1 in the circuit shown in Fig. 4.9a.

Solution: The circuit can be solved using the superposition theorem. Another way is to employ
the source transformation theorem. The corresponding steps are outlined in Fig. 4.9b, c. We use
the source transformation three times and end up with the parallel combination of two current
sources and three resistances. The three resistances in parallel are equivalent to the 0.75 kΩ
resistance; the voltage across every element in parallel is then 0.75 kΩ  2 mA ¼ 1.5 V.
Therefore, I1 ¼ 0.75 mA.

Example 4.6: The circuit in Fig. 4.10a includes a current-controlled voltage source with the
strength of 4000ix [V]. Find current ix by using source transformation.

Solution: The corresponding circuit transformation is shown in Fig. 4.10b. The circuit with the
current-controlled current source in Fig. 4.10b is solved using KCL and KVL. KCL applied to
the bottom node states that the current of 3 mA+3ix flows through the rightmost 1 kΩ resistance
(directed down). Since, by KVL, the voltages across both resistances must be equal, one has

IV-157
Chapter 4 Circuit Analysis and Power Transfer

Example 4.6 (cont.):

3 mA þ 3ix ¼ ix ) ix ¼
1:5 mA ð4:16Þ

Alternatively, one could convert the independent current source to an independent voltage
source. However, this method would hide ix.

a) 2.5 kW 2 kW

I1
0.5 kW 6 mA 2 kW + 2V
-

b)
0.5 kW 2.5 kW

I1
3V + 2 kW 2 kW 1 mA
-

c)
I1
1 mA 3 kW 2 kW 2 kW 1 mA

Fig. 4.9. Circuit modifications using the source transformation theorem.

a)
ix 1 kW
3 mA 1 kW + 4000ix
-

b)
ix

3 mA 1 kW 1 kW 4ix

Fig. 4.10. Using source transformation for a circuit with dependent sources.

IV-158
Chapter 4 Section 4.2: Generator Theorems and Their Use

4.2.4 Thévenin’s and Norton’s Theorems: Proof Without Dependent

Sources
The origin of Thévenin’s theorem is due to Léon Charles Thévenin, a French engineer (1857–1926).
The theorem is illustrated in Fig. 4.11a, b and can be expressed in the following form:
1. Any linear network with independent voltage and current sources, dependent linear sources, and
resistances as shown in Fig. 4.11a, can be replaced by a simple equivalent network: a voltage
source VT in series with a resistance RT.
2. The equivalent network in Fig. 4.11b is called the Thévenin equivalent.
3. Voltage VT is the open-circuit voltage VOC of the original network (with no current flowing into
either terminal).
4. When dependent sources are absent, Thévenin resistance RT is the equivalent resistance Req of
the original network with all independent sources turned off (voltage sources are replaced by
short circuits, current sources by open circuits).
5. When both dependent and independent sources are present, the independent sources are not
turned off. Resistance RT is given by RT ¼ VOC/ISC, where ISC is the short-circuit current of the
original network.
6. When only the dependent sources are present, a current (or voltage) source is connected to the
network terminals. Thévenin or equivalent resistance is given by RT ¼ V/I where I is the source
current and V is the voltage across the source. Thévenin voltage is strictly speaking not defined in
this case.
The Norton’s theorem is dual to the Thévenin’s theorem. It was named in honor of Edward
L. Norton (1898–1983), an engineer at Bell Labs in New Jersey.2 Norton’s theorem is illustrated in
Fig. 4.11c, d. The equivalent circuit (Norton’s equivalent) is now the current source in shunt with the
resistance as shown in Fig. 4.11d. Since the equivalence of both networks in Fig. 4.11b, d has
already been established, the Norton’s theorem will follow from the Thévenin’s theorem and vice
versa.

a) b)

a a
any linear network of RT
Thevenin
sources (independent VT +
and linear dependent) -
and resistances
b b
c) d)

a a
any linear network of
sources (independent Norton
and linear dependent) IN RN
and resistances
b b

Fig. 4.11. Thévenin’s and Norton’s theorems: replacing linear active circuits by its Thévenin and
Norton equivalents.

2
The first publication that discusses this equivalent circuit concept is actually due to Hans F. Mayer (1895–1980) who
made the discovery in 1926 while a researcher at Siemens Company.

IV-159
Chapter 4 Circuit Analysis and Power Transfer

Proof of Thévenin’s Theorem for Active Networks Without Dependent Sources

The proof is based on the assumption of circuit linearity. The υ-i characteristics of both networks is
established using a current source of strength I connected as seen in Fig. 4.12.

a) b)
a a
any linear network of RT

+
sources (independent VT +
I V I V
and linear dependent)
- - -
and resistances
b b

Fig. 4.12. Derivation of Thévenin’s theorem by establishing the υ-i characteristics.

Since the entire circuit is assumed linear, the υ-i characteristic of the current source in Fig. 4.12a
must have the form of linear function,
V ¼ AI þ B ð4:17Þ

where V is the voltage across the current source. A and B are some “constant” coefficients, which
do not depend on I, but do depend on the network parameters. Our goal is to find A and B,
respectively. First, we check the value I ¼ 0 when the external current source is turned off, i.e.,
replaced by an open circuit. From Eq. (4.17), voltage V equals B. Therefore, it is equals VOC of the
original network. Therefore,
B ¼ V OC ð4:18Þ

Next, let us turn off all internal sources. The network becomes an equivalent resistance Req. The
constant B (its open-circuit voltage) is zero. Equation (4.17) therefore yields V ¼ AI for any value
of I. On the other hand, for the current source I connected to the resistance Req, it must be V ¼ ReqI.
Comparing the two expressions we obtain,
A ¼ Req ð4:19Þ

The simpler network in Fig. 4.12b is also described by the υ-i characteristic in the form of
Eq. (4.17). In this case, B ¼ VOC ¼ VT, A ¼ Req ¼ RT. We finally compare the two υ-i
characteristics,
V ¼ Req I þ V OC for linear active network; V ¼ RT I þ V T for The venin equivalent
ð4:20Þ

and establish the Thévenin’s theorem. A test voltage source could be used in place of the current
source in Fig. 4.12, with the same result obtained. The physical background of Thévenin’s
theorem is thus the fact that the terminal response of any linear network has a linear υ-i
characteristic—a linear function with only two independent coefficients: A and B. A simpler
network with exactly two independent parameters, VT and RT, allows us to model this response.

IV-160
Chapter 4 Section 4.2: Generator Theorems and Their Use

Equivalence of Arbitrary Linear Networks with Identical VOC, ISC

On the one hand, linear active networks with only two elements (a source and a resistance) are
equivalent when their VOC, ISC are identical. On the other hand, any active linear network is
equivalent to a linear network with only two elements. Therefore, we conclude that two arbitrary
linear networks are equivalent when their VOC and ISC coincide.

4.2.5 Finding Thévenin and Norton Equivalents and Using Them

for Circuit Solution
Along with the superposition and source transformation theorems, Thévenin’s and Norton’s
theorems (or their outcome, Thévenin and Norton equivalent circuits or simply equivalents) are
powerful tools to solve more complicated electric circuits. The idea is to replace a larger part of the
circuit by its equivalent and, subsequently, find the required voltage/current/power for a remaining
small circuit part. However, a bit of practice is required to implement such a solution. We will first
establish Thévenin (and implicitly Norton) equivalents for some “large” circuits—one-port
networks—shown in Figs. 4.13 and 4.14, and then implement this idea using a few examples.

Example 4.7: Establish Thévenin equivalent circuits in the form of Fig. 4.11b for the one-port
networks shown in Fig. 4.13. This first network is a battery bank or a network consisting of series-
connected practical voltage sources, with RB1 ¼ RB2 ¼ RB3 ¼ 1 Ω, VB1 ¼ VB2 ¼ VB3 ¼ 9 V.
Solution:
Case (a): The Thévenin voltage is the open-circuit voltage between terminals a and b.
Since no current flows through any of the resistors, KVL gives VT ¼ Vab ¼ 27 V. The Thévenin
resistance is found by turning off the voltage sources, i.e., replacing them by short circuits.
Therefore, RT ¼ Rab ¼ 3 Ω.
Case (b): The Thévenin voltage is the open-circuit voltage between terminals a and b.
Current cannot flow through open terminals a and b, but it still can flow within the loop
containing the source and the two resistors. Therefore, VT ¼ Vab ¼ V3kΩ ¼ 9 V. The Thévenin
resistance is found by turning off the voltage source (replacing it by a wire). Since the two
resistors are combined in parallel, RT ¼ Rab ¼ 1.2 kΩ.
Case (c): Although the source transformation theorem immediately gives us VT ¼ 3 V,
RT ¼ 3 kΩ, we would like to solve the problem directly. The Thévenin voltage is the open-
circuit voltage between terminals a and b. Current cannot flow through open terminals a and b,
but it still can flow within the loop containing the current source and the resistor. Therefore,
VT ¼ Vab ¼ V3kΩ ¼ 3 V. The Thévenin resistance is found by turning off the current source
(replacing it by an open circuit). This gives RT ¼ Rab ¼ 3 kΩ.

IV-161
Chapter 4 Circuit Analysis and Power Transfer

Example 4.7 (cont.):

Case (d): The Thévenin voltage is the open-circuit voltage between terminals a and b.
Current cannot flow through open terminals a and b and through a 0.2 kΩ resistor, but it still
can flow within the loop containing the voltage source and both 2 kΩ and 3 kΩ resistors.
Therefore, VT ¼ Vab ¼ V3kΩ ¼ 9 V. The Thévenin resistance is found by turning off the voltage
source (replacing it by a short circuit). This gives us RT ¼ Rab ¼ 1.2 kΩ + 0.2 kΩ ¼ 1.4 kΩ.
Case (e): First, note that the 1 kΩ resistor placed in shunt with the voltage source has no
effect on the solution whatsoever; it is either shorted out by a wire (when finding Thévenin
resistance) or simply possesses the supply voltage (when finding Thévenin voltage). In
practice, such a combination only dissipates power and should be avoided when possible.
The rest of the solution follows case (b) with interchanged resistors and yields
VT ¼ Vab ¼ V2kΩ ¼ 6 V, RT ¼ Rab ¼ 1.2 kΩ.

Fig. 4.13. Active linear networks to be converted to their Thévenin equivalents as given in the form of
Fig. 4.11b.

Exercise 4.8: Along with the combination of a voltage source in parallel with a resistance as in
Fig. 4.13e, give another example of a “meaningless” circuit configuration.
Answer: A current source in series with a resistance.

IV-162
Chapter 4 Section 4.2: Generator Theorems and Their Use

Example 4.8: A two-terminal network shown in Fig. 4.14a is a two-bit R-2R ladder network
used for digital-to-analog conversion. Express
(i) Thévenin (or equivalent) voltage VT
(ii) Thévenin (or equivalent) resistance RT
in terms of (digital) voltages D0,D1, and resistance R.
Solution: One way to solve the problem is to find VT and RT directly from the circuit in
Fig. 4.14a. While the solution for RT is straightforward, finding VT requires more work. Yet
another method is to apply Thévenin equivalent to the leftmost section of the ladder network
first. The result is the circuit shown in Fig. 4.14b. The final Thévenin equivalent has the form
RT ¼ R, V T ¼ D21 þ D40 .
This method may be applied to ladder networks with multiple sections.

Exercise 4.9: Repeat the previous example for the ladder shown in Fig. 4.14c.

Answer: RT ¼ R, V T ¼ D22 þ D41 þ D80 .

a) R
a
2R
2R 2R
c) R R
D0 + D1 + a
- -
2R
b 2R 2R 2R
b) R
a D0 + D1 + D2 +
- - -
R 2R b

D0/2 + D1 +
- -
b

Fig. 4.14. Two-bit and three-bit R-2R ladder networks.

Using Thévenin and Norton Equivalents for Circuit Solution

In this section, we will use the method of Thévenin equivalent and conduct a circuit solution. Note
that, in contrast to a nodal analysis, this is a semi-analytical method which may not always work
out for every network. Consider two circuits shown in Fig. 4.15a, b. In either case, the goal is to
find the current I. We will implement the idea discussed above, i.e., replace a larger part (or parts)
of the circuit by its equivalent and then solve for the current in the remaining part.

IV-163
Chapter 4 Circuit Analysis and Power Transfer

Example 4.9: For the two circuits shown in Fig. 4.15a, b, find the current I.

Solution:
Case (a): One way to solve the problem is to apply the current division principle
multiple times. Another method is to replace the circuit to the left of the dashed line (also
called “reference plane” in more advanced applications) by its Thévenin equivalent as shown in
the figure. The remaining solution becomes trivial and yields a current value of 0.25 mA.
Case (b): One way to solve the problem is to apply the superposition theorem with or
without source transformation and then use the current division again. Yet another method is to
replace the circuits to the left/right of two dashed lines (two “reference planes”) by its Thévenin
equivalents as shown in the same figure. The remaining solution is straightforward; it yields the
current value of
3.214 mA.

Fig. 4.15. Two circuits solved using the method of Thévenin equivalent.

Thévenin and Norton Equivalents with Dependent Sources

Thévenin and Norton equivalents with dependent sources (amplifiers) are a fascinating subject
that can lead to unexpected concepts such as the negative equivalent resistance.

IV-164
Chapter 4 Section 4.2: Generator Theorems and Their Use

Example 4.10: Find Thévenin and Norton equivalent circuits for the network in Fig. 4.16a.

Solution: The network includes a voltage-controlled voltage source. Therefore, its analysis
should be performed in a general form, by finding the pair VOC, ISC. The short-circuit current
ISC is found straightforwardly. Since the rightmost resistance is shorted out, υx ¼ 0, and
ISC ¼ 10 mA. To find the open-circuit voltage, which is equal to υx, we use the source
transformation theorem and arrive at the equivalent circuit in Fig. 4.16b. Next, we solve this
circuit. Employing KVL, the voltage across the leftmost resistance is equal to 10 V
4υx. By
KCL, the currents through both resistances must be the same. Since the resistors are equal, we
obtain the equality 10 V
4υx ¼ υx so that υx ¼ 2 V. The open-circuit voltage has the same
value. The Thévenin and Norton equivalents are
V OC
V T ¼ V OC ¼ 2 V, RT ¼ ¼ 200 Ω ð4:21aÞ
I SC
V OC
I N ¼ I SC ¼ 10 mA, RN ¼ ¼ 200 Ω ð4:21bÞ
I SC

a)
3vx
- a
+

+
1 kW 10 mA 1 kW vx
-
b
b)
3vx
1 kW
- a
+

+
+ 10 V 1 kW vx
-
-
b

Fig. 4.16. A network with dependent sources to be converted to its equivalent forms.

4.2.6 Application Example: Generating Negative Equivalent Resistance

Consider a circuit shown in Fig. 4.17a. It includes only a dependent source: the voltage-controlled
voltage source; it does not have independent sources. Therefore, the circuit analysis has to be done

IV-165
Chapter 4 Circuit Analysis and Power Transfer

by connecting a current (or voltage) source between terminals a and b as shown in Fig. 4.16b. The
quantity of interest is the Thévenin (or equivalent) resistance. The KVL for the circuit in
Fig. 4.17b gives
Aυx
RI+υx ¼ 0. Therefore, by definition,
υx R
RT  ¼ ð4:22Þ
I 1
A

As long as the open-circuit voltage gain of the dependent source, A, is greater than one, Eq. (4.22)
states the negative equivalent or Thévenin resistance. Physically, this means that the Thévenin
equivalent circuit is delivering power instead of absorbing it.

Construction and the Use of Negative Equivalent Resistance

A circuit block, which is equivalent to the negative resistance, may be constructed using the
operational amplifier studied in the next chapter. This block may be used for different purposes
including signal generation. The difference between the negative resistance and the power source
is that the negative resistance may supply power of any type (DC, AC, or an arbitrary waveform),
i.e., support the self-oscillating circuits. Figure 4.17c summarizes Thévenin resistances generated
by their basic networks with only dependent sources. The same method of analysis (simultaneous
use of KCL and KVL) has been applied to every network. Although all of the networks in
principle generate the negative equivalent resistance values, the realization of some particular
circuits may be difficult.

a) b)
a a
R R
+ + + +
Avx + vx Avx + I vx
- -
- - - -
b b
c) RT=R/(1-A) RT=R/(1-GR)

+ R + + +
Avx + vx Gvx R vx
-
- - - -

RT=R-Z RT=R(1-A)
ix ix

+ R +
Zix + Aix R
-
- -

Fig. 4.17. Thévenin equivalent resistances for basic networks with dependent sources.

IV-166
Chapter 4 Section 4.2: Generator Theorems and Their Use

4.2.7 Short Summary of Circuit Analysis Methods

In summary, a linear circuit can be solved using any of the methods studied in this and in the
previous chapters:
- Superposition theorem
- Nodal/mesh analysis
- Source transformation theorem
- Method of Thévenin and/or Norton equivalent circuit(s)

or a combination of those. While the nodal/mesh analysis is always applicable, other methods may
turn out to be even more useful since they often provide a physical insight into the circuit behavior.

Exercise 4.10: How could you find the open-circuit voltage Vab in Fig. 4.14a?

Answer:
A. When the superposition theorem is applied, shorting out D0 gives Vab ¼ D1/2. Shorting
out D1 gives Vab ¼ D0/4.
B. When the nodal analysis is applied, we ground negative terminals of both sources and
find the unknown voltage of the upper left node via KCL. Only one equation needs to be
solved. This is perhaps the simplest solution method.
C. The source transformation theorem cannot easily be applied.
D. The method of Thévenin equivalent circuits is described in example 4.8.

Exercise 4.11: How could you find current I for the circuit in Fig. 4.15b?

Answer:
A. When the superposition theorem is applied, shorting out the 30 V source gives
I1 ¼ 3.214 mA. Shorting out the 15 V source gives I2 ¼
2I1(due to symmetry!).
The resulting current is I ¼ I1+I2 ¼
3.214 mA.
B. When the nodal analysis is applied, we ground the entire bottom node and combine both
0.2 kΩ resistors in series. There are two nontrivial nodes on top and two nodal
equations. After the solution is obtained (V1 ¼ 12.86 V for the left node 1 and
V2 ¼ 14.14 V for the right node 2), current I is found as the difference of two nodal
voltages divided by 0.4 kΩ.
C. The source transformation theorem followed by superposition could be applied too.
D. The method of Thévenin equivalent circuits is described in example 4.9.

IV-167
Chapter 4 Circuit Analysis and Power Transfer

Section 4.3 Power Transfer

4.3.1 Maximum Power Transfer
The principle of maximum power transfer from a source to a load will now be quantified. This
principle is also known as a maximum power theorem. The circuit under study is shown in
Fig. 4.18. It involves an arbitrary linear source (a battery, a generator, etc.), which is represented
by its Thévenin equivalent, and a load, which is characterized by its equivalent resistance RL. All
other load parameters (dynamic, mechanical, and thermal) are implicitly included into the load’s
resistance.

- + RT
VT + RL
-

Fig. 4.18. A battery (or another practical voltage source) connected to a load.

The key question you have to ask yourself is this: For a given ideal voltage source VT and a
given internal resistance RT, can the electric power delivered to the load be maximized, and at
which value of RL does the maximum occur? The answer is found by solving the circuit in
Fig. 4.18. First, the current is determined from the given voltage source VT and the total resistance
using the series equivalent,
VT
I¼ ð4:23Þ
RT þ RL

This allows us to compute the power at the load based on

RL V 2T
PL ¼ RL I 2 ¼ ð4:24Þ
ðRT þ RL Þ2

When VT and RT are fixed, the magnitude of the load resistance determines the delivered
power PL. This power tends to zero when either RL ! 0 or RL ! 1; moreover, it is always

IV-168
Chapter 4 Section 4.3: Power Transfer

positive. Therefore, according to Rolle’s theorem of calculus, the power must have a maximum at
a certain value of RL. For example, Fig. 4.19 shows a plot of the load power as a function of RL for
VT ¼ 9 V and RT ¼ 5 Ω.

Load power, W 3

0
0 10 20 30 40 50
Load resistance, W

Fig. 4.19. Load power as a function of the load resistance for fixed VT ¼ 9 V, RT ¼ 5 Ω.

We will find the maximum of the load power analytically. We treat PL in Eq. (4.24) as a function
of RL, i.e., PL ¼ PL(RL). It is known that a function has a maximum when its first derivative is set
to zero. Consequently, differentiating PLwith respect to RL gives
" # " #
dPL 1 RL R T
RL
¼ V 2T
2 ¼ V 2T ¼0 ð4:25Þ
dRL ðRT þ RL Þ2 ðRT þ RL Þ3 ðRT þ RL Þ3

The necessary and sufficient condition for Eq. (4.25) to hold is

RL ¼ RT ) PL ¼ 0:25V 2T =RT ð4:26Þ

This result is of great practical value despite, or maybe because of, its simplicity. The maximum
output power is achieved when the load resistance is equal to the internal resistance of the power
source. In other words, the load is matched to the source; it is called the matched load. In power
engineering and in RF and microwave engineering, the problem of load matching is very
important. However, it must be clearly stated that no more than 50% of the total circuit power
can be extracted even in the best case. This statement makes sense if we again examine the circuit
in Fig. 4.18 and assume two equal resistances. The power is divided equally; half of the total
power is spent to heat up the power source. The power maximum in Fig. 4.19 is relatively flat over
the domain RL > RT; however, the power drops sharply when RL < RT. This last condition should be
avoided if at all possible.

IV-169
Chapter 4 Circuit Analysis and Power Transfer

Example 4.11: An audio amplifier produces an rms output of 20 V. The amplifier’s output
resistance is rated at 4 Ω. You are given four 4 Ω speakers. How should you connect the
speakers for the maximum acoustic power—in series, parallel, or a single speaker only?

Solution: The rms voltage simply means the equivalent DC voltage that provides the same
power to the load as the average power of the primary AC voltage. Hence, the sophisticated AC
audio amplifier circuit is essentially replaced by its DC Thévenin equivalent with VT ¼ 20 V
and RT ¼ 4 Ω. Similarly, the dynamic speakers are replaced by a DC load with RL ¼ 16 Ω if
connected in series combination, RL ¼ 1 Ω if connected in parallel, or with RL ¼ 4 Ω if only a
single speaker is employed. The output (audio) powers are as follows:
1  400
PL ¼ ¼ 16 W four speakers in parallel ð4:27aÞ
ð4 þ 1Þ2
4  400
PL ¼ ¼ 25 W single speaker ð4:27bÞ
ð4 þ 4Þ2
16  400
PL ¼ ¼ 16 W four speakers in series ð4:27cÞ
ð4 þ 16Þ2

The best (loudest) choice would be surprisingly one single speaker.

4.3.2 Maximum Power Efficiency

A power analysis would be incomplete without discussing the efficiency of the power transfer.
Consider an electric boat driven by a marine battery. The optimization of the battery-motor system
for maximum power transfer implies that we will move fast, but perhaps not very far. Another
optimization is possible for maximum power efficiency. In this case, we could tolerate a smaller
speed in order to travel a longer distance. The circuit to be analyzed is again the circuit in
Fig. 4.18. The useful power delivered to the load is given by Eq. (4.24). The total power delivered
by the source is

V 2T
P ¼ ðRT þ RL ÞI 2 ¼ ð4:28Þ
RT þ RL

The power efficiency E is defined as the ratio of the useful power to the total power,
PL RL
E¼ ¼ ðRT þ RL ÞI 2 ¼ ð4:29Þ
P RT þ RL

Thus, the power efficiency is a simple function of the load resistance and the source resistance.
It does not depend on the source voltage. The efficiency is zero when the load resistance is zero.
It monotonically increases and approaches maximum (the maximum value is unity, which

IV-170
Chapter 4 Section 4.3: Power Transfer

corresponds to an efficiency of 100%) when the load resistance becomes large enough when
compared to the source resistance. The maximum power efficiency is thus achieved when RL > > RT.
For example, Fig. 4.20 augments the load power graph from Fig. 4.19 with the corresponding

4 power efficiency 80

Power efficiency, %
Load power, W
3 60

2 load power 40

1 20

0 0
0 10 20 30 40 50
Load resistance, W

Fig. 4.20. Load power and power efficiency for fixed VT ¼ 9V, RT ¼ 5 Ω.

efficiency curve.

Example 4.12: A battery with the stored energy of EB ¼ 0.1 MJ, and VT ¼ 12 V, RT ¼ 5 Ω,
delivers its entire energy during the time period 0  t  T and discharges with a constant output
voltage/current. Two loads are used: RL ¼ 5 Ω and RL ¼ 50 Ω. Determine discharge time T and
total energy delivered to the load in each case.

Solution: The discharge time, T ¼ EB/P, is determined first when the total power P follows
based on Eq. (4.28). Assuming constant battery discharge rate, we obtain
T  1:9 hours for the 5 Ω load; T  10:6 hours for the 50 Ω load ð4:30Þ

The total energy delivered to the load, E ¼ TPL, in each case is given by
E 5Ω ¼ 50 kJ for the 5 Ω load; E 50Ω ¼ 91 kJ for the 50 Ω load ð4:31Þ

Thus, the total energy extracted from the battery is nearly twice as high in the second case.
However, it takes about five times longer to extract this energy.

4.3.3 Application Example: Power Radiated by a Transmitting Antenna

A transmitting antenna in a radio handset features a monopole antenna. It is connected to a source
that has the same basic form as in Fig. 4.18 but with an AC generator instead of the DC source and
with an internal (generator) resistance of 50 Ω. The antenna represents a load with a “radiation”
resistance of 50 Ω. This resistance describes the power loss in terms of the electromagnetic

IV-171
Chapter 4 Circuit Analysis and Power Transfer

radiation from the antenna. Thus, the antenna, if properly matched to the power source, will
radiate 50% of the total power as electromagnetic waves into space. Now, an inexperienced
electrical engineer decides to “modify” the handset by cutting the monopole antenna to only one
third of its length. In this case, the antenna’s radiation resistance is reduced to one ninth of its
original value. How does this affect the radiated signal power? To answer this question, we find
the instantaneous load power, which according to Eq. (4.24), is
RL V 2T ðt Þ
PL ðt Þ ¼ ð4:32Þ
ðRT þ RL Þ2

The ratio of the power levels for the two antenna configurations does not depend on time and we
obtain
PL
short 50=9 50 0:0018
¼ 2
= 2
¼ ¼ 0:36 ð4:33Þ
PL
original ð50 þ 50=9Þ ð50 þ 50Þ 0:0050

Thus, for the shorter antenna, we will only achieve about 36% of the radiated power compared
to the original handset. In practice, this estimate becomes even lower due to the appearance of a
very significant antenna capacitance.

4.3.4 Application Example: Maximum Power Extraction from Solar Panel

Every solar panel reports the measured data for VOC and ISC listed on its backside. For linear
circuits, VT ¼ VOC, RT ¼ VOC/ISC. If the solar panel were a linear circuit, the maximum extracted
power would be exactly equal to 0.25VOCICS according to Eq. (4.26). Fortunately, this is not the
case. The maximum extracted power is significantly greater than this value. However, it is still less
than the “best” possible value of VOCICS. To quantify the maximum power output, every solar
panel has another set of measured data, VMP and IMP, also listed on its backside. VMP stands for
maximum power load voltage, and IMP stands for the maximum power load current. The
maximum extracted power is the product VMPIMP, which is always less than VOCICS. The ratio
of two powers
V MP I MP
F¼ <1 ð4:34Þ
V OC I SC

is known as the fill factor of the solar panel (or solar module). We will derive the theoretical value
of the fill factor in the next section. Table 4.2 lists some experimental data for crystalline (c-Si)
solar panels. The experimental fill factor not only accounts for the nonlinear physics of the cell but
it also includes some resistive losses in an individual cell and in the entire solar module. Equation
(4.34) approximates the fill factor of a solar cell as well.

IV-172
Chapter 4 Section 4.3: Power Transfer

Exercise 4.12: A REC SCM220 220 Watt c-Si solar panel has the following readings on its
back plane: a short circuit current of 8.20 A, an open circuit voltage VOC of 36.0 V, a maximum
power voltage of 28.7 V, and a maximum power current of 7.70 A. Estimate the load resistance
required for the maximum power transfer to the load.
Answer: 3.72 Ω.

Table 4.2. Manufacturer provided circuit parameters for 12 different c-Si solar panels from
5 different manufacturers (1 W to 230 W output power range, circa 2012–2015).
Cells VMP/ IMP/
Solar panel (series) VOC ISC F $/Watt (2012)
1 W BSPI-12 power up c-Si panel 36 0.81 0.86 0.70 24.00
10 W BSP-1012 power up c-Si panel 36 0.81 0.88 0.71 8.80
65 W BSP-1012 power up c-Si panel 36 0.80 0.94 0.75 6.35
80 W sharp NE-80EJEA c-Si panel 36 0.80 0.88 0.70 4.29
176 W sharp ND-176U1Y c-Si panel 48 0.80 0.91 0.73 4.68
230 W sharp ND- U230C1 c-Si panel 60 0.82 0.92 0.75 3.51
5 W BP solar SX-305 M c-Si panel 36 0.80 0.90 0.72 15.00
20 W BP solar SX-320 M c-Si panel 36 0.80 0.92 0.74 8.30
175 W BP solar SX3175N c-Si panel 72 0.83 0.92 0.76 4.19
65 W Kyocera KC65T c-Si panel 36 0.80 0.94 0.75 5.22
165 W SolarWorld SW230 c-Si panel 72 0.80 0.90 0.72 4.72
230 W SolarWorld SW230 c-Si panel 60 0.80 0.92 0.74 3.18
Average 0.806 0.908 0.73

Table 4.2 demonstrates that different c-Si solar cells have approximately the same values of
VMP/VOC, IMP/ISC and of the fill factor. Using the photocurrent density estimate and the open-
circuit cell voltage estimate given at the beginning of this section, we can assume the approximate
generic values for c-Si solar cells at normal irradiation conditions:

V MP ¼ 0:8V OC , I MP ¼ 0:9I SC , F ¼ 0:72, V OC ¼ 0:6 VðcellÞ, J P ¼ 0:03 A=cm2

ð4:35Þ

Unfortunately, these values are not exact; they are meant as a convenient tool for engineering
estimates. However, Eq. (4.35) can be used to address an important task: how to identify the
proper panel configuration and its approximate size in order to provide sufficient power for a
given load.

IV-173
Chapter 4 Circuit Analysis and Power Transfer

Example 4.13: A 3 Ω load (for instance, a hot plate in a camp) is rated at 23 V and is to be
powered by a solar panel. A c-Si photovoltaic sheet material is your material of choice. Outline
parameters of a solar module that is capable of powering the load and estimate the overall
module size.
Solution: First, we need to find the required load current. It is given by I ¼ 23 V/3 Ω ¼ 7.67 A.
Thus, the maximum power parameters of the module must be equal to VMP ¼ 23 V and
IMP ¼ 7.67 A. Next, we find the measurable parameters, VOC, ISC, of the module. According to
Eq. (4.35),
V I
V OC ¼ ¼ 28:75 V, I SC ¼ ¼ 8:52 A ð4:36Þ
0:8 0:9

Finally, we find the number of cells N and the area of an individual cell A, assuming a series
combination of individual cells:
V OC I SC
N¼ ¼ round ð47:9Þ ¼ 48, A¼ ¼ 284 cm2 ð4:37Þ
0:6 V 0:03 A=cm2

The overall module (panel) size for closely spaced cells is then 1.36 m2.

Example 4.14: Compare the theoretical design from Example 4.13 with a real solar module
having nearly the same output power (176 W) and nearly the same maximum power voltage
(23 V).

Solution: We choose a Sharp ND-176U1Y, 176 Watt solar panel from Table 4.2 for compar-
ison. Its maximum power voltage is 23.4 V. Table 4.3 lists the parameters of both panels. The
designs agree with the number of cells and with the size of the unit cell. The overall panel size
for closely spaced cells is also quite similar: 1.36 m2 versus 1.32 m2.

Table 4.3. Parameters of a theoretically designed 176 W solar panel versus the corresponding
176 W hardware prototype.
Example 4.3 (theory estimates) 176 Watt Sharp ND-176U1Y panel
VMP ¼ 23.0 V, PL ¼ 176 W VMP ¼ 23.4 V, PL ¼ 176 W
# of cells: 48 # of cells: 48
Unit cell area: 284 cm2 Unit cell area: 275 cm2

IV-174
Chapter 4 Section 4.3: Power Transfer

Exercise 4.13: A 9.6 W DC motor in an autonomous robot is rated at 17 V and is to be

powered by a solar panel. A c-Si photovoltaic sheet material is your material of choice. Outline
parameters of a solar module that is capable of powering the load and estimate the overall
module size.
Answer: The module should include 36 cells in series, with the area of A ¼ 21.0 cm2 each. The
overall module (panel) size for closely spaced cells is then 0.0756 m2.

IV-175
Chapter 4 Circuit Analysis and Power Transfer

Section 4.4 Analysis of Nonlinear Circuits: Load Line

Method
Consider a nonlinear passive circuit element which possesses a particular υ-i characteristic. It is
shown in Fig. 4.21 by a blue rectangle. Examples of such elements were given in Chap. 2. The
element’s polarity (direction of current inflow for a passive reference configuration) is labeled by a
“+” sign. Figure 4.21 presents the four basic nonlinear circuits (networks) encountered in practice:
a linear active network given by its Thévenin equivalent and connected to a nonlinear load, a
practical nonlinear voltage source connected to a linear load, a linear active network in the form of
the Norton equivalent connected to a nonlinear load, and a practical nonlinear current source
connected to a linear load. Interchanging the place of the nonlinear element and the resistance, if
necessary, and using the source transformation theorem, we can state that all four circuits in
Fig. 4.21 are topologically equivalent. Therefore, only one of them will be studied, for example,
the network shown in Fig. 4.21a. The υ-i characteristic of the practical voltage source between
terminals a and b in Fig. 4.21a is given, according to the KVL, by

VT
V
I¼ ð4:38Þ
RT

a) nonlinear load c) nonlinear source

RT I v-i I
a a
+ +
+

-
+

V
VT + V v-i VT + RL
- - -

b b

b) nonlinear load I d) nonlinear source I

a a
+

+
+

IN RN V v-i IN v-i V RL
- -
b b

Fig. 4.21. Four basic nonlinear circuits.

IV-176
Chapter 4 Section 4.4: Analysis of Nonlinear Circuits: Load Line Method

This linear function given by Eq. (4.38) is known as the load line. It is seen in Fig. 4.22; it
intersects the voltage axis at V ¼ VT and the current axis at I ¼ VT/RT. The υ-i characteristic of a
nonlinear element is plotted on the same figure. Both υ-i characteristics must give the identical
values of voltage and current. Thus, the intersection of the load line with the element’s υ-i
characteristic is the circuit solution. This is the essence of the load line method. Though primarily
graphical, the load line method provides important insight into the problem under study.

I, A

Load line
VT
RT
Solution

v-i characteristic

0
0 VT V, volts

Fig. 4.22. Load line method for a nonlinear circuit.

Example 4.15: The circuit in Fig. 4.21a is characterized by VT ¼ 3 V, RT ¼ 1 kΩ. The υ-i
characteristic of the nonlinear element (the ideal Shockley diode) is
  V  
I ¼ 1  10
9 exp 0:0257 V
1 ½A . The goal is to solve the circuit using the load line
method.

Solution: Figure 4.23 plots two dependencies: the load line of Eq. (4.38) and υ-i characteristic
of the ideal diode specified by the present example. Using the visual inspection, the intersection
is evaluated as I  2.6 mA, V  0.4 V. This is the solution for the circuit current and for the
load voltage, respectively. The solution accuracy improves when the scale of the plot is
adjusted. In particular, we usually do not have to extend the voltage axis all the way from
0 V to the supply voltage; only a small interval may be sufficient.

IV-177
Chapter 4 Circuit Analysis and Power Transfer

I, mA
5.0
v-i characteristic (Shockley equation)

Solution

3.0
2.5 Load line

0
0 0.5 1 1.5 2 2.5 3 V, volts

Fig. 4.23. Load line method applied to a nonlinear circuit with an ideal diode.

4.4.1 Iterative Method for Nonlinear Circuits

Assume that the nonlinear circuit element is characterized by an explicit function I ¼ I(V). Then,
according to Eq. (4.38), the load line method is equivalent to the graphical solution of the
transcendental algebraic equation in the form
VT
V
¼ I ðV Þ ) V ¼ V T
RT I ðV Þ ð4:39Þ
RT

An alternative is to solve this equation iteratively, starting with an initial guess V ¼ V0. The
iterative method for nonlinear circuits can be formulated as follows. Two iterative schemes
(explicit and implicit) are formally employed. The first (explicit) scheme follows from the second
Eq. (4.39) and the second (implicit) scheme from the first Eq. (4.39),
 n

1 V T
V
V nþ1
¼ V T
RT I ðV Þ, n ¼ 0, 1, 2, . . . or V
n nþ1
¼I , n ¼ 0, 1, 2, . . .
RT
ð4:40Þ

where I
1 denotes the inverse function of I(V). The first (explicit) scheme is simpler when I(V) is
given. However, only the second scheme is recommended in practice since the first scheme may
not converge for typical nonlinear circuit elements, which model semiconductor devices.

Example 4.16: Solve the previous example using the iterative solution of the transcendental
circuit equation.

Solution: We find the inverse υ-i characteristic of the nonlinear element first. It is good so

IV-178
Chapter 4 Section 4.4: Analysis of Nonlinear Circuits: Load Line Method

Example 4.16 (cont.):


I 3 V
V
V ¼ 0:0257  ln þ 1 ½V , I¼ ð4:41Þ
1  10
9 A 1 kΩ

The iterative scheme has the form (the second method of Eq. (4.40) is used)

3 V
Vn
V nþ1
¼ 0:0257  ln þ 1 , n ¼ 0, 1, 2, . . . ð4:42Þ
1  10
6 V

with the initial guess V0 ¼ 0 V. It converges very fast; the corresponding iterations are V0 ¼ 0 V,
V1 ¼ 0.3833 V, V2 ¼ 0.3798 V, and V3 ¼ 0.3798 V.
Therefore, only a few iterations are usually sufficient. The final result is I ¼ 2.62 mA,
V ¼ 0.380 V, which improves the solution obtained with the load line method—see the
previous example. The initial guess of the iterative solution may vary widely, but it should
not exceed the source voltage.

Application Example: Solving the Circuit for a Generic Solar Cell

Figure 4.24a shows a simplified physical composition of the solar cell in the form of a pn-junction
(a junction of two semiconductor materials), which essentially forms a semiconductor diode.

Sun light
a) b)
anode p n cathode

IP
I j(x) I
+
D ID
=
-
-
+

-R/2 +R/2 V
I I
RL b
RL
a
a b

Fig. 4.24. (a) Simplified physical composition of the solar cell in the form of a pn-junction—a
semiconductor diode and (b) simplified (lossless single-diode) equivalent circuit.

Free charge carries generated by sunlight are separated by a built-in voltage or potential ϕ(x)
within the diode, which is the cause of an equivalent current source: the photocurrent of the solar
cell IP. The photocurrent mostly flows through the load. At the same time, a certain portion of it,
ID, could still flow through the pn-junction diode itself as a forward diode current. Therefore, the

IV-179
Chapter 4 Circuit Analysis and Power Transfer

load current I in Fig. 4.24 is less than the photocurrent. Figure 4.24b shows a simplified equivalent
circuit of a solar cell. This circuit coincides with the nonlinear circuit in Fig. 4.21d and can be
solved in the same way once the υ-i characteristic of the equivalent diode is known. It is often
given by
  
V
I D ¼ I S exp
1 , V T ¼ 0:0257 V ð4:43Þ
nV T

with an effective ideality factor n and an effective saturation current IS of the corresponding diode.
The characteristic equation of the cell is the KCL in Fig. 4.24b,
  
V
I ¼ I P
I S exp
1 ð4:44Þ
nV T

Figure 4.25 plots the I(V) dependence of the characteristic Eq. (4.44). The horizontal straight
asymptote is the photocurrent IP or the short-circuit current ISC. The vertical straight asymptote is
the open-circuit voltage VOC, VOC  VTn ln (IP/IS) when IP/IS >> 1. The area of the shaded
rectangle is the load power; it is clearly maximized at a certain operating point Q, where V ¼ VMP,
I ¼ IMP.

I, A
IP
Q
IMP
Characteristic equation
PL of the cell I = I(V)

0 VMP VOC V, Volts

Fig. 4.25. Finding operating point Q of the solar cell for the maximum power transfer.

Finding maximum power parameters is a straightforward but lengthy procedure. The load power
is found as a function of V; its derivative must be set equal to zero to maximize the load power.
This is essentially the maximum power theorem for nonlinear circuits. The final expressions are
   
V OC nV T
V MP ¼ V OC
nV T ln 1 þ , I MP ¼ I SC 1
 90%of I SC ð4:45Þ
nV T V MP

For n ¼ 1.75 and VOC ¼ 0.6V, VMP  80 % of VOC, IMP  90 % of ISC which is close to the data in
Table 4.2 given that the fill factors for the cell and the panel are approximately the same. Note that
the load resistance is found as RL ¼ VMP/IMP.

IV-180
Chapter 4 Summary

Summary
Circuit analysis techniques: nodal/mesh analysis
Nodal Based on the KCL
analysis and Ohm’s law:

V 1
V S
R1 þ V 1
0
R3
V
þ V 1R
V
2
2
¼0

V 2
V S
R5 þ V 2
0
R4
V
þ V 2R
V
2
1
¼0

Supernode The KCL for the supernode:

V1
VS V1
þ
R1 R3
V2 V2
VS
þ þ ¼0
R4 R5

Plus the KVL:V2 ¼ V1+V0

Mesh Based on the KVL
analysis and Ohm’s law:

R1(I1
I3)+R3I1+R5(I1
I2) ¼ 0

R2(I2
I3)+R5(I2
I1)+R4I2 ¼ 0

VS+R1(I3
I1)+R2(I3
I2) ¼ 0
For meshes 1, 2, and 3.
Supermesh The KVL for the supermesh:

R1 ðI 1
I 3 Þ þ R3 I 1 þ R4 I 2 þ
R 2 ðI 2
I 3 Þ ¼ 0

Plus Eq. for mesh 3

And the KCL: I1
I2 ¼ IS
Circuit analysis techniques: source transformation theorem
Source Substitution of voltage source
transformation VT in series with resistance RT for
theorem current source IN with resistance
RN:

RN ¼ RT, I N ¼ VRTT

VOC ¼ VT, I SC ¼ VRTT

(continued)

IV-181
Chapter 4 Circuit Analysis and Power Transfer

Circuit analysis techniques: Thévenin/Norton theorems/equivalents

Thévenin Any linear network with indepen-
and Norton dent sources, dependent linear
theorems sources, and resistances can be
replaced by a simple equivalent
network in the form:
i. A voltage source VT in series
with resistance RT
ii. A current source IN in parallel
with resistance RN
Summary of major circuit analysis methods (linear circuits)
- Superposition theorem (pervious chapter)
- Nodal/mesh analysis (this chapter)
- Source transformation theorem (this chapter)
- Thévenin and Norton equivalent circuits (this chapter)
Linear networks: Measurements/equivalence
Method – Two arbitrary linear networks are
of short/ equivalent when their VOC and
open ISC coincide
circuit – This method is also used for
nonlinear circuits
Linear networks: maximum power theorem
Maximum – Power delivered to the load is
power theorem maximized when RL ¼ RT
(load – For high-frequency circuits it
matching) also means no “voltage/current
wave reflection” from the load

Linear networks: maximum power efficiency

Power Power transfer efficiency:
efficiency is
maximized E ¼ PPL ¼ RTRþR
L
L
when the load
resistance is
very high (load
bridging)

(continued)

IV-182
Chapter 4 Summary

Linear networks: dependent sources and negative equivalent resistance

Equivalent Case (a): RT ¼ 1
A
R

resistances Case (b): RT ¼ R
Z

of basic Case (c): RT ¼ 1
GR
R
linear networks Case (d): RT ¼ R(1
A)
with dependent Equivalent resistance may become
sources negative

Nonlinear networks: four basic topologies

Basic – Linear voltage source connected
nonlinear to a nonlinear load (diode cir-
circuits cuit)
– Linear current source connected
to a nonlinear load
– Nonlinear voltage source
connected to a linear load;
– Nonlinear current source
connected to a linear load (pho-
tovoltaic circuit)
Nonlinear networks: circuit analysis via load line method
Load line Solution: The load line (υ-i char-
method acteristic of the linear source,
which is I ¼ V TR
V
T
) intersects the
υ-i characteristic of the nonlinear
load, I(V )

Nonlinear networks: iterative solution

Finding the For the circuit in the previous row: Implicit scheme:
intersection  Initial guess V0may be 0 V
n
point VT
V VT
V
I ðV Þ ¼ ) V nþ1 ¼ I
1 ,
iteratively RT RT
n ¼ 0, 1, .. .
Nonlinear networks: finding resistive load for maximum power extraction
Finding Load power is computed as
maximum PL ¼ V  I(V ). Then, it is maxi-
load power mized, which is equivalent to
for the solving equation:
dV ðV Þ ¼ 0
equivalent dPL

model of a For unknown voltage V

solar cell
(continued)

IV-183
Chapter 4 Circuit Analysis and Power Transfer

Some useful facts about power extraction from solar cells/modules

Typical values of open-circuit voltage VOC and Crystalline silicon or c-Si cell:
photocurrent density JP of a c-Si cell VOC  0.6 V; JP  0.03 A/cm2
– Open-circuit module voltage is N times the cell
voltage, N  VOC
– Short-circuit module current ISC is the cell short-
circuit current ISC ¼ AJP where A is cell area

Typical values of maximum power parameters and Fjmodule ¼ VVMP I MP

OC I SC
 Fjcell (for low-loss
fill factor for c-Si cells/modules; the load resistance modules)
must be RL ¼ VMP/IMP VMP  0.8VOC, IMP  0.9ISC, F  0.72
Lossless single-diode model of a solar cell:
IP ¼ AJP—photocurrent (A)
A—cell area (cm2)
VT—thermal volt. (0.0257 V)
n—ideality
factor (1 < n < 2)
I S  I P exp
VnVOCT (A) h
 i
I ¼ I P
I S exp nVV T
1


Maximum power analytical solution for lossless V MP  V OC
nV T ln 1 þ VnVOCT ,
single-diode model of a solar cell

I MP ¼ I SC 1
VnVMPT

IV-184
Chapter 4 Problems

B. Show the current directions for every resis-

Problems tance in the circuit.
4.1 Nodal/Mesh Analysis
4.1.2 Nodal Analysis I 1 kW 1 kW
Problem 4.1. Using the nodal analysis, deter- 1 kW

mine the voltage V1 in the circuit shown in the 5V +

-
following figure.
3 kW 2 kW

Problem 4.4.
A. Using the nodal analysis, determine the
total circuit current I for the circuit shown
in the figure below.
B. Show the current directions for every resis-
tance in the circuit.
Problem 4.2. Using the nodal analysis, deter- 1 kW
mine the voltage V1 or current I in the circuits
shown in the following figure. 1 kW 1 kW

I
+ 10 V 5 kW 1 kW
-

Problem 4.5. Introduce the ground termination,

write the nodal equations, and solve for the node
voltages for the circuit shown in the figure below.
Calculate the current I shown in the figure.

1 kW

1 kW 1 kW

I
1 mA
10 V + 1 kW
-

Problem 4.6. Introduce the ground termination,

write the nodal equations, and solve for the node
voltages for the circuit shown in the figure. Cal-
Problem 4.3. culate the current through the resistance Rx and
A. Using the nodal analysis, determine the show its direction in the figure.
supply current I for the circuit shown in
the figure.

IV-185
Chapter 4 Circuit Analysis and Power Transfer

1W 1 kW

5V 1W 1 kW 1 kW
-
+
I
10 V
+ 5 kW 1 kW
1W Rx=1 W 1A -
1 kW 1 kW

Problem 4.7. For the circuit shown in the figure Problem 4.10. The figure below shows the DC
below equivalent of a residential three-wire system,
(i) Determine the current ix through resis- which operates at 240 V rms (do not confuse it
tance Rx. with the three-phase system, which carries a
(ii) Show its direction on the figure. higher current). Two 120 V rms power supplies
are connected as one dual-polarity power supply,
i.e., in series. The 10 Ω load and the 20 Ω are
1 kW 1 kW
those driven by the two-wire (and one ground)
0V Rx=1 kW
standard wall plug with 120 V rms—the lights, a
10 V +
- TV, etc. The 6 Ω load consumes more power, and
it is driven with 240 V rms using a separate bigger
3 kW 4 kW
wall plug (+/
and neutral-not shown)—the
stove, washer, dryer, etc. Determine the power
Problem 4.8. For the circuit shown in the figure delivered to each load. Hint: Use a calculator or
below software of your choice for the solution of the
A. Write nodal equations and solve for the system of linear equations (MATLAB is
node voltages. Then, find the value of i1. recommended).
B. Could this problem be solved in another
0.1 W
(simpler) way? hot

10 W
+ 120 V 10 W
-
i1 0.1 W neutral
1A 20 W 5 W 2A 6W

+ 120 V 20 W
-
0.1 W

hot
Problem 4.9.
A. Write the nodal equations and solve for the
node voltages for the circuit shown in the 4.1.3 Supernode
figure below. Then, find the value of i1. Problem 4.11. Introduce the ground terminal,
B. Use MATLAB or other software of your write the nodal equations, and solve for the node
choice for the solution of the system of
linear equations; attach the code to the
solution.

IV-186
Chapter 4 Problems

voltages for the circuit shown in the figure. Cal-

culate the current I shown in the figure. 1 kW 2 kW
1 kW

1 kW 1mA -

+
Vx

1 kW 5V 3 kW 2 kW
-

+
15 V +
-
1 kW I 1 kW Problem 4.15. For the circuit shown in the figure
A. Determine the circuit current I using either
the nodal analysis or the mesh analysis.
B. Explain your choice for the selected
Problem 4.12. Introduce the ground terminal, method.
write the nodal equations, and solve for the node
voltages for the circuit shown in the figure. Cal- 1 kW
culate the current I shown in the figure.
1 kW 1 kW

1 kW
I

1 kW 5V 5 kW + 10 V 1 kW
- -
+

10 V + 1 kW I 1 kW
-
Problem 4.16. For the circuit shown in the fig-
ure, determine its equivalent resistance between
5 kW terminals a and b. Hint: connect a power source
and use a mesh current analysis or the nodal
analysis.
4.1.4 Mesh Analysis
4.1.5 Supermesh 1 kW
Problem 4.13. For the circuit shown in the fig-
ure, determine the current i1 a 1 kW 1 kW
A. Using the mesh analysis
B. Using the nodal analysis 1 kW
1 kW
Which method is simpler?

5W b 1 kW 1 kW

2 mA
i1
20 W 5 W 3 mA
Problem 4.17. For the circuit shown in the fig-
ure, determine the current i1 of the 20 V voltage
source.

Problem 4.14. For the circuit shown in the fig-

ure, determine the voltage Vx
A. Using the mesh analysis
B. Using the nodal analysis
Which method appears to be simpler?

IV-187
Chapter 4 Circuit Analysis and Power Transfer

2W 9W 8W Problem 4.20. Given two networks shown in the

i1 figure below:
A. Determine their open-circuit voltage, VOC,
+ 20 V 10 W 4W 0.5 A
- and short-circuit current, ISC, for each
of them.
B. Are the networks equivalent?

Problem 4.18. Determine the voltage across the a)

current source for the circuit shown in the figure a
below using the mesh analysis and the supermesh 3V
1 kW
concept. +
-
1 kW b
b)
2 kW 1 kW a
6 mA
2 mA
10 V + 0.5 kW
-
3 kW 1 kW b

4.2.2 Application Example: Reading and

4.2 Generator Theorems Using Data for Solar Panels
4.2.1 Equivalence of Active One-Port Net- Problem 4.21. The area of a single cell in the
10 W BSP-1012 Power Up c-Si panel is ~22 cm2.
works: Method of Short/Open Circuit
Predict:
Problem 4.19. A linear active network with two
A. Open-circuit voltage of the solar cell, VOC
unknown circuit elements measures VOC ¼ 5 V,
B. Photocurrent density of the cell, JP
ISC ¼ 10 mA.
C. Short-circuit current of the cell, ISC
A. Determine the parameters VT, RT and IN, RN
of two equivalent networks shown in the Compare the above value with the value
figure below. ISC ¼ 0.66 A reported by the manufacturer.
B. Could you identify which exactly network
is it? Problem 4.22. The area of a single cell in the
175 W BP Solar SX3175 c-Si panel is
a) 156.25 cm2. Predict:
a A. Open-circuit voltage of the solar cell, VOC
RT
B. Photocurrent density of the cell, JP
VT + C. Short-circuit current of the cell, ISC
-
b
b)
a

IN RN

IV-188
Chapter 4 Problems

Compare the above value with the value

ISC ¼ 5.1A reported by the manufacturer. What
value should the photocurrent density have in
order to exactly match the short-circuit current
reported by the manufacturer?

Problem 4.23. Are the solar cells in the solar

module connected in parallel or in series? Why is
one particular connection preferred?

Problem 4.24. You are given three c-Si solar

cells shown in the figure below, each of area A/
3. Draw wire connections for a cell bank, which
has the performance equivalent to that of a large
solar cell with the area A. A. Estimate the area of the single solar cell
using the common photocurrent density
value for c-Si solar cells.
A/3 B. Estimate the open-circuit voltage for the
single cell.
a
n
Problem 4.27. A 20 W BSP2012 PV c-Si mod-
p
b ule shown in the figure has 36 unit cells connected
in series, the short-circuit current of 1.30 A and
the open-circuit voltage of 21.7 V.
A/3

c
n
p
d

A/3

e
n
p
f

Problem 4.25. Individual solar cells in the figure

to the previous problem are to be connected into a A. Estimate the area of the single solar cell
standard solar module. Draw the corresponding using the common photocurrent density
wire connections. value for c-Si solar cells.
B. Estimate the open-circuit voltage for the
single cell.
Problem 4.26. A 10 W BSP1012 PV c-Si mod-
ule shown in the figure has 36 unit cells connected
in series, the short-circuit current of 0.66 A and Problem 4.28. A c-Si solar module is needed
the open-circuit voltage of 21.3 V. with the open-circuit voltage of 10 V and short-

IV-189
Chapter 4 Circuit Analysis and Power Transfer

circuit current of 1.0 A. A number of individual 2 kW

solar cells are available, and each has the area of
34 cm2, open-circuit voltage of 0.5 V, and short- +
3 mA vx 1 kW + 2vx
circuit current of 1.0 A. Identify the proper mod- -
ule configuration (number of cells) and estimate -
module’s approximate size.

Problem 4.29. A c-Si solar module is needed Problem 4.33. Repeat the previous problem for
the circuit shown in the figure below.
with the open-circuit voltage of 12 V and short-
circuit current of 3.0 A. A number of individual 2 kW
solar cells are available, and each has the area of
34 cm2, the open-circuit voltage of 0.5 V, and +
short-circuit current of 1.0 A. Identify the proper 2 mA vx 1 kW - 2vx
+
module configuration (number of cells) and esti- -
mate module’s approximate size.
4.2.3 Source Transformation Theorem
Problem 4.30. Find voltage V in the circuit 4.2.4 Thévenin’s and Norton’s Theorems
shown in the figure below using source 4.2.5 Finding Thévenin and Norton Equiva-
transformation. lents and Using them for Circuit Solution
Problem 4.34. Find
1 kW 2 kW
(i) Thévenin (or equivalent) voltage
(ii) Thévenin (or equivalent) resistance
+
+
3 mA 1 kW V 2 kW
- 2V
for the two-terminal network shown in the fig-
-
ure below (three practical voltage sources or
batteries in series) when
RB1 ¼ 0.5 Ω, RB2 ¼ 0.5 Ω, RB3 ¼ 0.5 Ω,
Problem 4.31. The circuit shown in the VB1 ¼ 6 V, VB2 ¼ 6 V,VB3 ¼ 6 V.
figure below includes a current-controlled
voltage source. Find current ix using source a
transformation. RB3
VB3 +
- a
5 kW
RT
ix RB2
VB2 + VT +
3 mA 1 kW + 4000ix - -
-

RB1 b
VB1 +
-
Problem 4.32. The circuit shown in the figure b
below includes a voltage-controlled voltage
source. Find voltage υx using source Problem 4.35. Find
transformation. (i) Thévenin (or equivalent) voltage

IV-190
Chapter 4 Problems

(ii) Thévenin (or equivalent) resistance

for the two-terminal network shown in the fig-
ure below (three practical voltage sources or
batteries in parallel) when
RB1 ¼ 1 Ω, RB2 ¼ 1 Ω, RB3 ¼ 1 Ω,
VB1 ¼ 6 V, VB2 ¼ 6 V, VB3 ¼ 6 V.

Problem 4.38. For the circuits shown in the figure

below, determine their equivalent (Thévenin) cir-
Problem 4.36. Find cuit by finding VT and RT (show units).
(i) Thévenin (or equivalent) voltage
(ii) Thévenin (or equivalent) resistance

for the two-terminal network shown in the fig-

ure below. Assume three 9 V sources and resis-
tances of 1 Ω each.

a
RB1 RB2 RB3
+ VB1 + VB2 + VB3
- - -

b
a

RT
VT +
-

Problem 4.37. For the circuits shown in the figure Problem 4.39. Find
below, determine their equivalent (Thévenin) cir- (i) Thévenin (or equivalent) voltage
cuit by finding VT and RT (show units).

IV-191
Chapter 4 Circuit Analysis and Power Transfer

(ii) Thévenin (or equivalent) resistance Problem 4.41. A two-terminal network shown in
the figure is a starting section of a ladder network
for the two-terminal networks shown in the used for digital-to-analog conversion. Express
figure below. (i) Thévenin (or equivalent) voltage
(ii) Thévenin (or equivalent) resistance
a) 1 kW
a
in terms of (digital) voltage D0and resistance R.
+ 10 V 6 kW a
-
2R
2R
b
3 kW
b) D0 +
2 kW
a
-
b

+ 10 V 1 kW 6 kW
-
Problem 4.42. A two-terminal circuit network in
b the figure is a two-bit ladder network used for
2 kW
1 kW
digital-to-analog conversion. Express parameters
c) of the corresponding Norton equivalent circuit in
a
terms of (digital) voltages D0,D1 and resistance R.
1 mA 6 kW
R
a
b
2R
3 kW 2R 2R
d) 5W 4W
a
D0 + D1 +
- -
+ 48 V 20 W 16 W b
-
b
Problem 4.43. A two-terminal network shown in
the figure is a four-bit ladder network used for
Problem 4.40. Find digital-to-analog conversion. Express
(i) Thévenin voltage (i) Thévenin (or equivalent) voltage
(ii) Thévenin resistance (ii) Thévenin (or equivalent) resistance
for the two-terminal network shown in the fig- in terms of four (digital) voltages D0,D1,D2,D3
ure below when and resistance R.
R1 ¼ R2 ¼ R3 ¼ 1 kΩ, VS1 ¼ VS2 ¼ 10 V.

R2
a

VS1 + R1 R3
- VS2
-
+

IV-192
Chapter 4 Problems

R R R 1 a)
2R + R +
2R 2R 2R 2R Avx + vx
-
- -
D0 + D1 + D2 + D3 +
- - - - b)
2 ix

+ R
Zix +
Problem 4.44. Determine the Norton equivalent -
-
for the circuit shown in the figure. Express your
c)
result in terms of I and R.
+ +
a Gvx R vx
- -
I R I R d) ix

+
b Aix R
-
Problem 4.45. Establish Thévenin and Norton
equivalent circuits for the network shown in the
figure below. Problem 4.48. Each of three identical batteries is
characterized by its Thévenin equivalent circuit
with VT ¼ 9 V and RT ¼ 1 Ω. The batteries are
-2vx - a connected in series. The entire battery bank is
+

+ connected to a 5 Ω load. Find the power delivered

2 kW 0.1 mA 1 kW vx to the load.
-
b Problem 4.49. Each of three identical batteries is
characterized by its Thévenin equivalent circuit
with VT ¼ 9 V and RT ¼ 1 Ω. The batteries are
4.2.5 Application Example: Generating Neg- connected in parallel. The entire battery bank is
ative Equivalent Resistance connected to a 5 Ω load. Find the power delivered
Problem 4.46. Establish the equivalent to the load.
(Thévenin) resistance for the network shown in
the figure below. Carefully examine the sign of Problem 4.50. For two circuits shown in the
the equivalent resistance. figure below, find the current I using the method
of Thévenin equivalent.
a
+ R1 +
Avx + R2 vx
-
- -
b

Problem 4.47. Derive the equivalent (Thévenin)

resistance for the networks shown in the figure
below (confirm Fig. 4.16 of the main text).

IV-193
Chapter 4 Circuit Analysis and Power Transfer

Problem 4.54. Solve the circuits shown in the

figure below and determine the current I (show
units). You can use any of the methods studied in
class:
- Superposition theorem
- Nodal/mesh analysis
- Source transformation theorem
Problem 4.51. For the circuit shown in the figure - Thévenin and Norton equivalents
below, find the current I using the method of
or their combinations.
Thévenin equivalent.
a) I

1 mA 2 kW 1 kW 5 mA
3V
-
+
b) I

Problem 4.52. For the circuit shown in the figure

1 mA 2 kW 1 kW 4 mA
below, find the current I using the method of
3V
Thévenin equivalent. -
+

c) 6 kW 2 kW

6V + 3 kW + 3V
- -

Problem 4.55. Solve the circuits shown in the

figure below and determine unknown voltage Vor
4.2.6 Summary of Circuit Analysis Methods
current I (show units). You can use any of the
Problem 4.53. Solve the circuit shown in the figure
methods studied in class:
below and determine the current I (show units).
You can use any of the methods studied in class: - Superposition theorem
- Nodal/mesh analysis
- Superposition theorem
- Source transformation theorem
- Nodal/mesh analysis
- Thévenin and Norton equivalents
- Source transformation theorem
- Thévenin and Norton equivalents or their combinations.
or their combinations.
IV-194
Chapter 4 Problems

a) 1 kW 2 kW Power delivered to the load has a maximum at

exactly one value of the load resistance. Find that
+ 2 kW value and prove your answer graphically using
+ 4V V 6 mA software of your choice (MATLAB is
-
- recommended).
b) 0.5kW 1.5 kW

I 1W
+ 70V 1 kW + 14V
- - 24V + RL?
0.5kW 1.5 kW -

c) 1W

Problem 4.58. A power supply for an electric

10 V 1W
heater can be modeled by an ideal voltage source
-
+

I of unknown voltage in series with the internal

1W
resistance RT ¼ 4 Ω.
1W 2A
A. Can you still determine when the power
delivered to a load (a heating spiral with
resistance RL) is maximized?
B. Does the answer depend on the source
Problem 4.56. Solve the circuit shown in the
voltage?
figure below and determine the current I (show
units). You can use any of the methods studied in
class: Problem 4.59. For the circuit shown in the fig-
ure, when is the power delivered to the load
- Superposition theorem maximized?
- Nodal/mesh analysis
- Source transformation theorem
- Thévenin and Norton equivalents 1.5 W
12 V + RL ?
or their combinations. -
1.5 W

Problem 4.60. A battery can be modeled by an

ideal voltage source VT in series with a resistance
RT.

4.3 Power Transfer RT

VT + RL?
4.3.1 Maximum Power Transfer -
4.3.2 Maximum Power Efficiency
Problem 4.57. A deep-cycle marine battery is
modeled by an ideal voltage source of 24 V in
series with a 1 Ω resistance shown in the figure
below. The battery is connected to a load, and the
load’s resistance, RL, needs to be optimized.

IV-195
Chapter 4 Circuit Analysis and Power Transfer

(i) For what value of the load resistance RL B. How is the maximum load power different,
is the power delivered to the load compared to the previous problem?
maximized?
(ii) What percentage of the power taken from Problem 4.64. The heating element of an electric
the voltage source VT is actually deliv- cooktop has two resistive elements, R1 ¼ 50 Ω
ered to a load (assuming RL is chosen to and R2 ¼ 100 Ω, that can be operated separately,
maximize the power delivered)? in series, or in parallel from a certain voltage
(iii) What percentage of the power taken from source that has a Thévenin (rms) voltage of
the voltage source VT is delivered to a 120 V and internal (Thévenin) resistance of
load when RL ¼ 0.1RT? 30 Ω. For the highest power output, how should
the elements be operated? Select and explain one
Problem 4.61. A certain micro-power photovol- of the following: 50 Ω only, 100 Ω only, series,
taic device can be modeled under certain condi- and parallel.
tions as an ideal current power source and a
resistance in parallel—see the figure below. At Problem 4.65. You are given two speakers (rated
which value of the load resistance, RL, is the at 4 Ω and 16 Ω, respectively) and an audio
power delivered to the load maximized? amplifier with the output resistance(impedance)
equal to 8 Ω.

5 mA 50 kW RL
8W 4W 16 W

A. Sketch the circuit diagram that gives the max-

Problem 4.62. A low-cost polycrystalline Power imum acoustic output with the available
Up BSP1–12 1 W Solar panel lists ratings for the components. Explain your choice.
output voltage and current, which give maximum B. Sketch the circuit diagram for the maxi-
load power: VLmax¼17.28 V, ILmax¼0.06A. Based mum power efficiency. Explain your
on these cell specifications, which value of the choice.
equivalent resistance should the load to be
connected to the solar cell have for maximum 4.3.4 Application Example: Maximum
power output? Power Extraction from Solar Panel
Problem 4.66.
Problem 4.63. The solar panel from the previous A. Describe in your own words the meaning
problem generates a significant voltage of ~13 V of the fill factor of a solar cell (and solar
in a classroom without direct sun light, but the module).
resulting current is small, ~1 mA. B. A 200 W GE Energy GEPVp-200 c-Si
A. Now, which value of the equivalent resis- panel has the following reading on its
tance should the load have for maximum back: VOC ¼ 32.9 V, ISC ¼ 8.1 A,
power output?

IV-196
Chapter 4 Problems

VMP ¼ 26.3 V, and IMP ¼ 7.6 A. What is

the module fill factor? What is approxi-
mately the fill factor of the individual cell?

Problem 4.67. Using two Web links,

http://powerupco.com/site/
http://www.affordable-solar.com/,
identify the solar panel that has the greatest fill
factor to date.

Problem 4.68. A 10 W BSP1012 PV c-Si module

shown in the figure has 36 unit cells connected in
series, the short-circuit current of 0.66 A and the
open-circuit voltage of 21.3 V. The maximum power
Problem 4.70. A REC SCM220 220 Watt 20 V
parameters are VMP ¼ 17.3 V and IMP ¼ 0.58 A.
c-Si solar panel shown in the figure has the fol-
lowing readings on the back: the short circuit
current ISC of ~8.20 A, the open circuit voltage
VOC of ~36.0 V, the maximum power voltage VMP
of ~28.7 V, and the maximum power current IMP
of ~7.70 A. Repeat four tasks of problem 4.57.

A. Estimate the area of the single solar cell

using the common photocurrent density
value for c-Si solar cells (show units).
B. Estimate the open-circuit voltage for the
single cell. Problem 4.71. A 14.4 W load (a DC motor)
C. Estimate the fill factor of the module and of rated at 12 V is to be driven by a solar panel. A
the cell. c-Si photovoltaic sheet material is given, which
D. Estimate the value of the equivalent load has the open-circuit voltage of 0.6 V and the
resistance R required for the maximum photocurrent density of JP ¼ 0.03 A/cm2.
power transfer from the module to the load.

Problem 4.69. A 20 W BSP2012 PV c-Si mod-

ule shown in the figure has 36 unit cells connected
in series: the short-circuit current of 1.30 A, the
open-circuit voltage of 21.7 V, the maximum
power voltage VMP of 17.3 V, and the maximum
power current IMP of 1.20 A. Repeat four tasks of
the previous problem.
IV-197
Chapter 4 Circuit Analysis and Power Transfer

Outline parameters of a solar module (number of B. Compare your solution with the solution to
cells, cell area, and overall area) which is capable the previous problem.
of driving the motor at the above conditions and
estimate the overall panel size.

Problem 4.72. A custom 100 W load (a DC

motor) rated at 24 V is to be driven by a solar
panel. A c-Si photovoltaic sheet material is given,
which has the open-circuit voltage of 0.6 V and
the photocurrent density of JP ¼ 0.03 A/cm2. 4.4 Analysis of Nonlinear
Outline parameters of a solar module (number of Circuits: Generic Solar Cell
cells, cell area, and overall area) which is capable
of driving the motor at the above conditions and 4.4.1 Analysis of Nonlinear Circuits: Load
estimate the overall panel size. Line Method
4.4.2 Iterative Solution for Nonlinear
Problem 4.73. You are given the generic fill Circuits
factor F ¼ 0.72 for the c-Si solar panels, the Problem 4.75. A circuit shown in the figure
generic open-circuit voltage VOC ¼ 0.6 V of the below contains a nonlinear passive element.
c-Si cell, and the generic photocurrent density Using the load line method approximately, deter-
JP ¼ 0.03 W/cm2. mine the voltage across the element and the cur-
A. Derive an analytical formula that expresses rent through it for the two types of the υ-i
the total area Amodule in cm2 of a solar characteristic, respectively.
panel, which is needed to power a load, in
I
terms of the required load power PL. 10.55 kW a
B. Test your result by applying it to the previ-

+
9.5 V

+
ous problem.
+ V v-i
- -
Problem 4.74. You are given a low-cost
low-power flexible (with the thickness of b
a) I, mA
0.2 mm) a-Si laminate from PowerFilm, Inc., IA 1
with the following parameters: the fill factor of
F ¼ 0.61, the single-cell open-circuit voltage,
VOC ¼ 0.82 V, and the photocurrent density,
JP ¼ 0.0081 W/cm2. 0
V, volts
A. Derive an analytical formula that expresses
the total module area Amodule in cm2, which
is needed to power a load, in terms of the -1
required load power PL. 0 5 10
b) I, mA
1

0
V, volts

-1
0 5 10

IV-198
Chapter 4 Problems

Problem 4.76. A circuit shown in the figure υ-i characteristic of the nonlinear element (the
below contains a nonlinear passive element as a ideal Shockley diode) is
part of a current source. Using the load line
method approximately, determine the voltage
  

10 V
across the element and the current through it for I ¼ 5  10 exp
1 ½A :
0:025 V
the υ-i characteristic of the nonlinear element
shown in the same figure. Using the iterative solution, determine the voltage
across the element and the current through it.
a

I
+
9 mA 100 W
+

a
v-i V 667 W

+
- 1V
+ V v-i
b - -
I, mA
10 b

Problem 4.79. Repeat the previous problem for

0
the circuit shown in the figure below. The υ-i
V, volts characteristic of the nonlinear element is the
same.

-10 a
0 5 10

+
+
8 mA
v-i V 1 kW
Problem 4.77. Repeat the previous problem for -
the circuit shown in the figure that follows.
b
a
+

4.4.3 Application Example: Solving the

+

8 mA
v-i V 1 kW
Circuit for a Generic Solar Cell
-
Problem 4.80. The I(V )dependence for a resis-
b tive load in a circuit is shown in the figure below.
I, mA
10 I, A
16A

0
V, volts

PL
-10 V, Volts
0 10V
0 5 10

Problem 4.78. A circuit shown in the figure A. At which value of the load voltage is the
below contains a nonlinear passive element. The power delivered to the load maximized?

IV-199
Chapter 4 Circuit Analysis and Power Transfer

B. What is the related value of load

resistance?

Problem 4.81. A hypothetic thermoelectric

engine developed by US Navy has the I(V )
dependence shown in the figure below.

a) I, A
14.1A
1/2
10(2-V)

PL

V, Volts
0 2V

b) I, A
17.3A
1/2
10(3-V)

PL

V, Volts
0 3V

1. At which value of the load voltage is the

power, PL, delivered to the load
maximized?
2. What is the related value of load resistance for
maximum power transfer?

Problem 4.82. Estimate the values of VMP and

IMP versus VOC and ISC for a set of generic c-Si
solar cells. Every cell has VT ¼ 0.026V (room
temperature of 25 C) and VOC ¼ 0.6 V. The
ideality factor n in Eqs. (4.43)–(4.45) is allowed
to vary over its entire range as shown in the
Table below.

n VMP/VOC,% IMP/IOC,% F
1.00
1.25
1.50
1.75
2.00

IV-200
Chapter 5

Chapter 5: Operational Amplifier

and Amplifier Models

Overview
Prerequisites:
- Knowledge of major circuit elements (dependent sources) and their
-i
characteristics (Chapter 2)
- Knowledge of basic circuit laws (Chapter 3) and Thévenin equivalent (Chapter 4)

Objectives of Section 5.1:

- Learn and apply the model of an operational amplifier including principle
of operation, open-circuit gain, power rails, and input and output resistances
- Correlate the physical operational amplifier with the amplifier circuit model
- Establish the ideal-amplifier model
- Learn the first practical amplifier circuit—the comparator

Objectives of Section 5.2:

- Understand and apply the concept of negative feedback to an operational amplifier
circuit
- Construct three canonic amplifier circuit configurations with negative feedback: the
non-inverting amplifier, the inverting amplifier, and the voltage follower
- Understand the current flow in the amplifier circuit including the power transfer
from the power supply to the load

Objectives of Section 5.3:

- Choose the proper resistance values for the feedback loop and learn how to cascade
multiple amplifier stages
- Learn about input/output resistances of the amplifier circuit and establish load
bridging and load matching conditions important in practice
- Find ways to eliminate the DC imperfections of the amplifier that become very
apparent at high amplifier gains
- Use an amplifier IC with a single voltage supply (a battery)

Objectives of Section 5.4:

- Obtain the initial exposure to differential signals and difference amplifiers
- Build an instrumentation amplifier
- Connect an instrumentation amplifier to a resistive sensor

© Springer Nature Switzerland AG 2019 V-201

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_5
Chapter 5 Operational Amplifier and Amplifier Models

Objectives of Section 5.5:

- Learn a general feedback system including closed-loop gain and error signal
- Apply the general feedback theory to voltage amplifier circuits
- Construct current, transresistance, and transconductance amplifiers with the negative
feedback

Application Examples:
Operational amplifier comparator
Instrumentation amplifier in laboratory

Keywords:
Operational amplifier: (abbreviation op-amp, integrated circuit, dual in-line package, non-
inverting input, inverting input, output terminal, power terminals, offset-null terminals,
differential input voltage, open-circuit voltage gain, open-loop voltage gain, open-loop
configuration, closed-loop configuration, power rails, voltage transfer characteristic, rail-to-rail,
comparator, digital repeater, zero-level detector, circuit model, input resistance, output resistance,
ideal amplifier, ideal-amplifier model, marking, summing point, common-mode input signal,
differential input signal, summing-point constraints, first summing-point constraint, second
summing-point constraint, sourcing current, sinking current, DC imperfections, input offset
voltage, input bias current, input offset currents), Negative feedback, Feedback loop, Feedback
as a dynamic process, Non-inverting amplifier, Inverting amplifier, Voltage follower (buffer)
amplifier, Summing amplifier, Digital-to-analog converter, Binary counter, DC-coupled
amplifier, AC-coupled amplifier, Capacitive coupling of an amplifier, Gain tolerance of an
amplifier, Circuit model of a voltage amplifier, Input resistance of amplifier circuit, Output
resistance of amplifier circuit, Load bridging (impedance bridging), Load matching (impedance
matching), Cascading amplifier stages, Virtual-ground (integrated) circuit, Differential voltage of
a sensor, Common-mode voltage of a sensor, Differential sensor, Single-ended sensor, Difference
amplifier, Differential amplifier circuit gain, Common-mode amplifier circuit gain, Common-
mode rejection ratio (CMRR), Unity common-mode gain stage, Instrumentation amplifier, Load
cell, Current amplifier using op-amp, Transconductance amplifier using op-amp, Transresistance
amplifier using op-amp, Howland current source (Howland current pump)
Linear feedback system: (forward gain? open-loop gain, feedback gain, feedback factor,
summing node, difference node, closed-loop gain, error signal)

V-202
Chapter 5 Section 5.1: Amplifier Operation and Circuit Models

Section 5.1 Amplifier Operation and Circuit Models

The low-power amplifier integrated circuit (IC) is arguably the most widely employed
discrete circuit component encountered in common electronic audio, control, and com-
munication systems. Among amplifiers, the differential input, high-gain amplifier called
the operational amplifier (or simply op-amp) has become a popular choice in many
circuit applications. At this point, it is impossible for us to understand the internal
operation of the amplifier IC without basic knowledge of semiconductor electronics,
especially the junction transistor studied in the following courses. Fortunately, the circuit
model of an operational amplifier does not require knowledge of the IC fabrication steps,
nor does it require an understanding of the internal transistor architecture. Conceptually,
operational amplifiers can be introduced early in the book, which enables us to immedi-
ately proceed toward our goal of designing and building practical circuits.

5.1.1 Amplifier Operation

Symbol and Terminals
After the amplifier chip is fabricated as an integrated circuit and the bond wires are
attached, it is permanently sealed in a plastic package. Often the encasing is done in a
dual in-line (DIP-N) package with N denoting the number of IC pins. Figure 5.1 on the
right shows an example of a DIP package. One IC chip may contain several independent
individual amplifiers. We start analyzing the amplifier model by first labeling the termi-
nals and introducing the amplifier circuit symbol (a triangle) as shown in Fig. 5.1 on the
left. The amplifier is typically powered by a dual-polarity voltage power supply with
three terminals:
V CC and common (ground) port of 0 V, see Section 3.2. The index
C refers to the collector voltage of the internal transistors.

+VCC

v+
+
- + =
v- vout
- -VCC
v- v+
-VCC offset null
common common
NC +VCC vout

Fig. 5.1. Terminals of the operation amplifier (left); they also denote pins of the amplifier IC
package (see a common LM 741 chip on the right). All voltages are referenced with respect to a
common port of the dual-polarity voltage supply.

The amplifier has a total of five terminals, notably:

1. A non-inverting input with the input voltage υþ with respect to common
2. An inverting input with the input voltage υ with respect to common
3. An output terminal with the output voltage υout with respect to common

V-203
Chapter 5 Operational Amplifier and Amplifier Models

4. Power terminal þV CC with a positive voltage VCC (+9 V) with respect to common
5. Power terminal V CC with the negative voltage V CC (9 V) with respect to
common
Each of the five terminals corresponds to a particular metallic pin of the IC package. All
of the amplifier’s terminals are used in an amplifier circuit and none of them should be left
disconnected. However, the chip itself could have some not connected (NC) terminals
that maintain symmetry and which are used as heat sinks, see Fig. 5.1 on the right. Also
note that a number of amplifier ICs, including the LM74, may have extra terminals or
pins, the so-called offset or offset-null terminals. These terminals are used to control the
input offset voltage (an imperfection) of the amplifier.

Historical: The abbreviation for the operational amplifier is op-amp; this abbreviation is
not quite official but is used by most practitioners. The term operational amplifier first
appeared in a 1943 paper by John R. Ragazzini, an American electrical engineer and ECE
professor. One of his students introduced the terms inverting and non-inverting inputs. One
of his most notable students was Rudolf Kalman who became famous for the invention of
the Kalman filters.

Open-Circuit or Open-Loop Voltage Gain

Once the amplifier chip is properly powered, its operation is quite simple: the output
voltage is expressed through the two input voltages in the form

υout ¼ Aðυþ  υ Þ ð5:1Þ

which is identical to the operation of the voltage-controlled voltage source introduced in

Section 2.4. Here,υþ  υ is the differential input voltage to the amplifier. The dimensionless
constant A is called the open-circuit voltage gain of the amplifier. Quite frequently, the term
open-loop gain is used and A is replaced by AOL. Equation (5.1), which will be called the
amplifier equation, is always valid. It does not matter if the amplifier is in the open-loop
configuration, (i.e., no feedback loop is present) or in a closed-loop configuration (a feedback
loop is present; see the next section). The amplifier IC is intentionally built in such a way as to
provide the highest possible open-circuit gain; it is achieved using transistors connected in
series such as the Darlington pair. Typically,

A  105  108 ð5:2Þ

The exact gain value cannot be controlled precisely due to manufacturing tolerances. The
open-circuit gain is often measured in V/mV. For example, the value of 160 V/mV
corresponds to the gain value of 160,000. The open-circuit gain is difficult to measure.

V-204
Chapter 5 Section 5.1: Amplifier Operation and Circuit Models

Power Rails and Voltage Transfer Characteristic in the Open-Loop

Configuration
Two power interconnects of the amplifier are often called rails. The term “rail” appears
simply because the power interconnections are represented by two long horizontal wires
in the circuit diagram connected to þV CC and V CC , respectively, which resemble long
metal rails. The positive rail is þV CC , and the negative rail is V CC . The power rails are
interfaced to a laboratory dual-polarity voltage power supply that also provides a
common (ground) port to be used later. The output amplifier voltage can never exceed
the positive rail voltage or be less than the negative rail voltage. In other words,

V CC  υout  V CC ð5:3Þ

Should the output voltage found in Eq. (5.1) exceed VCC, it will be forced to VCC.
Likewise, should the output voltage drop to less than V CC , it will be forced to V CC .
In view of these physical constraints, Eq. (5.1) may be rewritten in the form

υout ¼ Aðυþ  υ Þ, jυout j < V CC

υout ¼ þV CC , Aðυþ  υ Þ > þV CC ð5:4Þ
υout ¼ V CC , Aðυþ  υ Þ < V CC

Example 5.1: Plot to scale the output voltage of an operational amplifier with an open-
circuit gain of A ¼ 105 when the non-inverting input voltage υþ changes from 1 mV to
+1 mV and the inverting input voltage υ is set to zero. The amplifier is powered by a
16-
V dual voltage supply. This plot will give us the voltage transfer characteristic of the open-
loop amplifier.
Solution: Amplifier Eq. (5.4) gives the result shown in Fig. 5.2 by a thick piecewise-linear
curve. Due to the extremely high open-loop gain, the amplifier output is almost always
saturated. This means that, except for a very narrow domain of input voltages on the order
of
0.2 mV, the output simply follows the power rail voltage, either positive or negative.
This is a very remarkable feature of the open-loop amplifier.

Power Rails in Practice

The power rail(s) of the amplifier or the supply voltage is specified in the datasheet. For
example, the LM358 amplifier IC operates using a single supply 3 V to 32 V or dual
supplies
1.5 V to
16 V. As we can see from this data, the amplifier does not necessarily
operate using a dual voltage supply; a single supply (“single rail”) can be used as well. The
same amplifier chip (e.g., LM358) can be used either with the single voltage supply or with
a dual supply. This question, although less important in theory, is very important in practice.
Also note that, in practice, the output never exactly reaches the positive or negative rail

V-205
Chapter 5 Operational Amplifier and Amplifier Models

voltages; there is always a voltage offset; it can vary from a minimum value of between
0.01 Vand 0.05 V for certain special ICs (called the rail-to-rail amplifiers) all the way up to
1.8 V for other amplifiers (e.g., LM741).

vout, V
20
+VCC positive power rail

10

-10
negative power rail
-VCC
-20
-0.8 -0.4 0 0.4 0.8
v+, mV

Fig. 5.2. Amplifier output voltage in the open-loop configuration. The open-loop gain is AOL
¼ 105 and the supply voltage is
16 V. Note that the scale for the input voltage is in mV.

5.1.2 Application Example: Operational Amplifier Comparator

A comparator is a circuit or a device that compares two input voltages and outputs a digital
voltage (e.g.,
10 V) as an indication of which input voltage is larger. Due to a very high
gain, the operational amplifier in the open-loop configuration shown in Fig. 5.3a may
operate as a basic comparator. Figure 5.3b shows one possible application of the compar-
ator: a digital repeater. We assume that υ ¼ V threshold ¼ 0. The input voltage to the
comparator υþ ðt Þ is a weak noisy digital signal shown in Fig. 5.3b. This signal is
compared to a threshold level of zero volts (the threshold voltage in Fig. 5.3).

a) b)
Output voltage vout(t), V Vthreshold=0 V
15

A =1,000,000 10
v+(t) vout 5
+
- 0
Vthreshold -5
-10
0V
-15
0 10 20 30 40 50
Input voltage v+(t), V time, µs

Fig. 5.3. A simple operational amplifier as a voltage comparator.

When the amplifier open-circuit gain tends to infinity (the transfer characteristic in Fig. 5.2
becomes a straight vertical line), Eq. (5.4) applied to the present case is reduced to

V-206
Chapter 5 Section 5.1: Amplifier Operation and Circuit Models

υout ¼ þV CC , υ þ ðt Þ > 0
ð5:5Þ
υout ¼ V CC , υ þ ðt Þ < 0
Figure 5.3b shows the resulting output voltage for V CC ¼ 10 V. The weak input digital
signal will thus be amplified and cleaned from noise, which is one major function of a
digital repeater. In practice, dedicated comparators are used instead of this simple setup,
which are much faster and have useful additional features. The comparator amplifier may
also be employed for other purposes such as a zero-level detector.

Exercise 5.1: In Fig. 5.3, the threshold voltage of the comparator amplifier is changed to
+5 V. What will be the output of the comparator circuit?
Answer: 10 V at any time instant.

5.1.3 Amplifier Circuit Model

Circuit Model
An equivalent circuit model of an amplifier is shown in Fig. 5.4. This circuit model is a
two-port electric network. It includes three single circuit elements: an ideal voltage-
controlled voltage source Aðυþ  υ Þ (Section 2.4), input resistance Rin of the amplifier,
and output resistance Rout of the amplifier.

Amplifier

v+
Rout +
+
v+ - v- Rin + A(v+- v-) vout= A(v+- v-)
-
-
-
v-

Fig. 5.4. Equivalent circuit model of an amplifier is in the shadow box as a two-port network. No
load is connected. The ground of the output terminal is the common port.

Analysis of the Amplifier Circuit Model: Effect of Input/Output Resistances

The input/output resistances of an amplifier in Fig. 5.4 impose rather severe limitations on
its desired operation. First, a large but finite input resistance always implies that some input
current,iin ¼ ðυþ  υ Þ=Rin , will flow into the amplifier as long as the input voltage signal
is different from zero. Consequently, the amplifier would require not only the input voltage
but also a certain amount of input power. As a result, the amplifier may appreciably load a
sensor connected to its input, i.e., require more power than (a tiny) sensor can actually
provide. Second, a finite output resistance limits the output current iout to the amplifier; this
resistance operates as a current limiting resistor which is studied in Chapter 3. Along with

V-207
Chapter 5 Operational Amplifier and Amplifier Models

this, it also leads to the fact that the voltage across any load connected to the amplifier’s
output will not be equal to the desired output voltage given by Eq. (5.1), except for an open
circuit. These limitations are quantified when we consider a circuit shown in Fig. 5.5. The
circuit includes the amplifier model, an arbitrary source represented by its Thévenin
equivalent υS, RS, and a load represented by its equivalent resistance RL.

Amplifier
iin iout

RS Rout
+ +
vS + vx Rin + Avx RL vout
- -
- -

Fig. 5.5. Amplifier circuit model with connected source and load resistances.

Using the voltage division principle twice, the output voltage in Fig. 5.5 is expressed as



Rin RL
υout ¼ υS  A  ð5:6Þ
Rin þ RS RL þ Rout
|fflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflffl}
υX

This result is quite different from the ideal behavior of the amplifier described by the
perfect amplification of the source signal

υout ¼ AυS ð5:7Þ

using the available open-circuit gain A of the amplifier.

Exercise 5.2: For the amplifier circuit in Fig. 5.5 with A ¼ 1000, determine the output
voltage given that υS ¼ 1 mV, RS ¼ 50 Ω, and RL ¼ 50 Ω for two cases:
A. Rin ¼ 1 MΩ and Rout ¼ 1 Ω.
B. Rin ¼ 50 Ω and Rout ¼ 50 Ω.
Answer:
Case A: υout ¼ 0:98 V (which is close to the ideal behavior, υout ¼ 1:00 V).
Case A: υout ¼ 0:25 V (three quarters of the voltage gain are lost).

According to Eq. (5.6), υout < AυS for any positive finite values of Rin and Rout. In order
to make use of the full available open-circuit gain A of the amplifier, we should:

V-208
Chapter 5 Section 5.1: Amplifier Operation and Circuit Models

1. Design Rin as large as possible, ideally an open circuit, that is,

Rin ¼ 1 ð5:8Þ
2. Design Rout as small as possible, ideally a short circuit, that is,
Rout ¼ 0 ð5:9Þ
In this and only this case, the equality υout ¼ AυS will be satisfied exactly.

5.1.4 Ideal-Amplifier Model and First Summing-Point Constraint

The amplifier IC can then be described with a high degree of accuracy by using the
so-called ideal-amplifier model. It is based on the best possible choices for input/output
resistances as described by Eqs. (5.8) and (5.9), respectively. It is also based on the
assumption that the open-loop gain in Eq. (5.1) is made as high as possible, i.e., equal to
infinity. The ideal-amplifier model is an important theoretical and practical tool for the
analysis of microelectronic amplifier circuits. This model will be used in the following
sections of this chapter and in subsequent chapters. We can summarize the model of an
ideal operational amplifier in concise form:
1. No current can flow into the amplifier (into either input terminal).
2. The open-loop gain A is infinitely high.
3. The input resistance Rin is infinitely high.
4. The output resistance Rout is zero.
Property 1 follows from property 3 and vice versa. One more condition of the ideal-
amplifier model could be added, namely, that the power rails
V CC are exactly reached
when operated in saturation. The ideal-amplifier model does not use the accurate internal
amplifier circuit shown in Fig. 5.4 or in Fig. 5.5, respectively. Instead, a simple triangle
symbol may be used for the ideal amplifier, which is shown in Figs. 5.1 and 5.3.

Exercise 5.3: Solve the previous exercise for the ideal-amplifier model.
Answer: Case A, B: υout ¼ 1:00 V.

First Summing-Point Constraint

The summing point of an amplifier is the connection of the two inputs to the amplifier.
The common-mode input signal is the half sum of the two input voltages, ðυþ þ υ Þ=2.
The differential input signal is the input voltage difference, υx ¼ υþ  υ . Conditions
applied to the amplifier’s input are called summing-point constraints. The first summing-
point constraint is applied to the ideal-amplifier model. It states that no current can flow
into either of the amplifier terminals as shown in Fig. 5.6.
This is consistent with an infinitely high input resistance. The condition of no input
current into the amplifier means that virtually no input power is necessary. For example,

V-209
Chapter 5 Operational Amplifier and Amplifier Models

summing point
i+
vout
v+
+
-
v-
i-

Fig. 5.6. The first summing-point constraint stipulates that no current flows into the ideal amplifier.

a tiny sensor, which does not deliver any appreciable power, could directly be connected
to the input. The voltage from the sensor will still be accepted as the input of the amplifier.
This condition is a convenient abstraction of the ideal-amplifier model. In reality, a very
small input current does exist, typically on the order of nanoamperes (nA) for common
amplifier ICs or picoamperes (pA) for ICs with an input JFET stage.

Realistic Values of Input/Output Resistances and Output Current

How far off from reality is the assumption of infinite input resistance? A review of the
datasheets reveals that the input resistance of the older amplifier IC (e.g., LM741,
LM1458) varies from 0.3 to 6 MΩ. The input resistance of JFET-input stage amplifiers
(TL082) is on the order of 1 TΩ (1012 Ω). Now, how realistic is the assumption of zero
output resistance? Note that if the output resistance were exactly zero, the amplifier would
be able to source an infinite current (power) into a low-resistance load. Clearly, we cannot
expect a large output power from a physically small amplifier IC. Therefore, we have to
introduce a small internal output resistance, which appears to be on the order of 1–100 Ω.
The corresponding output short-circuit current of the common amplifier ICs (LM741,
LM1458, LM358) cannot exceed 40–60 mA; the current into a load is smaller. The output
current of faster amplifier ICs (TL082) is even smaller. If the load requires more current
than the chip can provide, then the output voltage will notably be clipped.

V-210
Chapter 5 Section 5.2: Negative Feedback

Section 5.2 Negative Feedback

In most practical circuits, the amplifier IC is not used in open-loop configuration.
Engineers have modified the open-loop condition into a negative feedback loop in
order to set the gain to a desired value and ensure the amplifier’s stability. This section
provides you with all the essential knowledge needed to design an amplifier with negative
feedback. The mathematical model introduced in this section is based on two conditions
imposed at the amplifier’s input; we term them the summing-point constraints:
a) No electric current flows into or out of the amplifier inputs.
b) The differential voltage at the amplifier’s input is zero.
The first summing-point constraint has already been introduced in the previous section.
The second summing-point constraint has yet to be derived. We will show that the two
summing-point constraints, along with KCL and KVL, will enable us to solve any
amplifier circuit that involves a negative feedback, no matter what the specific nature
of the feedback loop is and regardless of whether it is DC, AC, or a transient circuit.

5.2.1 Idea of the Negative Feedback

The idea of the negative feedback goes way back—we may say almost to the Stone Age.
Take a wooden rod of 1–2 feet in length. Hold the rod in the vertical position at the tip of
your finger. You will probably succeed. Now, close your eyes and try to do the same. You
will most likely fail. The reason for the failure is a breakdown of the feedback loop. This
loop is created by visual control of the rod’s position; you automatically apply a
compensating acceleration to the bottom tip of the rod when it begins to fall. Another
good example is driving a car and trying to stay in the center of the lane. The negative
feedback for electronic amplifiers was first invented and realized by Harold S. Black
(1898–1983), a 29-year-old American electrical engineer at Bell Labs. To many electrical
engineers, this invention is considered perhaps the most important breakthrough of the
twentieth century in the field of electronics because of its wide applicability. We will
construct simple amplifier circuits of a given gain, using a resistive feedback loop. Being
able to perform this task is already critical from the practical point of view.

5.2.2 Amplifier Feedback Loop: Second Summing-Point Constraint

We construct the feedback loop, as shown in Fig. 5.7, by connecting the output to the
inverting input terminal. This was exactly the idea of Harold Black. The shadowed box in
the feedback loop may represent one or more circuit elements. The feedback loop may be
a simple wire, a resistance, a network of circuit elements (resistances, inductances,
capacitances), etc. The negative feedback simply means that the output voltage, or rather
a portion of it, is returned back to the inverting input.

V-211
Chapter 5 Operational Amplifier and Amplifier Models

v+ + vout
-
+
vx
- v-

Fig. 5.7. A feedback loop around an amplifier.

Feedback as a Dynamic Process

According to Eq. (5.1) of the previous section, the output voltage is proportional to

υx ¼ υþ  υ ð5:10Þ

where υx is the differential input voltage. Hence, υx or υx multiplied by a constant is

returned to the input during a very short period of time. The feedback effect is inherently a
very fast dynamic process, which leads to a static solution with quite remarkable
properties. In the example that follows, we will attempt to model the effect of the
feedback loop using several very simplifying assumptions.

Example 5.2: An amplifier with a feedback loop in Fig. 5.7 has υþ fixed at +10 V. υ is
equal to 0 V at t ¼ 0. We shall assume that 50 % of υþ is returned back to the input in 1 μs.
How does the differential voltage υx change with time?
Solution:
1. At t ¼ 0, υx ¼ 10 V  0 V ¼ 10 V. Next, 50 % of 10 V is returned in 1 μs. The
voltage υ becomes equal to 0 V þ 5 V ¼ 5 V after 1 μs.
2. At t ¼ 1 μs, υx ¼ 10 V  5 V ¼ 5 V. Next, 50 % of 5 V is returned in 1 μs.
The voltage υ becomes equal to 5 V þ 2:5 V ¼ 7:5 V after 2 μs.
3. At t ¼ 2 μs, υx ¼ 10 V  7:5V ¼ 2:5 V. Next, 50 % of 2.5 V is returned in 1 μs.
The voltage υ becomes equal to 7:5 V þ 1:25 V ¼ 8:75 V after 3 μs.
The process further continues so that voltage υx halves every microsecond. The process
dynamic is shown in Table 5.1 and visualized in Fig. 5.8.

Table 5.1. Dynamics of the differential input voltage as a function of

time for Example 5.2.
Time, μs υþ υ υx ¼ υþ  υ
0 10 V 0V 10 V
1 10 V 5V 5V
2 10 V 7.5 V 2.5 V
3 10 V 8.75 V 1.25 V
4 10 V 9.375 V 0.625 V

V-212
Chapter 5 Section 5.2: Negative Feedback

vx
10V

5V

2.5V

0V
0 1 2 3 time, µs

Fig. 5.8. Dynamics of the differential input voltage as a function of time.

Both Table 5.1 and Fig. 5.8 make clear that the differential voltage υx decays to zero
very rapidly, once the feedback loop is introduced. Hence we arrive at the second
summing-point constraint, which is valid only for the amplifiers with the negative
feedback loop: the differential input voltage to the amplifier is exactly equal to zero.
The second summing-point constraint is a close approximation to reality. Its accuracy
depends on the value of the open-loop gain of the amplifier. If the open-loop gain were
infinite, the second summing-point constraint would be exact.

5.2.3 Amplifier Circuit Analysis Using Two Summing-Point Constraints

Next, we will solve an amplifier circuit with negative feedback using the two summing-
point constraints (SPC): (i) no current into or out of the input amplifier terminals and
(ii) the differential input voltage is zero. The method of two summing-point constraints is
an accurate solution method for a wide variety of amplifier circuits with the negative
feedback. For amplifier circuits with a single input, we will denote the input voltage to the
amplifier circuit by υin. Voltage υin may be equal to υþ or to υ , depending on amplifier
type to be used.

Non-inverting Amplifier
The first amplifier configuration is the so-called non-inverting amplifier shown in
Fig. 5.9. The feedback loop contains one resistance R2. Another resistance R1 shunts
the inverting input to ground. The input voltage to the amplifier circuit is the voltage υin
with respect to ground, or common in this case, which implies the use of the dual-polarity
voltage power supply. The output voltage with respect to common is υout. We apply the
first summing-point constraint and KCL to the node “*” in Fig. 5.9 and obtain

i1 ¼ i2 ð5:11Þ

Equation (5.11) is further transformed using Ohm’s law in the form

υ*  0 υout  υ*
¼ ð5:12Þ
R1 R2

V-213
Chapter 5 Operational Amplifier and Amplifier Models

vin +
vx
+
- +
- vout
i2 -
v*
i1 R2
common
common
R1

common

Fig. 5.9. Circuit diagram of the non-inverting amplifier. A dual power supply is not shown.

The second summing-point constraint yields

υ* ¼ υin ð5:13Þ

since υx ¼ 0. Equation (5.12) thus reads

υin υout  υin υout υin υin
¼ ) ¼ þ ð5:14Þ
R1 R2 R2 R1 R2

As a result, we find that the voltage input-to-output relation becomes


R2
υout ¼ 1 þ υin ð5:15Þ
R1

The amplifier circuit is solved: we have expressed the output voltage in terms of the input
voltage and a resistor ratio. Equation (5.15) is the basic result in amplifier theory. It shows
that the feedback loop allows us to precisely control the gain with two arbitrary resis-
tances. One chooses the proper resistance combination to achieve any finite gain between
one (setting R2 ¼ 0) and the open-loop (infinite) gain (setting R1 ¼ 0). In the last case, the
negative input terminal becomes grounded; the feedback loop is irrelevant and can be
replaced by an open circuit so that the amplifier again becomes the comparator. The gain
expression


R2
ACL ¼ 1 þ 1 ð5:16Þ
R1

is called the closed-loop gain of the amplifier; it clearly relates the output voltage to the
input voltage. Equation (5.16) is a dramatic illustration of the negative feedback. We
started with an amplifier having a very large yet loosely predictable open-loop gain.
Through applying the negative feedback, we arrived at a gain that is much smaller than
the open-loop gain; however, it is controllable and stable. Equation (5.16) can be derived
more simply using the voltage divider concept. Namely, resistors R1, R2 form a voltage
divider between 0 V and the output voltage. Hence, the voltage at node (*) may be found.
Equating this voltage to the input voltage gives us Eq. (5.16).

V-214
Chapter 5 Section 5.2: Negative Feedback

Exercise 5.4: Solve the circuit shown in Fig. 5.10, i.e., find the output voltage υout with
respect to common.



R2 1  106
Answer: υout ¼ 1 þ υin ¼ 1 þ  1 mV ¼ 197 mV.
R1 5:1  103

+ +
1 mV - +
vout
- -
R2=1 MW
R1=5.1 kW

Fig. 5.10. A non-inverting amplifier circuit with an input voltage of 1 mV.

Inverting Amplifier
The next amplifier circuit configuration is the inverting amplifier shown in Fig. 5.11. Note
that the input terminals are now flipped. The negative feedback loop is still present; it involves
resistance R2. Another resistance, R1, forms a voltage divider.
R2

R1 v* i2
vin + -
vx +
+
i1 - vout
-
common common

Fig. 5.11. Circuit diagram of the inverting amplifier; a dual power supply is used (not shown).

The input voltage to the amplifier circuit is the voltage υin with respect to ground or
common. The output voltage with respect to common is υout. To solve the amplifier
circuit, we use the same solution procedure as for the non-inverting amplifier. However,
the final result will be quite different. We apply the first summing-point constraint and
KCL to the node labeled “*” in Fig. 5.11 and again obtain

i1 ¼ i2 ð5:17Þ

Equation (5.17) is transformed using Ohm’s law,

υin  υ* υ*  υout
¼ ð5:18Þ
R1 R2

V-215
Chapter 5 Operational Amplifier and Amplifier Models

The second summing-point constraint yields υ* ¼ 0 since υx ¼ 0. Equation (5.18) then

gives
R2
υout ¼  υin ð5:19Þ
R1

The amplifier circuit is solved: we have expressed the output voltage in terms of the input
voltage. Equation (5.19) is another key result in amplifier theory. The expression
R2
ACL ¼  ð5:20Þ
R1

is also called the closed-loop gain of the inverting amplifier; the gain again relates the
output voltage to the input voltage. It is now negative, which means that the output
voltage is inverted. This circumstance is hardly important for the AC signals where the
voltage inversion is equivalent to a phase shift of π radians or 180 degrees. The feedback
loop of the inverting amplifier also enables us to control the gain of the amplifier with two
standard resistors. We can choose the proper resistance combination to achieve any finite
gain between zero (R2 ¼ 0) and negative infinity (R1 ¼ 0). In Fig. 5.11 we clearly see how
the amplifier gain is controlled by the voltage divider with resistors R1 and R2.

Exercise 5.5: Solve the inverting-amplifier circuit shown in Fig. 5.12, i.e., find the output
voltage υout with respect to common.
R2 1  104
Answer: υout ¼  υin ¼   1 mV ¼ 196 mV.
R1 51

R2=10
-
+ +
1 mV R1=51 +v
out
- -

Fig. 5.12. An inverting amplifier circuit with an applied input voltage of 1 mV.

Voltage Follower or Buffer Amplifier

The third important member of the amplifier family is the voltage follower or buffer
amplifier whose circuit is shown in Fig. 5.13. The negative feedback loop is just a wire.

V-216
Chapter 5 Section 5.2: Negative Feedback

vin + +
vx +
-
- vout
-
common common

Fig. 5.13. Circuit diagram of the buffer amplifier; a dual power supply is used.

The use of the second summing-point constraint immediately leads to

υout ¼ υin , ACL ¼ 1 ð5:21Þ
so that the gain of this amplifier type is simply unity. Why do we need a unity-gain
amplifier? The reason is that, while the buffer amplifier in Fig. 5.13 passes the voltage
without change, it requires virtually no current at the input (virtually no input power) but,
at the same time, could source a significant current (on the order of 20–40 mA) at the
output, i.e., provide significant output power. In order words, it becomes in a certain sense
a power amplifier. A simple example would be a capacitive sensor that cannot deliver
currents on the order of 10 mA or even smaller currents; otherwise, the corresponding
capacitor would immediately discharge. Such a sensor cannot directly be connected to an
LED indicator that requires at least 10 mA. However, this sensor may deliver significant
voltages, on the order of 1–5 V, which do not need to be amplified. The use of a buffer
amplifier can nicely solve this connection problem. The above discussion directly leads
us to the concept of input resistance of the amplifier circuit studied in the next section.

Exercise 5.6: Solve the voltage-follower circuit shown in Fig. 5.13, i.e., find the output
voltage υout with respect to common when the input voltage with respect to common is a)
1 V and b) 10 V. The amplifier is powered by a
6-V dual supply.
Answer: a) υout ¼ 1 V; b) υout ¼ 6 V.

5.2.4 Mathematics Behind the Second Summing-Point Constraint

The second summing-point constraint might appear to be mysterious, at least at first
sight. How does the amplifier accept the input signal if there is no current at the input
and the differential input voltage is zero? Can we avoid using the second SPC, and at
what cost? We will show that the second SPC is nothing but a handy tool to solve the
amplifier circuit with the negative feedback, with a high degree of accuracy. Mathe-
matically, the second SPC gives a leading (and usually very accurate) term of what is
known as an asymptotic expansion with regard to a small parameter, here the inverse
open-loop gain A1 . Let us now ignore the second SPC and derive the gain equation for
the buffer amplifier exactly. A similar derivation for the non-inverting amplifier is given
as a homework problem. Looking at Fig. 5.13, we conclude that υ ¼ υout since the

V-217
Chapter 5 Operational Amplifier and Amplifier Models

negative amplifier terminal is directly connected to the output. According to the

amplifier equation, Eq. (5.1) of the previous section, we have
υout ¼ Aðυþ  υ Þ ¼ Aðυin  υout Þ ð5:22Þ
Solving Eq. (5.22) for the output voltage yields
A
υout ¼ υin ð5:23Þ
1þA
Using a Maclaurin series expansion, we obtain with A >> 1 the result
A 1
¼  1  1=A  1 ð5:24Þ
1 þ A 1 þ 1=A

which is consistent with Eq. (5.21) and is very accurate since typically A > 105. A similar
derivation holds for the non-inverting (or the inverting) amplifier configuration. With this
in mind, the second SPC is clearly optional. Instead, the amplifier definition Eq. (5.1) may
be used, along with the condition of the high open-loop gain. However, it is rather tedious
to repeat the asymptotic analysis every time; so we prefer to use the accurate and simple
summing-point constraint. The finite value of the open-circuit gain A becomes important
for high-speed amplifiers with the feedback loop; see Chapter 10.

5.2.5 Current Flow in the Amplifier Circuit

The current flow in the complete amplifier circuit is illustrated in Fig. 5.14. The output
current through the load resistance RL of the amplifier circuit in Fig. 5.14 is provided
by the dual-polarity power supply. In this sense, the amplifier is also a “valve” (similar
to its building block, the transistor), which “opens” the power supply in response to
the low-power (or virtually no-power) input voltage signal. In Fig. 5.14, you should
note that standard resistor values (5 % or 1 % tolerance) may be slightly different from
the values used in this figure for convenience. We consider the positive input voltage
of 100 mV in Fig. 5.14a first. The non-inverting amplifier has a closed-loop gain ACL
equal to 50. The output voltage is thus +5 V, which is the push mode. The load current
of 10 mA is found from Ohm’s law. The feedback current of 0.1 mA is found using the
second SPC and Ohm’s law. The power supply current is the sum of both. The
feedback current controls the gain of the amplifier, and the load current drives the
load. The overall amplifier circuit efficiency (neglecting the loss in the IC itself)
depends on the ratio of these two currents. Therefore, we should keep the feedback
current small. The power current path is shown in Fig. 5.14a by a thick trace. The current
at node A can only enter the upper power supply. Thus, it is the upper power supply
being used. The amplifier is operating in the “push” mode, i.e., the amplifier sources
the current. When the input voltage is negative as in Fig. 5.14b, the lower power supply is
delivering power. Now, the amplifier sinks the current; it is operating in the “pull” mode.

V-218
Chapter 5 Section 5.2: Negative Feedback

However, the ratio of the load current and feedback loop current remains the same, at least
for the ideal amplifier. Similar results are obtained for the inverting amplifier and the
voltage follower, respectively.

a) positive input voltage - “push”

10.1mA
vin
10.1mA vout=5 V
+
+ - 10.1mA
0.1mA
100 mV
0.1mA
- 100 mV
R2=49 k 10mA
+
RL=500 9V
common
-
R1=1 k
A 0V
common 0.1mA common 10.1mA +
9V
-
not used

b) negative input voltage - “pull”

vin
10.1mA vout=-5 V
+
- -
-100 mV
0.1mA
+ -100 mV
R2=49 k 10mA
0.1mA +
RL=500 9V
common
-
R1=1 k not used
A 0V
common 0.1mA common 10.1mA +
10.1mA 9V
-

Fig. 5.14. Current flow in the non-inverting amplifier circuit operating in the (a) push mode and
(b) pull mode. The path of the (relatively high) load current is marked in bold.

5.2.6 Multiple-Input Amplifier Circuit: Summing Amplifier

Figure 5.15 shows an important amplifier type on the basis of the inverting amplifier—the
summing amplifier. A summing amplifier performs a simple mathematical operation: it sums
several weighted input voltages. The summing amplifier is a prototype of the binary-
weighted-input digital-to-analog converter. According to KCL and to the first summing-
point constraint, one has with reference to Fig. 5.15

iF ¼ i1 þ i2 þ i3 ð5:25Þ

On the other hand, the second summing-point constraint (the differential voltage to the
amplifier is zero and the inverting input is the common or virtual ground) yields

V-219
Chapter 5 Operational Amplifier and Amplifier Models

υ1 υ2 υ3
i1 ¼ , i2 ¼ , i3 ¼ ð5:26Þ
R1 R2 R3
in terms of input voltages υ1, υ2, υ3. Therefore, voltage υout in Fig. 5.15 found from
Eq. (5.25) is now written in the form


0  υout υ1 υ2 υ3 RF RF RF
iF ¼ ¼ i1 þ i2 þ i3 ¼ þ þ ) υout ¼  υ1  υ2  υ3 ð5:27Þ
RF R1 R2 R3 R1 R2 R3

Example 5.3: An input to the amplifier circuit in Fig. 5.15a is a timing sequence shown in
Fig. 5.15b. Such a sequence is known as a binary counter; it represents all three-bit binary
numbers in an ascending order, with the time interval of 1 μs. The amplifier circuit is
characterized by RF ¼ 2 kΩ, R1 ¼ 40 kΩ, R2 ¼ 20 kΩ, and R3 ¼ 10 kΩ. Plot the absolute
output voltage to scale.
Solution: After plugging in the numbers, Eq. (5.27) is transformed to
jυout j ¼ 0:05υ1 þ 0:1υ2 þ 0:2υ3 ð5:28Þ
Figure 5.15c shows the result. This is a staircase approximation of the straight line.

a) v1 R1 i1 iF RF

v2 i2 0V
R2
-
i3
+ +v
v3 R3 out
-

b) c) Output
Input 2V
5V Vout=2.5e5*t
v1
0V
5V
v2 1V
0V
5V
v3
0V
0 1 2 3 4 5 6 7 8 0V
time, s 0 1 2 3 4 5 6 7 8
time, s

Fig. 5.15. (a) Circuit diagram of a summing amplifier. (b) and (c) Typical input and output
voltages.

A large collection of practical amplifier circuits with the negative feedback exists.
Some of them are DC-coupled amplifiers (considered here), some are intended for ampli-
fication of AC voltage signals with zero mean (the so-called AC-coupled amplifiers).

V-220
Chapter 5 Section 5.3: Amplifier Circuit Design

Section 5.3 Amplifier Circuit Design

Now that the theory of the negative feedback loop has been established, we can turn our
attention to the laboratory. Our hope is to be immediately successful with our designs.
However, a number of questions will arise almost instantly. They raise issues such as how
to choose the resistor values, how to connect the sensor as part of the input load, and how
to use an amplifier chip with a single power supply (a battery).

5.3.1 Choosing Proper Resistance Values

There are several rules regarding how to choose resistances R1, R2 controlling the
feedback loop in both non-inverting and inverting configurations. They are:
1. Resistances R1, R2 cannot be too small. Imagine that in Fig. 5.14 of Section 5.2, the
resistor values are changed to R1 ¼ 1 Ω, R2 ¼ 49 Ω. The same non-inverting gain
will be achieved and the same output voltage will be obtained. However, the feedback
loop current now becomes 100 mA instead of 0.1 mA. The general-purpose op-amp
chips are not capable of delivering such large currents. Furthermore, the ohmic losses
in the feedback loop become high. Therefore, one should generally use

R1 , R2  50  100 Ω ð5:29aÞ

2. Resistances R1, R2 cannot be too large. Let us assume that resistance R2 equals
100 MΩ. This means that this physical resistor and the feedback loop represent almost
an “open circuit.” Unwanted electromagnetic signals may couple into such a circuit
through the related electric field difference across its terminals. This effect is known as
capacitive coupling. Furthermore, the very large resistances increase the parasitic
effect of the input offset current. Plus, very large resistances are unstable—their values
depend on moisture, temperature, etc. Therefore, one should generally use

R1 , R2  1 MΩ ð5:29bÞ

3. When a precision design is not warranted, inexpensive 5 % tolerance resistors may be

used. Otherwise, 1 % or even 0.1 % tolerance resistors are employed. Moreover, in lieu
of fixed resistors, we may use one or two potentiometers to make the gain adjustable.
4. The load resistance should be sufficiently large in order not to overdrive the
amplifier. A good choice is

RL  100 Ω ð5:29cÞ

This requires an output current of 20 mA at υout ¼ 2 V when RL is exactly 100 Ω.

If RL < 100 Ω, the amplifier output voltage may decrease compared to the expected
value due to amplifier’s inability to source/sink sufficient current.

V-221
Chapter 5 Operational Amplifier and Amplifier Models

Example 5.4: A non-inverting amplifier circuit with a gain of 11 is needed in the

configuration depicted in Fig. 5.16. Identify one set of proper resistance values.
Solution:
To satisfy Eqs. (5.29), we simply choose the round numbers

R1 ¼ 1 kΩ, R2 ¼ 10 kΩ, RL ¼ 100 Ω ð5:30Þ

However, other choices are indeed possible. For example, the set

R1 ¼ 100 kΩ, R2 ¼ 1 MΩ, RL ¼ 100 Ω ð5:31Þ

will solve the problem too.

+ +
-100 mV<vin<+100 mV - +
- RL vout

R2 -
R1

Fig. 5.16. A non-inverting amplifier with unknown resistance values.

Discrete Resistance Values and Potentiometers

To achieve a proper gain, we sometimes have to use “strange” resistor values like 49 kΩ
(Fig. 5.14 of Section 5.2). Do such resistors really exist? For 5 % tolerance resistors, they do
not. However, for 1 % tolerance resistors, you can find the standard value of 48.7 kΩ, which
is close to the above value. When the exact resistor values are not available, a potentiometer
can be used as a variable resistor. Furthermore, an externally controlled potentiometer in the
feedback loop is also important when a variable-gain amplifier is needed, for example, for
applications that require automatic gain control.

Gain Tolerance
What about the gain tolerance? The feedback resistor tolerances indeed determine the
gain tolerance. If the resistor tolerance is X, then the gain tolerance is 2X. This result is
valid for both the inverting and the non-inverting amplifier. The corresponding proof uses
an asymptotic expansion for the gain about its unperturbed value. We consider the worst-
case scenario for the inverting amplifier and obtain

V-222
Chapter 5 Section 5.3: Amplifier Circuit Design

R2 ð1 þ X Þ R2 R2
ACL ¼    ð1 þ X Þð1 þ X Þ   ð1 þ 2X Þ f or X << 1 ð5:32Þ
R1 ð1  X Þ R1 R1

The non-inverting gain is treated similarly. For example, if two resistors of an inverting
amplifier circuit are R1 ¼ 1 kΩ, R2 ¼ 100 kΩ and both resistors have 5 % tolerances,
then the amplifier gain is equal to 100 with tolerance of 10 %. Similarly the gain of the
non-inverting amplifier circuit becomes 101 with tolerance of slightly less than 10 %.

5.3.2 Model of a Whole Voltage Amplifier Circuit

Any voltage amplifier circuit with the negative feedback loop may be modeled in a form
similar to Fig. 5.5. The corresponding model is shown in Fig. 5.17.

Amplifier circuit
iin with the feedback loop iout

RS Rout
+ +
vS + vin Rin + ACLvin RL vout
- -
- -

Fig. 5.17. Equivalent model of an amplifier circuit with a negative feedback loop.

First, the open-circuit gain A of the amplifier is replaced by the closed-loop gain
ACL << A. Second, the differential input voltage υx is replaced by υin. Third, resistances
Rin and Rout in Fig. 5.17 now become input and output resistances of the amplifier circuit,
not the amplifier itself. This difference may be quite important in practice.

Input/Output Resistances of Basic Amplifier Circuits Using

Ideal-Amplifier Model
The solution simplifies for the ideal-amplifier model; it is shown in Fig. 5.18. We assume
ideal operational amplifiers in all three cases. If there were no feedback loop, Rin would be
exactly equal to the input resistance of the amplifier itself; this is for an ideal operational
amplifier Rin ¼ 1. When the feedback loop is present, a more general definition should
be used, namely,
υin
Rin  ; ð5:33Þ
iin
where υin is the amplifier’s circuit input voltage and iin is now the current into the
amplifier’s circuit with the feedback loop. Figure 5.18 illustrates the corresponding
calculation for the three basic amplifier types. For both the non-inverting amplifier and
the voltage follower circuits, we have Rin ¼ 1, Rout ¼ 0. However, for the inverting
amplifier circuit, Rin ¼ R1 , Rout ¼ 0, since an input current can still flow into the

V-223
Chapter 5 Operational Amplifier and Amplifier Models

feedback loop but not into the amplifier itself. Thus, the inverting amplifier circuit
potentially provides greater flexibility in the input resistance simply by varying R1. To
ensure the necessary gain, R2 has to be chosen accordingly.

non-inverting op-amp
vin + vout Rout=0
- Rin=vin/iin=inf
iin

R2
R1

R2
inverting op-amp
Rout=0
R1
0V - Rin=vin/iin=R1
vin + vout

iin=Vin/Rin

voltage follower (buffer)

vin + vout
- Rout=0
iin Rin=vin/iin=inf

Fig. 5.18. Input and output resistances of amplifier circuits with the ideal operational amplifier.

5.3.3 Voltage Amplifier Versus Matched Amplifier

A sensor (input load to an amplifier circuit) “sees” an amplifier circuit as a simple
resistance Rin as depicted in Fig. 5.19. The sensor is represented by its Thévenin
equivalent circuit.

iin

RS
+ Amplifier circuit with the feedback
vS + vin Rin
- loop as seen from the sensor
-

Fig. 5.19. Sensor’s equivalent circuit and amplifier’s equivalent circuit as seen from the sensor.

Input Load “Bridging”

The general-purpose operational amplifier is a voltage amplifier (also called a signal
amplifier) and not a power amplifier. The input voltage matters, not the input power. For
the voltage divider circuit in Fig. 5.19, the input voltage υin to the amplifier circuit is
maximized when Rin is maximized. Therefore, Rin must satisfy the inequality

V-224
Chapter 5 Section 5.3: Amplifier Circuit Design

Rin >> RS ð5:34Þ

An appropriate value would be Rin ¼ 100RS , for example. Equation (5.34) is sometimes
called load bridging (or impedance bridging) condition, where the “load” resistance Rin
seen by the Thévenin source is much larger than the source resistance RS. The load
bridging is automatically satisfied for the non-inverting amplifier or for the voltage
follower. For the inverting amplifier, one should use large values of R1, for instance,

R1 ¼ 100RS ð5:35Þ

Bridging connections are used to maximize the voltage transfer from a sensor to an
amplifier. Even more importantly, the amplifier does not appreciably load the sensor.

Example 5.5: A sensor is given by its Thévenin equivalent circuit in Fig. 5.19 where the
sensor voltage υS is small. The sensor’s equivalent resistance RS is 100 Ω. An inverting
amplifier circuit is needed to generate an amplified version of the sensor’s voltage. The
output voltage should be  100υS .
Solution: The corresponding circuit is shown in Fig. 5.20. The input voltage to the
amplifier circuit is computed by voltage division:

R1
υin ¼ υS ð5:36Þ
R1 þ RS

If RS ¼ 100 Ω, R1 ¼ 10 kΩ, then υin  υS and there is almost no loss of voltage signal
strength across resistance RT. Therefore, a pair of resistors with R1 ¼ 10 kΩ, R2 ¼ 1 M
Ω will solve the problem, with the amplifier voltage gain of 100. If, however, we choose
R1 ¼ 0:25RS ¼ 25 Ω, then υin ¼ 0:2υS and 80 % of available voltage signal strength will
be lost! Even if the remaining voltage signal is still appreciable (above the sensitivity
threshold of the amplifier), the necessary amplifier gain becomes not 100, but 500. An
increase in gain leads to an increase in additive voltage noise at the output. Therefore, in
the best case, the amplified signal will be a noisier version of the corresponding signal in
the previous design.

R2
vin R1 vout
-
+
RS=100 0V

vS +
- RL

Fig. 5.20. A 100 Ω sensor connected to the inverting amplifier. Note the common connections.

V-225
Chapter 5 Operational Amplifier and Amplifier Models

Example 5.6: A sensor is given by its Thévenin equivalent in Fig. 5.19 where the sensor
voltage υS is small. The sensor’s equivalent resistance RS may vary in time but is always
less than 100 Ω. An inverting amplifier is needed that generates an amplified version of the
sensor’s voltage, which is  100υS .
Solution: With reference to Fig. 5.20, the input voltage to the amplifier is again given by
Eq. (5.36). This equation is further transformed to

R1
υin ¼ υS  υS ð5:37Þ
R1 þ RS ðt Þ

if we choose R1 ¼ 100maxðRS ðt ÞÞ ¼ 10 kΩ. In other words, not only have we provided

amplification but we also eliminated the effect of the sensor’s resistance variation by proper
load bridging. Therefore, a pair of resistors with R1 ¼ 10 kΩ, R2 ¼ 1 MΩ will solve the
problem, with the amplifier output of  100υS , irrespective of the specific value of RS.

Input Load Matching

And yet, in many modern applications related to radio-frequency (RF) circuits, the load
matching (but not the load bridging) may be a critical condition. RF amplifiers are internally
designed for matching to a precise 50 Ω load at both the input and the output—see Fig. 5.21.
The reason is that voltage and current signals in conductors behave like propagating
electromagnetic waves at high frequencies. If there is no matching, then multiple wave
reflections between the amplifier and its input and/or output loads can occur, resulting in
superimposing the previous signal onto the next signal.

Fig. 5.21. A RF amplifier to be matched to a 50 Ω at both the input and output.

Therefore, the amplifier circuit optimized for proper load matching (which also
achieves maximum power transfer from the input load to the amplifier circuit, see the
generator theorems) may still be critical in many high-frequency applications.

Example 5.7: Construct an amplifier circuit matched to an input source with RS ¼ 50 Ω.

The amplifier’s voltage gain is jACL j ¼ 20. The sign of ACL (either positive or negative) is
not important since an AC input signal is assumed.

V-226
Chapter 5 Section 5.3: Amplifier Circuit Design

Example 5.7 (cont.):

Solution: Two possible solutions are shown in Fig. 5.22. In the first case, we use the
inverting amplifier; in the second case, a smart trick is employed: a non-inverting amplifier
with a 50-Ω shunt resistor. You should note that the maximum power transfer will be
achieved for the entire amplifier circuit, including the shunt resistor.

a) b)
vin
1k vout
vin 50 +
vout
-
+
-

50
19 k
1k

Fig. 5.22. Two possible amplifier configurations matched to a 50-Ω input resistance.

5.3.4 Cascading Amplifier Stages

Assume that we need an inverting amplifier with an overall gain of 1000. The input to
the amplifier is a sensor as in Fig. 5.19 with the equivalent resistance given by RS ¼ 1 kΩ.
If we require load bridging according to Eq. (5.35), we arrive at

R1 ¼ 100RS ¼ 100 kΩ ð5:38Þ

This yields

R1 ¼ 100 kΩ ) R2 ¼ 100 MΩ ð5:39Þ

Such a resistance value is too large to satisfy Eq. (5.29b); it perhaps will not even be
included in your laboratory kit (although the ECE shop may still have such resistors).
What should we do? The answer lies in cascading the amplifier stages as shown in
Fig. 5.23. We use the non-inverting amplifier with a gain of 10 as the first stage; this
simultaneously provides the load bridging condition. We use the inverting amplifier with
a gain of 100 and with the reasonable resistor values as the second stage. The key point
of cascading is to realize that the overall gain of the cascade amplifier is given by the
product (not the sum!) of the individual stage gains, i.e.,

ACL ¼ ACL1  ACL2 ¼ 10  ð100Þ ¼ 1000 ð5:40Þ

V-227
Chapter 5 Operational Amplifier and Amplifier Models

The same result is valid for more than two stages. The proof for two stages is simple:

υout ACL2 υ1out ACL1 ACL2 υin

ACL  ¼ ¼ ¼ ACL1 ACL2 ð5:41Þ
υin υin υin

Stage 1 ACL=10 Stage 2 ACL=-100

1 k =R1 100 k =R2

1
+ -
vin - +
+
vout
vout
-
1 k =R1 9 k =R2

Fig. 5.23. Cascading two amplifier stages into a high-gain circuit. The first stage is a non-inverting
amplifier with a gain of 10; the second stage is an inverting amplifier with a gain of 100. The
overall gain is therefore 1000.

The cascading of individual amplifier stages is a simple and powerful tool to build high-
gain amplifier circuits. Cascading has a number of remarkable features, some of which
are studied here:
1. The gains of the individual stages multiply.
2. The gain per stage generally should not exceed 100 (absolute value) in order to
avoid instability.
3. The first stage in Fig. 5.23 sees R1 of the second stage as its output load resistance.
Therefore, R1 should be large enough.
4. The amplifier ICs usually include two (dual op-amp) or even four (quad op-amp)
individual amplifiers in one package. Therefore, they are ideally suited for building
multistage amplifiers.
5. The effect of an input offset voltage (an amplifier imperfection studied next) is
primarily important for the first stage, but it then loses its significance with every
subsequent stage.
Note that cascading is equivalent to a series combination of individual amplifiers.
Parallel configurations also exist, particularly in analog-to-digital converters.

Example 5.8: The input to the amplifier is a sensor in Fig. 5.19 with an equivalent
resistance given by RS ¼ 100 Ω and an equivalent sensor voltage, υS. An amplifier circuit
is needed that generates ~10,000 υS at its output. The load bridging condition must be
satisfied.

V-228
Chapter 5 Section 5.3: Amplifier Circuit Design

Example 5.8 (cont.):

Solution:
A. A single-stage, non-inverting amplifier with R1 ¼ 100 Ω, R2 ¼ 1 MΩ might do
the job, including load bridging. However, the gain per stage (10,000) is far too high
for stable operation.
B. A series combination of two inverting stages will do a much better job: the first
inverting op-amp assures load bridging, and it consists of
R1 ¼ 10 kΩ, R2 ¼ 1 MΩ. The second inverting amplifier has exactly the same
resistor values: R1 ¼ 10 kΩ, R2 ¼ 1 MΩ. And the overall gain is given by
ACL ¼ ð100Þ  ð100Þ ¼ 10, 000.
C. A series combination of two non-inverting stages will perform equally well: we
choose R1 ¼ 10 kΩ, R2 ¼ 1 MΩ for the first stage and R1 ¼ 10 kΩ, R2 ¼ 1 M
Ω for the second stage. The overall gain then yields ACL ¼
101  101 ¼ 10, 201  10, 000.

5.3.5 Amplifier DC Imperfections and Their Cancellation

In general, DC imperfections of the operational amplifier can have a severe influence on
its performance for high-gain amplifiers. Below, we study two types of imperfections, the
input offset voltage VOS and the input (bias and offset) currents, and provide a simple way
of how to cancel the corresponding output offset voltage.

Input Offset Voltage

The input offset voltage results in a nonzero output voltage when the two input terminals
of the amplifier are shorted out. It arises due to a small asymmetry in the input differential
transistor stage inside the amplifier chip. It is fixed for a certain chip but varies from chip
to chip. General-purpose amplifiers have the input offset voltage VOS in the range of
1 mV  6 mV. The offset can be modeled as a small DC source of strength, VOS, in
series with one of the input terminals to the amplifier as shown in Fig. 5.24. The input
offset voltage produces a similar effect for any amplifier configuration, including the
comparator where its effect becomes most dramatic. A large triangle in Fig. 5.24 indicates
the actual amplifier chip, whereas the small triangle is the ideal amplifier without DC
imperfections. The circuit in Fig. 5.24 is analyzed using the two summing-point con-
straints. Since the negative feedback is present, υx must be zero, which yields
R1 R2
υin þ V OS ¼ υ* ¼ υout ) υout ¼ ACL ðυin þ V OS Þ, ACL ¼ 1 þ ð5:42Þ
R1 þ R2 R1

V-229
Chapter 5 Operational Amplifier and Amplifier Models

actual op-amp
ideal op-amp
VOS
vin +
- +

+
Vx
- - +
vout
v* -
R2
R1

Fig. 5.24. Circuit for the non-inverting amplifier with an input offset voltage.

Exercise 5.7: The amplifier circuit in Fig. 5.24 has the closed-loop gain of ACL ¼ 100 and
an input offset voltage of V OS ¼ 5 mV. What is the general expression for the output
voltage?
Answer:
υout ¼ 100υin þ 0:5 V ð5:43Þ
where 0.5 V is the resulting output offset voltage to the amplifier.

Canceling the Output Offset Voltage

In some amplifiers like the LM741, special offset-null terminals are available for trim-
ming the output DC voltage to zero. Figure 5.25a shows the concept. The wiper of the
potentiometer is to be connected to the negative supply rail. Both inputs to the amplifier
should be connected to the common port during the adjusting procedure, which implies
that the output voltage is trimmed to zero. If the offset-null terminals are not available, the
voltage at the common port, which controls the feedback loop, might be subject to a small
offset. Figure 5.25b shows a circuit that can be used to eliminate the DC imperfections for
a particular non-inverting amplifier in the laboratory. A compensating voltage offset is
introduced by means of an adjustable voltage divider with a potentiometer. To achieve a
good degree of accuracy, one should set

RP << R ð5:44Þ

Furthermore, R1 should be considerably larger than R. This condition can be avoided by a

further modification of the present voltage divider. The output offset voltage is trimmed to
zero when the input to the amplifier in Fig 5.26b is connected to the common port (circuit
ground). This ensures that the effect of the input offset voltage will be eliminated entirely, for
any value of the input voltage. Note that such an adjustment has to be done for every discrete
temperature point, since VOS depends on temperature.

V-230
Chapter 5 Section 5.3: Amplifier Circuit Design

a) b)
LM741 vin +
- +
+VCC vout
-
R2
+
1
- 5
R1

-VCC +VCC
-VCC
10 k
R RP R

Fig. 5.25. (a) Output DC offset voltage for the LM741 is reduced to zero by adjusting the
potentiometer placed between its offset-null pins and (b) a similar operation performed with the
virtual ground of the feedback loop.

Input Bias and Offset Currents

In reality, for the op-amp to operate, there will be very small currents into the input
terminals (into transistor bases), typically on the order of 100 nA. When those currents
flow through the feedback resistances, they create corresponding voltages which appear
as an output offset voltage as well. Let a current of 100 nA flow into the negative terminal
of the amplifier in Fig. 5.24. If the input to the amplifier is grounded, this current must
flow through resistance R2. Therefore, it will create the extra output DC voltage of

υout ¼ þ 100 nA  R2 ð5:45Þ

when the input voltage to the amplifier υin is exactly zero. To appreciate its value, we can
use a resistance R2 ¼ 1 MΩ as an example. This yields

υout ¼ 0:1 V ð5:46Þ

at the output. Fortunately, the currents flowing into the amplifier are nearly the same for
either terminal. Therefore, their average (the input bias current) considerably exceeds their
difference (the input offset current). There is a way to eliminate the larger effect of the
input bias current. It consists of modifying the circuits
 for the non-inverting and inverting
amplifier by adding one extra resistance R ¼ R1 R2 as shown in Fig. 5.26.

a) b)
R=R1||R2 R2
R1
+ -
vin - + +
+
vout vin
vout
- -
R2
R=R1||R2
R1

Fig. 5.26. Cancellation of the effect of input bias currents.

V-231
Chapter 5 Operational Amplifier and Amplifier Models

From the amplifier’s gain point of view, the resistance R has a negligible, if any, effect.
The proof of the cancellation effect for the non-inverting amplifier circuit is as follows.
When both the input and output to the amplifier in Fig. 5.26a are grounded (connected to
the common port), the input current source at the non-inverting input sees resistance
R and the input current source at the inverting
 input sees the parallel combination of R1,
R2, respectively. Making R equal to R1 R2 yields an offset differential voltage that is zero.

5.3.6 DC-Coupled Single-Supply Amplifier: Virtual-Ground Circuit

A single voltage supply is a battery. We consider the non-inverting amplifier circuit
driven by a 9-V battery. To handle the problem of not being able to generate negative
voltages, a virtual-ground circuit may be used as shown in Fig. 5.27. We simply divide
the voltage of the battery by two, with two large, equal resistances R, and assign this
voltage of 4.5 V to the common port. The battery terminal voltages then formally become

4.5 V versus the common port. The power supply so constructed is unfortunately not
exactly the dual-polarity voltage power supply. Namely, both resistances R have to be
large to avoid ohmic losses in the virtual-ground circuit. If they are, we cannot source/
sink an appreciable output current into the common terminal since these resistances
simultaneously operate as current limiters. We can reference the output to ground
(which is the negative terminal of the battery), but not to the virtual ground of 4.5 V.
When referenced to ground, the circuit in Fig. 5.27 has one remarkable property. Namely,
if the input vs. ground is the (small) sensor voltage υin plus 4.5 V, then the output
vs. ground is the amplified sensor voltage, ACLυin, plus the same 4.5 V offset. In other
words, the offset voltage of the virtual ground is not amplified! This statement is proved
in the following example.

sensor 9 V=VCC
4.5 V+vin vout=4.5 V+ACLvin
+
RS -
0V
0V

vS + v*
- R2
R1 R

+ 9 V=VCC
-
4.5 V 4.5 V R

virtual-ground circuit
0V

Fig. 5.27. A non-inverting amplifier driven by a single voltage supply, a battery. Absolute voltages
versus ground (negative terminal of the battery) are shown.

V-232
Chapter 5 Section 5.3: Amplifier Circuit Design

Example 5.9: Solve the circuit in Fig. 5.27.

Solution: The solution is based on the two summing-point constraints. There is no current
into the input terminals so that resistances R1 and R2 again form a voltage divider but now
between υout and 4.5 V. By voltage division, the voltage at node (*) becomes

R1
υ* ¼ 4:5 V þ ðυout  4:5 VÞ ð5:47Þ
R1 þ R2

At the same time, using the second summing-point constraint for the amplifier with
negative feedback, we obtain:

υ* ¼ 4:5 V þ υin ð5:48Þ

Equating Eqs. (5.47) and (5.48), we arrive at the expected result:

R2
υout ¼ 4:5 V þ ACL υin , ACL ¼ 1 þ ð5:49Þ
R1

Feedback resistances R1, R2 should be much larger than resistance R in Fig. 5.27, in order
to assure a flawless circuit operation. Yet another solution is to use (Zener) diodes in the
bias circuit. Special virtual-ground integrated circuits exist that support single-supply
amplifier operation. They generate an output precisely midway between the two supply
rails.

V-233
Chapter 5 Operational Amplifier and Amplifier Models

Section 5.4 Difference and Instrumentation Amplifiers

5.4.1 Differential Input Signal to an Amplifier
Consider a sensing element that is a variable resistance. The sensor configuration is a
Wheatstone bridge (Section 3.2) as seen in Fig. 5.28. Since the amplifier is powered by a
dual supply with three terminals,
V CC , and common (ground) port, the same power
supply is connected to the bridge. We will use the positive rail and the ground rail in
Fig. 5.28 since the sensor (e.g., the strain gauge) may require lower voltages than the
amplifier chip itself.

+VCC

R1 R3
a b
va - vb
+

vD
R2 R(x)

0V

Fig. 5.28. A Wheatstone bridge sensor to be connected to an amplifier circuit.

Both voltages υa, υb have to be used when reading sensor information. They may be
written in terms of the other two voltages υD, υCM:
1 1
υa ¼ υCM þ υD , υb ¼ υCM  υD ð5:50aÞ
2 2
υD ¼ υa  υb , υCM ¼ 0:5ðυa þ υb Þ ð5:50bÞ

Here, υD is the differential component of the combined input signal or the differential
voltage and υCM is the sum component of the combined input signal or the common-mode
voltage. Only the differential voltage is usually important for sensor reading; the
common-mode voltage does not carry any information. When the bridge is exactly
balanced, i.e., when
R1 R3
¼ ; ð5:51aÞ
R2 RðxÞ

the differential voltage is exactly zero:

υD ¼ 0 ð5:51bÞ

Still, the common-mode DC voltage given by

V-234
Chapter 5 Section 5.4: Difference and Instrumentation Amplifiers


R2
υCM ¼ V CC ð5:51cÞ
R2 þ R1

can have any large positive value. For example, it is VCC/2 when all resistances are equal.
Even if the bridge in Fig. 5.28 uses
V CC power rails, it is hardly possible that the
common-mode voltage is set to zero since absolutely identical resistors do not exist.

Exercise 5.8: In the Wheatstone bridge in Fig. 5.28, we assume R2 ¼ 1:1R1 ,

RðxÞ ¼ 1:1R3 , and V CC ¼ 6 V. What are the differential and common-mode voltages?
Answer: υD ¼ 0, υCM ¼ 3:14 V.

5.4.2 Difference Amplifier: Differential Gain and Common-Mode Gain

The sensor in Fig. 5.28 is the differential sensor with three terminals: a, b, and ground.
How do we amplify the differential voltage? Reviewing the inverting and non-inverting
amplifier types reveals that they are not appropriate for this purpose: we simply do not
have two input terminals to be connected to nodes a and b in Fig. 5.28. Only one input
terminal referenced to common (ground) is available. Note that single-ended sensors
with two terminals (plus and ground) indeed exist and may be used. In that case, inverting
or non-inverting amplifiers will function well. However, the overall design accuracy
may deteriorate compared to the differential design. Therefore, a new amplifier type
with two input terminals should be introduced. It is the difference amplifier shown in
Fig. 5.29.

R1 R2
vb v* -
+
va +
R3 v* vout
-
R4
0V
0V
0V

Fig. 5.29. A difference amplifier.

Both inputs to this amplifier are referenced to common (ground) port. First, we see that
the difference amplifier is an inverting amplifier with the negative feedback loop. However,
the second input signal is now added to its positive terminal through a voltage divider. The
analysis of this amplifier type is done using two summing-point constraints. A shortcut is to
recognize, with the help of the first SPC, that in Fig. 5.29 we have two voltage dividers: one
between resistances R1, R2 and another between resistances R3, R4. Therefore, for the
voltage at node (*), one has, using the first voltage divider,

V-235
Chapter 5 Operational Amplifier and Amplifier Models

R1
υ* ¼ υb þ ðυout  υb Þ ð5:52aÞ
R1 þ R2

and, according to the second voltage divider,

R3
υ* ¼ υa þ ð0 V  υ a Þ ð5:52bÞ
R3 þ R4

Both expressions must be equal to each other due to the second SPC (the differential input
voltage in a negative feedback amplifier is zero). Therefore,
R1 R3
υb þ ðυout  υb Þ ¼ υa  υa )
R1 þ R2 R3 þ R4


 ð5:52cÞ
R1 R1 R3
υout ¼  1 υb   1 υa
R1 þ R2 R1 þ R2 R3 þ R4

To create a voltage difference, i.e., υa  υb , between the input voltages on the right-hand
side of Eq. (5.52c), we select
R2 R4
¼ ð5:52dÞ
R1 R3

as the necessary condition. Then, both factors in parentheses on the right-hand side of
Eq. (5.52c) become equal. This yields the basic equation of the difference amplifier,
R2 R2
υout ¼ ðυ a  υ b Þ ¼ υ D ð5:53Þ
R1 R1

Equation (5.53) is a simple, yet highly useful result for amplifier circuit design. Namely,
once the amplifier in Fig. 5.29 is connected to the sensor in Fig. 5.28, the differential
voltage υD ¼ υa  υb is amplified with the gain of R2/R1 (the differential amplifier circuit
gain). At the same time, the undesired common-mode voltage υCM ¼ 0:5ðυa þ υb Þ is
completely rejected, i.e., amplified with a gain of 0, no matter what specific values the
input voltages have versus ground. In other words, the common-mode amplifier circuit
gain is zero. Note that the ratio of two gains (differential gain versus common-mode gain)
is an important characteristic of the difference-amplifier circuit. It is called the common-
mode rejection ratio (CMRR). In our case, this ratio is clearly infinity. Unfortunately, in
reality, this value is finite though quite large. One obvious reason is a possible mismatch
in resistance ratios in Eq. (5.52d), which will not allow us to obtain Eq. (5.53) exactly.
A certain portion of υCM ¼ 0:5ðυa þ υb Þ will be present at the output.

V-236
Chapter 5 Section 5.4: Difference and Instrumentation Amplifiers

Example 5.10: Find the output voltage of the amplifier circuits shown in Fig. 5.30 below.
Assume the ideal amplifier and exact resistance values.
Solution: We check Eq. (5.52d) first and conclude that the circuit in Fig. 5.30 is a true
difference amplifier: it rejects the common-mode voltage. The differential voltage to the
amplifier is 0.01 V. Using Eq. (5.53) gives us an output voltage of υout ¼ 0:1 V. If the
resistance ratios were not equal to each other, a common-mode signal would be present at
the output. In that case, the complete amplifier equation (5.52c) should to be used.

10 k 100 k
1.01 V -
+
1V +
10 k
vout
-
100 k
0V
0V
0V

Fig. 5.30. Difference amplifier for Example 5.10.

To minimize the effect of bias currents (Section 5.3), we should choose

R1 ¼ R3 , R2 ¼ R4 ð5:54Þ

5.4.3 Application Example: Instrumentation Amplifier

Motivation for an Instrumentation Amplifier
Figure 5.31 shows a 700-Ω uniaxial strain gauge. The strain gauge is attached to an
aluminum slab. Figure 5.31 also shows a Wheatstone bridge intended for strain measure-
ment with the present device. The dual-polarity supply voltage is
7.5 V. You may build
the Wheatstone bridge according to Fig. 5.31 and connect it to a DMM, with the DMM
leads attached to terminals a and b, respectively. In this configuration, the DMM
measures the differential voltage υD. Using the potentiometer, you may balance the
bridge, i.e., reduce voltage υD at no strain to a minimum, which should be within the
range 0 V
3 mV. The highest DMM resolution should be used. Now, by applying a
strong bending force to the slab with two hands, you probably could obtain a maximum
voltage change of
2 mV.

V-237
Chapter 5 Operational Amplifier and Amplifier Models

7.5 V=VCC

10 k 1k
potentiometer
a b
va - vb

+
vD
10 k
700 strain gage

0V

Fig. 5.31. A Wheatstone bridge sensor where the strain gauge forms one of the four resistors.

The positive voltage change corresponds to one bending direction, the negative change,
to the opposite direction. Next, you may connect an oscilloscope instead of the DMM and
use the DC-coupled settings and the highest voltage resolution of 20 mV per division. You
will find that the noisy signal line on the screen hardly changes when you try your best. So
are you not strong enough? Well, no. These are the realistic differential voltages for strain
gauges which correspond to gauge resistance changes on the order of 1 Ω. Even smaller
voltage changes are often encountered in practice. Therefore, an accurate amplification of
an extremely small differential voltage should be done while rejecting the large common-
mode voltage (3.75 V in the present case). This nontrivial task is accomplished by an
instrumentation amplifier.

How to Build an Instrumentation Amplifier?

The initial guess is probably to use the difference amplifier from Fig. 5.30. For this
amplifier, R1 ¼ R3 ¼ 10 kΩ, R2 ¼ R4 ¼ 100 kΩ. However, we encounter two prob-
lems. The first one is that the amplifier should not perturb the sensor operation. In other
words, the amplifier’s input resistance must be large compared to any of the resistances in
the Wheatstone bridge. The differential-mode input resistance to the amplifier in Fig. 5.30
is Rin ¼ 2R1 ¼ 2R3 (check problem 5.84 at the end of this chapter). Therefore, we may
wish to increase R1 ¼ R3 by a factor of 10, i.e., choose R1 ¼ R3 ¼ 100 kΩ. The second
problem is the amplifier gain. An overall gain of 1000 is required in Eq. (5.53) in order to
obtain appreciable output voltages on the order of
2 V. This gain is too high;
prohibitively large resistor values R2 ¼ R4 ¼ 100 MΩ would be needed in Fig. 5.30.

Concept of an Instrumentation Amplifier

Thus, the solution for at least one problem is clear: we need to add an extra amplifier
stage. One way of doing so is shown in Fig. 5.32. Two non-inverting amplifiers are added
to both inputs of the difference amplifier. This design achieves two goals simultaneously.
First, it isolates the amplifier circuit from the Wheatstone bridge since the non-inverting
amplifiers have an infinite input resistance and do not sink any current from the bridge.
Second, it adds the extra gain; in other words, it eases the burden on the difference

V-238
Chapter 5 Section 5.4: Difference and Instrumentation Amplifiers

amplifier in the second stage. The difference amplifier becomes mainly responsible for
rejecting the common-mode signal and the amplification of the differential signal.

vb +
-
R2

R4 R1
R3
vb* -
+
va* +
R3 vout
R4 R1
0V -
R2
-
va + 0V

Fig. 5.32. An important step toward the instrumentation amplifier: we add non-inverting ampli-
fication stages at the input.

However, another problem arises there. The two non-inverting amplifiers amplify
voltages υa, υb (close to 3.75 V in the present case). Therefore, at any appreciable gain
(say, ACL ¼ 1 þ R4 =R3 ¼ 10), they simply saturate and will not function! To avoid this
issue, we use a simple yet critical change shown in Fig. 5.33 where we remove the
common-port connection from the non-inverting stage. The circuit in Fig. 5.33 behaves
completely differently compared to the original circuit in Fig. 5.32. We no longer have the
output voltages υ a , υ b given by
R4
υa * ¼ ACL υa , υa * ¼ ACL υa , ACL ¼ 1 þ ð5:55Þ
R3

Instead, those voltages now become

R4 R4
υa * ¼ υa þ ðυa  υb Þ, υb * ¼ υb  ðυ a  υ b Þ ð5:56Þ
2R3 2R3

The details of the derivation are seen in Fig. 5.33. The currents and voltages labeled in
this figure are obtained using two summing-point constraints. The key observation is that
the absolute voltages υa, υb are no longer amplified but are simply passed through. Only
the differential voltage υD ¼ υa  υb is amplified. The circuit in Fig. 5.33 is also a
“difference amplifier,” and it may be called the unity common-mode gain stage. The
final step in the construction of the instrumentation amplifier is to connect both stages
together. Figure 5.34 gives the final circuit that can be employed in conjunction with the
Wheatstone bridge for the strain gauge shown in Fig. 5.31. Here, a quad op-amp chip
(LM148 series) is used; it has four individual amplifiers inside the chip. For our circuit,
we need three of them. The circuit is powered by a
7:5 V dual supply. According to
Eqs. (5.53) and (5.56), the overall (differential) gain in Fig. 5.34 becomes

V-239
Chapter 5 Operational Amplifier and Amplifier Models

vb-R4i
vb +
-
+ -
vb
R4
R3
va-vb vb*
2R3 = i va*
R3
R4
va
- +
-
va +
va+R4i

Fig. 5.33. Next step toward the instrumentation amplifier: we convert the non-inverting stage to a
unity common-mode gain amplifier.




R2 R4 R2 R4 R2 R4
υout ¼ 1þ ðυ a  υ b Þ ¼ 1þ υD , ACL ¼ 1þ ð5:57Þ
R1 R3 R1 R3 R1 R3

The overall common-mode gain is exactly zero (ideal resistances). Choosing resistance
values from Fig. 5.34 gives us an overall differential gain of 1010. We retain a certain
gain (10) of the differential stage in order to have the gain of no more than 100 per stage.

3 pin assembly:
+ 1 LM148/LM248/LM348
7.5 V=VCC vb -
2
1k 100 k
10 k
potentiometer
100 k 10 k
va a b vb 1k 9
- vout
8
- +
+

vD
+
1k 10
10 k 700 straing gage
100 k 10 k -
100 k

0V va
6 - 7
+
5

Fig. 5.34. The complete instrumentation amplifier for the strain gauge testing.

5.4.4 Instrumentation Amplifier in Laboratory

The operation of the circuit in Fig. 5.34 is shown in Fig. 5.35. The oscilloscope resolution
is 1 V per division. At no applied strain, the oscilloscope connected to the amplifier
output (the ground terminal of the oscilloscope is connected to circuit ground) shows a
relatively small output voltage signal. When a bending moment is applied, as shown in
Fig. 5.35a, the output voltage rises to approximately 2 V. This voltage is sufficient to light
up a yellow LED (light-emitting diode) indicator connected between the output port and
common port. When the opposite bending moment is applied, as in Fig. 5.35b, the output
voltage drops to approximately –2 V. This voltage, taken with a negative sign, is again
sufficient to light up a red LED connected from the common port to the output port.

V-240
Chapter 5 Section 5.4: Difference and Instrumentation Amplifiers

The states in Fig. 5.35 have been achieved by a proper tuning of the potentiometer in
the Wheatstone bridge. Thus, we have built a simple, yet useful, uncalibrated, uniaxial,
stress-monitoring system. Frequently, the output of an instrumentation amplifier is
connected to an analog-to-digital converter (ADC) and then to a computer system.

Fig. 5.35. Operation of the instrumentation amplifier with the strain gauge attached to a metal slab.
(a) A “positive” bending moment is applied and (b) a “negative” bending moment is applied. The
oscilloscope resolution is 1 V per division in every case.

Load Cell and Other Uses

The circuit in Fig. 5.34 gives us the idea of a commercial load cell. Strain gauges are
commercially available in prefabricated modules such as load cells that measure force,
tension, compression, and torque. All four resistors of the Wheatstone bridge may be
strain gauges. Load cells typically use a full-bridge configuration and contain four leads
for bridge excitation and measurement. The manufacturers provide calibration and
accuracy information. However, the load cells do not normally include the instrumenta-
tion amplifier itself. Another mechanical engineering example where the differential
amplifier is quite useful is a thermocouple. When measuring a thermocouple in a noisy
environment, the noise from the environment appears as an offset on both input leads,
making it a common-mode voltage signal. Many other examples indeed exist, particularly
in biomedical engineering. The instrumentation amplifier is used to amplify an output
signal from virtually any analog differential sensor instrument. Also note that instrumen-
tation amplifiers with precision resistors are available as separate integrated circuits.
Those ICs have a much better performance than an instrumentation amplifier wired on
the protoboard. Other instrumentation amplifier types exist, which are different from the
topology of the instrumentation amplifier circuit in Fig. 5.34. In principle, it is possible
to design an instrumentation amplifier circuit with only two amplifier gain stages. The
summary to this chapter provides an example used in practice.

V-241
Chapter 5 Operational Amplifier and Amplifier Models

Section 5.5 General Feedback Systems

5.5.1 Signal-Flow Diagram of a Feedback System
Although the negative feedback was first quantified by electrical engineers, it is extensively
observed, studied, and employed in many mechanical, biomedical, chemical, and other
systems. Figure 5.36 shows a generic structure of a linear feedback system.

xin xe xout
+
Source A
-

xf xout

Fig. 5.36. A negative feedback loop for an arbitrary system (signal-flow diagram).

Figure 5.36 is a simplified signal-flow diagram or controls block diagram. The

variable x is a signal; it may be voltage, current (electrical engineering) or displacement,
velocity (mechanical engineering), etc. These three blocks have special names in control
theory:
1. The first block (basic voltage amplifier in terms of ECE) is called the forward or
open-loop gain A (often denoted by AOL). It operates according to the ideal-
amplifier rule

xout ¼ Axe , A ¼ const > 0 ð5:58Þ

2. The second block is called the feedback gain. It operates according to a linear rule

xf ¼ βxout , β ¼ const > 0 ð5:59Þ

where β is called the feedback factor.

3. The block where the input signal and feedback signal are compared and subtracted
from one another is called the summing (or difference) node.

5.5.2 Closed-Loop Gain and Error Signal

Mathematically, from Eq. (5.59), one has

xe ¼ xin  xf ¼ xin  β xout ð5:60Þ

According to Eqs. (5.58) and (5.60), we obtain

xout ¼ Axe ¼ Aðxin  β xout Þ ð5:61Þ

Solving for xout gives us the closed-loop gain ACL (sometimes denoted by G or Af)

V-242
Chapter 5 Section 5.5: General Feedback Systems

A A
xout ¼ xin ) ACL ¼ ð5:62Þ
1 þ Aβ 1 þ Aβ

If the open-loop gain A is made arbitrarily large (ideally approaching 1), then the closed-
loop gain approaches
1
ACL  ð5:63Þ
β

The significance of Eq. (5.63) cannot be overstated. It means that as long as the open-loop
gain A is large enough, the closed-loop gain ACL will approach the constant value 1/β,
which is precisely controlled by an external passive feedback network. In other words,
manufacturing uncertainties in A and the potential nonlinear behavior of A are eliminated
since A itself is eliminated. The price for this operation is a significant reduction of the
overall system gain. Equation (5.63) implies that
1
Aβ >> 1 ) ACL ¼ << A ð5:64Þ
β

Clearly, the feedback loop reduces the initial gain substantially. And yet, despite this
drawback, the closed-loop gain ACL may still be large enough and sufficient for ampli-
fication as long as A is made very large. Thus, the feedback loop in Fig. 5.36 is a simple
and powerful means to control the operation of an arbitrary high-gain system.

Exercise 5.9: The open-loop gain A in Fig. 5.36 varies between two extreme values of
A ¼ 1000
200 (
20 % gain variation) depending on the system parameters. The forward
gain block is used in the closed-loop configuration with the feedback factor β of 0.1.
Approximate the two extreme values of the closed-loop gain, ACL.
Answer: ACL ¼ A=ð1 þ AβÞ ¼ 9:90
0:02 or
0:2% closed-loop gain variation.

Error Signal
The second question of interest is finding the error signal, xe, in Fig. 5.36, which
corresponds to the differential input voltage for an amplifier circuit with the negative
feedback. Substitution of Eq. (5.62) into Eq. (5.60) yields
A 1
xe ¼ xin  β xin ¼ xin ð5:65Þ
1 þ Aβ 1 þ Aβ

When the open-loop gain A is large and furthermore Aβ >> 1, one obtains

xe  0 ð5:66Þ

V-243
Chapter 5 Operational Amplifier and Amplifier Models

As applied to the amplifier circuits, Eq. (5.66) is exactly the second summing-point
constraint or the condition of the zero differential input amplifier voltage under presence
of the negative feedback.

Exercise 5.10: The open-loop gain A in Fig. 5.36 is 10,000. The forward gain block is
used in the closed-loop configuration with the feedback factor β of 0.1. Determine the error
signal xe if the input voltage signal is 1 mV.
Answer: xe ¼ 0:999 μV  1 μV.

5.5.3 Application of General Theory to Voltage Amplifiers with Negative

Feedback
Two circuits shown in Fig. 5.37 are the buffer amplifier circuit and the non-inverting
amplifier circuit, respectively. The goal is to find the closed-loop gain ACL of the amplifier
circuit when the given open-circuit voltage gain A is large but finite. In this case, the
second summing-point constraint cannot be applied. Therefore, simple Eqs. (5.16) and
(5.21) obtained previously need to be modified.
a) b)
+ +
vin vout - vout vin vout - vout

R2
R1

Fig. 5.37. Two amplifier circuits with negative feedback networks indicated by a shaded rectangle.

Both circuits from Fig. 5.37 have the form of the feedback system as in Fig. 5.36. The
feedback network is indicated by a shaded rectangle. The signal x is now the voltage.
Since A is given, the only problem is to find the feedback factor, β. For the buffer
amplifier, the feedback factor is clearly one. For the non-inverting amplifier circuit, the
feedback loop is the voltage divider, where β is determined by the resistance ratio. Note
that the voltage divider model implies no current into amplifier’s input terminals.
Specifically, βυout is equal to R1 =ðR1 þ R2 Þυout . Therefore,
R1
β ¼ 1  buffer ampl: circuit; β¼  non-inv: ampl: circuit ð5:67Þ
R1 þ R2

Substitution into Eq. (5.62) gives us two expressions for the closed-loop gain:
A A
ACL ¼  buffer ampl: circuit; ACL ¼  non-inv: ampl: circuit ð5:68Þ
1þA 1 þ AR RþR
1
1 2

V-244
Chapter 5 Section 5.5: General Feedback Systems

The first equation (5.68) coincides with Eq. (5.23) obtained in Section 5.2 using the
accurate circuit analysis. So does second equation (5.68) when we repeat the same
analysis for the non-inverting amplifier configuration. If A ! 1, then the simple gain
expressions—Eqs. (5.16) and (5.21)—derived with the help of the second summing-point
constraint are obtained from Eqs. (5.68). The analysis of the inverting amplifier requires
more efforts since this amplifier type is not exactly the voltage amplifier but rather a
transresistance amplifier considered next.

Exercise 5.11: The open-loop (open-circuit) gain A of a non-inverting amplifier circuit

with R1 ¼ 1 kΩ, R2 ¼ 9 kΩ is 10,000. Determine the closed-loop gain.
Answer: ACL ¼ 9:99, which is by 0.1 % different from ACL ¼ 1 þ RR21 ¼ 10.

Last but not least, we emphasize another significant advantage of the negative feed-
back. When the input and output resistances of the amplifier model in Fig. 5.5 have finite
values (which occurs in practice), the negative feedback loop effectively increases the
input resistance and decreases the output resistance, i.e., makes the entire amplifier circuit
look closer to the ideal-amplifier model.

5.5.4 Voltage, Current, Transresistance, and Transconductance Amplifiers

with the Negative Feedback
At the end of this short section, we consider four basic amplifier circuits with negative
feedback, which correspond to the four basic dependent sources studied in Chapter 2:
voltage amplifier, transconductance amplifier, transresistance amplifier, and current
amplifier. Figure 5.38 shows the corresponding circuit diagrams. Load resistance RL is
introduced for the transconductance amplifier and the current amplifier, respectively,
where the output is the load current. Although every amplifier circuit may be represented
in the form similar to the feedback diagram in Fig. 5.36 and analyzed accordingly, only a
simplified treatment will be given here. It utilizes the condition A ! 1 and the resulting
second summing-point constraint. Using the both summing-point constraints, we may
obtain, with reference to Fig. 5.38,


R2
υout ¼ 1 þ υin voltage amplifier ð5:69aÞ
R1
iout ¼ GF υin , GF ¼ 1=RF transconductance amplifier ð5:69bÞ
υout ¼ RF iin transresistance amplifier ð5:69cÞ

V-245
Chapter 5 Operational Amplifier and Amplifier Models


R2
iout ¼ 1þ iin current amplifier ð5:69dÞ
R1

Note that other more elaborate circuits may be considered; some of them are analyzed in
the corresponding homework problems.

a) voltage amplifier b) transconductance amplifier

+ +
- vout -
RL iout
vin + R2 vin +
- -
R1
RF

c) transresistance amplifier d) current amplifier

R1
RF R2

- - RL iout
+ +
vout

iin iin

Fig. 5.38. Four basic amplifier types with negative feedback.

V-246
Chapter 5 Problems

Summary
Amplifier circuit Operation Formulas
Open-loop operational amplifier Operation with
V CC power rails:
(comparator)
υout ¼ Aðυþ  υ Þ, jυout j < V CC
Open-circuit (open-loop) gain A is
very high

Amplifier circuit model – Valid for any voltage amplifier or

voltage amplifier circuit, but Rin,
Rout, A are different in every case;
– Ideal amplifier model (useful sim-
plification):
Rin ¼ 1, Rout ¼ 0, A ¼ 1

Negative feedback for the ideal-amplifier model: differential input voltage is zero (2nd SPC)
Non-inverting amplifier For ideal-amplifier model:
R2
υout ¼ ACL υin , ACL ¼ 1 þ
R1
Rin ¼ 1, Rout ¼ 0

1
R1
Exact: ACL ¼ A 1 þ A
R1 þ R2

Voltage follower (buffer) For ideal-amplifier model:

amplifier υout ¼ ACL υin , ACL ¼ 1
Rin ¼ 1, Rout ¼ 0
A
Exact: ACL ¼
1þA

Inverting amplifier For ideal-amplifier model:

R2
υout ¼ ACL υin , ACL ¼ 
R1
Rin ¼ R1 , Rout ¼ 0
Exact:


R2 R2 1
ACL ¼ A Aþ1þ
R1 R1
(continued)

V-247
Chapter 5 Operational Amplifier and Amplifier Models

Cascaded amplifier – Gains of individual stages multiply;

– Input resistance of the amplifier
circuit is the input resistance of
stage 1;
– Individual stage gain should not
exceed 100

Input load bridging versus input load matching

Load bridging – Source (sensor) sees the amplifier as
an open circuit:
υout ¼ υS  ACL
– No current from the source can flow
into amplifier circuit

Load matching – Source (sensor) sees the amplifier as

resistance Rin ¼ R1 :
Rin
υout ¼ υS  ACL
RS þ Rin
– Matching condition RS ¼ Rin is
important for high-freq. circuits

DC imperfections and their cancellation

Non-inverting amplifier – Short-circuited output voltage may
be trimmed to zero using the offset
null terminal
– Short-circuited output voltage may
be trimmed to zero by adjusting
common-terminal voltage;
– Extra resistance R eliminates the
effect of the input bias current
Inverting amplifier The same as above

(continued)

V-248
Chapter 5 Problems

Multiple-input amplifier circuits

Summing amplifier – The summing amplifier sums sev-
eral weighted input voltages:
RF RF RF
υout ¼  υ1  υ2  υ3
R1 R2 R3
– Used as a prototype of the digital
to analog converter
Difference amplifier – True difference amplifier:
R2 R4 R2
¼ , υout ¼ ðυa  υb Þ
R1 R3 R1
– Rejects common-mode voltage
– For the general difference amplifier
circuit see Eq. (5.52c)
Unity common-mode gain stage – Differential gain:
υa *  υb * ¼ AD ðυa  υb Þ
R4
AD ¼ 1 þ
R3
– Common-mode gain:
υa * þ υb * ¼ ACM ðυa þ υb Þ
ACM ¼ 1
Instrumentation amplifier Output voltage:


R2 R4
υout ¼ 1þ ðυa  υb Þ
R1 R3
Closed-loop differential gain:


R2 R4
ACL ¼ 1þ
R1 R3
– Rejects common-mode voltage
Simpler instrumentation Output voltage:
amplifier


R2 R2
υout ¼ 2 þ 1 þ ðυa  υb Þ
RG R1
– Closed-loop differential gain:


R2 R2
ACL ¼ 2 þ 1 þ
RG R1
– Rejects common-mode voltage

(continued)

V-249
Chapter 5 Operational Amplifier and Amplifier Models

General feedback systems and amplifiers with negative feedback

Signal—flow diagram Closed-loop gain:
A 1
xout ¼ ACL xin , ACL ¼ 
1 þ Aβ β
Error signal:
1
xe ¼ xin  0
1 þ Aβ
Transconductance amplifier Closed-loop operation:
iout ¼ GF υin , GF ¼ 1=RF
Variations of this simple circuit are
possible

Transresistance amplifier Closed-loop operation:

υout ¼ RF iin
Variations of this simple circuit
are possible

Current amplifier Closed-loop operation:


R2
iout ¼ 1 þ iin
R1
Variations of this simple circuit
are possible

Howland amplifier (current pump) Closed-loop operation:

(R1 =R3 ¼ R2 =R4 )
iout ¼ GF ðυa  υb Þ, GF ¼ 1=R2
Variations of this clever circuit
are possible

V-250
Chapter 5 Problems

Problem 5.3. What is the minimum number of

Problems pins required for:
5.1 Amplifier operation A. The dual operational amplifier (the
corresponding IC package contains two
and circuit models separate operational amplifiers)?
5.1.1 Amplifier Operation B. The quad operational amplifier (the
Problem 5.1. An operational amplifier has five corresponding IC package contains four
terminals. separate operational amplifiers)?
A. Sketch the amplifier symbol.
B. Name each of the op-amp terminals and Problem 5.4. An operational amplifier has an
describe its function in one sentence per open-circuit gain of A ¼ 2  105 and is powered
terminal. by a dual source of
10 V. It is operated in the
C. Can the amplifier IC have more than five open-circuit configuration. What is the ampli-
terminals? Explain. fier’s open-circuit output voltage υout if
A. υþ ¼ 0V, υ ¼ 0V
Problem 5.2. You may wonder about the mean- B. υþ ¼ þ1V, υ ¼ þ1V
ing of the two letters preceding amplifier mark- C. υþ ¼ þ1V, υ ¼ 0V
ing, e.g., LM741. Each of the semiconductor D. υþ ¼ 0V, υ ¼ 1V
companies has its own abbreviation, e.g., LM E. υþ ¼ þ1mV, υ ¼ 0V
for an amplifier designed and manufactured by F. υþ ¼ 1mV, υ ¼ 0V
the National Semiconductor Corporation G. υþ ¼ 10μV, υ ¼ 0V
(acquired by Texas Instruments in 2011), AD H. υþ ¼ 0V, υ ¼ 10μV
for an amplifier manufactured by Analog
Devices, MC for STMicroelectronics, TL for Problem 5.5. Based on the solution to Problem
Texas Instruments, etc. The same chip, e.g., 5.4, why do you think the operational amplifier
LM741, may be manufactured by several semi- is seldom used in the open-loop configuration,
conductor chip makers. The part number is at least in analog electronics?
given by a numerical code that is imprinted on
the top of the package. An MC1458 amplifier Problem 5.6. Using the website of the National
IC chip is shown in the figure below. This IC is Semiconductor Corporation, determine the
a dual operational amplifier. In other words, maximum and minimum supply voltages (oper-
one such IC package contains two separate ating with the dual-polarity power supply) for
operational amplifiers. the following amplifier’s ICs:
A. LM358
B. LM1458
C. LM741
Which amplifier IC from the list may be
powered by two AAA batteries?
A. Download the amplifier’s datasheet from Problem 5.7. Plot to scale the output voltage of
http://www.datasheetcatalog.com the operation amplifier with an open-circuit gain
B. Redraw the figure to this problem in A ¼ 5  104 when the non-inverting input volt-
your notes and label the pins for the age υþ changes from 2 mV to +2 mV and the
non-inverting input, the inverting input, inverting input voltageυ is equal to 1 mV. The
and the output of the operational amplifier amplifier is powered by a
12-V dual voltage
#1. supply. Label the axes.
C. Label pins for þV CC and V CC .

V-251
Chapter 5 Operational Amplifier and Amplifier Models

Problem 5.8. Repeat the previous problem for Problem 5.10. Based on the solution to the
A ¼ 5  105 . previous problem, why do you think the oper-
ational amplifier in the open-loop configuration
5.1.2 Operational Amplifier may be useful for digital circuits?
Comparator
Problem 5.9. In a circuit shown in the figure Problem 5.11. Solve Problem 5.9 when the
below, an operational amplifier is driven by a input voltage is applied to the inverting input

10-V dual power supply (not shown). The and the threshold voltage is applied to the
open-circuit DC gain of the amplifier is non-inverting input.
A ¼ 1, 000, 000. Sketch to scale the output
voltage to the amplifier when Problem 5.12. The circuit shown in the figure
a) V threshold ¼ 30 mV is a zero-level detector. An operational ampli-
b) V threshold ¼ þ30 mV fier in the open-loop configuration is driven by
a
10-V dual power supply (not shown). The
Assume that the amplifier hits the power rails in open-circuit amplifier gain is 100,000. Sketch
saturation. the output voltage to scale. Assume that the
amplifier hits the power rails in saturation.
A=1,000,000
vout
v+
A =100,000
+ v out
- +
Vthreshold v in
-
0V
Input voltage v+, mV

80
Input and output voltages, V
60 15
v in
40 10

20 5

0 0

-20 -5
-10
-40
0 10 20 30 40 50 -15
time, ms 0 10 20 30 40 50
time, ms
a)
Output voltage vout, V Vthreshold=-30mV
15
Problem 5.13. In a circuit shown in the figure
10
5
below, an operational amplifier is driven by a

15-V dual power supply (not shown). The
0
-5
open-circuit gain of the amplifier is
-10 A ¼ 100, 000. Sketch to scale the output volt-
-15 age to the amplifier when
a) V threshold ¼ 0 mV
0 10 20 30 40 50
time, ms
b) b) V threshold ¼ þ4 mV
Output voltage vout, V Vthreshold=+30mV
15
10
5

0
-5
-10
-15
0 10 20 30 40 50
time, ms

V-252
Chapter 5 Problems

Assume that the amplifier hits the power rails in A. Rin ¼ 100 kΩ and Rout ¼ 2 Ω.
saturation. B. Rin ¼ 50 Ω and Rout ¼ 25 Ω.
C. Rin ¼ 1 and Rout ¼ 0.
A =100,000
vout
v+
+
Vthreshold
- iin iout

RS Rout RL
Input voltage v+, mV
0V + +
8 + vS vx Rin + Avx vout
- -
6 - -
4
2
0

-2
-4
Problem 5.16. Name one reason why we
0 10 20 30 40 50 should attempt to:
time, ms
a)
A. Make the input resistance (impedance)
Output voltage vout, V Vthreshold=0 mV to the amplifier as high as possible.
15
B. Make the output resistance (impedance)
10
to the amplifier as low as possible.
5

0
-5
Problem 5.17
-10
A. List all conditions of the ideal-amplifier
-15 model.
0 10 20 30 40 50
time, ms
B. What is the short-circuit output current
b) of the ideal amplifier?
Output voltage vout, V Vthreshold=+4 mV
15 Problem 5.18
10
A. What is the first summing-point constraint?
5
B. What is the equivalent formulation of the
0
first summing-point constraint in terms
-5
-10
of the input resistance (impedance) to the
-15 amplifier?
0 10 20 30 40 50
time, ms

Problem 5.19. An amplifier circuit is shown in

5.1.3 Amplifier Circuit Model the figure below. The first summing-point con-
straint applies. Determine current i2.
5.1.4 Ideal-Amplifier Model
Problem 5.14 R2=500 W

A. Draw the circuit model of an operational R1=100 W i2

amplifier. - v out
1V 0V
+
B. Describe the meaning of the amplifier as
the voltage-controlled voltage source in
your own words.
common

Problem 5.15. For an equivalent amplifier cir-

cuit with A ¼ 1500 shown in the figure below, Problem 5.20. An amplifier circuit is shown in
determine the output voltage given that the figure below. The first summing-point con-
υS ðt Þ ¼ 1 cos ω t ½mV
, RS ¼ 50 Ω, RL ¼ 50 straint applies. An ideal operational amplifier
Ω for three cases: has an open-circuit gain of A ¼ 2  105 .

V-253
Chapter 5 Operational Amplifier and Amplifier Models

Determine the output voltage, Vout. You are not 5.2 Negative Feedback
allowed to use any of the materials of the next
section! 5.2.2 Amplifier Feedback Loop. Second
Hint: Denote the unknown voltage at node * by Summing-Point Constraint
υ*, express υ* in terms of υout, and then solve 5.2.3 Amplifier Circuit Analysis Using
for υout. Two Summing-point Constraints
R2=500 Problem 5.23
A. Name the two summing-point con-
R1=100 i2
- v out straints used to solve an amplifier circuit.
1V
+ B. Which summing-point constraint remains
i1 valid without the negative feedback?

common
Non-inverting Amplifier
Problem 5.24
Problem 5.21. An ECE laboratory project uses A. Draw the circuit diagram of the basic
the LM358 amplifier IC. non-inverting amplifier configuration.
A. What semiconductor company has B. Accurately derive the expression for the
developed this chip? amplifier gain in terms of the resistances,
B. Is the chip from the lab project necessarily assuming an ideal operational amplifier.
manufactured by this company? (See
http://www.datasheetcatalog.com/ for Problem 5.25. Using the two summing-point
manufacturers’ datasheets related to this constraints, solve the ideal-amplifier circuit
product.) shown in the figure if the input voltage has the
C. Use the Digi-Key distributor’s website and value of 2 mV.
estimate average cost for this amplifier A. Label and determine the currents in the
chip (DIP-8 package) in today’s market. feedback loop.
B. Determine the output voltage of the
Problem 5.22. An ECE laboratory project uses amplifier versus the common port.
the TL082 amplifier IC.
A. What semiconductor company has vin pin 5 pin layout - LM1458(#2)
vout
+ + pin 7
developed this chip? + vx
2 mV - +
B. Is the chip from the lab project necessarily -
- pin 6
-
manufactured by this company? (See
http://www.datasheetcatalog.com/ for R2=51 kW
common
manufacturers' datasheets related to this R1=100 W
product.)
C. Use the Digi-Key distributor’s website to common

estimate the average cost for this amplifier

chip (DIP-8 package) in today’s market.

V-254
Chapter 5 Problems

Problem 5.26. Determine the output voltage R2=33 kW pin layout - LM1458 (#1)

of the ideal operational amplifier shown in the pin 2

vin
figure. The amplifier is driven by a
10-V dual
R1=1 kW vout
- pin 1
+ +
power supply. 0.5V +
- pin 3
-
pin layout - LM1458 (#1)
vin pin 3
vout
+ + pin 1
common
+ vx common
0.5 V - +
-
- pin 2
-
R2=100 kW Voltage Follower
common
R1=1 kW
Problem 5.30
A. Using only the first summing-point con-
common
straint (SPC), solve the circuit shown
Inverting Amplifier in the figure, i.e., determine the output
voltage of the amplifier versus the
Problem 5.27 common port.
A. Draw the circuit diagram of the basic B. What function does this amplifier have?
inverting amplifier configuration. Why is it important?
B. Give the expression for the amplifier
gain in terms of the resistances, assum- pin layout - LM1458 (#1)

vin pin 3 vout

ing an ideal operational amplifier. pin 1
+ +
1 mV - +
Problem 5.28. Using the two summing-point - pin 2
-
constraints, solve the ideal-amplifier circuit
shown in the figure that follows if the input common

voltage is 1 mV.
A. Label and determine the currents in the
feedback loop. Exercises on the Use of the Negative
B. Determine the output voltage of the Feedback
amplifier versus the common port. Problem 5.31. (A review problem) For three
basic ideal-amplifier circuits:
R2=10 kW pin layout - LM1458 (#2)

vin R1=100 W pin 6

- vout
Inverting amplifier
pin 7
+ +
+ Non-inverting amplifier
1 mV
- pin 5 Voltage follower
-
common
(each includes negative feedback) present
common 1. A circuit diagram
2. Expression for the amplifier gain
Problem 5.29. Determine the output voltage of
the ideal operational amplifier shown in the Problem 5.32. Determine the output voltage of
figure. The amplifier is driven by a
10-V amplifier configurations shown in the figure
dual power supply. that follows. The amplifier is powered by a

9-V dual-polarity voltage power supply.
Assume an ideal operational amplifier.

V-255
Chapter 5 Operational Amplifier and Amplifier Models

a) vin vout Problem 5.34. Each of the circuits shown in the

+ + figures below employs negative feedback. Find
- +
1 mV the output voltage Vout vs. ground
- - (or common). Hint: The ground symbol in the
amplifier circuit usually has the same meaning
common
common as the common port.
b) v vout
in

+ + a) 4 kW
-1 mV - +
1 kW
- - -
+ +
common
vout
+
common 1V
- -
Problem 5.33. Each of the circuits shown in the 1 kW
figures below employs negative feedback. Find b)
the output voltage υout vs. ground (or common). -
Hint: The ground symbol in an amplifier circuit + +
usually has the same meaning as the common + vout
2 mA
port. - 7V -
a) 3 kW c)
1V
+
-
- +
+ + 5 kW vout
vout -
2 mA
- 50 W 1 kW

d)
1 kW
b)
+ +
- - vout
+ + 3.5 mA 2 kW -
+ vout
3 mA
- 3V -
c) Problem 5.35. Each of the circuits shown in the
1V
+ figures below employs an inverting amplifier.
- + 1. Solve each circuit (find υout) with an
5 kW vout
- input voltage of 1 mV.
2 kW 1 kW 2. Based on this solution, find the closed-
loop voltage gain ACL of the
corresponding amplifier circuit.
d)
+ +
- vout
2 mA 3 kW -

V-256
Chapter 5 Problems

a) R2=5.1 kW A. Find the value of the output current iout if

the input current is 1 mA,
R1=100 W
vin
-
vout
R1 ¼ 9 kΩ, R2 ¼ 1 kΩ.
+ B. Why do you think this amplifier type is
known as the current amplifier? To
answer this question quantitatively, ana-
b)
lytically express the output current iout
R2=5.1 kW
(current through the load) in terms of
R1=100 W
the unknown input current iout and two
vin vout
- arbitrary resistor values, R1,2.
+
iin
RL=51 W
-
+

c) R2=5.1 kW RL iout
R1

vin R1=100 W vout

- R2
+

R1=100 W RL=51 W

5.2.4 Mathematics Behind the Second

d) R2=5.1 kW Summing-Point Constraint

vin R1=100 W vout Problem 5.38

-
+ A. Derive an expression for the closed-loop
gain of the non-inverting amplifier based
R1=100 W RL=51 W only on the
 þdefinition
 of the output voltage

υout ¼ A υin  υin , without using the sec-
ond summing-point constraint.
Problem 5.36. An inverting amplifier that B. Determine the exact gain value when
achieves high-gain magnitude with a smaller
A ¼ 2  105
range of resistance values is shown in the figure
below. Find its output voltage υout vs. ground R1 ¼ 1 kΩ,R2 ¼ 9 kΩ
(or common port) and the resulting amplifier
gain.
vin +
- +
10 kW 10 kW vout
1 kW
v*
* -
vin=1 mV R2
vout
- R1
+
1 kW

Problem 5.39
Problem 5.37. The amplifier circuit shown in A. Derive an expression for the closed-loop
the figure employs negative feedback. gain of the inverting amplifier based only
on the definition of the output voltage υout

V-257
Chapter 5 Operational Amplifier and Amplifier Models

 
¼ A υþ 
in  υin , without using the second
summing-point constraint.
+ vout
B. Determine the exact gain value when vin=1 V -
RL=100 W
A ¼ 2  105
R2=4 kW
R1 ¼ 1 kΩ,R2 ¼ 10 kΩ R1=1 kW
+
-
common 0 V +
5.2.5 Current Flow in the Amplifier
Circuit -
Problem 5.40. The amplifier circuit shown in
the figure below is powered by a
9-V dual-
polarity voltage power supply. Problem 5.42. The amplifier shown in the fig-
A. Redraw the amplifier schematic in your ure below is powered by a
9-V dual-polarity
notes. voltage power supply.
B. Show the current direction in every wire A. Redraw the amplifier schematic in your
of the circuit by an arrow and write the notes.
corresponding current value close to B. Show the current direction in every wire
each arrow. of the circuit by an arrow and write the
Hint: Change the polarity of the input voltage corresponding current value close to
and the voltage sign if you have trouble oper- each arrow.
ating with negative values. R2=10 kW

vout vin R1=1 kW

+ - vout
+ - +
-100 mV +100 mV
- RL=500 W
R2=49 kW
+ RL=100 W
+
R1=1 kW -
-
common 0 V
+
+
-
-

Problem 5.41. Repeat the previous problem for

Problem 5.43. Repeat the previous problem
the circuit shown in the figure below.
when the input voltage to the amplifier is
100 mV.

V-258
Chapter 5 Problems

5.2.6 Multiple-Input Amplifier Circuit: Problem 5.46. By solving the amplifier

Summing Amplifier circuit shown in the following figure,
Problem 5.44. By solving the amplifier circuit determine its output voltage in terms of resis-
shown in the figure, fill out the table that fol- tances R, RF, given the input voltages
lows. Assume that R0 ¼ RF =4, R1 ¼ RF =2, and D0 ¼ 0 V, D1 ¼ 5 V, D2 ¼ 0 V.
R2 ¼ RF .
D0 D1 D2

D0 R0 RF iF

D1 R1 RF
- vout 2R 2R 2R
0V + +
D2 R2 vout
0V -
- a b c +
2R R R
common common

D2, V D1, V D 0, V υout, V 0V 0V

0 0 0
0 0 5 5.3 Amplifier Circuit Design
0 5 0
5.3.1 Choosing Proper Resistance Values
0 5 5
Problem 5.47. State the limitations on the feed-
5 0 0 back resistances and the output load resistance
5 0 5 of an amplifier circuit.
5 5 0
5 5 5 Problem 5.48. The non-inverting amplifier
shown in the figure below has been wired in
Problem 5.45. The amplifier circuit shown laboratory.
in the figure below employs negative feedback. A. Do you have any concerns with regard to
This configuration is known as a three-bit dig- this circuit?
ital-to-analog converter (DAC) on the base of B. If you do, draw the corrected circuit
an R/2R ladder. By solving the amplifier cir- diagram.
cuit, determine its output voltage in terms of
resistances R, RF, given the input voltages D0 +
v in=100mV - +
¼ 0 V, D1 ¼ 0 V, D2 ¼ 5 V. 10 v out

D0=0 V D1=0 V D2=5 V

10 -
iF 1

RF common
2R 2R 2R

vout
0V - 5.3.2 Model of a Whole Amplifier Circuit
a b c +
2R R R 5.3.3 Input Load Bridging or Matching
Problem 5.49. For three basic amplifier circuits

0V 0V
Inverting amplifier
Non-inverting amplifier
Voltage follower

V-259
Chapter 5 Operational Amplifier and Amplifier Models

(each includes negative feedback), present R2=5.1 kW

1. A circuit diagram
v in R1=100 W v out
2. An expression for the closed-loop ampli- -
fier circuit gain +
3. An expression for the input resistance
R*=100 W RL=51 W
(impedance)
4. An expression for output resistance
(impedance)
Problem 5.50
A. Explain in your own words the concept 5.3.4 Cascading Amplifier Stages
of load bridging (impedance bridging). Problem 5.55. For the amplifier circuit shown
B. Which amplifier, the non-inverting or in the figure, find the output voltage and the
inverting, should be subject to load input resistance (show units).
bridging?
R2=10 k

Problem 5.51. An electromechanical sensor is

R1=100
given by its Thévenin equivalent wherein the +
-
sensor voltage υS is small. The sensor’s equiv-
+
1 mV=vin
alent resistance RS may vary in time; but it is -
always less than 1 kΩ. An inverting amplifier is
needed that generates an amplified version of
the sensor’s voltage. The output voltage should R4=10 k
be  100υS . Draw the corresponding circuit
R3=1 k
diagram and specify one possible set of resistor -
values. +
+ vout
-
Problem 5.52. Construct an amplifier circuit
matched to a 100-Ω load. The amplifier’s gain
is jACL j ¼ 100. The sign of the gain (positive or
negative) is not important and the input AC Problem 5.56. A sensor with Thévenin
signal. voltage (source voltage) υS ¼ 2:5 mV shown
Hint: Multiple solutions many exist. Present at in the figure is to be connected to an amplifier
least two solutions. circuit. An amplified replica of the sensor’s
voltage, υout  1000 υS , is needed at the cir-
Problem 5.53. Find the input resistance cuit’s output.
(impedance) to the amplifier circuit shown in A. Present one possible circuit diagram and
the figure below. specify all necessary resistor values.
B. Present another (distinct) circuit diagram
vin 200
+ and specify all necessary resistor values.
-
R2=51 k
200 RS=25 kW

R1=1 k vS +
-

0V

Problem 5.54. Find the input resistance

(impedance) to the amplifier circuit shown in Problem 5.57. An amplifier in the configura-
the figure below. tion shown in the figure below is connected to a
V-260
Chapter 5 Problems

sensor with Thévenin (source) voltage Problem 5.61. An amplifier circuit is needed
υS ¼ 25 mV. An amplified replica of the sen- with a positive gain of 5000
5 %. The input
sor’s voltage, υout  100 υS , is needed at the resistance (impedance) should be as high as
output. possible. Present one possible circuit diagram
A. Do you have any concerns with regard to and specify the necessary resistor values
this circuit? including tolerance.
B. If you do, draw an appropriate circuit
diagram. 5.3.5 Amplifier DC Imperfections and
Their Cancellation
Problem 5.62. Determine the output voltage to
vin
nonideal operational amplifier circuits (with
RS=50 k
the nonzero input offset voltage) shown in the
vS +
- figures below.

actual op-amp
0V a)
ideal op-amp
10 M
10mV -1mV vout
- +

+
100 k +
- vout - +
+
- -
1 5 kW=R2

1 kW=R1

b) actual op-amp
Problem 5.58. An amplifier circuit is needed ideal op-amp
with the closed-loop gain ACL ¼ þ1000. The
10mV
input resistance (impedance) to the circuit should vout
+ +
be 5 kΩ. Present two alternative circuit dia- - - +
+

grams and specify the necessary resistor values. - 5mV

-
The first circuit must use inverting amplifiers
5 kW=R2
and the second circuit-non-inverting amplifiers.
1 kW=R1
Problem 5.59. An amplifier circuit is needed
with the closed-loop gain ACL ¼ þ10, 000. The
input resistance (impedance) to the circuit
should be as high as possible. Present the Problem 5.63. Determine the output voltage to
corresponding circuit diagram and specify the nonideal operational amplifier circuits (with
necessary resistor values. nonzero input currents) shown in the figures
below. The input terminal is connected directly
Problem 5.60. An amplifier circuit is needed to the common terminal (grounded). The
with a positive gain of 1000
20 %. The input strength of every bias current source is 100 nA.
resistance (impedance) to the circuit should be Hint: The upper bias current source does not
1 kΩ. Present the circuit diagram and specify the contribute to the solution.
necessary resistor values including tolerance.

V-261
Chapter 5 Operational Amplifier and Amplifier Models

a) actual op-amp R2
ideal op-amp
5 mV=vin R1 vout
0V vout
-
+
+ +
- +
-
-
10 k 0V
4 mV 0V

2 kW
Problem 5.67. In the previous problem, denote
b) actual op-amp the terminal voltage of 4 mV by Voff, the input
ideal op-amp
voltage of 5 mV by υin, the output voltage by
0 V 5/3 kW vout υout, and the amplifier gain by ACL. Derive an
+ analytical formula that determines Voff in terms
- +
of υin given that the output voltage to the ampli-
- fier is exactly zero.
10 kW

Problem 5.68. For the circuit shown in the

2 kW
figure below:
A. Determine the output voltage of the
nonideal operational amplifier circuit
Problem 5.64. For the amplifier circuit shown (with the nonzero input offset voltage).
in the figure below, determine the output volt- B. Does the amplifier circuit really follow
age. Use R1 ¼ R2 ¼ 1 kΩ. the ideal-amplifier circuit law:
υout ¼ ACL υin ?
1 mV=v in
v out C. What happens if the input voltage
+ +
- + changes from 10 mV to 20 mV?
- - actual op-amp
R2
ideal op-amp
0V R1
0V
10 mV 5mV v out
2mV - +
+

+
- +
Problem 5.65. In the previous problem, denote - -
the terminal voltage of 2 mV by Voff, the input 5kW=R2
voltage of 1 mV by υin, the output voltage by 0V
0V
1 kW=R1
υout, and the amplifier gain by ACL. Derive an
analytical formula that determines Voff in terms 6 mV

of Vin given that the output voltage υout to the

amplifier is exactly zero. Problem 5.69. Solve the previous problem
with the offset voltage in the feedback loop
Problem 5.66. For the amplifier circuit changed from 6 mV to 5 mV.
shown in the figure below, determine
the output voltage. Use R1 ¼ 1 kΩ and Problem 5.70. In problem 5.68, denote the
R2 ¼ 4 kΩ. terminal voltage of 6 mV by Voff, the input

V-262
Chapter 5 Problems

voltage of 10 mV by υin, the input offset volt- 5.3.6 DC-Coupled Single-Supply

age by VOS, the output voltage by υout, and the Amplifier
amplifier gain by ACL. Derive an analytical Problem 5.72. For the single-supply amplifier
formula that determines Voff in terms of VOS circuit shown in the figure:
given that the output voltage υout to the ampli- A. Determine the output voltage versus cir-
fier must exactly follow the ideal-amplifier gain cuit ground (the negative terminal of the
law: υout ¼ ACL υin . voltage power supply).
B. What potential problem do you see with
R2 actual op-amp this circuit? How could you fix it?
ideal op-amp

vin VOS vout 9 V=VCC vout

-
- + -
+

+ + +
+ 100 =RS -
- -
0V

R1 + 0.9 M
- 2k
0V
Voff 0V 100 mV=vS 100 k
9 V=VCC
+
Problem 5.71. For two nonideal operational -
amplifier circuits (with the nonzero input offset virtual ground circuit
voltage) shown in the figures below, determine 1k

the necessary offset voltage, Voff, which ensures

0V
that the output voltage, υout, to the amplifier
exactly follows the ideal-amplifier gain law:
υout ¼ ACL υin . You need to express this voltage Problem 5.73. For the single-supply amplifier
in terms of other circuit parameters that are circuit shown in the figure:
given in figures a) and b). A. Determine the output voltage versus cir-
cuit ground (the negative terminal of the
a) voltage power supply).
R2 actual op-amp
B. Do you see any problem with this
ideal op-amp
circuit?
vin R1 VOS
- vout
-
+

+ + 12 V=VCC vout
+ + -
+
- - 100 =RS -
0V
0V Voff
0V + 0.9 M 100 k
b)
-
R2 1 mV=vS 1k
actual op-amp
12 V=VCC
ideal op-amp vx +
R1
-
vin vout
-
+ + virtual ground circuit
- + 100 k
+

VOS
- - 0V

0V Voff
0V

V-263
Chapter 5 Operational Amplifier and Amplifier Models

5.4 Difference and Problem 5.79. Find the output voltage to the
difference-amplifier circuits shown in the fig-
Instrumentation ures below. Assume the ideal-amplifier model
Amplifiers and exact resistance values.

5.4.1 Differential Input Signal to an a) 5 kW

Amplifier
1 kW
Problem 5.74. The model of an input signal 1V - v out
from a three-terminal sensing device is shown 0.5 V
+
+
in the figure below. What are the differential 10 kW
-
and common-mode voltages at terminals a and 50 kW
b?
0V
b)
a
5 kW
+ 0.1 V
- 1 kW
v out
1V -
+
1V +
1V + + 0.1 V 10 kW
-
- -
25 kW
b
0V
0V

Problem 5.75. The Wheatstone bridge in

Fig. 5.27 is connected to
V CC rails instead Problem 5.80. Your technician needs to control
of þV CC and ground. Furthermore, R2 ¼ 1:1R1 , a process using two sensors with output volt-
RðxÞ ¼ 1:1R3 , and V CC ¼ 6 V. What are the ages υ1 and υ2, respectively. The weighted dif-
differential and common-mode voltages? ference in sensor reading, υ ¼ 1υ1  0:5υ2 , is
critical for the product quality. The technician
Problem 5.76. The Wheatstone bridge in reads voltage υ1 and then voltage υ2 and then
Fig. 5.27 is connected to ground and V CC uses a calculator to find υ. Help the technician,
rails instead of þV CC and ground. Given that i.e., sketch for him a difference-amplifier circuit
R2 ¼ 1:05R1 , RðxÞ ¼ 1:05R3 , and V CC ¼ 6 V, that will directly output υ to the DMM. The
determine the differential and common-mode negative terminal of the DMM is always
voltages. grounded. Specify one possible set of resistor
values.
5.4.2 Difference Amplifier
Problem 5.77. Design a difference amplifier Problem 5.81. Repeat the previous problem
with a differential gain of 20. Present the circuit when the weighted difference in sensor reading,
diagram and specify one possible set of resistor υ ¼ 10υ1  5υ2 , needs to be processed.
values. In the circuit diagram, label the input
Problem 5.82. For the circuit shown in the
voltages as υa, υb and express the output voltage
figure, find the output voltage if the input volt-
in terms of υa, υb.
ages are υb ¼ 1 V and υa ¼ 1:01 V, respectively.
Problem 5.78. Repeat the previous problem for Assume the ideal-amplifier model and exact
a differential gain of 100. resistance values.

V-264
Chapter 5 Problems

vb
R 10R 5.4.3 Instrumentation Amplifier
- v out
Problem 5.85
+
+
A. Why is the original difference amplifier
not used as an instrumentation amplifier?
va - B. Why is the circuit in Fig. 5.31 not used as
R 10R the instrumentation amplifier?
0V

Problem 5.83 Problem 5.86

A. For the circuits shown in the figures A. Find the differential gain and the
below, find the output voltage if the common-mode gain for the amplifier cir-
input voltages are υa ¼ 1 V and cuit shown in Fig. 5.32. The differential
υb ¼ 1 V, respectively Assume the ideal output voltage is υa *  υb *, and the
amplifier and exact resistance values. common-mode output voltage is
B. What is the value of the common-mode 0:5ðυa * þ υb *Þ.
gain in every case? B. Find the differential gain and the
common-mode gain for the amplifier cir-
a) R 2R
cuit shown in Fig. 5.31.
vb
Problem 5.87. Design an instrumentation
- v out
+ amplifier with a differential gain of 210. Present
+ the corresponding circuit diagram and specify
va - one possible set of resistance values. In the
1.05R 2R circuit diagram, label the input voltages as
0V
υa, υb and express the output voltage in terms
b)
1.05R 2R of υa, υb.
vb

- v out
Problem 5.88. Design an instrumentation
+
+
amplifier with a differential gain of 1010. Present
the corresponding circuit diagram and specify
va - one possible set of resistor values. In the circuit
R 2R
0V
diagram, label the input voltages as υa, υb and
express the output voltage in terms of υa, υb.
Problem 5.84. For the difference-amplifier cir- Problem 5.89. The following voltages are mea-
cuit shown in the figure below, find the differ- sured: υa ¼ 3:750 V and υb ¼ 3:748 V. Find
ential-mode resistance (impedance) to the voltages versus circuit ground (common port of
amplifier. The differential-mode resistance is the dual supply) for every labeled node in the
defined as the ratio of a voltage of a power circuit shown in the figure below. The amplifier
supply placed between terminals a and b to circuit is powered by a
10  V dual supply.
the current that flows through this power Assume exact resistance values and the ideal-
supply. amplifier model.
R XR

- v out
+
+
-
R XR
0V

V-265
Chapter 5 Operational Amplifier and Amplifier Models

vb + 2 vb +
- -
100 kW 100 kW 100 kW 25 kW
2 4
1 kW 10 kW 0.95 kW 1 kW
4 -
- 6 6
1 + 1 +
1 kW 5 1.05 kW 1 kW
10 kW 3 5

100 kW 100kW 25 kW
100 kW
- -
+ +
va 3
va

Problem 5.90. Repeat the previous problem Problem 5.94

when node 1 is grounded. A. Find the output voltage for the amplifier
circuit shown in the figure below.
Problem 5.91. The following voltages are mea- B. Denote the input voltage of 0.1 V by υa,
sured: υa ¼ 5:000 V and υb ¼ 5:001 V. Find the input voltage of 0.08 V by υb, the
voltages versus circuit ground (common port of 10-kΩ resistor by R1, the 40-kΩ resistor
the dual supply) for every labeled node in the by R2, and the 10-kΩ resistor by RG.
circuit shown in the figure below. The amplifier Express the output voltage in the general
circuit is powered by a
10  V dual supply. form, in terms of two input voltages and
Assume exact resistance values and the ideal- the resistances.
amplifier model.
0.1 V + v out
vb -
+
- 40 kW
100 kW 25 kW
2 4
1 kW 1 kW 10 kW
- 6 4 kW
1 +
1 kW 1 kW
3 5 0.08 V +
-
100kW 25 kW
-
+ 40 kW 10 kW
va

Problem 5.92. Repeat the previous problem

when node 1 is grounded.
5.5 General Feedback
Problem 5.93. The following voltages are mea-
sured: υa ¼ 5:000 V and υb ¼ 5:001 V. Find
Systems
voltages versus circuit ground (common port of 5.5.1 Signal-flow Diagram of a Feedback
the dual supply) for every labeled node in the System
circuit shown in the figure below. The amplifier 5.5.2 Closed-Loop Gain and Error
circuit is powered by a
10  V dual supply.
Signal
Assume exact resistance values and the ideal-
Problem 5.95. The block diagram of Fig. 5.36
amplifier model.
is applied to a voltage amplifier.

V-266
Chapter 5 Problems

A. Given the input signal xin ¼ 10 mV, the Problem 5.100. The circuit shown in the figure
error signal xe ¼ 1 μV, and the output that follows is a feedback transconductance
signal xout ¼ 1 V, determine the open- amplifier. Express iout in terms of υin.
loop gain and the feedback factor.
B. Given the ratio of input to error signal xin transconductance amplifier

=xe ¼ 100 and the feedback factor of 0.1, +

-
determine the open-loop gain. RL iout
R2
v in +
Problem 5.96. The open-loop gain A in -
Fig. 5.36 varies between two extreme values R1 RF

of A ¼ 10, 000
2, 000 (
20 % gain variation)

depending on the system parameters. The for-
ward gain block is used in the closed-loop Problem 5.101. The amplifier circuit shown in
configuration with the feedback factor β of the figure that follows is the Howland current
0.1. Determine the two extreme values of the source widely used in biomedical instrumenta-
closed-loop gain, ACL. tion; its output is the current through the load
resistance.
Problem 5.97. The open-loop gain A in A. Classify the amplifier circuit in terms
Fig. 5.36 is 100,000. The forward gain of four basic amplifier topologies and
block is used in the closed-loop configuration mention the most important circuit
with the feedback factor β of 1. Determine features.
the error signal, xe, if the input voltage signal B. Derive its gain equation iout ¼ ðυa  υb Þ
is 1 mV. =R2 given that R1 =R3 ¼ R2 =R4 .

Problem 5.98. The closed-loop gain of a R3

non-inverting amplifier circuit with R1 ¼ 1 kΩ,
R1
R2 ¼ 100 kΩ is 99. Determine the open-circuit vb +
gain A of the amplifier chip. va -
R2
5.5.3 Application of General Theory to
Voltage Amplifiers with Negative iout =iL R L R4

Feedback
5.5.4 Voltage, Current, Transresistance,
and Transconductance Amplifiers with
the Negative Feedback

Problem 5.99. Derive the gain Eqs. (5.69) for

the amplifier circuits shown in Fig. 5.38.

V-267
 Part II
Transient Circuits
Chapter 6

Chapter 6: Dynamic Circuit Elements

Overview
Prerequisites:
- Knowledge of basic circuit theory (Chapters 2 and 3)
- Knowledge of operational amplifiers with negative feedback (Chapter 5)

Objectives of Section 6.1:

- Define types of capacitance encountered in electric circuits
- Define self-inductance and mutual inductance from the first principles
- Define field energy stored in a capacitor/inductor
- Be able to combine capacitances/inductances in series and in parallel
- Understand construction of practical capacitors/inductors
- Understand fringing effect and its use in sensor circuits

Objectives of Section 6.2:

- Derive dynamic equations for capacitance/inductance from the first principles
- Establish how the capacitance may create large transient currents
- Establish how the inductance may create large transient voltages
- Define instantaneous energy and power of dynamic circuit elements
- Establish the behavior of dynamic circuit elements in the DC steady state and at a
very high frequency

Objectives of Section 6.3:

- Obtain initial exposure to bypass/blocking capacitor and decoupling inductor
- Obtain initial exposure to amplifier circuits with dynamic circuit elements

Application Examples:
Electrostatic discharge and its effect on integrated circuits
How to design a 1-F capacitor? How to design a 1-mH inductor?
Capacitive touchscreens
Bypassing a DC motor

© Springer Nature Switzerland AG 2019 VI-271

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_6
Chapter 6 Dynamic Circuit Elements

Keywords:
Capacitance, Capacitance of two conductors, Self-capacitance, Capacitance to ground,
Capacitance of two equal conductors separated by large distances, Energy stored in a
capacitance, Electrostatic discharge (ESD), ESD effect on integrated circuits, Device under test
(DUT), Parallel-plate capacitor (base formulas, fringing effect, fringing fields), Capacitor
(absolute dielectric permittivity, relative dielectric permittivity, dielectric strength, normalized
breakdown voltage, electrolytic, tantalum, ceramic, marking, set of base values), Capacitive touch
screens (self-capacitance method, mutual-capacitance method), Magnetic flux density, Magnetic
field, Absolute magnetic permeability, Relative magnetic permeability, Magnetic induction,
Magnetic flux, Self-inductance, Inductance, Mutual inductance, Energy stored in an inductance,
Solenoid (air core, toroidal magnetic core, straight magnetic core, short, fringing fields), Inductor
(marking, set of base values, also see solenoid), Dynamic equation for capacitance (definition,
derivation, fluid mechanics analogy), Capacitance (instantaneous energy, instantaneous power,
behavior in the DC steady state, behavior at very high frequencies), Dynamic equation for
inductance (definition, derivation, fluid mechanics analogy), Inductance (instantaneous energy,
instantaneous power, behavior in the DC steady state, behavior at very high frequencies), Bypass
capacitor, Decoupling capacitor, Shunt capacitor, Snubber RC circuit, Decoupling inductor,
Inductor choke, Transient circuit, Amplifier circuits with dynamic circuit elements, Active
filters, Miller integrator (circuit, DC gain, compensation, time constant), Analog pulse counter,
Analog computer, Differentiator amplifier (circuit, gain at very high frequencies), Active
differentiator

VI-272
Chapter 6 Section 6.1: Static Capacitance and Inductance

Section 6.1 Static Capacitance and Inductance

6.1.1 Capacitance, Self-Capacitance, and Capacitance to Ground
Capacitance reflects the ability of arbitrary conductors to store electric charge and,
simultaneously, the store energy of the electric field in the surrounding space. When no
dielectric is present, capacitance is determined entirely by the geometry of conductors.
When a dielectric material is present, its permittivity becomes important. Capacitance
definitions will be given with reference to Fig. 6.1.

a) b) V=0 c) d) +Q
+V
1
+Q 1
V1 +Q +Q d
1 +V V V
- -Q 1 d
- - - - - - - - - - - 2
V2 -Q
V=0 -V
2 -Q

Fig. 6.1. Conductor geometry for capacitance definitions.

1. Capacitance, C, of Two Conductors. Two arbitrary insulated conductors near together

in Fig. 6.1a constitute a simple capacitor. Its capacitance, C, is found with the help of
electrostatic theory. Further, it is used in various dynamic models. Capacitance C of
two insulated conductors 1 and 2 is defined by the ratio
Q
C
>0 ð6:1aÞ
V
where Q > 0 is the (absolute) net charge of either conductor given that the net
charge of the system with both conductors is zero and V is the potential difference or
voltage between two conductors 1 and 2, i.e., V ¼ V 1  V 2 . This ratio does not
depend on V; it is always taken so as to make the capacitance positive.
2. Self-Capacitance, Cself, of a Conductor. When an electric charge Q is added to a
single isolated conductor in Fig. 6.1b, its surface will possess a certain absolute
voltage V versus 0 V at infinity. The ratio
Q
C self
>0 ð6:1bÞ
V
is the self-capacitance of the conductor. The self-capacitance is the capacitance
when the second conductor is a hollow conducting sphere of infinite radius subject
to 0 V.
3. Capacitance to Ground, C, of a Conductor. For conductor 1 in Fig. 6.1c with charge
þQ, its capacitance to ground is the capacitance when the second conductor is an
infinite conducting ground plane in Fig. 6.1c subject to 0 V (and charged to –Q).

VI-273
Chapter 6 Dynamic Circuit Elements

Capacitance to ground, C, is always greater than the self-capacitance, Cself; their

ratio becomes quite large when the separation distance d from the plane is small. On
the other hand,

C ! C self when d ! 1 in Fig: 6:1c: ð6:1cÞ

4. Capacitance, C, of Two Equal Conductors Separated by Large Distances. For
conductors 1 and 2 in Fig. 6.1d, the capacitance approaches
1
C ! C self when d ! 1 in Fig: 6:1d: ð6:1dÞ
2
The separation distance d must be large compared to the conductor’s size.
Equation (6.1d) will be proved shortly.
The capacitance is recorded in units of farads or F. This unit is named in honor of
Michael Faraday (1791–1867), a British physicist and chemist, who was known to many
as “the best experimentalist in the history of science.” Typical capacitance values in
electronics are pF (picofarad or simply paf), nF (nanofarad), and μF (microfarad). In
power electronic circuits, larger capacitances might be used. The capacitance unit is
linked to other MKS units as follows:
As J C
1 F¼1 ¼1 2¼1 ð6:2Þ
V V V
where C is the unit of coulomb. The total electric field energy stored between two
conductors and in the surrounding space is given by
1
E ¼ CV 2 ð6:3Þ
2
This result can be derived from the definition of the electric potential (voltage). The
energy is equal to work, which is necessary to put all charges of the capacitor in place.
Equation (6.3) is valid for any configuration shown in Fig. 6.1.

Example 6.1: Prove Eq. (6.1d).

Solution: For a conductor in Fig. 6.1b, the stored electric energy is given by

1
E self ¼ C self V 2 ð6:4aÞ
2
For two conductors in Fig. 6.1d separated by a very large distance d, the stored electric
energy is approximately given by

VI-274
Chapter 6 Section 6.1: Static Capacitance and Inductance

Example 6.1 (cont.):

1
E ¼ 2E self ¼ C ð2V Þ2 ð6:4bÞ
2
Comparing Eqs. (6.4a) and (6.4b) we obtain the necessary result.

Exercise 6.1: A metal circle or radius r ¼ 0:1 m has the self-capacitance C self ¼ 8ε0 r
where ε0 ¼ 8:85419  1012 F=m is the permittivity of vacuum. Estimate capacitance of a
capacitor formed by two coaxial circles separated by 1 m.
Answer: ~3.54 pF from Eq. (6.1d). A precise numerical solution predicts 3.77 pF.

Exercise 6.2: How large is the stored energy in a 100-μF laboratory capacitor at 10 V?
Answer: 0.005 J or, which is the same on the power basis, 5 mW of power delivered during
one second. However, this power will not be delivered uniformly.

Exercise 6.3: How large is the stored energy in a 20-F ultracapacitor charged to 25 V?
Answer: 6250 J or 6.25 kJ. This is certainly a significant value. At the same time, the
discharge rate (available current or power) is much less in this case than the current or
power delivered by laboratory electrolytic capacitors.

6.1.2 Application Example: ESD

Self-capacitance results may be applied for the prediction of ESD (electrostatic dis-
charge) effects on integrated circuits (ICs). One of the most common causes of electro-
static damage is the direct transfer of electrostatic charge through a significant series
resistor from the human body or from a charged material to the electrostatic discharge-
sensitive (ESDS) device. The concept is shown in Fig. 6.2a. A metal ground plane and the
highly conducting human body naturally form a capacitor. The body capacitance C in
Fig. 6.2b is defined as the capacitance between the body, assumed to be a conductor, and
the large (ideally infinite) ground plane. Its value depends significantly on the posture of
the body with respect to the ground surface. The typical separation distance is 2 cm. It may
be shown that, at such distances, C  2C self . Therefore, instead of calculating C directly,
we can find the self-capacitance of the human body, Cself, and then multiply it by 2.

VI-275
Chapter 6 Dynamic Circuit Elements

Typical self-capacitances of a 177-cm-tall male student are indicated in Fig. 6.3. These
values were obtained by the numerical method described in Chapter 1.

a) b)
DUT

t=0 t=0

R
series body
resistance
= charged body to C
ground capacitance

Fig 6.2. Equivalent circuit for understanding ESD and its effect on a device under test (DUT).

The simplifying assumption C  2C self and Fig. 6.3 predict body capacitances in the
range 86–95 pF. These values are in a good agreement with the generally accepted human
body model (HBM), which, with reference to Fig. 6.2b, uses

R ¼ 1:5 kΩ, C ¼ 100 pF ð6:5Þ

Cself=47.2 pF Cself=44.3 pF Cself=43.4 pF Cself=43.2 pF

Fig 6.3. Typical self-capacitance values for a 177-cm-tall male person. Note how the self-
capacitance changes when the body poses change.

6.1.3 Parallel-Plate Capacitor

Consider a parallel-plate capacitor shown in Fig. 6.4a, b. Both infinitely thin conducting
square plates with the side a and area A ¼ a2 are separated by distance d. The upper plate
has a total charge +Q; the lower plate has the opposite charge –Q; the net charge of the
capacitor is zero. Feeding conductors are implied to be disconnected; they are excluded
from consideration. Assuming that the entire electric field is concentrated only within the
capacitor and that it is uniform in space (equal to V/d), the approximate capacitance is
established as

VI-276
Chapter 6 Section 6.1: Static Capacitance and Inductance

ε0 A
C¼ ð6:6Þ
d
where ε0 is the dielectric permittivity of vacuum if the capacitor is situated in vacuum. For
Eq. (6.6) to hold, the plates do not have to be square. If the capacitor does not have a high-ε
dielectric inside, Eq. (6.6) is a good approximation only if d is very small compared to the
dimensions of the plates. Otherwise, the fringing effect must be taken into account. The
fringing effect is illustrated in Fig. 6.4a, b. Fringing means that the electric field extends
outside the physical capacitor. The electric field outside the capacitor possesses certain
extra energy. Therefore, according to Eq. (6.3) where voltage V is fixed, the capacitance
must increase compared to the non-fringing case. Figure 6.4c given below illustrates
numerically found capacitance values Cexact for the parallel-plate capacitor with fringing.
These values have been accurately computed using a rigorous numerical adaptive proce-
dure. Figure 6.4c predicts a nearly linear increase of the ratio Cexact/C as a function of
the separation distance. Therefore, the wrong result, C ! 0 when d ! 1, which
is predicted by Eq. (6.6), is corrected. Instead, one will have
C ! 0:5 C self when d ! 1.

a) b)
Potential (voltage) distribution, V
0.3 +1 V

0.4

0.5 lines of force

0.6
(E-field) -1 V

0.7
c)
0.8
0.9
Cexact/C
3.5
equipotential
lines
3

-0.9 2.5
-0.8
-0.7
-0.6 2
-0.5

-0.4
1.5

-0.3
1
0 0.2 0.4 0.6 0.8 d/a 1

Fig. 6.4. (a) Equipotential lines and lines of force for a capacitor with d=a ¼ 0:2 in the central
cross-sectional plane (the plates are at 1 V). (b) Fringing electric field for the same capacitor
observed in the central cross-sectional plane (the plates are at 1 V). (c) Ratio of the accurate
capacitance values (found numerically) to the values predicted by Eq. (6.6).

The fringing field of capacitors is utilized in capacitive touch screens. In this case, the
significant fringing field is a desired effect. Therefore, configurations other than the
parallel-plate capacitor are used. These configurations will be studied later in this section.

VI-277
Chapter 6 Dynamic Circuit Elements

When a dielectric material of relative permittivity εr > 1 is inserted between the

capacitor plates, a substitution ε0 ! εr ε0 has to be made in Eq. (6.6). The fringing effect
is less apparent for higher values of εr.

Exercise 6.4: Estimate the static capacitance of a parallel-plate capacitor with a ¼ 1 cm2
and d ¼ 1:57 mm using the basic formula. The substrate material is Rogers 4003 laminate
with εr ¼ 3:55.
Answer: 2.00 pF.

6.1.4 Circuit Symbol: Capacitances in Parallel and in Series

Figure 6.5 shows the capacitances in parallel and in series, along with the capacitance
circuit symbol. This symbol is reserved for the capacitance as a circuit element. Such an
element is an ideal capacitor excluding manufacturing imperfections (parasitic resistance
and inductance). In the following text, we will frequently employ both words—capaci-
tance and capacitor—to denote the same ideal circuit element. The parallel and series
connections of capacitances are opposite when compared to the resistances. To establish
this fact we consider two combinations in Fig. 6.5.

a) C1 b)
+ vC1 - + vC2 - + vC3 -
+Q1 -Q1 +Q2 -Q2 +Q3 -Q3
C2

A B A C1 C2 C3 B
C3

+ VC - + VC -

Fig. 6.5. Capacitances in parallel and in series: (a) capacitances in parallel are added; they behave
similarly to resistances in series and (b) capacitances in series are combined in the same way as
resistances in parallel.

For the parallel configuration in Fig. 6.5a, the same voltage VC is applied to every
capacitance. One has for the charges on the capacitor plates,

Q1 ¼ C 1 V C , Q2 ¼ C 2 V C , Q3 ¼ C 3 V C ) Q1 þ Q2 þ Q3 ¼ Qtotal
¼ ðC 1 þ C 2 þ C 3 Þ V C ð6:7Þ

VI-278
Chapter 6 Section 6.1: Static Capacitance and Inductance

Thus, the capacitances connected in parallel behave like a single capacitance Ceq,

C eq ¼ C 1 þ C 2 þ C 3 ð6:8Þ
Equation (6.8) also makes intuitive sense if we take a closer look at Fig. 6.5a where the
three individual capacitors visually form a bigger capacitor comprised of larger plate
areas. This clearly increases the capacitance accordingly. In Fig. 6.5b, however, the
situation is different. The thicknesses of each capacitor add together, which decreases
the overall capacitance since thickness varies inversely with capacitance. Assume that
every capacitor was initially uncharged and apply voltage VC between terminals A and B.
Since each pair of inner conductors in Fig. 6.5b has remained insulated,
Q1 ¼ Q2 ¼ Q3 ¼ Q. Next, by KVL,


Q1 Q2 Q3 1 1 1 Q
VC ¼ V1 þ V2 þ V3 ¼ þ þ ¼ þ þ Q¼ ð6:9Þ
C1 C2 C3 C1 C2 C3 C eq
and for the series combination of the capacitances, one has
1 1 1 1
¼ þ þ ð6:10Þ
C eq C 1 C 2 C 3

Exercise 6.5: Find the equivalent capacitance of the circuit shown in Fig. 6.6.
Answer: C eq ¼ 44 μF

33 F
a

33 F 33 F

b
33 F

Fig. 6.6. A capacitive network which includes series and parallel combinations of capacitances.

6.1.5 Application Example: How to Design a 1-μF Capacitor?

Let us to design a capacitor. Our modest goal is a 1-μF capacitor. We consider two
aluminum plates separated by 1 mm. Equation (6.6) then allows us to predict the required
plate area:

dC 103  106
A¼ ¼ ¼ 113 m2 ! ð6:11aÞ
ε0 8:854  1012

VI-279
Chapter 6 Dynamic Circuit Elements

Well, such a capacitor will certainly occupy a significant fraction of a lecture hall and is
hardly practical. How, then, do manufactures design a capacitor of 1 μF? The first step
is to use a dielectric material sandwiched within the capacitor. A dielectric medium
increases charges stored on the two metal capacitor plates depending on εr  1, the
relative dielectric permittivity of the dielectric medium. Table 6.1 gives us a list of
permittivities for a number of dielectric materials. For each material, a dielectric strength,
or normalized breakdown voltage, is also given. This is actually the maximum electric
field (notice the unit of V/m) that the capacitor can handle. It is for this reason that
capacitors carry a voltage rating that you should not exceed in your circuit. From a
practical point of view, the higher the capacitance, the lower the voltage rating. The well-
known dilemma with the capacitor is that a decrease in the separation distance increases
the capacitance and the stored energy. However, as already mentioned, it simultaneously
decreases the maximum applied voltage due to the dielectric breakdown effect. For our
capacitor, we will again use the mica dielectric material listed in Table 6.1. Equation (6.6)
now transforms to

dC 103  106
A¼ ¼ ¼ 16 m2 ð6:11bÞ
ε0 εr 8:854  1012  7

Table 6.1. Relative dielectric permittivity and dielectric strength of some common materials.
Material Relative permittivity Dielectric strength in V/m
Air 1.0 0.4–3.0  106
Aluminum oxide 8.5 Up to 1000  106
Fused silica (glass) 3.8 470–670  106 (or lower)
Gallium arsenide (GaAs) 13
Germanium (Ge) crystal 16 ~10  106
Mica 7.0 Up to 400  106
Nylon 3.8 ~20  106
Plexiglas 3.4 ~30  106
Polyester 3.4
Quartz 4.3 8  106 (fused quartz)
Rutile (titanium dioxide) 100–200 10–25  106
Silicon (Si) crystal 12 ~30  106
Styrofoam 1.03–1.05
Teflon 2.2 87–173  106
Water (distilled, deionized) ~80 65–70  106

Even though the result looks a bit better, it is still far from practical. However, what if we
try to make the dielectric layer very thin? An oxide is a dielectric, so could we just oxidize
one top aluminum plate with a very thin (i.e., d ¼ 10 μm) oxide layer and press-fit it to

VI-280
Chapter 6 Section 6.1: Static Capacitance and Inductance

the second plate? The result becomes (the relative dielectric constant of 8.5 is now that
for aluminum oxide from Table 6.1)
hC 105  106
A¼ ¼ ¼ 13 cm2 ð6:11cÞ
ε0 εr 8:854  1012  8:5

Electrolytic Capacitors
Once such a thin film is rolled into a cylinder, it will clearly become a compact design,
similar in size to a 1-μF electrolytic capacitor routinely used in the laboratory. Unfortu-
nately, one problem still remains: the permanent oxide layer is fragile and rough in shape.
A better idea is to chemically grow such a layer using a so-called anodization process.
This process occurs when the aluminum foil is in contact with an electrolyte as a second
conductor and an appropriate voltage is applied between them. This is the smart idea
behind an electrolytic capacitor. And this is also the reason why an electrolytic capacitor
is polarized. The term electrolytic capacitor is applied to any capacitor in which the
dielectric material is formed by an electrolytic method; the capacitor itself does not
necessarily contain an electrolyte. Along with aluminum capacitors, tantalum capacitors
(both wet and dry) are also electrolytic capacitors.

Ceramic Capacitors
A competitor to the electrolytic capacitor is a non-polarized ceramic capacitor. Ceramic
capacitors consist of a sandwich of conductor sheets alternated with ceramic material. In
these capacitors the dielectric material is a ceramic agglomerate whose relative static
dielectric permittivity, εr, can be changed over a very wide range from 10 to 10,000 by
dedicated compositions. The ceramic capacitors with lower εr values have a stable
capacitance and very low losses, so they are preferred in high-precision circuits and in
high-frequency and RF electronic circuits. Typically, these “fast” ceramic capacitors have
very small capacitances, on the order of pF and nF, and they can hold a high voltage. At
the same time, the “slow” ceramic capacitors may have values as high as 1 μF. Therefore,
the task of the above example can be solved with the ceramic capacitor as well.

Capacitor Marking
Figure 6.7 shows two examples of ceramic capacitors, with 100-pF and 1.0-μF capa-
citance from two different companies. To read the capacitance in the figure, we use the
following rule: 101 ¼ 10  101 pF ¼ 100 pF, and 105 ¼ 10  105 pF ¼ 1 μF. Indeed,
473 ¼ 47  103 pF ¼ 47 nF, and so forth. The tolerance letters may be present:
F ¼ 1 %, G ¼ 2 %, J ¼ 5%, K ¼ 10 %, and M ¼ 20 %. Also, the voltage rating should
be given.

VI-281
Chapter 6 Dynamic Circuit Elements

Fig. 6.7. Left, ceramic capacitors of 100 pF. Right, radial leaded ceramic capacitors of 1.0 μF.

A standardized set of capacitance base values is defined in the industry. The capacitance
of any (electrolytic or not) capacitor can then be derived by multiplying one of the base
numbers 1.0, 1.5, 2.2, 3.3, 4.7, or 6.8 by powers of ten. Therefore, it is common to find
capacitors with capacitances of 10, 15, 22, 33, 47, 68, 100, 220 μF, and so on. Using this
method, values ranging from 0.01 to 4700 μF are customary in most applications. The
value of the capacitance and the allowed maximum voltage are prominently written on the
case of the electrolytic capacitor so reading those does not constitute any difficulties.

6.1.6 Application Example: Capacitive Touchscreens

Capacitive touchscreens use the fringing field of a capacitor studied previously. Many
small capacitors with a significant fringing field are involved. If a conducting finger
(an extra conductor) is placed in the fringing field, the corresponding capacitance
changes. There are two possible solutions called the self-capacitance method and the
mutual-capacitance method, respectively. The difference is in the measurement nodes for
the capacitance. In the first case, the capacitance is measured between the touch pad
electrode and a ground. In the second case, the capacitance is measured between two pad
electrodes, neither of which is grounded. Both methods may be combined.

Self-Capacitance Method
Consider a human finger in the proximity of a touchscreen as shown in Fig. 6.8a. The
touchscreen itself may be a lattice of circular touch pads surrounded by a ground plane
and separated from it by an air-gap ring—see Fig. 6.8b. When the finger is not present,
each pad has capacitance CP to ground, which is called a parasitic capacitance. When the
(grounded) finger appears in the vicinity of the touchpad, there appears another capacitance,
CF, which is called the finger capacitance. Figure 6.8a indicates that both capacitances are
in parallel so that the resulting ground capacitance increases as

CP ! CP þ CF > CP ð6:12Þ

VI-282
Chapter 6 Section 6.1: Static Capacitance and Inductance

This change in capacitance is recorded. Physically, the presence of the finger (or hand)
increases the size of the ground conductor and thus increases the resulting capacitance.
Typical values of CP are on the order of 100 pF; CF is on the order of 1–0.1 pF. Now
assume that the desired resolution along one dimension of the screen is N. Then, N2
individual touch pads are needed including the corresponding sensing circuitry. This may
be a significant disadvantage of the self-capacitance method.
a) b)
CP CP CP CP
CF
CP

0V

pad
o
v
e
r
l
0V a
y
ground plane

Fig. 6.8. Self-capacitance method for a capacitive touchscreen. The touchscreen is enlarged.

Mutual-Capacitance Method
The electrodes are typically interleaving rows and columns of interconnected square
patches, which are shown in Fig. 6.9. Neither of them is connected to circuit ground (the
third conductor) or to each other. When a finger touches the panel, the mutual capacitance
CM between the row and column, which mostly concentrates at the intersection, decreases,
in contrast to the previous case. This change in capacitance is recorded. Assume again that
the desired resolution along one dimension of the screen is N. Then, only 2N individual
touch pads (electrodes) are needed including the corresponding sensing circuitry. This is a
significant advantage of the mutual-capacitance method.
sensing column

finger position CM

driven row

Fig. 6.9. Mutual-capacitance method for a capacitive touchscreen. Surface charge distribution is
illustrated when the driven row is subject to an applied voltage. Finger projection is a circle.

VI-283
Chapter 6 Dynamic Circuit Elements

6.1.7 Self-Inductance (Inductance) and Mutual Inductance

As long as the physical capacitor stores energy of the electric field, a physical inductor
stores energy of the magnetic field. The inductor stores magnetic-field energy only when
an electric current I flows through it. This is in contrast to a capacitor, which, once
charged, stores the electric field energy in vacuum indefinitely, even when disconnected
from the charging circuit. To use the inductor as an energy-storage element, one therefore
needs to maintain a current in the circuit. The primary quantity is the magnetic flux
density, ~
B, which is measured in webers per m2, or tesla (1 T ¼ 1 Wb=m2 ¼ 1 V  s=m2).
The magnetic flux density is related to the magnetic field, H, ~ by ~ B ¼ μH ~ where μ is
magnetic permeability. In older power electronics texts, ~ B, may be called magnetic
~
induction. In Fig. 6.10, the magnetic flux density B is created by circuit #1 (a closed loop
of current I). Instead of the vector field ~
B, it is convenient to use a simpler scalar quantity
known as magnetic flux or simply flux, Φ. For a constant ~ B, which is strictly perpendicular
to the plane of circuit #1 with area A, the magnetic flux would be equal to

Φ ¼ AB ð6:13Þ

where B is the magnitude (length) of vector ~

B. The flux is measured in webers or in V  s
(1 Wb ¼ 1 V  s).

circuit #1
circuit #2

lines of magnetic flux (B= H )

Fig. 6.10. Magnetic flux density generated by circuit #1.

Generally, Eq. (6.13) is only approximately valid for circuit #1 in Fig. 6.10. The exact
flux is given by a surface integral over the area of the circuit,
ðð
Φ¼ ~
B~nda ð6:14Þ
A

where ~n is a unit normal to the surface A. Flux is an algebraic quantity and hence could be
positive or negative depending on the chosen direction of ~ n. We assume that positive
current in the circuit produces positive flux—see Fig. 6.10. The self-inductance of circuit

VI-284
Chapter 6 Section 6.1: Static Capacitance and Inductance

#1 in Fig. 6.10 is its inductance; both terms have the same meaning. The inductance, L, of
circuit #1 is given by
Φ
L
>0 ð6:15Þ
I
Thus, the inductance is the magnetic flux through circuit #1 produced by a unit current
in the same circuit. The mutual inductance, M, between circuits #2 and #1 in Fig. 6.10 is
the magnetic flux, Φ0 , through circuit #2 produced by unit current in circuit #1, i.e.,
0
Φ
M
ð6:16Þ
I
Both L and M have the units of henry, or H. This unit is named in honor of Joseph Henry
(1797–1878), an American scientist. Typical inductance values in electronics are nH
(nanohenries) and μH (microhenries). In power electronics, larger inductances may be
used. Henry is converted to V, A, and energy, J, as follows:
Vs J
1 H¼1 ¼1 2 ð6:17Þ
A A
One may observe a close similarity between Eqs. (6.17) and (6.2). Both equations
become identical if we interchange V and A. Equation (6.17) also has a number of simple
and important implications related to energy and power. Total magnetic-field energy
stored in space surrounding circuit #1 in Fig. 6.10 is given by
1
E ¼ LI 2 ð6:18Þ
2
Equation (6.18) may be considered as another definition of self-inductance
(or inductance). As such, it is frequently used in practice.

Exercise 6.6: A flux linking the circuit is 0.1 Wb. Find the circuit’s inductance and
magnetic-field energy stored if the circuit current is 1 A.
Answer: L ¼ 100 mH, E ¼ 0:05 J.

6.1.8 Inductance of a Solenoid With and Without Magnetic Core

Consider a solenoid (a long helical coil of length l) with applied current I shown in Fig. 6.11.
The case of air-filled coil in Fig. 6.11a is studied first. Magnetic flux density ~ B within the
solenoid is nearly uniform and is directed along its axis. Therefore, the flux through one turn
of the coil (one loop) is given by Eq. (6.13). It is equal to AB, where A is the loop area. The
net flux Φ through the entire solenoid is AB times the number of turns comprising the coil, N.
The inductance is therefore obtained from Eq. (6.15) as

VI-285
Chapter 6 Dynamic Circuit Elements

AB
L¼N ð6:19Þ
I
The magnetic flux, B, within the solenoid is found in physics courses:
μ0 N I
B¼ ð6:20Þ
l
where the natural constant μ0 ¼ 4π  107 H=m is the magnetic permeability of
vacuum (or air, which is very close to a vacuum with regard to magnetic properties).
Substitution of Eq. (6.20) into Eq. (6.19) yields a simple equation for the inductance

μ0 AN 2
L¼ ½H ð6:21Þ
l
Thus, strong inductances can be created by a large number of turns (a quadratic depen-
dence), a large coil cross section, and a smaller coil length. Equation (6.21) also holds for
various bent solenoids (such as toroidal coils). Equation (6.21) makes clear that the
inductance, like capacitance and resistance, is independent of externally applied circuit
conditions.

magnetic core ( r) magnetic core ( r)

a) I B= 0H b) I B= 0 rH c) I

N N

A A

I I

Fig. 6.11. Three types of a solenoid.

The above derivation is only valid for a solenoid that is long compared to its diameter.
When this is not the case, a modification to Eq. (6.22) is made, namely,


μ0 AN 2 8w w2 w4 r
L¼ 1 þ  , w¼ <1 ð6:22Þ
l 3π 2 4 l
where r is the radius of the coil. Other handy formulas for short solenoids exist. Equation
(6.22) describes the fringing effect in practical inductors, which may also be used for
sensor purposes, similar to the fringing fields for the capacitor. The inductance of the
solenoid increases when it has a core with a magnetic material within the coil, as shown in
Fig. 6.11b, c. This material is called a magnetic core and inserting it into the coil increases
the magnetic-field energy stored in the inductor. When the magnetic core is closed, i.e.,
has the form of a toroid as in Fig. 6.11b, Eq. (6.21) is transformed to

VI-286
Chapter 6 Section 6.1: Static Capacitance and Inductance

μ0 μr AN 2
L¼ ð6:23Þ
l
where μr  1 is known as the relative magnetic permeability of a magnetic material.
Emphasize that Eq. (6.23) is not valid for the straight cylindrical core in Fig. 6.11c. The
calculation of inductance for the straight core becomes a nontrivial theoretical exercise.
We present here a useful theoretical result, which is only valid for a high-permeability
magnetic core, with approximately μr  100. The resulting inductance has the form


0:5πμ0 l * N 2 l
L h* i 1 * ½H ð6:24Þ
ln lr  1 2l

where l, l* are indicated in Fig. 6.11c and r is the radius of the coil (core). The resulting
inductance does not explicitly depend on the specific value of μr as long as this value is
sufficiently large. Equation (6.24) holds only for the situations where the core length-to-
diameter ratio is considerably smaller than the relative magnetic permeability, μr. It
predicts inductances that are much lower than those found by using Eq. (6.23).

Exercise 6.7: A solenoid coil in Fig. 6.11a, b, and c has r ¼ 0:45 cm, N ¼ 110,
l ¼ 2:15 cm. Determine the coil inductance in all three cases. In cases b and c, respectively,
μr ¼ 100. Furthermore, l* ¼ 8:90 cm in Fig. 6.11c.
Answer: L ¼ 45 μH (air-core coil); L ¼ 4500 μH (toroidal coil); L ¼ 640 μH (straight-
magnetic-core coil).

6.1.9 Circuit Symbol: Inductances in Series and in Parallel

Figure 6.12 shows the inductances in series and in parallel, along with inductance circuit
symbol. This symbol is reserved for the inductance as a circuit element. Such an element
is an ideal inductor excluding manufacturing imperfections (parasitic resistance and
capacitance). In the following text, we will frequently employ both words—inductance
and inductor—to denote the same ideal circuit element. The series and parallel connec-
tions of inductances are identical when compared to the connections of resistances. To
establish this fact we consider two combinations shown in Fig. 6.12.

a) b) L1
I I I I
L2

A L1 L2 L3 B A B
L3

Fig. 6.12. Inductances in series and in parallel.

VI-287
Chapter 6 Dynamic Circuit Elements

For the series configuration in Fig. 6.12a, the same current I is applied to every
inductance. Given that the equivalent inductance is also subject to current I and must
possess the same magnetic-field energy defined by Eq. (6.18), one has
1 1 1 1
E ¼ Leq I 2 ¼ L1 I 2 þ L2 I 2 þ L3 I 2 ) Leq ¼ L1 þ L2 þ L3 ð6:25Þ
2 2 2 2
The parallel configuration in Fig. 6.12b may be analyzed given the condition of equal
magnetic flux through each inductance. Since this condition is related to Faraday’s law of
induction, we postpone the corresponding discussion until the next section.

6.1.10 Application Example: How to Design a 1-mH Inductor?

Again, after this theoretical excursion, let us design a practical inductor. Our goal is to
construct a 1-mH inductance. According to Eq. (6.21) the required number of turns is
sffiffiffiffiffiffiffiffi
Ll
N¼ ð6:26Þ
μ0 A

We will select a coil length of l ¼ 5 cm and an (average) coil radius of r ¼ 1 cm.

Equation (6.26) then yields
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
103  5  102
N¼ ¼ 356 ð6:27Þ
4π 2  107  104
Such a coil can be wound on a former in the laboratory with a sufficiently thin wire, say
AWG 28. A different approach to reducing the number of turns while maintaining, or
even increasing, the inductance is to use a magnetic material within the coil, as shown in
Fig. 6.11. Table 6.2 lists the magnetic permeability for a number of magnetic materials.
The simplest magnetic core is an iron core. However, it is lossy since an alternating
magnetic field creates so-called eddy currents in the conducting core, which are dissi-
pated into heat. One solution to this problem is to use thin insulated sheets of iron, or
laminations. Various ferrites (oxides of iron, or other metals) are an alternative to iron,
which are ceramics and known as good electric insulators. Other types of losses may
occur there, explanations of which go beyond the scope of our text. Once a soft ferrite
with μr
100 is used in the design of the 1-mH inductor, the number of turns necessary
pffiffiffiffi
to achieve the same inductance decreases by μr ¼ 10. In our example, it becomes
equal to 36 turns instead of over 300 turns. However, the magnetic core cannot be a short
rod like that shown in Fig. 6.11c; it must form a closed loop shown in Fig. 6.11b.

VI-288
Chapter 6 Section 6.1: Static Capacitance and Inductance

Table 6.2. Relative permeability of some common materials.

Material Relative static permeability
Air 1.0
Magnetic iron 200
Iron powder 2–75
Nickel 100
Permalloy (78.5 % nickel + 21.5 % iron) 8000
Soft ferrites with low losses at frequencies up to 100 MHz 20–800
Hard ferrites with low losses up to 1 MHz 1000–15,000

Inductor Marking
Leaded inductors have color codes, similar to resistors. A standardized set of inductance
base values is defined in the industry. The inductance of any inductor can then be derived
by multiplying one of the base numbers 1.0, 1.1, 1.3, 1.5, 1.9, 2.2, 2.7, 3.3, 3.9, 4.7, 5.8,
6.8, or 8.2 by powers of ten. Therefore, it is common to find inductances with values of
1.0, 2.7, 6.8 μH, and so on. Using this method, values ranging from 0.01 to 100 μH are
customary in most applications.

VI-289
Chapter 6 Dynamic Circuit Elements

Section 6.2 Dynamic Behavior of Capacitance

and Inductance
6.2.1 Set of Passive Linear Circuit Elements
The three elements, resistance, capacitance, and inductance, constitute the fundamental
set of passive circuit elements for any linear electric circuit. This is very similar to a
mechanical system consisting of dashpot, spring, and mass, which form the basic set of
any linear kinematic system. Having discussed the underlying DC concepts of capaci-
tance and inductance, we now turn our attention to their dynamics. A simple example of
dynamic behavior is given by a vacuum cleaner. If one manually unplugs the working
vacuum cleaner from the wall outlet (please avoid doing so), a profound spark may
appear. On the other hand, turning off the vacuum cleaner normally produces no spark.
The reason for the spark is that breaking the current through a dynamic circuit element—
an inductance—which models the coil of the motor, creates very large transient voltages.
One reason for studying transients is the wish to avoid such sparks and to properly design
the electric switch. The use of two dynamic circuit elements—the capacitance and
inductance—is enormous, especially in power systems. Every electric motor is basically
an inductance; most power motors need a power correction circuit that in turn requires a
shunt capacitance. Some motors need starting capacitors or surge capacitors for large
motors. On the other hand, the capacitance of logic gates is responsible for the so-called
propagation delay. This delay determines a very important measure of the performance of
a digital system, such as a computer, which is the maximum speed of operation. Thus, the
capacitances and inductances are just everywhere, like mass and spring systems present
everywhere in mechanical engineering. However, they are becoming most apparent when
we consider a transient behavior, an alternating current, or high-frequency digital and
communication circuits.

6.2.2 Dynamic Behavior of Capacitance

Both capacitance and inductance are passive circuit elements, which means that, like
resistance, they do not deliver a net power increase to the circuit. Indeed, after charging,
the capacitor is able to power a circuit, usually for a short period of time. At the end of the
discharge cycle, it needs to be recharged. Therefore, we use the passive reference
configuration for the capacitance in Fig. 6.13a. Lowercase letters denote time-varying
voltage and current. The dynamic behavior of the capacitance is described by the well-
known voltage-to-current relation (dynamic equation), which plays the role of “Ohm’s
law” for the capacitance
dυC
iC ¼ C ð6:28Þ
dt

VI-290
Chapter 6 Section 6.2: Dynamic Behavior of Capacitance and Inductance

Equation (6.28) follows from the capacitance definition, qC ¼ CυC , given in the previous
section. It is obtained after differentiation and using the equality
dqC
¼ iC ð6:29Þ
dt

a) b) flexible membrane
capacitance C
+ vC -

iC

Fig. 6.13. Passive reference configuration for capacitance and its fluid mechanics analogy.

A fluid mechanics analogy of the dynamic capacitance effect corresponds to alternating

fluid motion in Fig. 6.13b. A capacitance is represented by a flexible membrane. The
capacitance value, C, corresponds to the inverse stiffness, 1/k, of the membrane. When
k ! 1 and C ! 0 (a rigid membrane), the capacitance value tends to zero. The
membrane becomes a solid wall, which blocks the alternating flow entirely. In another
limiting case (k ! 0 or C ! 1), the membrane moves with fluid and has no effect on the
fluid flow. Intermediate cases correspond to a partial blocking. The electric current in
Eq. (6.28) is not the DC conduction current, but the displacement current which was first
introduced by Sir James Clerk Maxwell to complete Maxwell’s equations. This current
can flow through empty (or free) space between two capacitor plates. It is not supported by
directional motion of free charges, in contrast to the DC conduction current. The capacitor
itself was invented yet in 1745 by Ewald Georg von Kleist (1700–1748), German lawyer
and physicist, and by Dutch scientist Pieter Van Musschenbroek (1692–1761). Some also
believe the Biblical Ark of the Covenant was protected by a first capacitor—the Leyden
jar—capable of producing thousands of volts of static electricity.

Example 6.2: The voltage across a 100-μF capacitor is shown in Fig. 6.14 that follows by
a solid curve. At t ¼ 0, the voltage is zero. Sketch the current through the capacitor to scale
versus time.
 
Solution: We use Eq. (6.28) to find the current. In Fig. 6.14, υC ðt Þ ¼ 105 t  106 V
when t changes from 1 to 2 μs, and υC ðtÞ ¼ 0 at t < 1 μs. At t > 2 μs, the voltage is
100 mV. Therefore, the current is found in the form: iC ðtÞ ¼ 104  105 ¼ 10 A when
t changes from 1 to 2 μs and iC ðtÞ ¼ 0 otherwise. The result is shown in Fig. 6.14 by a dashed
curve. We observe a strong current spike when the voltage across the capacitor changes rapidly,
and we observe no current flow when the voltage across the capacitor remains constant.

Note that the relatively small voltage on the order of 100 mV in Fig. 6.14 leads to a very
large current spike of 10 A(!) through the capacitance. The key point here is that the current

VI-291
Chapter 6 Dynamic Circuit Elements

increase is due to the rapid change in voltage. Such a change can be created when the
capacitor discharges through a small resistance. This is the reason why capacitors are
routinely employed to deliver large currents, or high power levels, for a very short period
of time. The high currents are common in motor starting circuits, in electronic flashes, in
solenoids, and in various electromagnetic propulsion systems.
100 F capacitance
vC
+ -
vC , V iC, A
iC
0.2 10

0.1 5

0 1 2 3
t, s

Fig. 6.14. Applied voltage across the capacitor (solid curve) and resulting current (dashed curve).

The capacitor is charged with an electric current. The voltage across the capacitor, from
Eqs. (6.28) and (6.29), is given by
ðt
1 q ðt Þ
υC ¼ iC ðt 0 Þdt 0 ¼ C ð6:30Þ
C C
0

Equation (6.30) tells us that the capacitor voltage is equal to zero at the initial time, i.e., at
t ¼ 0. Once the current iC(t) is known, the voltage at any point in time is obtained by
carrying out the integration in Eq. (6.30). At any time instant, the voltage is equal to the
instantaneous stored charge qC(t) divided by capacitance. The current in Eq. (6.30) is
either predefined or found from circuit considerations. The example that follows illus-
trates voltage calculations.

Example 6.3: A 1-μF capacitor is charged with an electric current, iC ðt Þ ¼ 2  t ½mA . The
capacitance voltage is equal to zero at the initial time instance t ¼ 0. When will the
capacitor be charged to 10 V?
Solution: The integration yields

ðt
1 0:001 2
υC ¼ iC ðt0 Þdt 0 ¼ t ¼ 10 ) t ¼ 0:1 s ð6:31Þ
C C
0

VI-292
Chapter 6 Section 6.2: Dynamic Behavior of Capacitance and Inductance

6.2.3 Dynamic Behavior of Inductance

We use the passive reference configuration for the inductance in Fig. 6.15a. Lowercase
letters denote time-varying voltage and current. The dynamic behavior of the inductance
is described by the well-known voltage-to-current relation (dynamic equation), which
plays the role of “Ohm’s law” for the inductance:
d iL
υL ¼ L ð6:32Þ
dt
Equation (6.32) follows from the inductance definition, Φ ¼ LiL , given in the previous
section. We obtain Eq. (6.32) after differentiation and using Faraday’s law of induction
for the time derivative of the magnetic flux (the plus sign is used in Faraday’s law to be
consistent with the passive reference configuration):

dΦðt Þ
¼ υL ð6:33Þ
dt

massive wheel
a) b) of mass m
inductance L
+ vL -

iL

Fig. 6.15. Passive reference configuration for inductance and its fluid mechanics analogy.

A fluid mechanics analogy of the dynamic inductance effect is given here in terms
of alternating current, which corresponds to alternating fluid motion in Fig. 6.15b.
A massive wheel with rotational inertia in Fig. 6.15b represents inductance.
The inductance value, L, corresponds to the mechanical mass m of the wheel. When
m ! 1 or L ! 1, the wheel does not responds to fluid oscillations and blocks
the alternating fluid flow entirely. In the opposite case (m ! 0 or L ! 0), the wheel
has no effect on the fluid flow. Intermediate cases correspond to a partial blocking.

Example 6.4: The current through a 2-mH inductor is shown in Fig. 6.16 by a solid curve.
At t ¼ 0, the current is zero. Sketch the voltage across the inductance to scale versus time.
Solution: We use Eq. (6.32) to find the voltage across the inductance. In Fig. 6.16, iL ðt Þ
 
¼ 103 t  106 A when t changes from 1 to 2 μs and iL ðt Þ ¼ 0 at t < 1 μs. At t > 2 μs,
the current is 1 mA. Therefore, the voltage is found in the form: υL ðt Þ ¼ 2  103  103
¼ 2 V when t changes from 1 to 2 μs and υL ðt Þ ¼ 0 otherwise. The result is shown in
Fig. 6.16 by a dashed curve.

VI-293
Chapter 6 Dynamic Circuit Elements

2 mH inductance
vL
+ -
iL, mA v L, V
iL
2 2

1 1

0 1 2 3
t, µs

Fig. 6.16. Impressed current through an inductance (solid curve) and resulting voltage across the
inductance (dashed curve).

Note that relatively small current, on the order of 1 mA, leads to a large voltage spike
of 2 V across the inductance in Fig. 6.16. The key point here is again the rapid change in
the current. If the current in the present example were on the order of 1 A, a voltage spike
of 2000 V would be observed. This is the reason why inductors are routinely used to
boost the voltage to a higher level. These high voltages are common in electric and
electronic ignition systems including the most common car ignition plug. From
Eqs. (6.32) and (6.33) the current through the inductance is given by
ðt
1 Φðt Þ
iL ¼ υL ðt 0 Þdt 0 ¼ ð6:34Þ
L L
0

Equation (6.34) implies that the current is equal to zero at the initial time, i.e., at t ¼ 0.
Once the voltage is known as a function of time, the current through the inductance at any
time moment is obtained by the calculation of the integral in Eq. (6.34). At any time
instant, the current is equal to the instantaneous magnetic flux Φ(t) divided by inductance.

Example 6.5: A 1-mH inductor is subject to applied voltage, υL ðt Þ ¼ 2  t ½mV . The

inductance current is equal to zero at the initial time instance t ¼ 0. When will the
magnetic-field energy stored in the inductance reach 1 J?
Solution: The integration in Eq. (6.34) and using Eq. (6.18) for the energy stored in the
inductor yield

ðt
1 0:001 2 1  2
iL ¼ υL ðt 0 Þdt 0 ¼ t ¼ t 2 ½A ) L t2 ¼ 1 J ) t ¼ 6:7 s ð6:35Þ
L L 2
0

VI-294
Chapter 6 Section 6.2: Dynamic Behavior of Capacitance and Inductance

So, if the capacitance is associated with a spring, the inductance is associated with the
mass, and the resistance is associated with a dash pot (damping element), then the entire
electric circuit containing dynamic elements is nothing else but a mechanical system. Is
this correct? Clearly the same analysis methods are applicable to both systems, electrical
and mechanical! The model of an entire building in terms of lumped mechanical elements
is in theory the same as the model of a complicated electric circuit. Both models can be
analyzed by using the theory of linear systems, and both models follow the same control
theory. A more difficult issue is related to nonlinear circuit elements.

6.2.4 Instantaneous Energy and Power of Dynamic Circuit Elements

An elegant derivation of the energy stored in a capacitance can also be obtained by
integrating the power delivered (or taken) by the capacitance. The instantaneous electric
power pC(t) can be written in the form

dυC 1 dυ2C
pC ðt Þ ¼ υC iC ¼ υC C ¼ C ð6:36Þ
dt 2 dt
The stored energy is then the time integration of the power, i.e.,
ðt ðt
1 dυ2C 0 1  2 
EC ðt Þ ¼ pC ðt 0 Þdt 0 ¼ C 0 dt ¼ C υC ðt Þ  υ2C ð0Þ ð6:37Þ
2 dt 2
0 0

where the lower limit, υC(0), is the initial state of the capacitance. Suppose that υC ð0Þ ¼ 0,
i.e., the capacitance is initially uncharged and has zero stored energy. Then,
1
EC ðt Þ ¼ Cυ2C ðt Þ ð6:38Þ
2
Equation (6.38) is the formal proof of the corresponding static result, Eq. (6.3), postulated
in the previous section. We can derive the energy stored in the inductance using the same
method—by integrating the power. The instantaneous power supplied to or obtained from
the inductance has the form

diL 1 di2L
pL ðt Þ ¼ υL iL ¼ iL L ¼ L ð6:39Þ
dt 2 dt
The energy stored in the inductance is the integral of Eq. (6.39), i.e.,
ðt ðt
0 0 1 di2L 0 1  2 
EL ðt Þ ¼ pL ðt Þdt ¼ L 0 dt ¼ L iL ðt Þ  i2L ð0Þ ð6:40Þ
2 dt 2
0 0

Suppose that iL ð0Þ ¼ 0, i.e., the inductance is initially “uncharged” or has no stored
energy. Then,

VI-295
Chapter 6 Dynamic Circuit Elements

1
EL ðt Þ ¼ Li2L ðt Þ ð6:41Þ
2
Equation (6.41) is the formal proof of the corresponding static result, Eq. (6.18), postu-
lated in the previous section. The series and parallel combinations of inductances and
capacitances may also be analyzed using the dynamic element equations; the laws
obtained in the previous section will be confirmed.

Exercise 6.8: Determine instantaneous power supplied to the capacitance in Fig. 6.14 at
A. t ¼ 2 μs and B. t ¼ 1 μs.
Answer: 1 W and 0 W, respectively.

Exercise 6.9: Repeat the previous exercise for the inductance shown in Fig. 6.16.
Answer: 0.002 W and 0 W, respectively.

6.2.5 DC Steady State

According to Eq. (6.28) when voltage across the capacitance does not change with
time, the capacitance becomes an open circuit (no current) under DC steady-state
condition, i.e.,
dυC dυC
¼ 0 ) iC ¼ C ¼0 ð6:42Þ
dt dt
This is to be expected since a DC current cannot flow through empty space between two
capacitor plates. Similarly, according to Eq. (6.32) the inductance becomes a short circuit
for the DC steady state when current across the inductance does not change in time, i.e.,
diL diL
¼ 0 ) υL ¼ L ¼0 ð6:43Þ
dt dt
In other words, there is no voltage drop across a (long) bent piece of wire, which is the
inductor, for DC currents. Equations (6.42) and (6.43) allow us to establish the behavior
of any transient electric circuit in the long run, after the circuit behavior has been
stabilized. The transient circuit is a circuit with dynamic elements and a switch.
Figure 6.17 shows one such circuit that consists of a number of dynamic (and static)
elements and a switch. The switch connects the voltage source to the rest of the circuit as
the switch closes at t ! 0.

VI-296
Chapter 6 Section 6.2: Dynamic Behavior of Capacitance and Inductance

t=0 10 H

40
5V + 1 F
- I
1 H

Fig. 6.17. A circuit is used to study the DC steady state. The switch closes at t ¼ 0. The behavior
of the circuit at t ! 1 is sought after the circuit has stabilized.

Example 6.6: Find current I in Fig. 6.17 at t ! 1, i.e., under DC steady state.
Solution: Immediately after the switch in Fig. 6.17 closes, the voltages and currents in the
circuit may be subject to a complicated response. In particular the voltage across certain
dynamic elements may be higher than the voltage of the power supply of 5 V. However, in
the long run as t ! 1, the circuit behavior stabilizes and we reach the DC steady state. The
capacitance in Fig. 6.17 becomes an open circuit and may be ignored. Both inductances
can be replaced by a wire (short circuit). The resulting DC circuit is shown in Fig. 6.18.
Thus, we obtain

5 V
I¼ ¼ 125 mA ð6:44Þ
40 Ω
Other more complicated circuits can be analyzed in exactly the same way.

5V + I 40
-

Fig. 6.18. A DC equivalent of the circuit in Fig. 6.17 under steady-state conditions.

6.2.6 Behavior at Very High Frequencies

At very high frequencies, the behavior of the two dynamic circuit elements is exactly the
opposite: the capacitance becomes a short circuit, whereas the inductance becomes an
open circuit. To establish this fact, we can either use the fluid mechanics analogies or the
dynamic equations themselves. For example, the inertia of a massive wheel (inductance)
will prevent any very fast movements of it so that the oscillating fluid flow will be entirely
blocked when the oscillation frequency tends to infinity. On the other hand, the forces on
a flexible membrane of zero mass will be so high for a rapidly oscillating fluid flow that
its finite stiffness no longer matters. The membrane will simply be moving along with
the fluid, which means the full transmission through its counterpart—the capacitance.

VI-297
Chapter 6 Dynamic Circuit Elements

The behavior of dynamic elements at very high frequencies is exactly as important as the
behavior at DC; it will be studied quantitatively in Chapter 9.

Example 6.7: Illustrate how is the capacitance becoming a short circuit at very high
frequencies using the capacitor’s dynamic equation as a starting point.
Solution: Assume that there is a periodic current with the amplitude of 1 A,
iC ðtÞ ¼ 1 A cos ωt through a 10-μF capacitance. The resulting capacitor voltage is given
by Eq. (6.30), υC ðt Þ ¼ 1=ðωCÞ sin ω t. When ω ¼ 108 rad=s, the capacitor voltage has the
amplitude of 1 mV. This small voltage approximately corresponds to a short circuit. When
ω increases, the voltage amplitude is reduced even further.

VI-298
Chapter 6 Section 6.3: Application Circuits Highlighting Dynamic Behavior

Section 6.3 Application Circuits Highlighting Dynamic

Behavior
6.3.1 Bypass Capacitor
Let us consider the circuit shown in Fig. 6.19a. It includes a voltage source represented by
its Thévenin equivalent and a load represented by its equivalent resistance RL. The source
generates a voltage in the form of a (large) DC component VS and superimposed (small)
AC signal υS(t). This setup could model a nonideal DC voltage power supply, which does
not create the exactly DC voltage. In fact, a weak AC component may be present. This
AC component (also called the noise component) has a frequency of either 60  n Hz,
where n is an integer (USA, Canada, parts of South America, Saudi Arabia, etc.), or
50  n Hz (the rest of the world) and appears due to a not quite perfect rectification of the
primary AC power. As an aside, switching power supplies create noise spikes at much
higher frequencies. The weak AC component may lead to circuit oscillations, especially
when dealing with high-gain amplifiers. It therefore should be removed from the load, or
“filtered out” as engineers often say. The idea is to use a capacitor C in parallel with the
(imperfect) power supply and in parallel with the load, the so-called bypass capacitor.
This capacitor ideally becomes a short circuit for the high-frequency noise component of
the source and shorts it out (or bypasses). The corresponding circuit diagram is shown in
Fig. 6.19b. An electrolytic capacitor is typically used as the bypass capacitor.

a) b)
source source
load load
RS RS

vS(t) + vS(t) +
- DC - DC C
AC RL AC RL
VS + VS +
- -

Fig. 6.19. Model of a voltage source connected to a load with a bypass capacitor.

In many cases, the undesired noise source in a circuit is not a low-frequency noise
source of a nonideal power supply, but rather a high-frequency noise generator. Examples
include high-speed DC motors, analog-to-digital converters, and other digital circuits.
Radio-frequency (RF) high-speed amplifiers are also very sensitive to RF noise that is
created by connectors and wires which can act like antennas. By placing a bypass
capacitor as closely as possible to the power supply pins of every chip, such RF noise
sources may be eliminated. Bypass capacitors are so prevalent that they are encountered
in virtually every working piece of electronic equipment.

VI-299
Chapter 6 Dynamic Circuit Elements

Example 6.8: Explain the operation of bypassing a DC motor.

Solution: A DC motor with a bypass capacitor is also described by the model shown in
Fig. 6.19b. In this case, the source voltage V S þ υS ðt Þ becomes the induced electromotive
force (emf), E, of the motor, The induced emf is still a DC voltage but with quite a significant
high-frequency noise component created by the spinning rotor comprised of a finite number
of individual switched coils. The source resistance becomes the armature and brush resis-
tance RM, that is, RS ¼ RM . The load resistance RL may, for example, be the oscilloscope
resistance. We consider a small DC fan motor directly connected to a 5-V power supply
shown in Fig. 6.20. The motor creates a substantial high-frequency noise seen on the
oscilloscope in Fig. 6.20, left (with 100 mV per division resolution). The oscilloscope
measures the voltage across the motor, which is the 5-V DC component plus the noise
component. Once a 1000-μF capacitor is connected in parallel with the motor (one may call it
a shunt capacitor), the resulting voltage becomes a highly stable 5-V DC (see Fig. 6.20,
right).

Fig. 6.20. Effect of bypass capacitor on the high-frequency noise created by a DC motor.

The bypass capacitor in Fig. 6.20 may be considered as a part of the snubber RC
circuit, which includes a capacitance and a small series resistance. The snubber circuits
are used to suppress high-voltage spikes in inductive switching systems like electric
motors.

VI-300
Chapter 6 Section 6.3: Application Circuits Highlighting Dynamic Behavior

6.3.2 Blocking Capacitor

Quite often, an opposite scenario is desired—we want to block a DC component at the
load—see Fig. 6.21. An example is an audio amplifier (the source) connected to a speaker
(the load). The audio amplifier may generate an unwanted DC component, which may
overheat the speaker coil made of a very thin wire. The idea is to use a capacitor C in
series with the (imperfect) amplifier and in series with the load. This blocking or
decoupling capacitor will block the DC current at the load as shown in Fig. 6.21b.

a) b)
source source
load C load
RS RS

vS(t) + DC vS(t) +
- -
AC RL AC RL
VS + VS +
- -

Fig. 6.21. Model of a voltage source connected to a load with a blocking capacitor.

6.3.3 Decoupling Inductor

A decoupling inductor is the complement of the bypass capacitor. Consider the circuit
shown in Fig. 6.22a. It includes a source represented by its Norton equivalent I S þ iS ðt Þ,
RS and a load with an equivalent resistance RL. The source generates an electric current
in the form of a DC component and a superimposed AC signal. Assume that we would
like to have only the direct current at the load. The idea is to use an inductor L in series
with the load, the so-called decoupling inductor as shown in Fig. 6.22b. At a sufficiently
high frequency, this inductor will block the AC component at the load.

a) source load b) source load

L

iS(t) AC iS(t) AC
RS RL RS RL
IS IS

Fig. 6.22. Model of a current source connected to a load with a decoupling inductor.

A common application of the decoupling inductor is the so-called radio-frequency

(RF) inductor choke. Here, we’d rather intend to redirect the alternating current. The
inductor choke prevents the very weak alternating current received by an antenna to be
lost in the low-resistance DC power supply, which powers an amplifier. Instead, it forces
the current to flow directly into the input port of the amplifier. In order to model the choke
effect in Fig. 6.22, we should in fact interchange the role of two resistances: we consider
the load resistance as the DC supply resistance and the source resistance as the desired
input resistance of the amplifier.
VI-301
Chapter 6 Dynamic Circuit Elements

6.3.4 Amplifier Circuits With Dynamic Elements: Miller Integrator

Amplifier circuits with the dynamic elements in the negative feedback loop can serve
different purposes. In particular, they operate as active filters. Here, we will introduce the
operation concept and present simple examples. The Miller integrator circuit is an
inverting-amplifier circuit considered in Chapter 5, but with the feedback resistance R2
replaced by a capacitance C—see Fig. 6.23. Given a time-varying input voltage signal, the
capacitor will conduct a current. Therefore, the negative feedback is still present, even
though we now have a capacitance instead of the resistance in the feedback loop.

+ vC -

vin R i2
+ v* - vout
+ vx +
i1 +
- -
-

Fig. 6.23. Miller integrator circuit.

The circuit analysis uses two summing-point constraints: no current into the amplifier
and zero differential input voltage. Therefore, the node voltage υ * in Fig. 6.23 is also
zero. The currents i1 ¼ iR and i2 ¼ iC in Fig. 6.23 are equal to each other. This yields
iC
zfflffl}|fflffl{ iR
z}|{
dυC d ðυ*  υout Þ dυout υR υin  υ* υin
C ¼C ¼ C ¼ ¼ ¼ )
dt dt dt R R R
ðt
dυout υin 1
C ¼ ) υout ¼  υin ðt 0 Þdt 0  V C ð6:45Þ
dt R RC
0

where VC is a constant (the initial voltage across the capacitor at t ¼ 0). Thus, an integral
of the input voltage (weighted by 1=ðRC Þ) is provided at the output. Interestingly, the
time constant τ of the integrator, τ ¼ RC, has the unit of seconds.

Example 6.9: The analog pulse counter is an integrator circuit shown in Fig. 6.23 that
counts monopolar voltage pulses simply by integrating the input voltage as time pro-
gresses. Assume that the input to the amplifier is the voltage shown in Fig. 6.24, where
every rectangular voltage pulse of 10-ms duration corresponds to a car passing through a
gate. Given that R ¼ 10 kΩ, C ¼ 0:1 μF, and that the initial value of the output voltage is
reset to zero, how many cars should pass the gate in order to reach the output voltage
threshold of 6 V?
Solution: The time constant τ of the integrator, τ ¼ RC, is equal to 1 ms.

VI-302
Chapter 6 Section 6.3: Application Circuits Highlighting Dynamic Behavior

Example 6.9 (cont.):

According to Eq. (6.45),

X
N  
υout ¼ 1000 s1 0:06 V  0:01 s ¼ N  0:6 V ð6:46Þ
n¼1

where N is the number of pulses (cars). Equating the above expression to 6 V gives
N ¼ 10. The time interval between passing cars is not important.

Input voltage, mV
80
60
40
20
0
0 10 20 30 40 50
time, ms

Fig. 6.24. Input signal to the amplifier in Example 6.9.

Along with Example 6.9, other applications of the Miller integrator include various
wave-shaping circuits.

6.3.5 Compensated Miller Integrator

The circuit in Fig. 6.23 will not function in the laboratory, when a realistic amplifier chip
is used that is different from the ideal-amplifier model. The reason is that the capacitance
is equivalent to an open circuit at DC. Therefore, the feedback loop is simply missing in
the Miller integrator at DC, and the entire amplifier circuit becomes a comparator with a
very high open-loop gain. A small random input offset voltage, VOS, which is present for

open circuit actual op-amp

ideal op-amp

- vout
+
- +
+

R
VOS
-

Fig. 6.25. Effect of the input offset voltage on the integrator circuit at DC.

VI-303
Chapter 6 Dynamic Circuit Elements

any realistic amplifier IC, will saturate the amplifier toward one of the power rails
(depending on the sign of VOS) even if its input is at zero volts (grounded)—see Fig. 6.25.
Note that the input offset voltage source, VOS, may be added to either amplifier
terminal. Hence, the voltage across the capacitance will approach the rail voltage and
the capacitance itself will become permanently charged. The dynamics of this process
can be analyzed explicitly, starting with some initial voltage value, say υC ¼ 0 V.
In this case, we are allowed to use the negative feedback. For example, given that
R ¼ 10 kΩ, C ¼ 0:1 μF, V OS ¼ 5 mV; it takes exactly 1 s to reach the output voltage
of 5 V! A similar effect is created by input bias currents to the amplifier. To overcome
this issue, a large resistance, RF, is introduced in parallel with C in order to maintain the
negative feedback at DC and discharge the capacitance as needed—see Fig. 6.26. If, for
example, RF ¼ 10 MΩ, then the capacitance will discharge over time on the order of
RF C ¼ 1 s. This estimate is comparable with the estimate for the charging time. As a
result, a balance will be established that results in a certain nonzero υout with the output
of the amplifier grounded. A further quantitative discussion may be carried out.

RF

vin R C
v* - vout
+ +
+
- -

Fig. 6.26. Miller integrator improved with a large resistance, RF, in the feedback loop.

6.3.6 Differentiator and Other Circuits

When the resistance and the capacitance in Fig. 6.23 are interchanged, a differentiator
amplifier circuit (or active differentiator) is obtained; its output signal is a derivative of
the input signal. The corresponding solution is studied in one of the homework problems.
The differentiators are rarely used in practice since they attempt to amplify any input
noise (they become “noise magnifiers”). The reason for this is an infinitely high gain of
the amplifier circuit at high frequencies, when the capacitance becomes a short circuit.
A small resistance added in series with the capacitance reduces this effect and assures the
finite gain similar to the standard inverting amplifier.

Exercise 6.10: Draw an integrator circuit with an inductance instead of the capacitance.
Answer: The circuit in Fig. 6.23 but with the resistance replaced by an inductance and with
the capacitance replaced by the resistance.

VI-304
Chapter 6 Section 6.3: Application Circuits Highlighting Dynamic Behavior

Inductances might be used instead of capacitances; the amplifier circuits so constructed

will be either a differentiator or an integrator. We again pass the corresponding analysis to
the homework exercises. However, the physical inductors tend to have a significant series
resistance and are more bulky. Last but not least, we may ask ourselves a question: as
long as the amplifier circuits can perform multiplication, addition (or subtraction), and
integration (or differentiation), can we now build an analog computer, which operates
with analog voltages and replaces its digital counterpart at least for simple computational
tasks? The answer is yes, we can. In fact, this was done a long time ago, in the mid-1960s.

VI-305
Chapter 6 Dynamic Circuit Elements

Summary
Static capacitance and inductance
Property Capacitance Inductance

Definition Q Φ
C¼ >0 L¼ >0
V I
Q—charge of either conductor; Φ—magnetic flux through the
V—volt. between two conductors circuit; I—circuit current
Units: F ¼ C=V Units: H ¼ ðV  sÞ=A
Physical meaning Charge on either conductor Magnetic flux through the
produced by 1 V voltage difference circuit produced by 1 A of
between the two conductors current in the same circuit
Stored energy, 1 1 1 1
E ¼ CV 2 or E ðt Þ ¼ Cυðt Þ2 E ¼ LI 2 or E ðt Þ ¼ Liðt Þ2
J (static or dyn.) 2 2 2 2
Series/parallel 1 1 1 1 Leq ¼ L1 þ L2 þ L3 in series
¼ þ þ in series
combinations C eq C 1 C 2 C 3 1 1 1 1
¼ þ þ in parallel
C eq ¼ C 1 þ C 2 þ C 3 in parallel Leq L1 L2 L3

Basic models
(no fringing fields)
ε0 A εr ε0 A μ0 AN 2 μ μ AN 2
C¼ or C ¼ (diel. L¼ or L ¼ 0 r
d d l l
material inside) (closed magnetic core )
A ¼ ab, ε0 ¼ 8:854187  1012 F=m μ0 ¼ 4π  107 H=m
Models with fringing
Capacitance of a
parallel-plate
square capacitor
(Table 6.1)


Inductance of a 8w w2 w4 r
finite-radius Lexact ¼ L 1  þ  , w ¼ < 1; r is the radius of the coil
3π 2 4 l
solenoid


Inductance of a 0:5πμ0 l * N 2 l
solenoid with a L  h* i 1  * , μr  100, see Fig. 6.12c
ln lr  1 2l
finite core
(continued)

VI-306
Chapter 6 Summary

Dynamic behavior
Property Capacitance Inductance
Dynamic model-
Passive ref. conf.

Dynamic model- dυC diL

iC ¼ C υL ¼ L
υ–i characteristic dt dt
Dynamic model- dqC dΦðt Þ
¼ iC ¼ υL (passive ref. conf.)
charge and flux dt dt
Behavior at DC
Behavior at very
high frequencies
Bypassing/Decoupling

Bypass capacitor

Blocking
capacitor

Decoupling
inductor

Amplifier circuits with capacitor/inductor

Property Capacitance Inductance
Miller Integrator
(open-loop
amplifier at DC)

Time-domain dυout υin dυout υin L

¼  , τ ¼ RC ¼ , τ¼
operation dt τ dt τ R
(continued)

VI-307
Chapter 6 Dynamic Circuit Elements

Compensated
integrator with a
finite gain at DC

Time-domain dυout υout υin dυout Rin υin L

þ ¼  , τ ¼ RC þ υout ¼  , τ ¼
operation dt Rf C τ dt L τ R

Differentiator cir-
cuit (infinite gain
at very high
frequencies)

Time-domain dυin dυin L

υout ¼ τ , τ ¼ RC υout ¼ τ , τ¼
operation dt dt R

Differentiator with
a finite gain at very
high frequencies

Time-domain dυout dυin L dυout dυin

Rin C þ υout ¼ τ , τ ¼ RC þ υout ¼ τ , τ ¼ L=R
operation dt dt Rf dt dt

VI-308
Chapter 6 Problems

Problems +
vC
b

6.1 Static Capacitance a

0 r

h
and Inductance
E
6.1.1 Capacitance, Self-capacitance, and
Capacitance to Ground
6.1.2 Application Example: ESD
Problem 6.1. A. Determine capacitance of the
A. Describe in your own words the physical capacitor, C.
meaning of capacitance. B. Determine the electric field strength,
B. Suggest a way to memorize the expres- E (in V/m), within the capacitor volume
sion for the capacitance of two and total charge, Q, on either capacitor
conductors. plate if the applied voltage is 25 V.
C. What is approximately the self- C. Determine the electric field energy stored
capacitance of a human body? in the capacitor if the applied voltage is
D. What is approximately the capacitance 25 V.
of a human body (to ground)?
E. How does the human-body self-capaci- Problem 6.5. Solve the previous problem when
tance change in embryo pose (yoga)? the separation distance, h, between the plates is
reduced to 100 μm.
Problem 6.2. A metal square plate with the
side of 10 mm has the self-capacitance Problem 6.6. For the enclosed-cylinder capac-
C self ¼ 0:41 pF. itor shown in the figure, a ¼ 10 cm, b ¼ 9:99
A. Estimate the capacitance of a capacitor cm, H ¼ 5 cm, εr ¼ 16, the electrodes are the
formed by two such parallel plates sepa- inner and outer cylinder surfaces, respectively.
rated by 30 mm. Compare this value to A. Determine the capacitance of the
the exact result of 0.23 pF. capacitor, C.
B. Estimate the capacitance of the plate to B. Determine the electric field strength,
ground when the separation distance is E (in V/m), within the capacitor volume
30 mm. Compare this value to the exact and total charge, Q, on either capacitor
result of 0.43 pF. plate if the applied voltage is 50 V.
C. Determine the electric field energy stored
Problem 6.3. Draw the basic electric-circuit in the capacitor if the applied voltage is
model of a human body and specify the generic 25 V.
element values.
Hint: The capacitance per unit area of the
6.1.3 Parallel-Plate Capacitor device is that of the parallel-plate capacitor.
6.1.4 Capacitances in Parallel and
+ V - C
in Series 0 r
a
Problem 6.4. For the parallel-plate capacitor b
schematically shown in the figure,
H

a ¼ 10 cm, b ¼ 20 cm,
h ¼ 1 mm, εr ¼ 12

VI-309
Chapter 6 Dynamic Circuit Elements

Problem 6.7. Solve the previous problem when 47 F

a
the separation distance between the two elec-
trodes is reduced to 25 μm.
47 F 47 F

Problem 6.8. A cross section of the rolled b

capacitor is approximated by a rational spiral 47 F
shown in the figure. Here,
a ¼ 0:25 cm, εr ¼ 10, the separation distance Problem 6.11. Find the equivalent capacitance
between the two conductors is 20 μm, and the for each circuit shown in the figure below.
height of the entire roll is 1.5 cm.
A. Determine capacitance of the a) 10 F 10 F
a
capacitor, C.
B. Determine the electric field strength,
15 F
E (in V/m), within the capacitor and 47 F 15 F
total charge, Q, on either capacitor plate
if the applied voltage is 12 V. b
C. Determine the electric field energy stored
in the capacitor if the applied voltage is b) 10 F 10 F
a
12 V.
15 F 15 F
Hint: The capacitance per unit area of the 47 F
device is that of the parallel-plate capacitor. 15 F

a 0 r b
VC
+ -
Problem 6.12. Find the equivalent capacitance
for each circuit shown in the following figure.

a)
33 nF 33 nF

a b

Problem 6.9. Determine the capacitance of the

three leaded capacitors shown in the figure 33 nF 33 nF

(from left to right). b) a

2.2 pF 2.2 pF

2.2 pF 2.2 pF

Problem 6.13. Two 1-μF capacitances have an

Problem 6.10. Find the equivalent capacitance initial voltage of 10 V and 0 V, respectively
for the circuit shown in the following figure. (before the switch is closed), as shown in the
figure. Find the total electric energy stored in
the system before the switch is closed. Find the

VI-310
Chapter 6 Problems

voltage across each capacitance and the total surrounding space if the applied current,
stored energy after the switch is closed. What IL, is 0.5 A.
could have happened to the missing energy?

t=0

+ +
10 V 1 F 1 F 0V
- -

Problem 6.17.
6.1.7 Self-inductance (Inductance) and A. Obtain more accurate answers to the pre-
Mutual Inductance vious problem using the precise expres-
6.1.8 Inductance of a Solenoid With and sion for the inductance of a solenoid,
Without Magnetic Core which is Eq. (6.22).
B. Estimate the relative error of the com-
6.1.9 Circuit Symbol. Inductances in
mon expression for solenoid’s induc-
Series and in Parallel tance, Eq. (6.21).
Problem 6.14.
A. Describe in your own words the physical Problem 6.18.
meaning of inductance. A. Design a 100-μH air-core inductor (deter-
B. Do you think a straight wire has a certain mine the necessary number of turns) with
inductance per unit length? You might a radius of 1 cm and a length of 10 cm.
want to ask the TA and/or browse the B. Determine the magnetic field energy
Web and present the corresponding stored in the inductor if the applied cur-
expression (if any). rent, IL, is 1.0 A.
Problem 6.15. Three air-core inductors of the Problem 6.19.
same cross section are shown in the following A. Design a 0.5-mH air-core inductor
figure. The inductor length is proportional to (determine the necessary number of
the number of turns. Find the ratios of induc- turns) having a radius of 0.5 cm and a
tances: L2/L1, L3/L2, L3/L1. length of 10 cm.
B. Determine the magnetic field energy
+
stored in the inductor if the applied cur-
+ rent, IL, is 1.0 A.
+

L1 L2 L3
Problem 6.20. The solenoid shown in the
figure has diameter, d ¼ 1 cm, length,
- - l ¼ 10 cm, and a closed magnetic core of the
same circular cross section with μr ¼ 1000.
- A. Find the solenoid’s inductance, L, using
the common assumption d=l << 1.
Problem 6.16. The solenoid shown in the fig- B. Determine the magnetic field energy
ure has a diameter d ¼ 1 cm and a length stored in the inductor if the applied cur-
l ¼ 10 cm. rent, IL, is 0.5 A.
A. Find the solenoid’s inductance, L, using C. Compare two of your answers to the
the common assumption d=l << 1. answers in Problem 6.16.
B. Determine the magnetic field energy D. Determine voltage across the inductor
stored within the inductance and in the assuming the ideal (zero-resistance) wire.

VI-311
Chapter 6 Dynamic Circuit Elements

+ a)
0 r
A IL 10 nH 10 nH

N vL
a b

B= 0 rH

20 nH 20 nH

b)
Problem 6.21. a
A. Obtain a more accurate solution to 10 nH 20 nH
the previous problem using the precise
expression for the inductance of a
solenoid, Eq. (6.22). To do so, propose
your own modification of this expression 20 nH 10 nH
to include the effect of the
b
magnetic core.
B. Estimate the relative error of the simpli- Problem 6.25. For the inductor shown in
fied expression for the inductance, Fig. 6.11, l * ¼ 2l ¼ 10 cm and r ¼ 0:75 cm.
Eq. (6.21). How do the inductances with the magnetic core
(with μr  100) and without the core compare
Problem 6.22. Find the equivalent inductance
to each other?
for the circuit shown in the following figure.
Problem 6.26. Repeat the previous problem
10 H
a with l * ¼ 15 cm, but still 2l ¼ 10 cm. The
other parameters are the same.
10 H 10 H

b
10 H
6.2 Dynamic Behavior
of Capacitance
Problem 6.23. Find the equivalent inductance
for circuits shown in the figure that follows.
and Inductance
6.2.2 Dynamic Behavior of Capacitance
a) 5 H 5 H
6.2.3 Dynamic Behavior of Inductance
a 6.2.4 Instantaneous Energy and Power
5 H 10 H 5 H of Dynamic Circuit Elements
b
Problem 6.27.
A. Sketch the circuit symbol for the capac-
b) 5 H 5 H
itance and the inductance; label the
a corresponding voltages and currents.
5 H 2000 nH 15 H 5 H B. Write two dynamic equations for the
capacitance and the inductance which
b
relate the voltages and currents.
C. Express the units for the inductance and
Problem 6.24. Find the equivalent inductance capacitance through volts and amperes.
for circuits shown in the figure.

VI-312
Chapter 6 Problems

D. Which capacitance and inductance the capacitance voltage to scale versus

values are typical in electronic circuits? time over the interval from 0 to 0.05sec.
B. How much time is approximately neces-
Problem 6.28. sary to charge the capacitance to 1.5 V?
A. Sketch the hydrodynamic analogies for C. Solve the same problem analytically.
the capacitance and the inductance.
B. Which mechanical quantities are associ- Problem 6.32. Repeat the previous problem
ated with the capacitance and inductance? with the capacitance changed to 1 μF. Does
the 10-V charging time increase or decrease?
Problem 6.29. The voltage across a 10-μF
capacitance is shown in the following figure. Problem 6.33. The current through a 5-mH
At t ¼ 0, the voltage is zero. Sketch the current inductance is shown in the following figure.
through the capacitance to scale versus time. At t ¼ 0, the current is zero. Sketch the voltage
across the inductance to scale versus time.
vC, V
iL, mA
2
10 F capacitance
vC 2
5 mH inductance
+ - vL
1
+ -
1
iC
iL
0 1 2 t, s
0 1 2 3
Problem 6.30. Repeat the previous problem for t, s
the voltage shown in the following figure.
A. How is the solution different from the Problem 6.34. Repeat the previous problem for
previous problem? the current shown in the figure.
B. For creating large currents, should the A. How is the solution different from the
voltage across the capacitance change previous problem?
slowly or quickly? B. For creating large voltages, should the
current through the inductance change
vC, V slowly or quickly?

iL, mA
2
10 F capacitance
vC
+ - 5 mH inductance
2
1
vL
+ -
iC 1

0 1 2 t, ms iL

0 1 2 3
Problem 6.31. A 10-μF capacitance is charged t, ms
by the current iC ¼ 1 mA  cos 2 ð1000t Þ. At
t ¼ 0, the capacitance voltage is zero. Problem 6.35. The voltage across a 3-mH
A. Using software of your choice inductance is given by υL ðt Þ ¼
(MATLAB is recommended), sketch 10 mV  cos 2 ð1000t Þ. The current through the

VI-313
Chapter 6 Dynamic Circuit Elements

inductance is equal to zero at the initial time t=0 10 F

t ¼ 0. Using software of your choice
(MATLAB is recommended), plot the current
through the inductance for t  50 ms. 50
10 V
+ I 50
6.2.5 DC Steady State -
Problem 6.36. 10 H

A. Draw the equivalent circuit for the 10 mH

inductance shown in the figure at DC.
B. Repeat for the capacitance.
Problem 6.40.
a b A. Find current I in the circuit shown in the
figure at t ! 1, in the DC steady state.
a b
The switch closes at t ¼ 0.
B. Find voltage across the capacitance in
Problem 6.37. Find current I in the circuit the DC steady state and label its polarity
shown in the figure that follows at t ! 1, in in the figure.
the DC steady state. The switch closes at t ¼ 0.
t=0

t=0 10 H
1 F

3A 50

100 I 50
10 V
+ 10 H
- I 100

10 H 50

10 F
0.01 H

Problem 6.41. Find the voltage across the

Problem 6.38. Find current I in the circuit capacitance in the DC steady state and label
shown in the figure at t ! 1, in the DC steady its polarity in the figure at t ! 1, in the DC
state. The switch closes at t ¼ 0. steady state. The switch opens at t ¼ 0.
t=0
t=0

1 nF
50
10 V 100
3A
+ I 50
- I 50
10 H
10 H
10 mH
50

0.01 H
Problem 6.39. Find current I in the circuit
shown in the following figure at t ! 1, in
the DC steady state. The switch closes at t ¼ 0.

VI-314
Chapter 6 Problems

Problem 6.42. Determine the equivalent resis- 6.3.4 Amplifier Circuits with Dynamic
tance between terminals a and b for the circuit Elements: Miller Integrator
shown in the following figure in the DC steady 6.3.5 Compensated Miller Integrator
state.
6.3.6 Differentiator and Other Circuits
a
Problem 6.45. The input voltage to the Miller
integrator circuit with the ideal amplifier
10 F shown in the figure is a series of rectangular
50 voltage pulses. Each is 50 mV tall and 8 ms
50 wide. Given that the initial value of the output
voltage is zero, how many voltage pulses are
10 H
necessary to reach the negative output voltage
50 threshold of 8 V?
b
10nF

Problem 6.43. Determine the equivalent resis- vin(t) 100k

vout(t)
tance between terminals a and b for the circuit -
+
shown in the following figure in the DC steady
+
+
- -
state.

a
Problem 6.46. The input voltage to the Miller
10 F
integrator circuit with the ideal amplifier shown
100
10 H in the figure is a series of rectangular voltage
100 pulses. Each is 50 mV tall and 16 ms wide.
10 H Given that the initial value of the output voltage
is zero, how many voltage pulses are necessary
100 to reach the negative output voltage threshold
b of 9.2 V?

10nF
6.3 Application Circuits vin(t) 200k
vout(t)
Highlighting Dynamic + +
-
+
-
Behavior -
6.3.1 Bypass Capacitor
6.3.2 Blocking Capacitor Problem 6.47. How would you modify the
6.3.3 Decoupling Inductor circuit to the previous problem when the posi-
Problem 6.44. Describe the purpose of a tive threshold voltage of +9.2 V should be
A. Bypass capacitor reached at the output?
B. Blocking capacitor
C. Decoupling inductor Problem 6.48. For the circuit shown in the
following figure, express the output voltage,
in your own words. Specify the placement of υout(t), as a function of time in terms of the
each component: in series or in parallel with the input voltage, υin(t), and circuit parameters R,
source. C. Assume ideal amplifiers.

VI-315
Chapter 6 Dynamic Circuit Elements

R Problem 6.51. For the circuit shown in the

vin(t) C
following figure, express the output voltage,
- υout(t), as a function of time in terms of the
+ +
- input voltage, υin(t), and circuit parameters R,
L. Assume ideal amplifier.

L
R

vin(t) R
C vout(t) - vout(t)
-
+ +
+
+
+ -
- -

Problem 6.52. Explain why is the Miller inte-

Problem 6.49. For the circuit shown in the grator typically used with a shunt resistance,
following figure, express the output voltage, RF.
υout(t), as a function of time in terms of the
input voltage, υin(t), and circuit parameters Problem 6.53. For the circuit shown in the
R, C. Assume the ideal amplifier. figure, assume that the voltage across the
capacitance at t ¼ 0 is zero. Also assume that
R
the negative feedback is present. Derive the
vin(t) C
- vout(t) dynamic expression for υC(t) at any positive
+ + time instance.
+
- -
+ vC - actual op-amp
ideal op-amp
VOS/R
vout
Problem 6.50. For the circuit shown in the VOS -
following figure, express the output voltage,
+
- +
+

R
υout(t), as a function of time in terms of the 0V
VOS -
input voltage, υin(t), and circuit parameters R, 0V 0V
L. Assume ideal amplifier.
Problem 6.54. Establish the gain of amplifier
R circuits shown in the figure that follows.
vin(t) L
- vout(t)
+ +
+
- -

VI-316
Chapter 6 Problems

a) 0, DC Problem 6.55. For two circuits shown in the

figure that follows, obtain an analytical expres-
Rf sion for the output voltage as a function of time
vin C and circuit parameters when the input voltage
R vout
-
+
has the form υin ðt Þ ¼ 1expðαt Þ ½mV .

a) Rf
b) R 0, DC
C
vin Rin L vout vin
- R - vout
+ +

c) R
b) R

vin Rin L vout

vin Rin C vout -
- +
+

d)
Rf

vin L
R - vout
+

VI-317
Chapter 7

Chapter 7: Transient Circuit Fundamentals

Overview
Prerequisites:
- Knowledge of first-order ordinary differential equations (calculus)
- Knowledge of Thévenin/Norton equivalent circuits (Chapter 4)
- Knowledge of constitutive relations for dynamic circuit elements (Chapter 6)
- Knowledge of basic amplifier theory (Chapter 5)

Objectives of Section 7.1:

- Demonstrate the universal character of the KVL/KCL as applied to any electric
circuit including transient circuits
- Establish the general character of the time constant τ ¼ RC for RC circuits
- Establish the continuity of the capacitor voltage and its role in circuit ODEs
- Solve any first-order transient RC circuit configuration and understand the
practical meaning of the RC circuit using different application examples

Objectives of Section 7.2:

- Demonstrate the universal character of the KVL/KCL as applicable to any
electric circuit including transient circuits
- Establish the general character of the time constant τ ¼ L=R for RL circuits
- Establish the continuity of the inductor current and its role in circuit ODEs
- Solve any first-order transient RL circuit configuration and understand the
practical meaning of the RL circuit using an application example
Objectives of Section 7.3:
- Obtain initial exposure to a bistable amplifier circuit with positive feedback
- Understand the principle of operation of a relaxation oscillator—RC timer—on
the base of the bistable amplifier circuit
- Establish oscillation frequency and voltage amplitudes from the relaxation
oscillator; demonstrate the corresponding laboratory setup
- Briefly discuss the 555 timer IC

© Springer Nature Switzerland AG 2019 VII-319

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_7
Chapter 7 Transient Circuit Fundamentals

Objectives of Section 7.4:

- Define the single-time-constant (STC) transient circuit
- Be able to classify any transient circuit with dynamic elements of the same type
as either a single-time-constant circuit or a more complicated circuit
- Solve an example of a non-STC circuit
- Convert an arbitrary transient circuit with one capacitance or one inductance to
the basic RC/RL first-order circuit
- Solve a first-order transient circuit with a harmonic forcing function

Objectives of Section 7.5:

- Understand topology and classification for the second-order transient circuits
- Convert a transient circuit with a series/parallel LC block to the standard second-
order RLC series/parallel transient circuits
- Introduce two major RLC circuit parameters: damping coefficient and undamped
resonant frequency
- Introduce the step response of a second-order transient circuit as a solution with a
DC source and a switch. Understand the general value of the step response
- Properly select the independent function (capacitor voltage or inductor current)
for the standard form of the step response with zero initial conditions

Objectives of Section 7.6:

- Use the method of characteristic equation for second-order transient circuits
- Understand the meaning of overdamped, critically damped, and underdamped
circuits
- Use the value of damping ratio ς to distinguish between three different cases of
circuit behavior
- Obtain the complete analytical solution for the step response of the RLC circuit
- Apply this solution for modeling a nonideal (realistic) digital waveform

Application Examples:
- Electromagnetic railgun
- Electromagnetic material processing
- Digital memory cell
- Laboratory ignition system
- RC timer or clock circuit in laboratory
- Transient circuit with a bypass capacitor
- Modeling and origin of the nonideal digital waveform

VII-320
Chapter 7 Transient Circuit Fundamentals

Keywords:
Transient RC circuit, Transient RL circuit, Energy-release RC/RL circuit, Energy-accumulating
RC/RL circuit, Time constant of RC circuit, Time constant of RL circuit, Relaxation time, Voltage
continuity across the capacitor, Fluid mechanics analogy of transient RC circuit, Lorentz force,
Self-induced Lorentz force, Railgun, Electromagnetic material processing, Electromagnetic
forming, Current continuity through the inductor, Fluid mechanics analogy of transient RL
circuit, Forced response, Electronic ignition system, Piezoelectric effect, Clock frequency,
Clock signal, Positive feedback, Linear oscillators, Switching oscillators, Switching RC
oscillator, Astable multivibrator, Relaxation oscillator, Bistable amplifier circuit (operation,
threshold voltage, mechanical analogy, triggering, trigger signal), Digital memory element,
Inverting Schmitt trigger, Non-inverting Schmitt trigger, 555 timer IC, Single-time-constant
circuits (definition, classification of, examples of, with general sources), STC circuits, Non-
STC circuits (definition, examples of), Series RLC circuit (generic representation, qualitative
description, mechanical analogy, step response, duality), Parallel RLC circuit (generic
representation, qualitative description, mechanical analogy, step response, duality), Second-
order ODE (homogeneous, nonhomogeneous, initial conditions, in terms of current, in terms of
voltage, forcing function, general solution, forced response, particular solution, complementary
solution natural response, step response), Damping coefficient, Neper, Time constant of the decay
envelope, Undamped resonant frequency, Step response, Impulse response, Damping ratio,
Natural frequency, Overdamped circuit, Critically damped circuit, Underdamped circuit,
Overshoot, Undershoot, Rise time, Fall time, Ringing, Nonideal digital waveform

VII-321
Chapter 7 Transient Circuit Fundamentals

Section 7.1 RC Circuits

The first-order RC circuits explored in this section involve the process of discharging or
charging a capacitor. This is a time-dependent, or transient, circuit behavior, and to
understand it, we are required to solve dynamic circuit equations. Mathematically, this
implies the solution of first-order ordinary differential equations (ODEs) with time as one
independent variable. Fortunately, KVL and KCL remain valid for any static or dynamic
circuit. These laws can be employed to derive the circuit equations. After that, it is either
solved analytically for simple circuits or numerically for realistic RC circuits.

7.1.1 Energy-Release Capacitor Circuit

The circuit in Fig. 7.1 depicts a capacitor, C, that has been charged to a certain voltage

V 0 ¼ υC ðt  0Þ ð7:1Þ

prior to use. Through a switch, the capacitor is connected to a load, represented by a

resistor R ¼ 10 Ω. The switch shown in Fig. 7.1 may be a transistor switch. We assume
that the switch closes and thereby connects the load to the capacitor at t ¼ 0. Our goal is
to find all circuit parameters, plus the power delivered to the load as functions of time.

t=0

+ +
C=10 F vC iC iR vR R=10
- -

Fig. 7.1. Discharging a capacitor through a load resistor.

The solution to this dynamic circuit is based on applying KVL and KCL, which are
valid for all electric circuits. Using KCL gives the result:

iC ¼ iR ð7:2Þ

at any instance of time, t. Since both circuit elements in Fig. 7.1 are passive, we can apply the
constitutive relations between currents and voltages without changing the sign:
iC
zfflffl}|fflffl{ iR
z}|{
dυC υR υC
C ¼ ¼ ð7:3Þ
dt R R
This is true because KVL states for any positive time, t > 0,

VII-322
Chapter 7 Section 7.1: RC Circuits

υR ¼ υC ð7:4Þ

Equation (7.3) therefore yields

dυC υC dυC υC
C þ ¼0) þ ¼ 0, τ ¼ RC ð7:5Þ
dt R dt τ
This is the famous first-order transient circuit equation. Here, τ carries units of seconds
since R is recorded in Ω and C is given in F ¼A  s/Vand is called the time constant or the
relaxation constant of the circuit. It is the only constant that is present in the first-order
differential equation. The solution of an ODE of this type has the generic form
 t
υC ðt Þ ¼ Kexp  ð7:6Þ
τ
This fact is proven by direct substitution. The constant K is determined from the initial
condition, Eq. (7.1), which yields

K ¼ V0 ð7:7Þ

Thus, the circuit voltages have the same form

 t
υC ðt Þ ¼ υR ðt Þ ¼ V0 exp  ; t  0 ð7:8aÞ
τ
for nonnegative values of t. However, although the capacitor voltage is equal to V0 at
t < 0, the resistor voltage is exactly zero at t < 0, since the switch was open. The current
through the load resistor is

υR ðt Þ V0  t
i R ðt Þ ¼ ¼ exp  ; t0 ð7:8bÞ
R R τ
and is zero for negative t. We recall that the capacitor current is the negative of the load
current. The instantaneous power delivered to the load resistance is expressed in the form

V0 2  t
pR ðt Þ ¼ υR ðt ÞiR ðt Þ ¼ exp 2 ; t0 ð7:8cÞ
R τ
Equations (7.8a–c) provide the complete solution of the circuit shown in Fig. 7.1. What is
the most remarkable and perhaps most important property of the solution? The answer to
this question is linked to the amount of power that can be discharged in a finite amount of
time. Let us examine Eq. (7.8c) more closely. When the load resistance, R, becomes
small, the delivered power can reach an arbitrarily high value at small positive t.
Expressed in another way, when discharged through a small resistance, the (ideal)
capacitor delivers an extremely high power pulse during a short period of time!

VII-323
Chapter 7 Transient Circuit Fundamentals

This conclusion is not affected by the specific capacitance value; the capacitance value
only affects the discharge duration. In reality, however, an infinitely small resistance
cannot be achieved. How can we use the ability of the charged capacitor to create a large
current and, consequently, supply a large power for a short period of time? There are a
number of well-known applications such as an electronic photoflash or drivers for the
light-emitting diodes (LEDs) or even electromagnetic material processing.

7.1.2 Time Constant of the RC Circuit and Its Meaning

To appreciate the value of the time constant as a fundamental property of an RC circuit, we
consider two examples with explicit component values. Our objective is to find the
dynamic voltage and current responses of the circuit as the capacitor discharges. You
should note that the time constant τ determines the duration at which the capacitor voltage
will have dropped to 1/e or 0.368 (36.8 %) of the initial voltage V0. This number arises
from the fact that, at the time instance t ¼ τ, we obtain from Eq. (7.8a)
 τ
υC ðt Þ ¼ υR ðt Þ ¼ V0 exp  ¼ V0 e1 ¼ V0 =e ¼ 0:368 V0 ð7:9aÞ
τ
It is sometimes useful to study the dynamic response at t ¼ 2τ, in which case we obtain

υC ðt Þ ¼ υR ðt Þ ¼ V0 e2 ¼ 0:135 V0 ð7:9bÞ

or 13.5 % of its original value, V0. At 3 τ, we already see the voltage drop less than 5 %.

Example 7.1: In Fig. 7.1, a 10-μF capacitor discharges into a 10-Ω load. The capacitor is
initially charged to V 0 ¼ 10 V. Plot the capacitor voltage υC, load current iR, load voltage
υR, and load power pR, over the interval from 0.2 ms to 0.5 ms.
Solution: First, we determine the time constant τ. According to Eq. (7.5),

τ ¼ RC ¼ 105 F  10 Ω ¼ 104 s ¼ 0:1 ms ð7:9cÞ

The solution then relies on Eqs. (7.8a) through (7.8c) based on V 0 ¼ 10 V. Figure 7.2
shows the behavior of voltage, load current, and load power. The vertical line is the time
constant τ. This constant determines how fast the capacitor discharges. At t ¼ τ, the voltage
is equal to 1/e or 0.368 of the initial capacitor voltage, V0. Note that a rather low
capacitance value of 10 μF is used. We can purchase a 10-μF electrolytic capacitor of
5 mm diameter and 12 mm height and rated at 25 V or 50 V. As seen in Fig. 7.2c, an
appreciable load power of 10 W (!) can be created. Unfortunately, it is created for only a
very short period of time, on the order of τ.

VII-324
Chapter 7 Section 7.1: RC Circuits

a) capacitor voltage v , V b) load current i , A

t<0 C t<0 R
10 1

8 0.8

6 0.6

4 0.4

2 0.2

0
τ 0
τ
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5

c) load voltage v , V d) load power p , W

t<0 R t<0 R
10 10

8 8

6 6

4 4

2 2

τ τ
0 0
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
time, ms time, ms

Fig. 7.2. (a) Capacitor voltage, (b) load current (c) load voltage, and (d) load power for a 10-μF
capacitor discharging into a 10-Ω load resistor.

Exercise 7.1: The capacitor in Fig. 7.1 is initially charged to V 0 ¼ 20 V. Determine the
capacitor voltage and the instantaneous power delivered to the load resistance at (i) t ¼ 50
μs and (ii) t ¼ 1 ms.
Answer: (i) 12.13 V and 14.72 W; (ii) 0.91 mV and 0.08 μW.

7.1.3 Continuity of the Capacitor Voltage

Energy Consideration
The voltage across the discharging capacitor remains a continuous function of time over
the breakpoint t ¼ 0. On the other hand, all other quantities in Fig. 7.2 such as the circuit
current, the load voltage, and the load power are subject to a sudden jump when the
switch closes. Why is that so? The electric field energy stored in the capacitor is given by
1
EC ¼ Cυ2C ðt Þ ð7:10aÞ
2

VII-325
Chapter 7 Transient Circuit Fundamentals

at any time instant. Any energy cannot be released instantaneously. For instance, a
vehicle with mass m and speed υ possesses the kinetic energy
1
ET ¼ mυ2 ð7:10bÞ
2
and cannot be stopped instantaneously. The kinetic energy must be a continuous function
of time, as does the vehicle speed. Similarly, the capacitor energy must be a continuous
function of time and so does the capacitor voltage. Such an effect might be called the
“capacitor inertia” in reference to mechanical inertia. Thus, the capacitor voltage υC(t) is
the only variable which is always a continuous function of time in an RC circuit.
Therefore, it must be used as an independent function in the ODEs for RC circuits.
Using any other function (circuit current or resistor voltage) is prohibited since we cannot
specify the initial conditions for a noncontinuous function.

Fluid Mechanics Analogy

The continuity of the capacitor voltage may be illustrated by a fluid-flow analogy of the
discharging capacitor shown in Fig. 7.3. The voltage corresponds to the fluid level in the
water-filled tank, which gradually decreases, but cannot jump instantaneously. On the
other hand, the fluid acquires a certain velocity (the equivalence to electric current)
immediately after the switch in Fig. 7.3 opens. Interestingly, the value of the load
resistance in Fig. 7.1 is the reciprocal of the cross section of the pipe in Fig. 7.3. The
smaller the cross section of the pipe (i.e., the greater the resistance), the slower the
observed fluid flow from the tank (i.e., the smaller load current). At the same time, the
leakage time (or the discharge time) increases accordingly. In Fig. 7.3, we actually need to
open the mechanical valve, whereas in Fig. 7.1 we close the electric switch. There is no
real contradiction though since both operations really enable the flow of a substance:
either the flow of electric current in Fig. 7.1 or the water flow in Fig. 7.3.

t=0

Fig. 7.3. A fluid-flow analogy for the circuit shown in Fig. 7.1.

7.1.4 Application Example: Electromagnetic Railgun

Figure 7.4 shows a generic structure of an electromagnetic accelerator, sometimes called
an electromagnetic railgun. Apart from potential high-power applications, this setup
helps us to visualize the operation of linear motors and generators. The discharging

VII-326
Chapter 7 Section 7.1: RC Circuits

capacitor is connected via a resistor to the rails, as shown on the left in Fig. 7.4. Resistor
R models ohmic losses in the metal rails and a (typically small) series resistance of the
capacitor. The capacitor current flows through two metal rails and through a sliding or
rolling metal rod to be accelerated. Also shown in Fig. 7.4 are two permanent magnets
responsible for creating a magnetic flux emanating from the north pole (N) and termi-
nating at the south pole (S). When this perpendicular magnetic flux density B, measured
in tesla (T), is applied between the rails, the Lorentz force will act on the moving object of
 
length l ¼ ~ l  and accelerate this object in the direction of the rails. This force is given by
 
F ¼ iC ~
~ l~ B ½N  ð7:11Þ

N
B

F
iC
l
iC
iC S metal rails

conductive moving metal rod

t=0
R
V0
+

Fig. 7.4. An electromagnetic accelerator based on the Lorentz force effect in the magnetic field
created by two permanent magnets and the capacitor discharge current.

The Lorentz force was named after Hendrik Antoon Lorentz (1853–1928), a Dutch
physicist and Nobel Prize laureate. Only 24 years of age, Lorentz was appointed to the
newly established chair in theoretical physics at the University of Leiden, the oldest
university in the Netherlands founded by William, Prince of Orange. Lorentz made
significant contributions to field theory ranging from hydrodynamics to general relativity.
The Lorentz force is the driving force of electric motors/generators you are using. The
cross product in Eq. (7.11) is consistent with the right-hand rule: the current direction of
the moving object represents the fingers of your right hand, and they are turned into the
magnetic flux direction so that the thumb points in the direction of the Lorentz force.
Another way is to picture a screw whose body points along the force and which is turned
in the plane spanned by ~ B such that ~
l and ~ l is rotated into ~
B. From a circuit point of view,

VII-327
Chapter 7 Transient Circuit Fundamentals

the construction in Fig. 7.4 could be replaced by an RC circuit, with the capacitor
connected to the rails through a switch. Resistance R in Fig. 7.4 is formed by the rail
resistance combined with the object’s resistance and with the contact resistance between
the object and the rails. The force is directly proportional to the discharge current iC.

Example 7.2: The capacitor in Fig. 7.4 is initially charged to a voltage V 0 ¼ 100 V and
has a capacitance of 1000 μF. The total system resistance R is 1 Ω. For the above example,
what force in N is to be expected and for how long? The field is 0.1 T; the rod is 6 cm long.
Solution: According to Eq. (7.8b), the maximum current value, which occurs when the
switch has just closed, is V 0 =R ¼ 100 A. Using Eq. (7.11), we calculate the initial force
value, which is 0.6 N.
As time progresses, the current and the force both quickly decrease. Over a time duration
of τ ¼ RC ¼ 1 ms, both of these values decrease to 36.8 % of their initial values. For
simplicity, we assume that an average force acts over the time duration τ. Its value is
estimated as approximately 60 % of the initial force value. We then obtain an average of
0.36 N over the time interval τ ¼ 1 ms, which is a rather modest result. Realistic capacitors
used for electromagnetic (EM) acceleration are the so-called pulsed capacitors. They have
a high charge voltage of V 0  10,000 V and capacitances on the order of 100 μF.
Therefore, a high-voltage power supply is needed. A number of capacitors are put in
parallel to increase the overall capacitance. Large currents, on the order of 10,000 A, into
the 1-Ω load may then produce much higher force values.

7.1.5 Application Example: Electromagnetic Material Processing

Electromagnetic Forming
The moving object in Fig. 7.4 may be implemented in various forms. For example, it
could be replaced by a liquid metal such as molten aluminum. In principle, an electro-
magnetic “die casting” machine could be constructed that creates a high-speed liquid
metal jet. The key point is the small mass of the object in order to enable a fast
acceleration. Electromagnetic forming is used to accelerate solid metal sheets at velocities
up to a few hundred meters per second, which are 100–1000 times greater than the
deformation rates of conventional forming such as sheet metal stamping. The noncontact
electromagnetic forming of metals is a process that has been applied since the 1960s but
has not seen extensive use. Its common application is to expand, or compress, axisym-
metric metal parts as shown in Fig. 7.5a. It has been commercially applied for the joining
and assembly of concentric parts and compression crimp seals. Figure 7.5b shows a more
recent experiment at the Ohio State University and made with aluminum car door panels.

VII-328
Chapter 7 Section 7.1: RC Circuits

Fig. 7.5. (a) Electromagnetic forming of metal joints. The current in the windings generates a
magnetic flux that induces eddy currents in the metal. Their product is the Lorentz force.
(b) Electromagnetically reformed door panel compared with the production geometry.

Self-Induced Lorentz Force

Electromagnetic forming processes shown in Fig. 7.5 do not use permanent magnets.
Instead, the so-called self-induced Lorentz force is employed. The idea is to generate the
magnetic flux B with the same current iC. Figure 7.5a shows the related concept used in
noncontact electromagnetic forming of metal joints. The high discharge current, iC, creates a
strong time-varying magnetic field, both inside and outside of the coils in Fig. 7.5a. In turn,
the time-varying magnetic field induces so-called eddy currents in the metal sample. The
product of these eddy currents and the magnetic field gives rise to a Lorentz force according
to Eq. (7.11). This force is strong enough to deform the joints.

7.1.6 Application Example: Digital Memory Cell

This completely different example investigates a digital circuit that stores binary infor-
mation. Figure 7.6 shows a schematic of a dynamic random-access memory (DRAM)
memory cell. The cell stores its bit of information as charge deposited on the cell
capacitor C. When the cell is storing a logic 1, the capacitor is charged to a positive
voltage V0; when a logic zero is stored, the capacitor is discharged to a zero voltage.
Because of leakage effects, there is always a nonzero resistance R to ground (not shown in
the figure). Thus, the cell circuit becomes that of Fig. 7.1. The capacitor will discharge
and must be refreshed periodically. During refresh, the capacitor voltage is restored to V0
if necessary. The refresh operation is in fact performed every 5–10 ms!

VII-329
Chapter 7 Transient Circuit Fundamentals

Word line

C Bit line

0V Cell

Fig. 7.6. Dynamic RAM memory cell. The bit line carries either logic 1 or 0 information.

7.1.7 Energy-Accumulating Capacitor Circuit

Charging is the inverse process of discharging a capacitor and it involves a power supply.
The corresponding circuit is shown in Fig. 7.7. The switch closes at t ¼ 0. The resistor
R can be either the Thévenin resistance of the practical voltage source or the series
parasitic resistance of the capacitor itself or even a combination of both.

R=10
vR - t=0
+

iR +
VS + iC vC C=10 F
- -

Fig. 7.7. Charging a capacitor with a DC voltage source as an example of another RC circuit.

To support this conclusion, we note that the positions of the switch and the resistor in
Fig. 7.7 may be interchanged without affecting the circuit analysis. Similarly, two resistors
may be placed on both sides of the switch; the circuit solution will display their series
equivalent. The solution to the circuit is once again based on KVL and KCL. By KCL,

iC ¼ iR ð7:12Þ

at any time instance t. Since both R and C in Fig. 7.7 are passive elements, we can apply
the constitutive relations between currents and voltages without changing the sign:
iC
zfflffl}|fflffl{ z}|{
iR
dυC υR V S  υC
C ¼ ¼ ð7:13Þ
dt R R
since KVL states

υR ¼ V S  υC ð7:14Þ

VII-330
Chapter 7 Section 7.1: RC Circuits

at any positive time instance t. Therefore, Eq. (7.13) yields

dυC υC V S dυC υC V S
C þ ¼ ) þ ¼ , τ ¼ RC ð7:15Þ
dt R R dt τ τ
Equation (7.15) has now an excitation term (the power supply voltage) on its right-hand
side. Consequently, the solution to this equation is called a forced response. Equation
(7.15) is known as an inhomogeneous first-order differential equation. This is in contrast to
the homogeneous differential equation (7.5). Nonetheless, Eq. (7.15) still remains a first-
order transient equation with the same time constant τ. The solution to any first-order
ordinary differential equation of that type has the generic form
 t
υC ðt Þ ¼ K 1 exp  þ K 2 ð7:16aÞ
τ
This fact can be checked by direct substitution. The two terms containing the exponential
factor will cancel out after differentiation. The remaining terms in Eq. (7.15) yield
K2 V S
¼ ) K2 ¼ V S ð7:16bÞ
τ τ
The constant parameter K1 can be determined from the initial condition, υC ðt ¼ 0Þ ¼ 0.
Since exp(0) ¼ 1 in Eq. (7.16a), we conclude

K 1 þ K 2 ¼ 0 ) K 1 ¼ V S ð7:16cÞ

Thus, the circuit voltages have the form

h  t i  t
υC ðt Þ ¼ V S 1  exp  , υR ðt Þ ¼ V S exp  ð7:16dÞ
τ τ
The resistor voltage has the same form as the resistor voltage in Eq. (7.8a) for the
discharging capacitor. However, the capacitor voltage has not. The capacitor current is

dυC ðt Þ V S  t
iC ðt Þ ¼ C ¼ exp  ð7:16eÞ
dt R τ
It is equivalent to Eq. (7.8b), the discharge current for the RC circuit. What is the most
remarkable property of the solution given? According to Eq. (7.16d), we always need a
certain amount of time to charge the capacitor. It is clear that this time will be on the order
of the time constant τ. Moreover, from a formal point of view, the capacitor voltage will
never exactly reach the source voltage (the exact equality only occurs at t ! 1), see
Fig. 7.8.

VII-331
Chapter 7 Transient Circuit Fundamentals

Example 7.3: A 10-μF capacitor in Fig. 7.7 is charged by a 10-V voltage source. The
switch closes at t ¼ 0, and the system resistance is 10 Ω. Plot capacitor voltage υC and
capacitor current iC to scale over the time interval from 0.2 ms to 0.5 ms.
Solution: First, we find the time constant τ. According to Eq. (7.15) or (7.5),
τ ¼ RC ¼ 105 F  10 Ω ¼ 0:1 ms. The solutions for this example are given by
Eqs. (7.16d) and (7.16e) with V S ¼ 10 V. Solutions for the capacitor voltage and capacitor
current are plotted in Fig. 7.8. The vertical line denotes the time constant τ so that you can
see how fast the capacitor charges. At one time constant, i.e., τ ¼ 0.1 ms, the capacitor is
charged to ð1  1=eÞV S or to 63.2 % of the source voltage VS.

a) t<0 capacitor voltage v C, V b) t<0 capacitor current i C, A

10 1

8 0.8

6 0.6

4 0.4

2 0.2

0
τ 0
τ
-0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5
time, ms time, ms

Fig. 7.8. Voltage/current plots for charging a 10-μF capacitor in series with a 10-Ω resistor. We
again observe continuity of the capacitor voltage (the “capacitor inertia”).

Exercise 7.2: The source voltage in Fig. 7.7 is 20 V. Determine the capacitor voltage and
the circuit current at (i) t ¼ 5 μs and (ii) t ¼ 1 ms.
Answer: (i) 0.98 V and 1.90 A; (ii) 19.999 V and 91 μA.

VII-332
Chapter 7 Section 7.2: RL Circuits

Section 7.2 RL Circuits

The first-order RL circuits studied in this section lay the foundation of understanding the
behavior of any transient circuit containing inductances. The transient circuit response is
a dynamic process; we are once again required to solve dynamic circuit equations that can
be formulated as ordinary differential equations. You can again rely on KVL and KCL
since they apply to any static (DC) or dynamic (transient and AC) circuit. Along with the
inductor’s voltage/current relation, they are used to derive the circuit ODE.

7.2.1 Energy-Release Inductor Circuit

The inductor stores the magnetic-field energy created by an electric current. Thus, in
order to be “charged,” the inductor must carry some current. A natural choice is therefore
a circuit with the current source shown in Fig. 7.9a. This is in contrast to the charged
capacitor, which does not need a voltage supply to stay charged. If the switch in Fig. 7.9a
is open (t < 0), the entire current IS flows through the inductor. The inductor is thus
“charged.” When the switch closes at t ¼ 0, the current source still generates the same
current IS at its terminals. However, the supply is now shorted out, i.e., no current flows
into the circuit. In other words, the current supply is effectively disconnected so that the
RL circuit becomes a stand-alone circuit in Fig. 7.9b, with the initial current IS still
flowing in the inductor. As time progresses, the inductor releases its energy to the load.

a) t<0 b) t 0
IS iR iL
t=0
+ + +
IS v R L R vR L vL
- - -

Fig. 7.9. The concept of “discharging” the previously charged inductor through a load resistor R.

The solution of the dynamic circuit in Fig. 7.9b is again based on KVL and KCL. With
the voltage polarities shown in Fig. 7.9b, the use of KVL yields

υL ¼ υR ð7:17Þ

at any time instance, t. Since both circuit elements in Fig. 7.9b are passive, we can directly
apply the constitutive relations between voltages and currents without changing the sign:
υL
zffl}|ffl{ υR
diL z}|{
L ¼ R iR ¼ RiL ð7:18Þ
dt

VII-333
Chapter 7 Transient Circuit Fundamentals

According to KCL, iR ¼ iL , at any positive time instance t, and Eq. (7.18) yields
diL diL iL
L þ RiL ¼ 0 ) þ ¼ 0, τ ¼ L=R ð7:19Þ
dt dt τ
Here again we encounter a first-order transient equation. The constant τ carries the unit of
seconds (since R is in ohms and L is in henrys) and is known as the time constant, or
relaxation constant, of the RL circuit. It is the only constant that is present in the first-
order transient differential equation. We can observe a remarkable similarity between the
transient RL circuit and the RC circuit of discharging a capacitor. The mathematics is
the same, but the capacitor voltage is replaced by the inductor current, and the value of the
time constant changes from RC to L/R. The initial condition iL ðt ¼ 0Þ ¼ I S includes the
past inductor current instead of the past capacitor voltage. The solution of Eq. (7.19) is
 t
iL ðt Þ ¼ Kexp  , t  0 ð7:20Þ
τ
The validity of Eq. (7.20) is seen by direct substitution. The constant K is determined
from the initial condition. Setting t ¼ 0 yields K ¼ I S . Both currents in Fig. 7.9b are
 t
iL ðt Þ ¼ iR ðt Þ ¼ I S exp  t0 ð7:21aÞ
τ
At t < 0, the inductor current maintains its value IS but the resistor current is zero, as
shown in Fig. 7.10c. The resistor (or load) voltage is given by
 t
υR ðt Þ ¼ R iR ðt Þ ¼ RI S exp  ð7:21bÞ
τ
At t < 0, the load voltage is zero. The instantaneous power delivered to the load is
 t
pR ðt Þ ¼ υR ðt ÞiR ðt Þ ¼ RI S2 exp 2 ð7:21cÞ
τ
Equations (7.21a)–(7.21c) provide the complete solutions for the circuit depicted in
Fig. 7.9.

Example 7.4: A 1-mH inductor in Fig. 7.9 is connected to a 1-kΩ load. The supply current
(disconnected at t ¼ 0) is 1 A. Plot inductor current iL, inductor (or load) voltage υR, load
current, iR, and load power, pR, over the interval from –2τ to 5τ.

VII-334
Chapter 7 Section 7.2: RL Circuits

Example 7.4 (cont.):

Solution: First, we find the time constant τ. According to Eq. (7.19),
τ ¼ L=R ¼ 103 H=1000 Ω ¼ 106 s ¼ 1 μs ð7:22Þ
which is a rather small value. The solution for this example is given by Eqs. (7.21a)
through (7.21c), with I S ¼ 1 A, and is shown in Fig. 7.10. The vertical line in all plots is
the time constant τ. One can see that the time constant determines how quickly the inductor
current and the load voltage decrease. At 1τ, the load voltage is equal to 1/e or 0.368
(36.8 %) of the initial voltage RIS.

a) inductor current i L , A b) load voltage v , V

t<0 t<0 R
1 0

0.8 -200

0.6 -400

0.4 -600

0.2 -800

0 -1000
-2 -1 0 1 2 3 4 5 -2 -1 0 1 2 3 4 5

c) load current iR, A d) ,W

R
t<0 t<0
0 1000

-0.2 800

-0.4 600

-0.6 400

-0.8 200

-1.0 0
-2 -1 0 1 2 3 4 5 -2 -1 0 1 3 4 5
time, s time, s

Fig. 7.10. (a) Inductor current, (b) load voltage, (c) load current, and (d) load power for a 1-mH
inductor connected to a 1000-Ω load resistor.

VII-335
Chapter 7 Transient Circuit Fundamentals

Exercise 7.3: The supply current in Fig. 7.9 is 0.5 A. Given L ¼ 1 mH and R ¼ 1 kΩ,
determine the load voltage and the instantaneous power delivered to the load resistance at
(i) t ¼ 0:2 μs and (ii) t ¼ 10 μs.
Answer: (i) –409.4 V and 167.6 W; (ii) –23 mV and 0.52 μW.

What is the most remarkable property of the solution? According to Fig. 7.10, the high
voltage spike across the inductor is created in an RL switching circuit when the load
resistance, R, is large. For example, an air gap has a very high resistance. When used as a
load, it may possess a very high voltage drop of several kVand more. This is the idea behind
any medium-to-high-power electronic ignition system, including the 12-V-powered car
ignition system, a missile ignition system, etc. Such a circuit must include at least three
basic elements: (a) a voltage or current power supply, (b) a switch, and (c) an inductor (coil).
The switch can be a transistor switch controlled by a sensor.

7.2.2 Continuity of the Inductor Current

The current through the inductor remains a continuous function of time over the
breakpoint t ¼ 0. However, all other quantities in Fig. 7.10, the load voltage, the load
current, and the load power are subject to a sudden jump when the switch closes. The
reason for such a continuity is the finite magnetic-field energy stored in the inductor:
1
EL ¼ Li2L ðt Þ; ð7:23Þ
2
at any time instant. This energy cannot be released instantaneously. Such an effect might
be called “inductor inertia” in reference to mechanical inertia of a vehicle with mass
m and speed υ and kinetic energy ET ¼ 0:5 mυ2, which cannot be stopped instantaneously.
The kinetic energy is a continuous function of time, as is the vehicle speed. Similarly, the
inductor energy is a continuous function of time, as is the inductor current. The inductor
current is the only variable which is always a continuous function. Therefore, it must be
used as an independent function in the ODEs for RL circuits. Using any other function is
prohibited since we cannot state the initial conditions for a noncontinuous function.

Fluid Mechanics Analogy

The continuity of the inductor current may be illustrated by a fluid-flow analogy of the
energy-releasing inductor circuit shown in Fig. 7.11. The electric current corresponds to
the velocity of the fluid. The inductance is a massive wheel of mass m. When subject to a
DC current (constant water flow), it acquires a certain angular velocity. This is the case of
Fig. 7.9a at t < 0. When the water pump is suddenly turned off, the wheel inertia will still
support the same water flow, at least at the initial time moment. After that, the wheel
slowly decelerates. This is the case of Fig. 7.9b.

VII-336
Chapter 7 Section 7.2: RL Circuits

a) DC circuit massive wheel b) Transient current

of mass m

DC current DC current DC current

Fig. 7.11. A fluid-flow analogy for the two circuits shown in Fig. 7.9.

7.2.3 Energy-Accumulating Inductor Circuit

The circuit behavior is now exactly the opposite of the circuit shown in Fig. 7.9. For
instance, when the switch in Fig. 7.12 is closed, the current supply is shorted out. No
current flows into the circuit. However, when the switch opens, the supply current IS starts
to flow into the circuit, and as time approaches infinity, the entire supply current IS flows
through the inductor. Thus, the inductor becomes “charged.”

iR iL
t=0
+ + +
IS v R vR L vL
- - -

Fig. 7.12. The concept of “charging” an inductor using a current source.

The solution of the dynamic circuit in Fig. 7.12 is based on KVL and KCL. With the
voltage polarities in Fig. 7.12, the use of KVL yields

υL ¼ υR ð7:24Þ

at any time instance, t. Since both circuit elements in Fig. 7.12 are passive, we can apply
the constitutive relations between voltages and currents without changing the sign:
υL
zffl}|ffl{ υR
diL z}|{
L ¼ R iR ¼ RI S  RiL ð7:25Þ
dt
because, according to KCL, iR ¼ I S  iL , at any positive time instance, t. Equation (7.25)
yields
diL iL I S
þ ¼ , τ ¼ L=R ð7:26Þ
dt τ τ
which is the inhomogeneous first-order transient equation with the forcing function
(right-hand side) equal to IS/τ. Here, τ is the generic time constant of the RL circuit.
Once again, there is a close similarity between the present RL circuit and the series RC
circuit for charging the capacitor. The mathematics is the same, but the capacitor voltage
is replaced by an inductor current and the voltage supply is replaced by the current supply.

VII-337
Chapter 7 Transient Circuit Fundamentals

The initial condition for Eq. (7.26) now includes the past inductor current of 0 A instead
of the past capacitor voltage, which is 0 V. Equation (7.26) has the solution
 t
iL ðt Þ ¼ K 1 exp  þ K 2 , K 2 ¼ I S ð7:27Þ
τ
This fact is seen by direct substitution. The constant K1 is found from the initial condition
of zero inductor current at t ¼ 0, which yields K 1 ¼ K 2 . Therefore,
h  t i  t
iL ðt Þ ¼ I S 1  exp  , iR ðt Þ ¼ I S exp  , t  0 ð7:28aÞ
τ τ
Both currents are zero at t < 0. However, the inductor current is continuous over the
breakpoint while the resistor current is not. The inductor/resistor voltages are given by

diL ðt Þ  t
υ L ðt Þ ¼ L ¼ RI S exp  , υ R ðt Þ ¼ υ L ðt Þ t0 ð7:28bÞ
dt τ
Both voltages are zero at t < 0. This completes our circuit analysis.

Example 7.5: A 1-mH inductor in Fig. 7.12 is connected to a 1-A current power supply.
The resistor value is R ¼ 1 kΩ. Plot the inductor current, iL, and the inductor voltage, υL, to
scale versus time over the interval from –2τ to 5τ.
Solution: First, we find the time constant τ. According to Eq. (7.26), we get
τ ¼ L=R ¼ 103 H=1 kΩ ¼ 106 s ¼ 1 μs. The solution to the example is given by
Eqs. (7.28a, b) with I S ¼ 1 A; see Fig. 7.13. The vertical line in both plots is the time
constant τ. One can see that this time constant determines how fast the circuit stabilizes.
At 1τ, the inductor current reaches ð1  1=eÞI S , i.e., 63.2 % of the expected DC value.
Note that Fig. 7.13 of this section and Fig. 7.8 of the previous section are identical to within
interchanging voltage and current terms!

a) inductor current i L , A b) inductor voltage v L , V

t<0 t<0
1 1000

0.8 800

0.6 600

0.4 400

0.2 200

0 0
-2 -1 0 1 2 3 4 5 -2 -1 0 1 2 3 4 5
time, s time, s

Fig. 7.13. Voltage/current plots for “charging” the 1-mH inductor in parallel with the 1-kΩ resistor.

VII-338
Chapter 7 Section 7.2: RL Circuits

Exercise 7.4: The supply current in Fig. 7.12 is 0.5 A. Given L ¼ 1 mH and R ¼ 1 kΩ,
determine the inductor voltage and the inductor current at (i) t ¼ 0:2 μs and (ii) t ¼ 10 μs.
Answer: (i) 409.4 V and 91 mA; (ii) 23 mV and 0.49999 A.

The final question to ask is what is the most remarkable property of the solutions
given by Eqs. (7.28a, b)? According to Eq. (7.28a), we always need a certain period
of time to create a given current through the inductor. The elapsed time will be on the
order of the time constant τ. Moreover, the inductor current will never exactly reach
the supply current (the exact equality only occurs as t ! 1). In practice, this effect is
masked by noise and by other factors. Interestingly, the resistor carries most of the circuit
current when the solution changes rapidly, i.e., close to the initial time t ¼ 0. At the
same time, when the circuit stabilizes, i.e., when t becomes large compared to τ, the
influence of the inductor dominates. This observation leads us to the concept of imped-
ance (the “resistance” of dynamic circuit elements) that is considered next. The imped-
ance is similar to a resistance (and has the same unit), but it depends on how fast circuit
current and voltage change. When the changes are very fast, the inductor exhibits a
much greater “resistance” than the resistor; it becomes virtually an open circuit with no
current flow.

7.2.4 Energy-Release RL Circuit with the Voltage Supply

The combination of the current supply IS and resistor R in Figs. 7.9 and 7.12 is in fact the
Norton equivalent circuit of any network of power supplies and resistors. The RL circuit
may be modeled a Thévenin equivalent too. The concept is shown in Fig. 7.14.

a) original circuit b)
R0=10 Ω t=0 at t>0

L=1 mH R=1 kΩ
+ +
+ VS L=1 mH R=1 kΩ vL iL iR vR
- - -

Fig. 7.14. “Discharging” an inductor with a voltage supply, as an example of an RL circuit.

Prior to opening the switch in Fig. 7.14a, the inductor current is found to be VS/R0.
When the switch opens, the supply is disconnected from the RL circuit; see Fig. 7.14b. As
time progresses, the inductor releases its energy into the load. The circuit in Fig. 7.14b is
identical to the circuit in Fig. 7.9b. Therefore, all prior results related to the energy-release
RL circuit will remain valid if we replace the initial inductor current IS by VS/R0.
According to Eq. (7.21b), the load voltage is given by

VII-339
Chapter 7 Transient Circuit Fundamentals

R  t
υR ðt Þ ¼ RiR ðt Þ ¼  V S exp  ð7:29Þ
R0 τ
If the ratio R/R0 is large, the initial voltage spike of the inductor is large too. The
magnitude of the initial voltage spike for the circuit in Fig. 7.14 is a hundred times the
supply voltage VS! Can we model an electronic ignition system in the laboratory? Yes,
according to Fig. 7.14 this can be accomplished relatively easily. The key, however, is the
construction of a fast switch. A proper choice may be a power transistor switch.

7.2.5 Application Example: Laboratory Ignition Circuit

A circuit rated at 6 V for safety purposes is shown in Fig. 7.15. The laboratory DC voltage
source usually delivers up to 3 A of current, if not current limited. The electric step-up
transformer with two coils is a 6-V car ignition coil. Instead of a simple coil, a transformer
is used to further boost the inductor voltage spike of the RL circuit. The small resistance R0
is the transistor/wire resistance. The very large resistance R is the resistance of the spark
plug in series with the large resistance of the spark plug cable RCABLE, which is a carbon
core wire. This carbon core wire is used to prevent higher EM radiation and its possible
influence (we would hear it as noise) on the car audio receiving equipment. Figure 7.16
shows the operating circuit. The sparking frequency ranges from 2 Hz to 100 Hz.

transistor switch spark plug cable

R0 R

spark plug

6V=VS +
- ignition coil
(transformer)

0V

Fig. 7.15. Modeling the ignition system as in a laboratory. The spark plug voltage is about 3–10 kV.

555 timer and control circuitry

Transistor switch
Spark
plug
6V DC supply
Ignition coil

Fig. 7.16. A constructed ignition system.

VII-340
Chapter 7 Section 7.2: RL Circuits

Spark Gap Radio

The spark gap in the circuit in Fig. 7.15 and in Fig. 7.16 is a powerful and broadband
source of electromagnetic radiation. One favorite story in science fiction novels is that,
after crashing his or her spaceship on an inhabited planet, the commander can quickly
construct a spark generator out of remaining parts of the ship and send an SOS signal out
to space. An AM radio can “listen” to the circuit shown in Fig. 7.16 over the entire AM
band from 540 to 1610 kHz used in the USA. Even better results are observed for the
long-wave AM band from 153 to 279 kHz used in Europe, Africa, and parts of Asia.

Exercise 7.5: The supply voltage in Fig. 7.14 is 10 V. Determine the inductor voltage and
the inductor current at (i) t ¼ 0:2 μs and (ii) t ¼ 10 μs.
Answer: (i) –818.7 V and 0.82 A; (ii) –45 mV and 45 μA.

VII-341
Chapter 7 Transient Circuit Fundamentals

Section 7.3 Switching RC Oscillator

The time constant τ of an RC circuit provides a natural time scale. It is widely employed
for timing purposes. Some of you may have already used microcontroller starter kits. The
microcontroller manual discussing the various settings will likely feature a topic entitled
“RC oscillator.” Here you will discover that the microcontroller clock frequency can be
controlled by an external resistor R and capacitor C. How is this possible? We have just
seen that the RC circuit discharges the capacitor through the resistor, but how can it be
used to create a periodic clock signal at a given frequency? The present section aims to
augment an RC circuit with an amplifier circuit and establish a clock circuit.

7.3.1 About Electronic Oscillators

An electronic oscillator is a circuit that has an output, but no input in the common sense.
It generates a certain periodic waveform at the output. The period, amplitude, and shape
of this waveform are determined by the circuit topology. The “heart” of any oscillator
circuit is an amplifier block with some sort of a positive feedback. The positive feedback
is the opposite of the negative feedback. A part of the output amplifier’s voltage is fed
back into the input with the sign plus. All oscillator circuits may be divided into linear
oscillators, which create sinusoidal waveforms, and switching oscillators, which create
square and other periodic nonharmonic waveforms. The subject of this section is a
switching oscillator circuit, which is called an astable multivibrator or a relaxation
oscillator. This circuit uses the comparator amplifier but with a positive feedback loop
and a transient RC block. It is perhaps the simplest and yet efficient oscillator circuit.

7.3.2 Bistable Amplifier Circuit with the Positive Feedback

Saturation Mechanism
Consider the circuit shown in Fig. 7.17. At first sight, it is similar to the inverting
amplifier configuration. However, the amplifier polarity is interchanged, which means
that the feedback is now positive. The circuit has no input: both potential inputs are
grounded. Since there is no current into the amplifier itself (the first summing-point
constraint still applies), two resistors of the feedback loop form a voltage divider between
the output voltage υout and ground. Therefore, the voltage at node (+) becomes
R1
υþ ¼ υout ð7:30Þ
R1 þ R2

VII-342
Chapter 7 Section 7.3: Switching RC Oscillator

R1 R2
a + + +
- vout
0V
v+=
R1 v -
R1+R2 out
b 0V

Fig. 7.17. A bistable amplifier circuit.

To analyze the circuit in Fig. 7.17, we again consider the feedback as a fast dynamic
process with a very short delay in the feedback loop. We assume that R1 ¼ R2 for
simplicity. It means that 50 % of υout is returned back to the non-inverting input in, say,
1 μs. The open-loop amplifier’s gain will be A ¼ 106 ; the amplifier hits the power rails
υout ¼ V CC in saturation. The initial value of υþ will be 0 V, and the initial value of υout
will be 1 μV (at the noise level). Table 7.1 shows the dynamics of the feedback process
where the amplifier operates as υout ¼ Aυþ , but it takes 1 μs to return 50 % of υout. It
follows from Table 7.1 that the amplifier will be very quickly saturated; its output will be
the positive rail voltage υout ¼ þV CC . All other positive initial values of υout will lead to
the same result. Simultaneously, all negative initial values of υout will lead to the
saturation at the negative rail υout ¼ V CC .

Table 7.1. Dynamics of the output voltage for the bistable amplifier
circuit.
Time, μs υþ υout
0 0V 106 V
1 0:5  106 V 0.5 V
2 0.25 V þV CC

Two Stable States

The key point is that once the saturation state
R1
υout ¼ þV CC , υþ ¼ þ V CC ð7:31Þ
R1 þ R2
has been reached, the amplifier circuit will exist in this state indefinitely, despite all the
subsequent electric noise. To prove this fact, we may introduce a small perturbation in υout
and/or in υþ ; the circuit will quickly return to the solution given by Eq. (7.31). Quite
similarly, once the opposite saturation state
R1
υout ¼ V CC υþ ¼  V CC ð7:32Þ
R1 þ R2

VII-343
Chapter 7 Transient Circuit Fundamentals

has been reached, the amplifier circuit will exist in this state indefinitely. Thus, the
positive feedback always forces the comparator to operate in saturation, i.e., in either of
the two stable states, υout ¼ V CC , where V CC is the supply voltage of the amplifier.
This result is valid for any pair of the resistances R1, R2. The resistance pair specifies
values of υþ in Eqs. (7.31) and (7.32), respectively, known as threshold voltages.
Figure 7.18 shows the corresponding mechanical analogy of the bistable amplifier
circuit. Note that a grounded comparator amplifier without the positive feedback loop
would also be always saturated due to inherent electric noise. However, there are no
stable states whatsoever; the switching between the rails is random; it is controlled by
random noise.

vout=-VCC vout= 0 V vout=VCC

Fig. 7.18. Mechanical analogy of two stable states for the bistable amplifier circuit.

7.3.3 Triggering
The amplifier circuit in Fig. 7.17 can be in either of two stable states defined by the initial
conditions. As such, it is useless as long as we do not have a mean to change the state.
A trigger signal (an input voltage signal) may be applied to switch between the states.
After introducing an external trigger signal in the form of short pulses, the bistable
amplifier circuit becomes a basic digital memory element capable of saving and retrieving
one bit of data. When the input voltage signal is applied to node (a) in Fig. 7.17 instead of
grounding it, the corresponding circuit becomes the non-inverting Schmitt trigger. When
the input voltage is applied to node (b) in Fig. 7.17 instead of grounding it, the
corresponding circuit becomes the inverting Schmitt trigger. The Schmitt triggers are
used as zero-level detectors in analog electronics and for many other purposes. When
triggered, the bistable amplifier circuit operates as a comparator.

Exercise 7.6: The bistable amplifier circuit with R1 ¼ R2 in Fig. 7.17 exists in the positive
stable state with υout ¼ þV CC . A trigger signal is applied to node (b) in Fig. 7.17.
Determine output voltage when the applied trigger signal is (i) V CC , (ii) þ0:4V CC , and
(iii) þ0:6V CC , where V CC is the supply voltage of the amplifier.
Answer: (i) þV CC ; (ii) þV CC ; (iii) V CC .

VII-344
Chapter 7 Section 7.3: Switching RC Oscillator

7.3.4 Switching RC Oscillator

The idea of the switching RC oscillator (or the relaxation oscillator) is not to use an
external trigger, but rather to derive the trigger signal from an RC circuit connected to the
output of the amplifier itself. The circuit diagram of the relaxation oscillator is shown in
Fig. 7.19. The RC circuit with the source voltage υout forms a negative feedback loop. The
capacitor may either charge or discharge depending on the source voltage.

R1 v
v+=
R1+R2 out
R1 R2
+
+ +
- vout
0V R
-
+
vC C
- trigger
0V signal

Fig. 7.19. Bistable amplifier circuit with an RC circuit in the negative feedback loop.

To analyze the amplifier circuit in Fig 7.19, we assume an infinitely high open-loop DC
gain A and use basic amplifier equations (5.4) with υ ¼ υC . This yields
R1
υout ¼ þV CC if υout > υC ð7:33aÞ
R1 þ R2
R1
υout ¼ V CC if υout < υC ð7:33bÞ
R1 þ R2
Thus, the circuit in Fig. 7.19 becomes equivalent to a simple RC circuit given in Fig. 7.20
where the dependent (or rather switching) voltage source is defined by Eqs. (7.33a, b).
The corresponding transient analysis is performed starting with some initial conditions.

R
+
vout + C vC
- -

Fig. 7.20. Equivalent circuit for finding oscillation behavior. The dependent voltage source is
controlled by the capacitor voltage.

VII-345
Chapter 7 Transient Circuit Fundamentals

Example 7.6: Given initial conditions υout ¼ 10 V and υC ¼ 0 V at t ¼ 0, solve the

circuit in Fig. 7.19 and Fig. 7.20. Assume that V CC ¼ 10 V and R1 ¼ R2 .
Solution: Step 1. The source voltage in Fig. 7.20 is υout ¼ 10 V and the capacitor voltage
is zero, i.e., υC ¼ 0 V at the initial time moment t ¼ 0.
Step 2. The capacitor starts to charge. When its voltage reaches υC ¼ 5 V, the differential
voltage at the amplifier’s input becomes negative so that Eq. (7.33a) is no longer valid. The
source voltage in Fig. 7.20 switches to
υout ¼ 10 V according to Eq. (7.33b).
Step 3. The capacitor starts to discharge. When it reaches υC ¼ 5 V, the differential
voltage at the amplifier’s input becomes positive, and the output voltage therefore switches
back to υout ¼ 10 V according to Eq.(7.33a).
After Step 3, the circuit returns to Step 1, and the process continues periodically. This
results in the output voltage shown in Fig. 7.21.

Voltage, V
10
vout
5
discharge charge discharge charge
0
vC
-5

-10

0 0.1 0.2 0.3 0.4 0.5

time, s

Fig. 7.21. Relaxation oscillator operation for the idealized amplifier model (amplifier hits the rails
in saturation); the amplifier supply voltage is 10 V.

Exercise 7.7: The relaxation oscillator circuit in Fig. 7.19 uses R1 ¼ 1 kΩ and
R2 ¼ 4 kΩ. The supply voltage of the amplifier is 15 V. Determine the amplitude
(peak value) of the oscillating capacitor voltage and the oscillating output voltage.
Answer: 3 V and 15 V, respectively.

7.3.5 Oscillation Frequency

Consider the positive half cycle in Fig. 7.21. The solution for the RC circuit in Fig. 7.20 is
given by Eqs. (7.16a, b) with V S ¼ V CC . The constant K1 is found from the initial
condition of υC ðt ¼ 0Þ ¼ β V CC where β ¼ R1 =ðR1 þ R2 Þ is the amount of the positive

VII-346
Chapter 7 Section 7.3: Switching RC Oscillator

feedback. We assume that the initial time instance has been switched to the start of the
half cycle. Therefore, during the entire positive half cycle
 t
υC ðt Þ ¼ V CC  ð1 þ βÞV CC exp  , τ ¼ RC ð7:34Þ
τ
At the end of the positive half cycle, the capacitor voltage becomes β VCC. This allows us
to find the half cycle duration T/2 and the oscillation period T. Solving Eq. (7.34) with
υC ¼ β V CC and t ¼ T =2, one has ( f is the oscillation frequency in hertz)
 
1þβ 1 1 1 þ β 1
T ¼ 2τln , f ¼ ¼ ln ð7:35Þ
1β T 2τ 1β

Exercise 7.8: For the relaxation oscillator with R1 ¼ R2 , express the oscillation frequency
in terms of its time constant τ.
Answer: f ¼ 0:455
τ ½Hz.

7.3.6 Circuit Implementation: 555 Timer

Figure 7.22 shows the relaxation oscillator circuit implemented in a laboratory and its
output voltages when the supply voltages are 5 V. Since the realistic amplifier might not
reach the supply rails, the peak-to-peak (Pk-Pk) value of the output voltage might be less
than 10 V. The output current limitations of the amplifier IC may severely affect circuit
performance. Also, the oscillation frequency only approximately follows Eq. (7.35).

a) b)
Lm348 quad op-amp LED C: Two 10 F
capacitors
in parallel

R: potentiomer in series
with a fixed resistor

Fig. 7.22. (a) Timer circuit operation in laboratory. The square waveform is the output voltage
υout(t). The curved waveform is the capacitor voltage υC(t). A variable resistance R makes it
possible to visually control the oscillation frequency using an LED connected to the output
through a buffer amplifier . The oscillation frequency changes from 0.5 to about 100 Hz.

VII-347
Chapter 7 Transient Circuit Fundamentals

555 Timer Integrated Circuit

Rectangular pulse forms at lower frequencies are routinely created by the well-known
555 timer IC (integrated circuit). The 555 timer operates conceptually similarly to the
relaxation oscillator circuit described above; it is more versatile though. The 555 timer
creates a waveform of relatively sharp and clean rectangular voltage pulses whose
frequency is controlled by an external capacitor and resistor. The duty cycle (ratio of
the positive phase duration to the signal period) can also be controlled. The 555 timer is
perhaps one of the most popular integrated circuits ever built.

VII-348
Chapter 7 Section 7.4: Single-Time-Constant (STC) Transient Circuits

Section 7.4 Single-Time-Constant (STC) Transient Circuits

7.4.1 Circuits with Resistances and Capacitances
Consider a transient circuit with an arbitrary number of capacitances and resistances. The
circuit has an independent voltage source (or sources) and a switch. Instead of the voltage
source, some capacitors may be charged prior to closing the switch. A single-time-
constant transient circuit (STC circuit) is that which solution has the form
 t
υðt Þ ¼ K 1 exp  þ K 2 ð7:36Þ
τ
for any branch voltage in the circuit. In other words, only one exponential function is
involved, similar to the basic RC circuits studied previously. Here, τ is the only time constant
of the circuit. The STC transient circuits are frequently encountered in practice, in particular,
in the study of transistor amplifiers. The STC transient circuits include:
1. Transient circuits with only one capacitance C. According to Thévenin’s theorem,
the network of resistances and source(s) seen by the capacitor is reduced to its
Thévenin equivalent. As a result, we obtain the circuit shown in Fig. 7.7. Its time
constant is given by

τ ¼ RT C ð7:37Þ

where RT is Thévenin resistance – the equivalent resistance of the network with the
independent voltage source(s) shorted out.
2. Transient circuits with only one resistance R. Thévenin’s theorem may be applied
again, this time to the network of capacitances and source(s) seen by the resistance.
As a result, we again obtain the circuit from Fig. 7.7. Its time constant is given by

τ ¼ RC T ð7:38Þ

where CT is the equivalent capacitance of the network with the independent voltage
source(s) shorted out.
3. Transient circuits with an arbitrary number of capacitances and resistances given
that the solutions for different capacitor voltages obtained by the simultaneous use
of KVL and KCL are all linear functions of each other. Consider a circuit with
multiple capacitances. Assume that N is the final number of capacitances after all
possible series/parallel combinations. For the STC condition to hold, there should
be N  1 independent closed loops that include only capacitances (and possibly
independent voltage source(s)) but do not include resistances. This useful result has
been confirmed by the authors based on an extensive circuit analysis.

VII-349
Chapter 7 Transient Circuit Fundamentals

As an example, we consider here simple transient circuits shown in Fig. 7.23.

a) t=0 R

VS + C
-

b) t=0 R1

VS + R2 C
-

c) t=0 C1

VS + R C2
-

t=0 R1 C1
d)

VS + R2 C2
-

Fig. 7.23. Transient circuits with multiple resistances and capacitances.

Exercise 7.9: For the circuits in Fig. 7.23, establish the number of independent closed
loops that include only capacitance(s) and independent voltage source but do not include
resistances.
Answer: (a) Zero; (b) zero; (c) one; (d) zero.

The first case in Fig. 7.23 is the transient circuit of Fig. 7.7. The positions of the switch
and the resistor may be interchanged without affecting the circuit solution, which has the
form of Eq. (7.36)—see also Eqs. (7.16). The second case is again the STC circuit; the
solution for the capacitor voltage (and any other voltage in the circuit) is given by
 
 
Eq. (7.36) with τ ¼ R1 R2 C. The constants K1 and K2 in this equation will be different
for voltages across different circuit elements. The third case is also the STC circuit; the
solution for the capacitor voltage (and any other voltage in the circuit) is still given by
Eq. (7.36) with τ ¼ RðC 1 þ C 2 Þ. This case requires extra care since two capacitances and
the source form a closed loop. Therefore, according to KVL, capacitor voltages cannot be
both equal to zero at the initial time moment (and at any other time moment). Finally,
consider the last case in Fig. 7.23. For this circuit with two capacitances and two
resistances, there is no closed loop that includes only the capacitances but does not

VII-350
Chapter 7 Section 7.4: Single-Time-Constant (STC) Transient Circuits

include resistances. Therefore, this circuit is not a STC circuit. This result will be
confirmed shortly.

Exercise 7.10: Given initially uncharged capacitor(s), write solutions for the capacitor
voltage for the circuits shown in Fig. 7.23a–c. For the circuit in Fig. 7.23c, assume that that
C1 was initially uncharged but C2 was initially charged to VS and connected to the rest of
the circuit at t ¼ 0 via a second switch.
Answer:
h  t i
υC ðt Þ ¼ V S 1  exp  , τ ¼ RC in Fig: 7:23a ð7:39Þ
τ
R2 h  t i  
υC ðt Þ ¼ V S 1  exp  , τ ¼ R1 R2 C in Fig: 7:23b ð7:40Þ
R1 þ R2 τ
h  t i  t
υC1 ðtÞ ¼ V S 1  exp  , υC2 ðt Þ ¼ V S exp  ,
τ τ ð7:41Þ
τ ¼ RðC 1 þ C 2 Þ in Fig: 7:23c

7.4.2 Circuits with Resistances and Inductances

The forthcoming analysis is quite similar to the analysis performed previously for the
capacitances and resistances. The single-time-constant transient circuit (STC circuit) with
resistances and inductances is that which solution has the form
 t
iðt Þ ¼ K 1 exp  þ K 2 ð7:42Þ
τ
for any branch current in the circuit. The list of the corresponding STC circuits also
becomes identical to the previous case with a few modifications:
1. Capacitances are replaced by inductances.
2. Independent current source(s) will be used. They may be converted to voltage
sources according to the source transformation theorem.
3. In condition #3, instead of loops, we use circuit nodes. This condition is now
formulated as follows. Consider a circuit with multiple inductances. Assume that
N is the final number of inductances after all possible series/parallel combinations.
For the STC condition to hold, there should be N  1 independent (single) nodes,
every branch of which is either an inductance or an independent current source.

VII-351
Chapter 7 Transient Circuit Fundamentals

As an example, we consider here simple transient circuits shown in Fig. 7.24.

a)
t=0
IS R L

R2
b)
t=0
IS R1
L

R
c)
t=0
IS L1
L2

R2
d)
t=0
IS R1 L1
L2

Fig. 7.24. Transient circuits with multiple resistances and inductances.

Exercise 7.11: For the circuits in Fig. 7.24, establish the number of independent nodes,
every branch of which is either an inductance or an independent current source.
Answer: (a) Zero; (b) zero; (c) one; (d) zero.

The first case in Fig. 7.24 is the transient circuit of Fig. 7.12. The corresponding solution
for the inductor current is given by Eq. (7.42); see also Eqs. (7.28). The second case is
again the STC circuit; the solution for the inductor current (and any other current in the
circuit) is given by Eq. (7.42) with τ ¼ L=ðR1 þ R2 Þ. The constants K 1 and K 2 in this
equation will be different for voltages across different circuit elements. The third case is
also the STC circuit; the solution for the inductor current (and any other current in the
circuit) is still given by Eq. (7.42) with τ ¼ ðL1 þ L2 Þ=R. This case requires extra care
since two inductances and the current source are three branches of the same node.
Therefore, according to KCL, inductance currents cannot be both equal to zero at the
initial time moment (and at any other time moment). We may assume, for example, that
inductance L2 carried initial current I S and employs a second switch. Finally, consider
the last case in Fig. 7.24. For this circuit with two inductances and two resistances, there is
no nontrivial single node that includes only the inductances but does not include resis-
tances. Therefore, this circuit is not the STC circuit.

VII-352
Chapter 7 Section 7.4: Single-Time-Constant (STC) Transient Circuits

Exercise 7.12: Given zero inductor current at the initial time moment, write solutions for
the inductor current for the circuits shown in Fig. 7.24a–c. For the circuit in Fig. 7.24c,
additionally assume that that inductor L1 was carrying zero initial current but inductor L2
was initially carrying current IS and connected to the rest of the circuit at t ¼ 0 via a second
switch.
Answer:
h  t i
iL ðtÞ ¼ I S 1  exp  , τ ¼ L=R in Fig: 7:24a ð7:43Þ
τ
R1 h  t i
iL ðtÞ ¼ I S 1  exp  , τ ¼ L=ðR1 þ R2 Þ in Fig: 7:24b ð7:44Þ
R1 þ R2 τ
h  t i  t
iL1 ðt Þ ¼ I S 1  exp  , iL2 ðtÞ ¼ I S exp  ,
τ τ ð7:45Þ
τ ¼ ðL1 þ L2 Þ=R in Fig: 7:24c

7.4.3 Example of a Non-STC Transient Circuit

Figure 7.25a shows the last circuit from Fig. 7.23 to be analyzed here in detail. With
reference to Fig. 7.25a, KCL and KVL give
dυ1 dυ2 υ2
i ¼ i1 þ i2 ) C 1 ¼ C2 þ ;
dt dt R2 ð7:46Þ
dυ1
 V S þ υ1 þ υ2 ¼ 0 ) R1 C 1 ¼ V S  υ1  υ2
dt
Expressing either υ1 in terms of υ2 or vice versa, we obtain from Eq. (7.46) a second-
order ODE for either of the capacitor voltages. For simplicity, we will assume that
C 1 ¼ C 2 ¼ C, R1 ¼ R2 ¼ R, and τ0 ¼ RC. Then, the corresponding ODE for υ1 has
the form

d 2 υ1 3 dυ1 υ1 V S
þ þ 2¼ 2 ð7:47Þ
dt 2 τ0 dt τ0 τ0

The analysis of the second-order ODEs like Eq. (7.47) is thoroughly explained in the
last section of this chapter. Here, we present its succinct version suitable for our
immediate purposes. The solution of the homogeneous second-order ODE is sought in
the form expðαt=τ0 Þ with α being a dimensionless constant. Substitution of
this expression into the homogeneous ODE gives a quadratic equation for α,
α2 þ 3α þ 1 ¼ 0, with the two positive roots α1 ¼ 2:62 and α2 ¼ 0:38. Therefore, the
solution for Eq. (7.47) should have the form

VII-353
Chapter 7 Transient Circuit Fundamentals

a) t=0 R1 C1

i1 i2
i + v1 - +
VS + R2 C 2 v2
- -

b) capacitor voltages v1(t) and v2(t) c) circuit current i(t) and current I0
1.0VS

v1
2I0

0.0VS
v2

I0

-1.0VS

-1.5VS 0
0 5 0 10 0 0 0 5 0 10 0
time in terms of 0 time in terms of 0

Fig. 7.25. A non-STC circuit with two resistances and two capacitances and its solution behavior.

   
α1 t α2 t
υ1 ðt Þ ¼ V S þ K 1 exp  þ K 2 exp  ð7:48Þ
τ0 τ0

where K1, K2 are two constants determined by the initial conditions The solution for
voltage υ2(t) is found using the second of equations (7.46):
   
α1 t α2 t
υ2 ðt Þ ¼ ð1 þ α1 ÞK 1 exp  þ ð1 þ α2 ÞK 2 exp  ð7:49Þ
τ0 τ0

The initial conditions imply that both capacitors are uncharged prior to closing the switch.
This gives K 1 ¼ ð1 þ α2 Þ=ðα1  α2 ÞV S and K 2 ¼ ð1 þ α1 Þ=ðα2  α1 ÞV S . With this in
mind, the solution is complete. Figure 7.25b, c shows the behavior of the two capacitor
voltages. A truly remarkable point is that instantaneous voltage across the second capacitor
in Fig. 7.25b exceeds the (absolute) source voltage. Moreover, the instantaneous circuit
current in Fig. 7.25c exceeds the initial circuit current I 0 ¼ V S =R1 by 2.5 times. Those
distinct features are observed for other second-order transient circuits studied further.

7.4.4 Example of an STC Transient Circuit

In Fig. 7.26, we present an example of a rather complicated circuit, which still follows the
STC circuit model. The proof is based on the observation that node (*) in Fig. 7.26
connects three branches: two of which are exactly the inductances and the remaining one
is the current source.

VII-354
Chapter 7 Section 7.4: Single-Time-Constant (STC) Transient Circuits

t=0
L1 L2
IS

R1 R2

Fig. 7.26. An STC circuit with two inductances and two resistances.

7.4.5 Method of Thévenin Equivalent and Application Example:

Circuit with a Bypass Capacitor
Let us consider a transient circuit with a bypass capacitor shown in Fig. 7.27a. It includes
a voltage source represented by its Thévenin equivalent υS(t), RS and a load represented
by its equivalent resistance RL. The source is turned on at t ¼ 0 and it generates a voltage
in the form of a (large) DC component VS and a superimposed (small) AC signal:

υS ðt Þ ¼ V S þ V m cos ωt, ω ¼ 2π f ð7:50Þ

where f is frequency in Hz. Generally, V S >> V m .

a) source load b)
t=0 RT t=0

RS +
+
vS(t) + C vC RL vT(t) + C vC
- - -
-

Fig. 7.27. Model of a voltage source connected to a load with a bypass capacitor.

We first interchange positions of capacitance C and resistance RL in Fig. 7.27. With the
switch closed, the voltage source υS(t) with two resistances RS, RL is then converted to its
Thévenin equivalent circuit with
RL 
υT ðt Þ ¼ υS ðt Þ, RT ¼ RS RL ð7:51Þ
RS þ RL
The resulting circuit is the simple RC circuit shown in Fig. 7.27b. The method of a
Thévenin (or Norton) equivalent circuit is common for all transient circuits with one
capacitor or inductor as explained in the previous section. According to this method, a
circuit with one capacitance or one inductance is always converted to the basic RC or RL
transient circuit. Therefore, it is always a first-order transient circuit which is described
by a first-order ODE. Once the solution for the capacitor voltage υC(t) in Fig. 7.27a is
found, the voltage across the load resistor is then simply expressed as υL ðt Þ ¼ υC ðt Þ.

VII-355
Chapter 7 Transient Circuit Fundamentals

Forced Response and Natural Response

Using KVL, KCL, and the capacitor equation, we obtain the first-order ODE for the
circuit in Fig. 7.27b in the form
dυC ðt Þ υC ðt Þ 1 RL
þ ¼ f ðt Þ, f ðt Þ ¼ ðV S þ V m cos ωt Þ, τ ¼ RT C ð7:52Þ
dt τ τ RS þ RL
at any positive time instance, t > 0. The initial condition is υC ðt ¼ 0Þ ¼ 0. The time-
dependent source term f(t) is called a forcing function, whether or not the switch is
present. The action of the switch may always be included into f(t) by setting f ðt Þ ¼ 0 at
t < 0. A general solution of the homogeneous Eq. (7.52), let us call it xc(t), is known
as the complementary solution (or natural response). The natural response is given by
xc ðt Þ ¼ Kexpðt=τÞ where K is an arbitrary constant. The natural response carries
information about the circuit. A particular solution of the inhomogeneous Eq. (7.52),
let us call it xp(t), is known as the forced response. The forced response carries informa-
tion about the excitation f(t). It does not need to contain an arbitrary integration constant.
By linearity, the complete solution for Eq. (7.53) is the sum of both responses, i.e.,
υC ðt Þ ¼ xp ðt Þ þ xc ðt Þ ð7:53Þ

We already know the natural response. The forced response is sought in the form
xp ðt Þ ¼ a cos ωt þ b sin ωt þ c ð7:54Þ

where constants a, b, c are to be uniquely determined. If the function f(t) is sinusoidal, a

combination of sine and cosine would suffice under certain conditions for any linear
differential equation, not necessarily of the first order. The particular solution in the form
of Eq. (7.54) is even useable for second-order circuits considered next.

Example 7.7: Find the forced response for Eq. (7.52).

Solution: Equation (7.54) is substituted into Eq. (7.52); then all terms are pulled to the left-
hand side, and all terms with cos ωt and cos ωt are combined. We obtain:




a V m RL b c V S RL
cos ω t þ bω  þ sin ω t  aω þ  ¼0 ð7:55Þ
τ τ RS þ RL τ τ τ RS þ RL
In order to satisfy Eq. (7.55), we require all three expressions in the square brackets to be
zero. This operation yields a system of two equations for a and b, while the constant c is
found directly. Working out the algebra produces the results
RL V m 1 RL V m ωτ RL V S
a¼ , b¼ , c¼ ð7:56Þ
RS þ RL 1 þ ðωτÞ2 RS þ RL 1 þ ðωτÞ2 RS þ RL

VII-356
Chapter 7 Section 7.4: Single-Time-Constant (STC) Transient Circuits

Example 7.8: Obtain the complete solution for Eq. (7.52) and plot it to scale.
Solution: The final solution is based on Eqs. (7.54)–(7.56). The initial condition υC ðt ¼ 0Þ
¼ 0 is satisfied if K ¼ a  c. The required load voltage has the form

RL V S
υL ðt Þ ¼ υC ðt Þ ¼ ð1  expðt=τÞÞ
RS þ RL
ð7:57Þ
RL V m 1
þ ½ cos ωt þ ωτ sin ωt  expðt=τÞ
RS þ RL 1 þ ðωτÞ2

This solution is compared with the load voltage without bypass capacitor:

RL V S RL V m
υL ðt Þ ¼ þ cos ωt ð7:58Þ
RS þ RL RS þ RL
Given that ωτ and t/τ are both large, the dominant AC term in Eq. (7.57) is the sine
function. Comparing Eqs. (7.57) and (7.58) with each other, we can therefore state that
the bypass capacitor reduces the amplitude of the unwanted AC component at the load by a
factor 1/(ωτ) while keeping the DC component unchanged! Figure 7.28 plots the load
voltage with and without the bypass capacitor. The circuit parameters are RS ¼ 5 Ω,
RL ¼ 1 kΩ, C ¼ 1000 μF. The source parameters are V S ¼ 10 V, V m ¼ 1 V,
f ¼ 500 Hz. One hidden yet critical solution parameter is the source resistance RS.
When this parameter is very small, the bypass capacitor has little if any effect on the
solution.

Load voltage, V
12

10

load voltage without bypass capacitor

6

load voltage with bypass capacitor

4

0
0 5 10 15 20 25 30 35 40 45 t, ms

Fig. 7.28. Load voltage with and without the bypass capacitor predicted by Eqs. (7.57) and (7.58).

VII-357
Chapter 7 Transient Circuit Fundamentals

Section 7.5 Description of the Second-Order

Transient Circuits
7.5.1 Types of Second-Order Transient Circuits
A first-order transient circuit is described by a first-order differential equation. All
single-time-constant (STC) circuits considered thus far are the first-order transient cir-
cuits. A second-order transient circuit is described by a second-order differential equa-
tion. Figure 7.29 outlines two major types of the second-order order transient circuits.

a) Circuits with a capacitance and an inductance in series or in parallel and

an arbitrary number of resistances- second-order transient circuits
t=0 R

t=0

VS + L IS R L C
-
C

b) Non-STC circuits with two capacitances or two inductances and

at least two resistances - second-order transient circuits
t=0 R1 C1 R2

t=0

VS + R2 C2 IS R1 L1
-
L2

Fig. 7.29. Classification of second-order transient circuits.

The first type involves two nonidentical dynamic elements (capacitance and induc-
tance) assembled as a series or parallel two-terminal LC network—see Fig. 7.29a. This
figure shows two such connections. The number of resistances may be arbitrary. As long
as the LC network sees only the combination of source(s) and resistances, the equi-
valent circuit (obtained with the help of Thévenin or Norton equivalents) will have the
form shown in Fig. 7.29a. The second type includes non-STC circuits with two capac-
itances or two inductances—see Fig. 7.29b. In this section, we will study the second-
order circuits on the base of the LC networks.

7.5.2 Series-Connected Second-Order RLC Circuit

Generic Representation of a Series RLC Circuit and Qualitative Operation
Consider a series LC network shown in the shaded box of Fig. 7.30. Using Thévenin’s
theorem, any network of resistances/independent DC sources connected to this block may
be represented as the series combination of the voltage source and the resistance R. This is

VII-358
Chapter 7 Section 7.5: Description of the Second-Order Transient Circuits

why we arrive at the very generic series RLC circuit shown in Fig. 7.30. The switch
implies that the ideal voltage source is to be connected to the circuit at time t ¼ 0. Prior to
t ¼ 0, this source has no effect, it is disconnected. Qualitatively, just after closing the
switch the entire voltage drop will be acquired by the inductance. Then, the capacitor
voltage will increase. Finally, the capacitor voltage will assume the supply voltage VS,
and no current will flow in the circuit under DC steady-state condition.

t=0 R
i(t)
+
VS + L vL
- -
i(t) C

- vC + LC network

Fig. 7.30. Series RLC second-order transient circuit; the switch closes at t ¼ 0.

Mechanical Analogy
Figure 7.31 shows an intuitive analogy between a mechanical mass-spring-damping
system and an electric (or electronic) RLC transient circuit depicted in Fig. 7.30.
a) b)

k viscous
damping viscous
damping
m

x(t=0)=0
m t
x(t) equilibrium

x transient behavior

Fig. 7.31. Mechanical counterpart of an RLC circuit; at t ¼ 0 the gravity force is applied.

The inductance L corresponds to the mechanical mass m. The capacitance C is the

inverse stiffness, 1/k, of the spring. The resistance R corresponds to viscous damping. At
t ¼ 0, the gravity force is applied to mass m in Fig. 7.31 initially located at x ¼ 0. As a
result, the mass reaches a new equilibrium position xðt ! 1Þ with or without interme-
diate oscillations depending on damping. The following correspondences may be
established between mechanical and electrical quantities (q(t) is the capacitor charge):

i ðt Þ ! dxðt Þ=dt, qðt Þ ! xðt Þ, mg ! VS ð7:59Þ

VII-359
Chapter 7 Transient Circuit Fundamentals

Solution in Terms of Circuit Current

A solution to the dynamic circuit in Fig. 7.30 is based on KVL and KCL. KCL prescribes
the same current i(t) through every circuit element at any time instant t. KVL states

υL þ υR þ υC ¼ V S ð7:60Þ

at any positive time t > 0. Indeed, υL, υR, υC are all functions of time. Using constitutive
relations for the inductance, resistance, and capacitance (integral form), we obtain
ðt
diðt Þ 1
L þ Riðt Þ þ iðt 0 Þdt 0 ¼ V S ð7:61Þ
dt C
0

where we assumed that the initial capacitor voltage is zero. Differentiation over time
yields the expected homogeneous second-order ODE:

d 2 i ðt Þ diðt Þ 1
L 2
þR þ i ðt Þ ¼ 0 ð7:62Þ
dt dt C
which, after division by L, can be cast in the standard form:

d 2 i ðt Þ diðt Þ
2
þ 2α þ ω20 iðt Þ ¼ 0 ð7:63Þ
dt dt
The two constants present in this equation are given by
R 1
α¼ , ω0 ¼ pffiffiffiffiffiffiffi ð7:64Þ
2L LC
Both constants in Eq. (7.64) have a general mathematical meaning that should be
remembered. The first constant α with the units of neper/sec is the damping coefficient.
It generally characterizes how fast oscillations in Fig. 7.31 decay and reach a steady state.
The neper (Np) is a dimensionless unit named after John Napier (1550–1617), a Scottish
mathematician. The constant α is similar to the inverse time constant 1/τ for the first-order
transient circuits. The second constant ω0 with the units of rad/s is the undamped resonant
frequency of the RLC circuit. This constant characterizes frequency of oscillations—see
again the mechanical analogy in Fig. 7.31. The meaning of the undamped resonant
frequency remains the same for any LC circuit block, either in transient analysis or in
the AC circuit analysis, either for series or parallel configurations.

Exercise 7.13: When is iðtÞ ¼ K cos ω0 t a solution to Eq. (7.63)?

Answer: At R ¼ 0.

VII-360
Chapter 7 Section 7.5: Description of the Second-Order Transient Circuits

7.5.3 Initial Conditions in Terms of Circuit Current and Capacitor Voltage

In contrast to the first-order transient circuits, there are two equally possible choices of the
independent function for the RLC circuit in Fig. 7.30:
- Circuit (or inductor) current i(t)
- Capacitor voltage, υC(t)
The circuit current remains continuous over time (inductor “inertia”) and so does
the capacitor voltage (capacitor “inertia”). We have chosen the circuit current and
obtained the second-order ODE Eq. (7.63). You may wonder if this is really the best
choice. The answer is nontrivial and is hidden in the initial conditions. Any first-order
ODE needs one initial condition. Any second-order ODE needs two initial conditions.
Let us establish these initial conditions for the circuit current first. Following the current
continuity through the inductor, the circuit current must be zero at t ¼ 0, that is,

iðt ¼ 0Þ ¼ 0 ð7:65aÞ

Hence, the first initial condition is established. The second one is that of the initial
capacitor voltage equal to zero. According to Eq. (7.60) and Eq. (7.65a),
di di VS
υL ðt ¼ 0Þ þ υR ðt ¼ 0Þ þ υC ðt ¼ 0Þ ¼ V S ) L ðt ¼ 0Þ ¼ V S ) ðt ¼ 0Þ ¼
|fflfflfflfflfflffl{zfflfflfflfflfflffl} |fflfflfflfflfflffl{zfflfflfflfflfflffl} dt dt L
Riðt¼0Þ¼0 0

ð7:65bÞ

Thus, the initial circuit current is zero, whereas its first derivative is not. This is a
drawback of the electric current formulation given by Eq. (7.63) for the series RLC
circuit. On the other hand, the capacitor voltage and its derivative (which is proportional
to the capacitor/inductor/circuit current) are both zero at t ¼ 0, which leads to a simpler
“universal” homogeneous formulation of the initial conditions, i.e.,

dυC ðt ¼ 0Þ
υC ðt ¼ 0Þ ¼ 0, ¼0 ð7:66Þ
dt
At the same time, the second-order ODE for the capacitor voltage becomes inhomoge-
neous, i.e., at t  0

d 2 υ C ðt Þ dυC ðt Þ
þ 2α þ ω20 υC ðt Þ ¼ ω20 V S ð7:67Þ
dt 2 dt
It is worth noting that this equation has exactly the same form as Eq. (7.63), but with the
nonzero right-hand side. The derivation of Eq. (7.67) is similar to the derivation of
Eq. (7.63); it is suggested as a homework problem.

VII-361
Chapter 7 Transient Circuit Fundamentals

7.5.4 Step Response and Choice of the Independent Function

The selection between current and voltage formulations reflects our desire to convert the
circuit differential equation into a standard form, which will allow us to use powerful
tools of signals and systems theory. What should be preferred: the homogeneous second-
order ODE Eq. (7.63) for the circuit current augmented with inhomogeneous initial
conditions Eqs. (7.65a, b), or the inhomogeneous second-order ODE Eq. (7.67) for the
capacitor voltage augmented with the homogeneous initial conditions Eq. (7.66)? The
answer is as follows. If the initial conditions are all zero, the only remaining excitation is
the forcing function: the right-hand side of Eq. (7.67). We consider a unit step function,
u(t), defined by (see Fig. 7.32a)

0 t<0
uðt Þ ¼ ð7:68Þ
1 t0

Equation (7.67) may be conveniently written in the form

d 2 υ C ðt Þ dυC ðt Þ
2
þ 2α þ ω20 υC ðt Þ ¼ Auðt Þ ð7:69Þ
dt dt
at any time instant where A ¼ ω20 V S . The forcing function is thus the product of the
constant A and the unit step function u(t), as seen in Fig. 7.32b. The solution to Eq. (7.67)
or Eq. (7.69), after division by A, is the normalized step response of a second-order
system. We call it the response to a unit step voltage excitation. It is generally accepted in
signals and systems theory that for the unit step response the initial conditions should be
homogeneous or zero. Therefore, Eq. (7.67) or Eq. (7.69) with zero initial conditions
Eq. (7.66) is preferred when dealing with future applications of the unit step response.
a) u(t) b) Au(t)
A
1

t t
0 0

Fig. 7.32. (a) Unit step function u(t) and (b) the scaled right-hand side of Eq. (7.69).

The step response is the “business card” of the circuit, which actually contains the
complete information about its behavior. If the circuit in Fig. 7.30 with an arbitrary time-
varying voltage source V S ! υS ðt Þ is considered, this source may be represented as a
number of “steps” in time. Hence, the complete solution may be constructed as a sum
(or integral) of the elementary unit step response solutions, properly scaled and shifted in
time. As an example, we consider a voltage source υS(t) in Fig. 7.30, which generates a
pulse (think of one bit) with the duration T and a 5-V peak value, as depicted in Fig. 7.33.

VII-362
Chapter 7 Section 7.5: Description of the Second-Order Transient Circuits

vS(t) 5u(t) -5u(t-T)

5V 5V
= +
t t t
0 T 0 0 T

-5V

Fig. 7.33. A one-bit excitation voltage constructed as the sum of two step responses.

The switch in Fig. 7.30 is replaced by a short circuit. According to Fig. 7.33, υS ðt Þ ¼
5uðt Þ  5uðt  T Þ. By linearity, the solution for the bit excitation is simply the sum 5υC ðt Þ
5υC ðt  T Þ in Fig. 7.33 where υC(t) is the solution of Eq. (7.69) with V S ¼ 1 V.

7.5.5 Parallel Connected Second-Order RLC Circuit

Generic Representation of the Parallel RLC Circuit and Qualitative Operation
Consider a parallel connected LC block in the shaded box in Fig. 7.34. Using Norton’s
theorem, any network of resistors and power supplies connected to this block is
represented as the parallel combination of a current source and a resistor R, as seen in
Fig. 7.34. This is why we arrive at the very generic parallel RLC circuit shown in
Fig. 7.34. The switch implies that the ideal current power supply is to be connected to
the circuit at time instant t ¼ 0. Qualitatively, just after closing the switch, the entire
circuit current will flow through the capacitor. As time progresses, the capacitor current
will decrease. Finally, the entire current will flow through the inductor, which becomes
the short circuit under DC conditions.
IS
iL
t=0
+ +
IS v R L C v
- -

LC network

Fig. 7.34. Generic representation of any network of sources and resistances connected to the
parallel LC circuit: the standard parallel RLC circuit.

Circuit Equation in Terms of Voltage

The same reasoning as mentioned for the series RLC circuit applies. There are two
possible choices of the independent function for the RLC circuit in Fig. 7.34:
- Circuit (or capacitor) voltage, υ(t)
- Inductor current iL(t)

VII-363
Chapter 7 Transient Circuit Fundamentals

The capacitor voltage remains continuous over time (inductor “inertia”) and so does the
inductor current (inductor “inertia”). We use for derivation υ(t) first. KCL yields
ðt
dυðt Þ υðt Þ 1
C þ þ υðt 0 Þdt 0 ¼ I S ð7:70Þ
dt R L
0

at any positive time t > 0. Differentiation over time and division by C yield the expected
homogeneous second-order ODE:

d 2 υ ðt Þ 1 dυðt Þ 1
2
þ þ υ ðt Þ ¼ 0 ð7:71Þ
dt RC dt LC
Eq. (7.71) is written in the form of Eq. (7.63):

d 2 υ ðt Þ dυðt Þ
2
þ 2α þ ω20 υðt Þ ¼ 0 ð7:72Þ
dt dt
if we define the damping coefficient α and the undamped resonant frequency ω0 as
1 1
α¼ , ω0 ¼ pffiffiffiffiffiffiffi ð7:73Þ
2RC LC
Initial Conditions and Choice of Independent Function
The initial conditions for Eq. (7.72) are that the voltage across the capacitor and the
inductor current must be continuous. Therefore, they must have the form
dυ IS
υðt ¼ 0Þ ¼ 0, ðt ¼ 0Þ ¼ ð7:74Þ
dt C
The voltage derivative is not zero at the initial time moment. Eq. (7.74) is similar to
Eqs. (7.65a, b). It has been stated that the step response of the second-order system is
generally calculated with the homogeneous (zero) initial conditions. The second-order
circuit ODE written in terms of the inductor current

d 2 i L ðt Þ diL ðt Þ
2
þ 2α þ ω20 υL ðt Þ ¼ ω20 I S ð7:75Þ
dt dt
possesses zero initial conditions, i.e.,

diL ðt ¼ 0Þ
iL ðt ¼ 0Þ ¼ 0, ¼0 ð7:76Þ
dt
Those are the preferred conditions for the step response calculations. The derivation of
Eq. (7.75) is similar to the derivation of Eq. (7.72); it is left as a homework problem.

VII-364
Chapter 7 Section 7.5: Description of the Second-Order Transient Circuits

Duality
Comparing Eqs. (7.66) and (7.67) for the series LCR circuit with Eqs. (7.76) and (7.75)
for the parallel LCR circuit, we can establish the following substitutions:
L
V S $ RI S , υC $ RiL , $ RC ð7:77Þ
R
These substitutions make both sets of equations including all the constants mathemati-
cally identical. This fact reflects the duality of series/parallel RLC electric circuits. A
similar duality is established for the steady-state RLC resonator circuits. Since the initial
conditions are also the same, we conclude that the step response of the parallel RLC
circuit is equivalent to the step response of the series RLC circuit.

Exercise 7.14: The damping coefficient of a second-order RLC circuit (i) does not depend
on capacitance, (ii) decreases when resistance increases, and (iii) equals zero.
Determine the circuit topology in every case.
Answer: (i) Series RLC circuit, (ii) parallel RLC circuit, (iii) series LC circuit.

VII-365
Chapter 7 Transient Circuit Fundamentals

Section 7.6 Step Response of the Series RLC Circuit

7.6.1 General Solution of the Second-order ODE
Solution for Step Response
The starting point is Eq. (7.69) of the previous section for the capacitor voltage of the
series RLC circuit augmented with the homogeneous initial conditions, i.e.,

d 2 υ C ðt Þ dυC ðt Þ
2
þ 2α þ ω20 υC ðt Þ ¼ Auðt Þ, A ¼ ω20 V S ð7:78aÞ
dt dt
dυC ðt ¼ 0Þ
υC ðt ¼ 0Þ ¼ 0, ¼0 ð7:78bÞ
dt
Similar to the first-order transient circuits with arbitrary sources, the general solution is
also given by the sum of two parts: a particular solution of the inhomogeneous equation
(7.78a), let us call it xp(t), and a complementary solution, let us call it xc(t), of
the homogeneous equation (7.78a). Homogeneous implies that the right-hand side of the
ODE equals zero. The particular solution is known as the forced response and the
complementary solution is known as the natural response. As a result, the total solution is

υC ðt Þ ¼ xp ðt Þ þ xc ðt Þ ð7:79Þ

For the circuit with the DC voltage source and the switch acting as a step excitation, the
particular solution is trivial. It is proved by direct substitution:

xp ðt Þ ¼ V S ) υC ðt Þ ¼ V S þ xc ðt Þ ð7:80Þ

The complementary solution carries information about the entire circuit and requires care.

Solution in Arbitrary Case

What if the right-hand side of Eq. (7.78a) is an arbitrary function of time? How is the
solution obtained? We have already established that any such solution can be obtained on
the basis of the step response. In general, the solution is expressed in terms of a
convolution integral, which involves an arbitrary right-hand side of the second-order
ODE and the time derivative of the step response. This interesting and fundamental
question is studied further in signals and systems theory.

7.6.2 Derivation of the Complementary Solution: Method

of Characteristic Equation
Similar to the first-order transient circuits, we seek a complementary solution (natural
response) of the homogeneous Eq. (7.78a) in the most general exponential form

VII-366
Chapter 7 Section 7.6: Step Response of the Series RLC Circuit

xp ðt Þ ¼ Kexpðst Þ ð7:81Þ

where K and s are two arbitrary constants. The substitution yields

2
s þ 2α s þ ω20 Kexpðst Þ ¼ 0 ð7:82Þ

For a nontrivial solution, the characteristic equation s2 þ 2αs þ ω20 ¼ 0 must be satis-
fied, that is,
8 sffiffiffiffiffiffiffiffiffiffiffiffi!
>
> ς2  1
>
> α 
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi >
< 1
α þ pα 2  ω2 ς2
s2 þ 2αs þ ω20 ¼ 0 ) s1, 2 ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0 or s
1, 2 ¼ sffiffiffiffiffiffiffiffiffiffiffiffi! ð7:83Þ
α  α2  ω20 >
> ς 2

>
> 1
: α 1 þ
>
ς2

where the new constant ζ ¼ α=ω0 is the damping ratio of the RLC circuit. Formally, this
constant has units of 1/rad; it is often considered dimensionless. We must distinguish
between three separate cases depending on the value of the damping ratio:
Case A This situation (overdamping) corresponds to ς > 1. In this case, s1,2 are both real and
negative. Since the original ODE is linear, the general solution is simply the combination of two
independent decaying exponential functions:

xc ðt Þ ¼ K 1 expðs1 t Þ þ K 2 expðs2 t Þ ð7:84aÞ

Case B This case (critical damping) corresponds to ς ¼ 1. Both roots s1,2 become identical.
Therefore, a solution in the form of Eq. (7.84a) with two independent constants can no longer be
formed. Only one independent constant may be available. Fortunately, another solution in the
form t exp(s1t) exists in this special case. This fact is proved by direct substitution. Thus, the
general solution becomes

xc ðt Þ ¼ K 1 expðs1 t Þ þ K 2 texpðs1 t Þ ð7:84bÞ

Case C This case (underdamping) corresponds to ς < 1. Both roots s1,2 become complex. This
means that our initial simple guess Eq. (7.81) is no longer correct. One can prove by direct
substitution that the general solution now has the oscillating form

xc ðt Þ ¼ K 1 expðα t Þ cos ωn t þ K 2 expðα t Þ sin ωn t ð7:84cÞ

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
where ωn ¼ ω20  α2 is the (radian) natural frequency of the circuit. 1/α is also called the time
constant or the time constant of the decay envelope. The complementary solution in the form of
Eqs. (7.84) always contains two independent integration constants. They should be used to satisfy
the initial conditions, which complete the solution.

VII-367
Chapter 7 Transient Circuit Fundamentals

7.6.3 Finding Integration Constants

According to Eqs. (7.79) and (7.80), the capacitor voltage is υC ðt Þ ¼ xc ðt Þ þ V S where
xc(t) is given by Eqs. (7.84). The integration constants may be found using Eq. (7.78b),
which dictates that both the capacitor voltage and its derivative must vanish at the initial
time t ¼ 0. We then have from Eqs. (7.84)

Case A: K 1 þ K 2 þ V S ¼ 0, s1 K 1 þ s2 K 2 ¼ 0 ð7:85aÞ
Case B: K 1 þ V S ¼ 0, s1 K 1 þ K 2 ¼ 0 ð7:85bÞ
Case C: K 1 þ V S ¼ 0,  αK 1 þ ωn K 2 ¼ 0 ð7:85cÞ

The solution of Eqs. (7.85) has the form

s2 V S s1 V S
Case A: K1 ¼ , K2 ¼ ð7:86aÞ
s1  s2 s2  s1
Case B: K 1 ¼ V S , K 2 ¼ s1 V S ð7:86bÞ
α
Case C: K 1 ¼ V S , K 2 ¼  V S ð7:86cÞ
ωn
Equations (7.83) through (7.86) complete the step response solution for the series RLC
circuit. The circuit may behave quite differently depending on the value of the damping
ratio ς.

7.6.4 Solution Behavior for Different Damping Ratios

We consider the series RLC circuit shown in Fig. 7.35. We will choose round numbers
L ¼ 1 mH, C ¼ 1 nF. These values approximately correspond to an RLC transient circuit
operating in the 100 kHz–1 MHz frequency band.

t=0
R
+ -
vR
i(t)
+
10 V=VS + 1 mH vL
- -
1 nF

- vC +

Fig. 7.35. RLC series circuit; the resistance value R may be varied.

VII-368
Chapter 7 Section 7.6: Step Response of the Series RLC Circuit

Example 7.9: Determine the solution for the capacitor voltage for the circuit shown in
Fig. 7.35 over a time interval from 0 to 25 μs for R ¼ 200 Ω, 1 kΩ, 2 kΩ, and 20 kΩ.
Solution: Equations (7.83) through (7.86) give
R ¼ 200 Ω ) α ¼ 105 , ω0 ¼ 106 ) ζ ¼ 0:1 (Case C—underdamped circuit)
 
) υC ðt Þ ¼ 10  10exp 105 t cos 9:95  105 t  1:005exp 105 t sin 9:95
 105 t ð7:87aÞ

R ¼ 1 kΩ ) α ¼ 5  105 , ω0 ¼ 106 ) ζ ¼ 0:5 (Case C—underdamped circuit)


) υC ðt Þ ¼ 10  10exp 5  105 t cos 8:66  105 t

 5:77exp 5  105 t sin 8:66  105 t ð7:87bÞ

R ¼ 2 kΩ ) α ¼ 106 , ω0 ¼ 106 ) ζ ¼ 1 (Case B—critically damped circuit)

 
) υC ðt Þ ¼ 10  10exp 106 t  107 texp 106 t ð7:87cÞ

R ¼ 20 kΩ ) α ¼ 107 , ω0 ¼ 106 ) ζ ¼ 10 (Case A—overdamped circuit)

 
) υC ðt Þ ¼ 10  10:0252exp  5:013  104 t þ 0:0252exp  1:995  107 t ð7:87dÞ

Equations (7.87) satisfy both the initial conditions to within numerical rounding error.
Figure 7.36 shows the solution behavior for four distinct cases.

The first person who discovered and documented the oscillatory transient response of
an electric circuit similar to that depicted in Fig. 7.36 was probably Félix Savary (1797–
1841). A renowned astronomer and French academician, he worked with Ampère and
discovered an oscillatory discharge of a Leyden jar (an early prototype of the battery) in
1823–1826. Some fifteen years later, the similar observation has been made by Joseph
Henry.

7.6.5 Overshoot and Rise Time

One important result seen in Fig. 7.36a, b is the so-called dynamic overshoot caused by a
sudden application of a voltage pulse and the associated voltage ringing. The dimension-
less overshoot (overshoot percentage after multiplying by 100) Mp is the maximum
voltage value minus the supply voltage divided by the supply voltage. For a slightly

VII-369
Chapter 7 Transient Circuit Fundamentals

damped circuit, the overshoot may be quite large—see Fig. 7.36a, b. The rise time tr
(which is sometimes called the “transition time”) in digital circuits is the time taken for
the voltage to rise from 0.1VS to 0.9VS (T. L. Floyd, Digital Fundamentals, 9th, p. 8);
see Figs. 8.14a–c. The circuit designer typically attempts to minimize both the rise
time and the overshoot. An important example considered later in our text is a pulse
train to be transmitted at a maximum speed (which requires minimum rise time) and
with minimum distortion (which requires minimum overshoot). Figure 7.36 indicates
that those goals are in fact conflicting. Decreasing the rise time increases the overshoot
and vice versa. Designing the damping ratio close to unity, or slightly below it, is a
reasonable compromise to quickly achieve the desired voltage level without a significant
overshoot and ringing. The overshoot and rise time may be estimated analytically. We
present here the estimates found in common control theory textbooks:

expðπζ Þ
M p ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi f or ζ < 1, M p ¼ 0 f or ζ  1 ð7:88aÞ
1  ζ2

t r ¼ 1  0:4167ζ þ 2:917ζ 2 =ωn f or ζ < 1 ð7:88bÞ

Exercise 7.15: The damping coefficient of 15,000 neper/s and the natural frequency of
10 kHz are measured for an unknown series RLC circuit in laboratory via its step response.
Given R ¼ 10 Ω, determine L and C.
Answer: L ¼ 0:333 mH, C ¼ 0:719 μF.

As to mechanical engineering, it is interesting to note that an automotive suspension

system is described by the same step response model and behaves quite similarly to the
RLC circuit in Fig. 7.36. Well, driving a car everyday should certainly be a motivation for
studying this topic.

7.6.6 Application Example: Nonideal Digital Waveform

Modeling Circuit
The series RLC block is an appropriate model to study the voltage pulse as it realistically
occurs in digital circuits and in power electronic circuits involving pulse width modula-
tion (PWM). The ideal square voltage pulse, shown in Fig. 7.33 of the previous section, is
a crude approximation of reality. Parasitic capacitance, resistance, and inductance are
always present in the circuit. As a result, the pulse form is distorted. To model the pulse
form distortion, we consider the circuit shown in Fig. 7.37.

VII-370
Chapter 7 Section 7.6: Step Response of the Series RLC Circuit

20
overshoot

Capacitor voltage, V
16

R=200 , =0.1 12
Case C VS
underdamped circuit
8

rise time
4

0
20
Capacitor voltage, V

16

R= k , =0.5 overshoot
12
Case C VS
underdamped circuit
8
rise time
4

0
20
Capacitor voltage, V

16

R=2k , =1.0 12 no overshoot

Case B VS
critically damped
circuit 8

rise time
4

0
20
Capacitor voltage, V

16

R=20k , =10 12 no overshoot

Case A VS
overdamped circuit
8
rise time
4

0
0 5.0 10 15 20 t, s 25

Fig. 7.36. Circuit responses in terms of capacitor voltages for different damping factors.

VII-371
Chapter 7 Transient Circuit Fundamentals

a) b)
R vS(t)
VS
Falling (trailing)
vS(t) + L edge
- Rising (leading)
edge
C

0 T t
- vC +

Fig. 7.37. The RLC circuit for studying the nonideal digital (pulse) waveform.

The switch is now removed and a time-varying voltage source υS(t) is introduced; it
generates the voltage pulse (one bit) of duration T with amplitude VS as seen in Fig. 7.37b.
One may think of the source voltage as an “ideal” digital waveform and, for example, of
the capacitor voltage as a nonideal (realistic) waveform influenced by parasitic capaci-
tance, resistance, and inductance.

Solution
The solution to the pulse problem is derived as described at the end of the previous
section. We know that Eqs. (7.83) through (7.86) determine the step response—the
capacitor voltage υC(t) for the circuit with the DC voltage source shown in Fig. 7.35
after closing the switch. To obtain the solution υpulse
C (t) for the voltage pulse shown in
Fig. 7.37, we simply combine two such step responses, i.e.,

υCpulse ðt Þ ¼ υC ðt Þ  υC ðt  T Þ ð7:89Þ

This operation again underscores the importance of the fundamental step response
solution. Close inspection of Eq. (7.89) shows that the pulse will possess the dynamic
overshoot and the nonzero rise time similar to the step response solution. This happens at
the rising (or leading) edge of the pulse. At the same time, a dynamic undershoot and a
nonzero settling time will happen at the falling or trailing edge as depicted in Fig. 7.38.

Example 7.10: Determine the solution for the capacitor voltage, υpulse
C (t), for the circuit
shown in Fig. 7.37 with L ¼ 1 μH, C ¼ 1 nF, V S ¼ 10 V, T ¼ 0:5 μs over the time
interval from 0 to 1 μs for R ¼ 15 Ω, 30 Ω, and 60 Ω.
Solution: We find the step response υC(t) following Eqs. (7.83) through (7.86) first and
then obtain the final solution using Eq. (7.89). For the step response, we obtain
R ¼ 15 Ω ) ζ ¼ 0:24 (Case C—underdamped circuit) )

υC ðt Þ ¼ 10  10exp 7:5  106 t cos 3:07  107 t

 2:44exp 7:5  106 t sin 3:07  107 t ð7:90aÞ

VII-372
Chapter 7 Section 7.6: Step Response of the Series RLC Circuit

Example 7.10 (cont.):

R ¼ 30 Ω ) ζ ¼ 0:47 (Case C—underdamped circuit) )

υC ðt Þ ¼ 10  10exp 1:5  107 t cos 2:78  107 t

 5:39exp 1:5  107 t sin 2:78  107 t ð7:90bÞ

R ¼ 60 Ω ) ζ ¼ 0:95 (Case C—underdamped circuit) )


υC ðt Þ ¼ 10  10exp 3:0  107 t cos 1:0  107 t

 30exp 3:0  107 t sin 1:0  107 t ð7:90cÞ

Figure 7.38a–c shows the distorted pulse forms for three particular cases. Figure 7.38a
outlines the major pulse parameters: rise time, fall time, overshoot, undershoot, and pulse
width. One can see that there is again a conflict between the desire to simultaneously
decrease the rise time and the overshoot.

The overshoot and undershoot in Fig. 7.38 approximately coincide, and so do the rise
time and the fall or settling time. Note that this is not always the case. The voltage pulse
may be very significantly and unsymmetrically distorted when the initial pulse width, T,
is comparable with the rise time. A good illustration is the previous example solved for
R ¼ 15 Ω when T ¼ 0:25 μs or less.

Exercise 7.16: Using a theoretical approximation, find the overshoot for the case of
Fig. 7.38a and compare this value with value observed on the figure.
Answer: 48 % (theory) versus 50 % (observation).

VII-373
Chapter 7 Transient Circuit Fundamentals

a) overshoot
ringing VS
10

Capacitor voltage, V
90% 90%

R=15W, ␨=0.24 5 50% 50%

Case C pulse width
ringing
underdamped circuit 0 10% 10%

-5 undershoot
rise time fall time
-10
b)
overshoot VS
Capacitor voltage, V

10
90%

R=30W, ␨=0.47 5 50%

Case C
underdamped circuit 0 10%
undershoot
-5
rise time
-10

c)
VS
10
Capacitor voltage, V

90%

R=60W, ␨=0.95 5 50%

Case C
underdamped circuit 0 10%

-5
rise time
-10
0 0.2 0.4 0.6 0.8 t, ms

Fig. 7.38. Distorted pulse forms for three different values of the damping ratio. When the damping
ratio increases, the overshoot decreases but the rise time increases.

The solution for the second-order circuits with arbitrary sources and arbitrary initial
conditions can quite simply be obtained numerically. A straightforward finite-difference
second-order method may be implemented in MATLAB or in other software packages
with a few lines of the code. This method is the extension of the Euler method used for
first-order transient circuits. Interestingly, the same method may be applied to radio-
frequency pulse propagation in transmission lines and in free space, including problems
such as signal penetration through walls.

VII-374
Chapter 7 Summary

Summary
General facts about transient circuits
Voltage across the capacitor(s) remains continuous for all times—use this voltage as an unknown
function
Current through the inductor(s) remains continuous for all times—use this current as an unknown
function
When discharged through a small resistance, the (ideal) capacitor is able to deliver an extremely
high power (and current) during a very short period of time
When disconnected from the source, the (ideal) inductor in series with a large resistance is able to
deliver an extremely high power (and voltage) during a very short period of time
Transient circuit Generic circuit diagram Solution plot
Energy-release RC circuit
 t
υC ðt Þ ¼ V 0 exp 
τ
τ ¼ RC
ODE:
dυC υC
þ ¼0
dt τ
Energy-accumulating RC circuit
h  t i
υC ðt Þ ¼ VS 1  exp 
τ
τ ¼ RC
ODE:
dυC υC VS
þ ¼
dt τ τ
Energy-release RL circuit
 t
iL ðt Þ ¼ I S exp 
τ
τ ¼ L=R
ODE:
diL iL
þ ¼0
dt τ
Energy-accumulating RL circuit
h  t i
iL ðt Þ ¼ I S 1  exp 
τ
τ ¼ L=R
ODE:
diL iL I S
þ ¼
dt τ τ
Energy-release RL circuit
 t
iL ðt Þ ¼ I 0 exp 
τ
τ ¼ L=R, I 0 ¼ VS =R0
ODE:
diL iL VS
þ ¼
dt τ R0 τ
(continued)

VII-375
Chapter 7 Transient Circuit Fundamentals

Voltage or current Power

Generic energy-release curves

for either dynamic element

Switching RC oscillator (relaxation oscillator)—RC timer

Bistable amplifier circuit
with positive feedback.
Two stable states:
υout ¼ þVCC
υout ¼ VCC
Threshold voltage(s):
R1
υ* ¼ υout
R1 þ R2
Relaxation oscillator with
positive feedback:
υout ¼ þVCC if υ* > υC
υout ¼ VCC if υ* < υC
Threshold voltage(s):
R1
υ* ¼ υout
R1 þ R2
1þβ  
Period and frequency of the
T ¼ 2τln 1 1 1 þ β 1
relaxation oscillator 1β f ¼ ¼ ln
T 2τ 1β
ðβ ¼ R1 =ðR1 þ R2 ÞÞ

Single-time-constant (STC) transient circuits

τ ¼ RC
or
L
τ¼
R
 
τ ¼ R1 R2 C or

L
τ¼
R1 þ R2

τ ¼ RðC 1 þ C 2 Þ
or
L1 þ L2
τ¼
R
(continued)

VII-376
Chapter 7 Summary

L1 þ L2
τ¼
R1 þ R2

τ ¼ R1 R2 
ðC 1 þ C 2 Þ Arbtr. initial conditions are not allowed
Arbtr. initial conditions are not
allowed
STC circuits with general sources

Bypass capacitor
and decoupling
inductor
υS ðt Þ ¼ VS þVm cos ω t, τ ¼ iS ðt Þ ¼ I S þ I m cos ω t, τ ¼
RS RL C L=ðRS þ RL Þ
RL VS   t  RS I S   t 
υL ¼ 1  exp  iL ¼ 1exp 
RS þ RL τ RS þRL τ
Solution for load RL V m 1 RS I m 1
voltage or load þ  þ 
RS þ RL 1 þ ðωτÞ2 RS þRL 1þ ðωτÞ2
current h  t i
h  t i
cosωt þ ωτ sinωt  exp  cosωtþωτsinωtexp 
τ τ
Second-order transient circuits

With two
identical dynamic
elements

• Damping coefficient: α ¼ R=ð2LÞ

pffiffiffiffiffiffiffi
• Undamped res. freq.: ω0 ¼ 1= LC
With series LC • Damping ratio: ζ ¼ α=ω0
network pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
• Step response with zero initial con- • Natural freq.: ωn ¼ ω20  α2
ditions: use capacitor voltage υC(t) Characteristic Eq.: s2 þ 2αs þ ω20 ¼ 0
• Damping coefficient: α ¼ 1=ð2RC Þ
pffiffiffiffiffiffiffi
• Undamped res. freq.: ω0 ¼ 1= LC
With parallel LC
• Damping ratio: ζ ¼ α=ω0
network • Step response with zero initial con- pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ditions: use inductor current iL(t) • Natural freq.: ωn ¼ ω20  α2
Characteristic Eq.: s2 þ 2αs þ ω20 ¼ 0
(continued)

VII-377
Chapter 7 Transient Circuit Fundamentals

• Overdamped (ζ > 1):

xc ðt Þ ¼ K 1 expðs1 t Þ þ K 2 expðs2 t Þ
s2 V S s1 V S
K1 ¼ , K2 ¼
s1  s2 s2  s1
Overdamped, • Critically damped (ζ ¼ 1):
critically-damped, xc ðt Þ ¼ K 1 expðs1 t Þ þ K 2 t expðs1 t Þ
and underdamped
RLC circuits K 1 ¼ V S , K 2 ¼ s1 V S
• Underdamped (ζ < 1):
xc ðt Þ ¼ K 1 expðα t Þ cos ωn tþ
K 2 expðα t Þ sin ωn t
K 1 ¼ VS , K 2 ¼ ωαn VS
• Overshoot
expðπζ Þ
M p ¼ pffiffiffiffiffiffiffiffiffiffiffiffiffi f or ζ < 1
1  ζ2
• Undershoot is approximately over-
Non-ideal digital shoot for rise times small compared
waveform: to pulse width
second-order • Rise time

circuit t r ¼ 10:4167ζþ2:917ζ 2 =ωn
f or ζ<1
Fall time is approximately rise time
for rise times small compared to
pulse width

VII-378
Chapter 7 Problems

Problems C=10 F
t=0
R=100

7.1 RC Circuits +
vC
-
vR
iC iR
7.1.1 Energy-Release Capacitor Circuit - +

7.1.2 Time Constant of an RC Circuit

and Its Meaning
7.1.3 Continuity of the Capacitor Problem 7.3. Prove that Eq. (7.6) is the solu-
tion to Eq. (7.5) (both from Section 7.1) using
Voltage direct substitution and the differentiation that
Problem 7.1. For the capacitor as a dynamic follows.
circuit element, develop:
1. Equivalent circuit at DC Problem 7.4.
2. Relation between voltage and current A. Show that the time constant, τ, of an RC
3. Expression for the time constant of a circuit has the units of seconds.
transient circuit that includes the dynamic B. To obtain the slow discharge rate of lesser
element and a resistor R instantaneous power into the load, should
the load resistance be small or large?
Dynamic circuit element vC
+ - Problem 7.5. A 100-μF capacitor discharges
into a load as shown in the following figure.
iC The load resistance may have values of 100 Ω,
Equivalent circuit at DC (short or 10 Ω, and 1 Ω. The capacitor is charged to 20 V
open) prior to t ¼ 0.
t=0
Relation between voltage and
current (passive reference C=100 μF

configuration) +
vC RL
Expression for the time constant τ¼ -
of a transient circuit that includes
the dynamic element (C) and a
resistor R. A. Find time constant τ and the maximum
instantaneous power delivered to the load
resistor in the very first moment for every
Problem 7.2. Using KCL and KVL, derive the
resistor value—fill out the Table that
differential equation for the circuit shown in the
follows.
following figure, keeping the same labeling for
the voltages and the currents. Instantaneous load power right after the switch
A. Is the final result different from Eq. (7.5) closes
of Section 7.1?
B. Could you give an example of a certain RL τ, s pL ðt ¼ þ0Þ, W
voltage and/or current labeling 100 Ω
(by arbitrarily changing polarities and 10 Ω
directions in the figure) that causes the 1Ω
differential equations to change?
B. Do the instantaneous power values from
the Table depend on the capacitance
value?

VII-379
Chapter 7 Transient Circuit Fundamentals

Problem 7.6. A 100-μF capacitor, shown in the vC(t), V

following figure, discharges into a 10-Ω load 20

resistor. The capacitor is charged to 15 V prior 15

to t ¼ 0.
10
A. Find the time constant of the circuit
(show units). 5

B. Express the voltage across the capacitor

as a function of time and sketch it to
-1.0 -0.5 0 0.5 1.0 1.5 2.0 2.5 t, ms
scale versus time over time interval pR(t), W
from 2 ms to 5 ms.
80

t=0
60
C=100 F R=10
40
+ +
vC vR
- - 20

-1.0 -0.5 0 0.5 1.0 1.5 2.0 2.5 t, ms

vC(t), V
20
Problem 7.8. In the circuit shown in the fol-
15 lowing figure, the capacitor is charged to 20 V
prior to t ¼ 0.
10
A. Find an expression for the voltage across
5 the capacitor as a function of time and
sketch it to scale versus time over the
-2 -1 0 1 2 3 4 t, ms
interval from –2τ to 5τ.
B. Repeat for instantaneous power deliv-
Problem 7.7. A 100-μF capacitor, shown in the ered to the rightmost resistor.
following figure, discharges into a 5-Ω load
t=0
resistor. The capacitor is charged to 20 V prior
to t ¼ 0. +
C=1 F
vC
A. Find an expression for the voltage across - R=50
the capacitor as a function of time and R=50
sketch it to scale versus time over the
interval from –2τ to 5τ.
B. Repeat the exercise for instantaneous Problem 7.9. Present the text of a MATLAB
power delivered to the resistor. script (or of any software of your choice) in
order to generate Fig. 7.2d of Section 7.1.
t=0
Attach the figure so generated to the homework
report.
+ C=100 F R=5 +
vC vR
- -

VII-380
Chapter 7 Problems

Problem 7.10. Prove that the integral of the B ¼ 0:3 T. The accelerating object has a length
load power in Fig. 7.2d given by Eq.(7.8c) is of 2 cm. Plot to scale the Lorentz force as a
exactly equal to the energy stored in the function of discharge current over the interval
charged capacitor, E C ¼ 12 CV 20 prior to t ¼ 0. 0 < iC < 1000 A.

Problem 7.11. Problem 7.14. An electromagnetic capacitor

A. Create the generic capacitor voltage dis- accelerator needs to create an average force of
charge curve similar to Fig. 7.2a but for an 5 N over 2 ms on a moving object with length
arbitrary capacitor powering an arbitrary of 1 cm. The load (armature plus object) resis-
load resistor over the time interval from – tance is 1 Ω, and the external magnetic field is
2τ to 5τ. The capacitor is charged to V0 B ¼ 0:25 T. Determine:
prior to t ¼ 0. To do so, find the capacitor A. The required capacitor voltage prior to
voltage as a fraction of V0 for every unit discharge
of τ and fill out the Table that follows. B. The required capacitance of the capacitor
(bank of capacitors)
Capacitor voltage in terms of V0 Hint: Assume that the average force acts over
t 2τ -τ 0 τ 2τ 3τ 4τ 5τ the time interval τ. Its value is approximately
υC(t)
equal to 60 % of the initial force value.
Problem 7.15. Solve the previous problem
B. Repeat the same task for Fig. 7.2d related
when:
to load power. Find the load power in
A. The average force increases to 50 N
terms of the maximum power just after
B. The average force increases to 500 N
closing the switch.
Problem 7.16. The world's largest capacitor
Problem 7.12. For an unknown energy-release
bank is located in Dresden, Germany. The
RC circuit, capacitor voltage and capacitor cur-
pulsed, capacitive power supply system was
rent were measured in laboratory before and
designed and installed for studying high mag-
after closing the switch at t ¼ 0 as shown in
netic fields by experts from Rheinmetall Waffe
the figure that follows. Approximate R and C.
Munition. The bank delivers 200 kA of dis-
25 5
charge current in the initial time moment (just
after the switch closes). The time constant is
Capacitor current, A
Capacitor voltage, V

20 4
100 ms. Estimate the bank capacitance if the
3
15
charging voltage is 200 kV.
10 2

5 1

0 0
-2 -1 0 1 2 3 4 5
time, ms

7.1.4 Application Example:

Electromagnetic Railgun
7.1.5 Application Example:
Electromagnetic Material Processing
Problem 7.13. An electromagnetic capacitor
accelerator with permanent magnets has

VII-381
Chapter 7 Transient Circuit Fundamentals

7.1.7 Energy-Accumulating Capacitor Problem 7.19. For the circuit shown in the
Circuit following figure:
Problem 7.17. A 1-μF capacitor shown in the A. Find an expression for the capacitor volt-
following figure is charged through the 1-kΩ age, υC, and the capacitor current, iC,
load resistor. The initial capacitor voltage is including the value of time constant.
zero. B. Sketch the capacitor voltage, υC, and the
A. Find the time constant of the circuit capacitor current, iC, to scale versus time
(show units). over the time interval from –2τ to 5τ.
B. Express the voltage across the capacitor
as a function of time and sketch it to t=0
R=0.5
scale versus time over the interval from
2 ms to 5 ms.
+
+ 50 V=VS iC iC vC
t=0 R=1 k - -
vR - C=33 F C=33 F
+

15 V=VS iR
+
+
vC
Problem 7.20. Sketch your own fluid-flow
C=1 F iC
- - counterpart of the charging circuit shown in
the figure and establish as many analogies
between electrical (R, C, VS) and mechanical
vC(t), V parameters of your drawing as possible.
20

t=0 R
15

10

VS + C
5
-

-2 -1 0 1 2 3 4 t, ms
Problem 7.21. For the circuit shown in the
Problem 7.18. For the circuit shown in the following figure:
following figure: A. How much time does it take to charge
A. Find an expression for the capacitor volt- the capacitor to 10 V?
age, υC, and the capacitor current, iC, B. To 25 V?
including the value of time constant.
B. Sketch the capacitor voltage, υC, and the t=0 R=50
capacitor current, iC, to scale versus time
over the interval from –2τ to 5τ. +
+ 25 V=VS C=47 F iC vC
- -
t=0 R=1

25 V=VS
+
+ C=47 F iC vC Problem 7.22. For the circuit shown in the
- - figure, how much time does it take to charge

VII-382
Chapter 7 Problems

the capacitor, C1, to 10 V? Assume that the

t=0
initial voltages of both capacitors are zero. 1 A=IS iR + +
vR vC
t=0
- -
R=50 R=100 C=10 F

C1=47 F
+ 25 V=VS
Problem 7.25. Repeat the previous problem for
- C2=47 F the circuit shown in the following figure.

R=100
1 A=IS iR iC
Problem 7.23. For an unknown energy- + +
vR vC
accumulating RC circuit, capacitor voltage - -
t=0 C=47 F
and capacitor current were measured in labora-
tory before and after closing the switch at t ¼ 0
as shown in the figure that follows. Approxi- Problem 7.26. Obtain an analytical solution
mate R and C. for the capacitor voltage in the circuit shown
in the following figure at any time and express
25 5
it in terms of Is, R1, R2, C. Find
Capacitor voltage, V

Capacitor current, A

4
20
the time constant of the circuit when
15 3 R1 ¼ R2 ¼ 100 Ω, C ¼ 47 μF.
10 2

5 1
1 A=IS t=0
0 0 C
-2 -1 0 1 2 3 4 5
time, ms
R1 R2

Problem 7.24.
A. Obtain an analytical solution for the
Problem 7.27. For the circuit shown in the
capacitor voltage in the circuit shown
figure:
in the following figure. When the switch
A. Derive and solve the dynamic circuit
is closed, the current source still gener-
equation after the switch opens. Assume
ates current IS at its terminals. However,
the initial capacitor voltage equal to zero.
the supply is shorted out – no current
B. Plot the capacitor voltage to scale versus
flows into the circuit. When the switch
time over the time interval from –2τ to 5τ.
is open, the current flows into the circuit.
B. Could you convert this circuit to an t=0
equivalent RC transient circuit with the
voltage source?
C. Plot the voltage across the resistor versus R=100
C=10 F
time over the time interval from -2τ to 5τ. + 10 V=VS
-
R=100

VII-383
Chapter 7 Transient Circuit Fundamentals

Problem 7.28. In the circuit that follows, the

Dynamic circuit element vL
capacitor, C1, is initially uncharged. The switch + -
is closed at t ¼ 0.
iL
t=0
Equivalent circuit at DC (short or
open)
R1 Relation between voltage and
+ VS R2 current (passive reference
-
C1 configuration)
Expression for the time constant τ¼
of a transient circuit that includes
the dynamic element (L) and a
Give answers to the following questions based resistor R
on known circuit parameters C1, R1, R2, VS:
A. What is the current through resistor R2 as Problem 7.30.
a function of time? A. Using KCL and KVL, derive the differ-
B. What is the maximum current through ential equation for the inductor current in
resistor R1? the circuit shown in the figure that fol-
C. What is the current through resistor R1 at lows, keeping the same labeling for the
a time long after the switch closes? voltages and the currents.
D. What is the charge, Qþ ðt Þ, of the capac- B. Is the final result different from
itor, C1, as a function of time? Eq. (7.19) of Section 7.2?
The switch is then opened a very long time after
it has been closed – reset the time to t ¼ 0.
R=1 kΩ
A. What is the charge Qþ ðt Þ of the capaci-
L=1 mH
+ +
tor, C1, as a function of time? vL iL iR vR
- -
B. What is the current through resistor R2 as
a function of time? Specify the current
direction in the figure.
Problem 7.31. Prove that Eq. (7.20) is the
solution to Eq. (7.19) using direct substitution
7.2 RL Circuits and the corresponding differentiation.
7.2.1 Energy-Release Inductor Circuit
Problem 7.32.
7.2.2 Continuity of the Inductor Current A. Show that the time constant τ has units
Problem 7.29. For the inductor as a dynamic of seconds for the RL circuit.
circuit element, present: B. To ensure a slower energy release rate of
1. Equivalent circuit at DC the inductor, should the load resistance
2. Relation between voltage and current be small or large?
3. Expression for the time constant of a C. To ensure a faster energy release rate of
transient circuit that includes the dynamic the inductor, should the load resistance
element and a resistor, R be small or large?

VII-384
Chapter 7 Problems

Problem 7.33. A 6.8-μH inductor releases its B. Express the current through the inductor
energy into a load resistor as shown in the as a function of time and sketch it to
following figure. The load resistance may scale versus time over time interval
have values of 10 Ω, 100 Ω, and 1 kΩ. The from 2 μs to 5 μs.
inductor current is 1 A prior to t ¼ 0.
t=0 R L
t=0 R L IS
IS

iL(t), A

A. Find time constant τ and the maximum 1

instantaneous power delivered to the 0.8

load resistor in the very first moment
for every resistor value—fill out the 0.6

Table that follows. 0.4

Instantaneous load power just after the switch 0.2

closes
time, ms
τ, s pR ðt ¼ þ0Þ, W
-2 -1 0 1 2 3 4
R
10 Ω
Problem 7.36. A 270-μH inductor, shown in
100 Ω
the following figure, releases its energy into the
1 kΩ 510-Ω load resistor. The supply current is
0.8 A.
B. Do those instantaneous power values
A. Find the time constant of the RL circuit
from the Table depend on the inductance
(show units).
value?
B. Express the current through the inductor
Problem 7.34. Prove that the integral from 0 to as a function of time and sketch it to
1 of the load power in Fig. 7.10d is exactly scale versus time over time interval
from 1 μs to 2.5 μs.
equal to the energy stored in the inductor E L
C. Express the resistor voltage as a function
¼ 12 LI 2S prior to t ¼ 0. Hint: The proof should of time and sketch it to scale versus time
include analytical integration of the instanta- over time interval from 1 μs to 2.5 μs.
neous power in Eq. (7.21c).

Problem 7.35. A 2-mH inductor, shown in the t=0 R L

following figure, releases its energy into the IS
2-kΩ load resistor. The supply current is 0.8 A.
A. Find the time constant of the RL circuit
(show units).

VII-385
Chapter 7 Transient Circuit Fundamentals

iL(t), A B. Express the current through the inductor

1
as a function of time and sketch it to
0.8 scale versus time over time interval
from 2 μs to 10 μs.
0.6

0.4 t=0 1 kW 2 mH 1 kW

0.2
IS

-1 -0.5 0 0.5 1 1.5 2 time, ms

iL(t), A
0.1
vR(t), V
0 0.08

-100 0.06

-200 0.04

-300 0.02

-400
-2 0 2 4 6 8 time, ms

-1 -0.5 0 0.5 1 1.5 2 time, ms

Problem 7.39.
A. Obtain the solution for the inductor cur-
Problem 7.37. For an unknown energy-release rent in the circuit shown in the figure at
RL circuit, inductor current and resistor voltage any time.
were measured before and after closing the B. Plot to scale the current through the
switch at t ¼ 0 as shown in the figure that inductor versus time over the interval
follows. Approximate R and L. from –2τ to 10τ.
0
R=1kW
0.2 -10
inductor current iL, A

load voltage vR, V

-20 R=1kW
1 A=IS L=1 mH
0.1 -30
t=0
-40

0 -50
-0.1 0 0.1 0.2
7.2.3 Energy-Accumulating Inductor
time, m s

Problem 7.38. A 2-mH inductor, shown in the Circuit

following figure, releases its energy into two Problem 7.40. In the energy-accumulating RL
1-kΩ load resistors. The supply current is circuit shown in the following figure, R ¼ 2 kΩ
100 mA. and L ¼ 2 mH. The supply current is 1 A.
A. Find the time constant of the RL circuit
(show units).

VII-386
Chapter 7 Problems

t=0 R iR iL t=0 R iR iL
L L
+ +
IS IS vR vL
- -

A. Find the time constant of the RL circuit i L (t), A

(show units). 1
B. Express the current through the inductor
as a function of time and sketch it to 0.8

scale versus time over time interval 0.6

from 2 μs to 5 μs.
0.4

iL(t), A
1 0.2

0.8
-1 -0.5 0 0.5 1 1.5 2 time, ms

0.6

vR(t), V
0.4
1000

0.2
800

-2 -1 0 1 2 3 4 time, ms 600

400
Problem 7.41. In the energy-accumulating RL
circuit shown in the following figure, R ¼ 510 200

Ω and L ¼ 270 μH. The supply current is 1 A.

A. Find the time constant of the RL circuit -1 -0.5 0 0.5 1 1.5 2 time, ms

(show units).
B. Express the current through the inductor Problem 7.42.
as a function of time and sketch it to A. Obtain the solution for the inductor cur-
scale versus time over time interval rent in the circuit shown in the figure at
from 1 μs to 2.5 μs. any time.
C. Express the resistor voltage as a function B. Plot the voltage across the rightmost
of time and sketch it to scale versus resistor versus time over the interval
time over time interval from 1 μs to from –2τ to 5τ.
2.5 μs.
R=100

R=100
1 A=IS L=1 mH
t=0

VII-387
Chapter 7 Transient Circuit Fundamentals

7.2.4 Energy-Release RL Circuit with 7.3 Switching RC Oscillator

the Voltage Supply
7.3.2 Bistable Amplifier Circuit with the
7.2.5 Application Example: Laboratory
Positive Feedback
Ignition Circuit
Problem 7.43. A 270-μH inductor shown in the 7.3.3 Triggering
following figure releases its energy into the Problem 7.45. The bistable amplifier circuit
510-Ω load resistor. The power supply voltage shown in the following figure (inverting
is 20 V. The switch opens at t ¼ 0. Schmitt trigger) exists in the positive stable
A. Present an expression for the inductor state. Amplifier’s power supply rails are
current as a function of time and sketch 12 V. Determine output voltage when the
it to scale versus time over the interval applied trigger signal is
from 1 μs to 2.5 μs. 1. υin ¼ 6 V
B. Repeat the same task for the resistor 2. υin ¼ 2 V
voltage. 3. υin ¼ 4 V
Assume that the amplifier hits the power rails in
R0=51 t=0 saturation.
3 kW
L=270 H R=510
+ 20 V=VS 1 kW
-
+ +
- vout
0V
-
Problem 7.44. The circuit for the previous vin
problem is converted to the energy-
accumulating RL circuit by inversing the Problem 7.46. Repeat the previous problem
switch operation. Assume that the switch was when the initial stable state of the amplifier
open prior to t ¼ 0. The switch closes at t ¼ 0. circuit is negative.
A. Derive an expression for the inductor
current as a function of time. Problem 7.47. The bistable amplifier circuit
B. Repeat the same task for the voltage shown in the following figure (non-inverting
across resistor R. Schmitt trigger) exists in the positive stable
C. Could this circuit generate large voltage state. Amplifier’s power supply rails are
spikes, similar to the circuit from the 15 V. Determine output voltage when the
previous problem? applied trigger signal is
A. υin ¼ 1 V
R0=51 t=0 B. υin ¼ 2 V
C. υin ¼ 4 V
L=270 H R=510 Assume that the amplifier hits the power rails in
+ 20 V=VS saturation.
-
3 kW

vin 1 kW
+ +
- vout
-
0V

VII-388
Chapter 7 Problems

Problem 7.48. Repeat the previous problem 7.4 Single-Time-Constant

when the initial stable state of the amplifier
circuit is negative. (STC) Transient Circuits
7.3.4 Switching RC Oscillator 7.4.1 Circuits with Resistances and
7.3.5 Oscillation Frequency Capacitances
Problem 7.49. A clock circuit (relaxation oscil- 7.4.2 Circuits with Resistances and
lator circuit) shown in the following figure is Inductances
powered by a 10-V power supply. Problem 7.52. Determine whether or not the
transient circuits shown in the following figure
300 kΩ
are the STC circuits. If this is the case, express
100 kΩ the corresponding time constant in terms of the
+ circuit parameters.
- +
vout
+ R - A) R1 t=0 R3
vC C
- 0V
Vout
10V + VS R2 R4 C
-
7.5V

5.0V

2.5V t=0
B)
0V
t
-2.5V
C1
-5.0V
+ VS
-7.5V -
C2 R1
-10V

Sketch to scale the capacitor voltage, υC, as a t=0 R1

function of time. Assume that υC ðt ¼ 0Þ ¼ 0. C)
Assume the ideal amplifier model. The specific C1
values of R and C do not matter; they are + VS
already included in the time scale. -
C2 R2

Problem 7.50. An RC clock circuit is needed t=0

with the oscillation frequency of 1 kHz and D)
amplitude of the capacitor voltage of 4 V. C1 R1
Determine one possible set of circuit parame- + VS
ters R1, R2, R given that the capacitance of -
C2 R2
100 nF is used. The power supply voltage of
the amplifier is 12 V. Assume that the ampli-
fier hits the power rails in saturation. Problem 7.53. Repeat the previous problem for
the circuits shown in the following figure.
Problem 7.51. A relaxation oscillator circuit
may generate nearly triangular waveforms at
the capacitor. Which values should the feed-
back factor β ¼ R1RþR
1
2
attain to make it possible?

VII-389
Chapter 7 Transient Circuit Fundamentals

A) R2 7.4.4 Example of a STC Circuit

t=0 Problem 7.56. For the circuit shown in
R3 Fig. 7.26, derive ODEs for inductor currents
IS R1 i1, i2.
L

R2 7.4.5 Method of Thévenin Equivalent.

B)
t=0
Application Example: Circuit with a
Bypass Capacitor
IS L1 Problem 7.57. In the circuit from Fig. 7.27a,
R1 L2 another resistance R0 is present in series with
the capacitance C. Determine the natural
response of the circuit and find the
corresponding time constant, τ.
7.4.3 Example of a Non-STC Transient
Circuit Problem 7.58. A transient circuit with the DC
Problem 7.54. Using software of your choice, voltage source VS is shown in the following
generate the solution for the non-STC circuit figure. Given that V S ¼ 10 V and
in Fig. 7.25a over the time interval from 0 to R1 ¼ 1 kΩ, R2 ¼ 1 kΩ, and C ¼ 1 μF:
15τ0 where τ0 ¼ RC with C 1 ¼ C 2 ¼ C, A. Present the ODE for the capacitor volt-
R1 ¼ R2 ¼ R. Use R ¼ 1 kΩ, C ¼ 1 μF, and age υC(t).
V S ¼ 10 V. Plot two capacitor voltages and B. Determine the value of the time constant
the circuit current as functions of time to scale. τ and the ODE right-hand side (the forc-
ing function).
Problem 7.55. C. Present the solution for the capacitor
A. Derive the general ODE for the non-STC voltage as a function of time assuming
circuit in Fig. 7.25a in terms of υ1 for an initially uncharged capacitor
arbitrary circuit parameters.
B. Present its particular form when C 1 ¼ C 2 t=0
¼ C and R1 ¼ 2R2 ¼ R. Express all
coefficients in terms of τ0 ¼ RC. R1
+
C. Given that the solution for the homoge- VS + R2 C vC
-
neous ODE has the form expðαt=τ0 Þ, -
determine two possible solutions for the
dimensionless coefficient α.
D. Using software of your choice, generate Problem 7.59. A transient circuit with the DC
the circuit solution over the time interval voltage source VS is shown in the following
from 0 to 15τ0. Use R ¼ 1 kΩ, figure. Given that V S ¼ 10 V and
C ¼ 1 μF, and V S ¼ 10 V. Plot two R1 ¼ R2 ¼ 1 kΩ, R3 ¼ 2 kΩ, and C ¼ 1 μF:
capacitor voltages and the circuit current A. Present the ODE for the capacitor volt-
as functions of time to scale. age υC(t).
B. Determine the value of the time constant
t=0 R1 C1 τ and the ODE right-hand side (the forc-
i1 i2 ing function).
i + v1 - + C. Present the solution for the capacitor
VS + R2 C 2 v2 voltage as a function of time assuming
- - an initially uncharged capacitor.

VII-390
Chapter 7 Problems

t=0 Problem 7.62. A transient circuit with the DC

voltage source VS is shown in the figure below.
R1
R3 Given that V S ¼ 10 V and R1 ¼ R2 ¼ 1 kΩ,
VS + R2 + L ¼ 1 mH:
- C vC
- A. Present the ODE for the inductor current
iL(t).
B. Determine the value of the time constant
Problem 7.60. A transient circuit with the τ and the ODE right-hand side (the forc-
DC voltage source VS is shown in the ing function).
following figure. Given that V S ¼ 10 V and C. Present the solution for the inductor cur-
R1 ¼ R2 ¼ 1 kΩ, R3 ¼ R4 ¼ 2 kΩ, and C ¼ 1 μF: rent as a function of time assuming the
A. Present the ODE for the capacitor volt- initial current equal to zero.
age υC(t).
B. Determine the value of the time constant t=0
τ and the ODE right-hand side (the forc-
ing function). R1
iL
C. Present the solution for the capacitor VS + R2 L
-
voltage as a function of time assuming
an initially uncharged capacitor.

t=0 Problem 7.63. Consider the circuits in two

previous problems at arbitrary values of R1, R2,
R1
R4 L, IS, VS. What should be the relation between
VS + R2 + these parameters to guarantee the same solution
- C vC
R3
- for the inductor current in every case?

Problem 7.64. Describe the mathematical

Problem 7.61. A transient circuit with the cur- meaning of
rent source IS is shown in the following figure. A. Natural response
Given that I S ¼ 10 mA and R1 ¼ R2 ¼ 1 kΩ B. Forced response
and L ¼ 1 mH: for a first-order transient circuit in your own
A. Present the ODE for the inductor current words. Do you think that this concept can be
iL(t). applied to any transient circuit?
B. Determine the value of the time constant
τ and the ODE right-hand side (the forc- Problem 7.65. If a transient circuit uses a DC
ing function). supply (either voltage or current) and a switch,
C. Present the solution for the inductor cur- what is a general form of the forcing function?
rent as a function of time assuming the
Problem 7.66. If the forcing function of a first-
initial current equal to zero.
order transient circuit is a combination of sine/
L
cosine function and a constant, what is the
general form of the forced response?
t=0
IS R1 R2
Problem 7.67. In the transient circuit shown in
the figure below, υS ðt Þ ¼ V S þ V m sin ωt.
A. Write the solution for the voltage across
the load resistor RL in terms of the circuit

VII-391
Chapter 7 Transient Circuit Fundamentals

parameters assuming an initially 7.5 Description

uncharged capacitor.
B. Write the solution for the voltage across of the Second-Order Transi-
the load resistor RL when the bypass ent Circuits
capacitor is absent.
7.5.1 First-order Transient Circuits Ver-
t=0 sus Second-order Transient Circuits
Problem 7.72. A transient circuit is shown in
RS
+ the following figure. Is it a first- or second-
vS(t) + C RL vL order transient circuit? Justify your answer.
- -
t=0 R1 C1

Problem 7.68. Plot to scale the load voltage for

the circuit shown in Fig. 7.27a with and without VS + R2
-
the bypass capacitor over time interval from 0 to C2
50 ms. The circuit parameters are RS ¼ 5 Ω,
RL ¼ 1 kΩ, C ¼ 500 μF. The source given
by Eq. (7.50) is the superposition of the DC Problem 7.73. A transient circuit is shown in
and AC components. The source parameters are the figure below. Is it a first- or second-order
V S ¼ 10 V, V m ¼ 1V, f ¼ 250 Hz. transient circuit? Justify your answer.

Problem 7.69. What is the asymptotic form of R1

the solution given by Eqs. (7.57)–(7.58) when

the source resistance, RS, tends to zero? t=0
IS L1 R2 L2

Problem 7.70. In the transient circuit shown in

the figure below, assume iS ðtÞ ¼ I S þ I m sin ωt.
A. Write the solution for the current through
the load resistor RL in terms of the circuit Problem 7.74. Establish the ODE for the tran-
parameters assuming that the initial sient circuit shown in the figure below. Both
inductor current is equal to zero. capacitors have zero voltage prior to closing the
B. Write the solution for the current through switch.
the load resistor RL when the decoupling A. Assume R1 ¼ R2 ¼ R, C 1 ¼ C 2 ¼ C.
inductor is absent B. Assume arbitrary values of R1,2, C1,2.

L t=0

R1
iS(t) t=0
RS RL VS + R2 C2
-
C1

Problem 7.71. In the circuit for the previous Problem 7.75. Establish the ODE for the tran-
problem, another resistor R0 is present in par- sient circuit shown in the figure below. Inductor
allel with the inductance L. Determine the nat- currents are zero prior opening the switch.
ural response of the circuit and find the time A. Assume R1 ¼ R2 ¼ R, L1 ¼ L2 ¼ L.
constant, τ. B. Assume arbitrary values of R1,2, L1,2.

VII-392
Chapter 7 Problems

R1 L1 similar to Eq. (7.63) of this section. Present

your result in terms of damping coefficient α
t=0
and undamped resonant frequency ω0.
IS R2 L2

R
+ -
vR
i(t)
+
Problem 7.76. In the previous problem, add vS(t) + L vL
- -
resistance R3 as shown in the figure that follows i(t) C

and solve task B.

- vC +

R1 L1

t=0 7.5.3 Choice of Independent Function:

IS R2 L2
Initial Conditions
R3 7.5.4 Step Response
Problem 7.81. The RLC circuit shown in
Fig. 7.30 is described by dynamic equation
(7.63) written in terms of the electric current.
7.5.2 Series Connected Second-order
How do the initial conditions to this equation
RLC Circuit change if the capacitor was charged to VS/2
Problem 7.77. Describe in your own words the prior to t ¼ 0?
mechanical counterpart of the series RLC
circuit. Problem 7.82. For the circuit shown in
Fig. 7.30:
Problem 7.78. An RLC circuit in Fig. 7.30 has A. Derive the dynamic circuit equation
R ¼ 1 kΩ, C ¼ 1 μF, L ¼ 1 mH. (7.67) in terms of the capacitor voltage
A. Find the value of the damping coeffi- υC(t).
cient, α (show units). B. How does this equation change if
B. Find the value of undamped resonant the capacitor was charged to VS/2 prior
frequency, ω0. to t ¼ 0?
Problem 7.79. How does the second-order Problem 7.83. For the series RLC circuit with
ODE Eq. (7.63) for the circuit current in the the switch and the DC supply shown in
series RLC circuit from Fig. 7.30 change if the Fig. 7.30, we know the solution, υC(t), of
capacitor was charged to VS/2 prior to t ¼ 0? Eq. (7.67) or Eq. (7.69) for V S ¼ 1 V. The
circuit shown in the following figure is now
Problem 7.80. In the circuit shown in Fig. 7.30,
considered, with the voltage source in the
the switch is replaced by a short circuit. The
form of a voltage pulse (one bit) centered
constant voltage source is replaced by an arbi-
about t ¼ 0. Express the solution to the present
trary time-varying voltage source V S ! υS ðt Þ
problem in terms of υC(t).
as shown in the figure that follows. Derive the
dynamic circuit equation for the circuit current

VII-393
Chapter 7 Transient Circuit Fundamentals

Problem 7.87. Describe the duality between

R series and parallel RLC circuits in your own
vS(t) + L words.
-
C
Problem 7.88. For the parallel RLC circuit
- vC + with the switch and the DC current source
vS(t) shown in Fig. 7.34, we know the solution,
10V iL(t), of Eq. (7.75) for I S ¼ 1 mA. The circuit
shown in the following figure is now consid-
ered, with the current source in the form of a
pulse (one bit) centered about t ¼ 0. Express
-T/2 0 T/2 t the solution to in terms of iL(t).

Problem 7.84. Repeat the previous problem for iL

R L C
the voltage source shown in the figure that
iS(t)
follows (a voltage pulse train of two bits).

vS(t)
iS(t)
5V
10mA

T T T t
0
-T/2 0 T/2 t
Problem 7.85. Repeat Problem 7.83 for the
voltage source shown in the figure that follows
(a bipolar voltage pulse). 7.6. Step Response of the
Series RLC Circuit
vS(t)
10V 7.6.1 General Solution of the
Second-order ODE
t 7.6.2 Derivation of Complementary
-T/2 T/2 Solution
Problem 7.89.
-10V A. The complete solution to a second-order
homogeneous ODE is a sum of two dis-
tinct components. Describe each
7.5.5 Parallel Connected Second-Order of them.
RLC Circuit B. Write three forms of the complementary
Problem 7.86. For the circuit shown in solution (natural response) for the second-
Fig. 7.34: order homogeneous ODE.
A. Derive the dynamic circuit equation C. What is a new parameter to be intro-
(7.75) written in terms of the inductor duced for the underdamped circuit?
current iL(t).
Problem 7.90. For the series RLC circuit
B. How does this equation change if
shown in the following figure:
the switch in Fig. 7.34 was open prior
A. Find the value of the damping coeffi-
to t ¼ 0 and closes at t ¼ 0?
cient, α (show units).
C. How do the initial conditions change in
this case?
VII-394
Chapter 7 Problems

B. Find the value of the undamped resonant Given fixed L and C, which values of resistance
frequency, ω0 (show units). (large or small) lead to the overdamped circuit?
C. Find the value of the damping ratio, ζ
(show units). Problem 7.93. For the series RLC circuit
D. Find the particular solution (forced shown in the figure below, fill out the table of
response). circuit parameters.
E. Outline the form of the complementary
t=0
solution (natural response). 100
F. Which value should the circuit resistance
have for a critically damped circuit?
10V=VS + 1 mH
-
t=0 300 C

5V=VS + 1 mH
- Table of circuit parameters
1 nF
C, ζ Circuit type (overdamped, criti-
μF cally damped, underdamped)
Problem 7.91. Repeat the previous problem for 0.01
the series RLC circuit shown in the figure 0.1
below. 0.4
t=0
1.0
300

Given fixed L and R, which values of capaci-

+ tance (large or small) lead to the overdamped
5V=VS 10 mH
- circuit?
1 F

Problem 7.94. For the series RLC circuit

Problem 7.92. For the series RLC circuit shown in the following figure, fill out the
shown in the following figure, fill out the table of circuit parameters.
table of circuit parameters.
t=0
100
t=0
R

10V=VS + L
-
10V=VS + 1 mH 1 nF
-
1 F

Table of circuit parameters

Table of circuit parameters
L, ζ Circuit type (overdamped, criti-
R, ζ Circuit type (overdamped, critically μH cally damped, underdamped)
Ω damped, underdamped)
0.1
25
1
50
2.5
75
10
100

VII-395
Chapter 7 Transient Circuit Fundamentals

Given fixed R and C, which values of induc- Problem 7.98. Repeat the previous problem for
tance (large or small) lead to the overdamped the circuit shown in the following figure.
circuit?
t=0 1 mH
Problem 7.95. Show that underdamped solu-
tion and critically damped solutions coincide
with each other when ς ! 1. 10V=VS + 6
-
100 F
7.6.3 Finding Integration Constants
7.6.4 Solution Behavior for Different
Damping Ratios Problem 7.99. In the circuit shown in the figure
Problem 7.96. For the circuit shown in the below, the capacitor was charged to 10 V prior
figure below: to closing the switch.
t=0 60
t=0 6

5V=VS + 1 H
-
1 nF 1 mH

100 F

A. Determine damping coefficient α, - vC +

undamped resonant frequency ω0, and
damping ratio ζ. A. How are the circuit equation and initial
B. Determine constants K1, K2. conditions different from Eqs. (7.78a, b)?
C. Write solution for the capacitor voltage B. Determine damping coefficient α,
with all constants defined. undamped resonant frequency ω0, and
D. Calculate and plot to scale capacitor damping ratio ζ.
voltage at 0, 0.05, 0.1, 0.2, and 0.3 μs. C. Determine constants K1, K2.
D. Write the solution for capacitor voltage
Problem 7.97. For the circuit shown in the with all constants defined.
following figure: E. Calculate the capacitor voltage at 0, 1,
A. Determine damping coefficient α, 2, 3, 4, and 5 ms and plot it to scale
undamped resonant frequency ω0, and versus time.
damping ratio ζ.
B. Determine constants K1, K2. 7.6.5 Overshoot and Rise Time
C. Write the solution for the capacitor volt- 7.6.6 Application: Non-ideal Digital
age with all constants defined. Waveform
D. Calculate capacitor voltage at 0, 1, 2, 3, Problem 7.100. The following figure shows
4, and 5 ms and plot it to scale the underdamped step response for a series
versus time. RLC circuit. The DC source has the voltage
of 10 V. Using the figure:
t=0 100 F A. Estimate the overshoot percentage.
B. Estimate the rise time.
C. Do these estimates (approximately)
1V=VS + 10 agree with Eqs. (7.88a, b)?
-
1 mH

VII-396
Chapter 7 Problems

Capacitor voltage, V Problem 7.103. For the circuit shown in the

16
following figure:
14
12
a)
10 R
8
vS(t) + L
6 -
4 C
2
0
vS(t)
- vC +
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 t, ms
b)
VS
Problem 7.101. Capacitor voltage is measured
in a series RLC circuit as shown in the figure to
t
the previous problem. Given R ¼ 2 Ω, estimate
circuit inductance L and circuit capacitance C. 0 T
A. Determine the step response υC(t)
Problem 7.102. The figure that follows shows for the circuit shown in figure (a) given
the distorted rectangular waveform (capacitor that L ¼ 1 μH, C ¼ 1 nF; V S ¼ 10 V,
voltage) for the circuit shown in Fig. 7.33. The and R ¼ 75 Ω.
DC source has the voltage of 10 V. B. Express the solution υpulse
C (t) for the volt-
A. Using the figure, estimate the overshoot age pulse shown in figure (b) in terms of
and undershoot percentages. the step response.
B. Using the figure, estimate the rise time C. Given T ¼ 0:5 μs, calculate the solution
and the fall time. for the voltage pulse over the time inter-
C. Do these estimates (approximately) val from 0 to 0.7 μs in steps of 0.1 μs and
agree with Eqs. (7.88a, 7.88b)? plot it to scale.
Capacitor voltage, V
15 Problem 7.104. Repeat the previous problem
assuming T ¼ 0:2 μs. Calculate the solution
10
for the voltage pulse over the time interval
5 from 0 to 0.5 μs in steps of 0.05 μs and plot it
to scale versus time.
0

-5

-10
0 1 2 3 4 5 6 7 8 9 t, s

VII-397
 Part III
AC Circuits
Chapter 8

Chapter 8: Steady-State AC Circuit

Fundamentals

Overview
Prerequisites:
- Knowledge of DC circuit analysis (Chapters 2, 3, and 4)
- Knowledge of dynamic circuit elements (Chapter 6, optionally Chapter 7)
- Knowledge of complex arithmetic and calculus

Objectives of Section 8.1:

- Apply and work with the major parameters of the steady-state AC signals:
amplitude, frequency, and phase
- Establish the concept of phase leading or lagging for AC voltages and currents
- Become familiar with the major function of the oscilloscope—measure periodic
(AC) voltages in a circuit
- Understand the meaning of the phasor as a representation of the real signal
- Convert real signals to phasors and vice versa
- Perform basic operations with phasors
- Be able to construct the phasor diagram for real signals and restore the real signals
from the phasor diagram
- Become familiar with the phasor (angle) notation

Objectives of Section 8.2:

- Provide the complete mathematical derivation of complex impedances
- Apply the impedance concept to resistor, capacitor, and inductor
- Understand the meaning of magnitude and phase of the complex impedance

Objectives of Section 8.3:

- Understand and apply the AC circuit analysis with phasors and impedances
- Appreciate the value of the phasor diagram as a tool for AC circuit analysis
- Transfer major circuit theorems to steady-state AC circuits
- Be able to solve multifrequency AC circuits using superposition principle

Application examples:
- Measurements of amplitude, frequency, and phase
- Impedance of a human body

© Springer Nature Switzerland AG 2019 VIII-401

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_8
Chapter 8 Steady-State AC Circuit Fundamentals

Keywords:
Steady-state AC circuit, Signals, Harmonic, Amplitude, Pk-Pk value, Angular frequency, Phase,
Period, Steady-state AC voltage, Steady-state alternating current, Leading signal, Lagging signal,
Oscilloscope, Phasor, Phasor voltage, Phasor current, Phasor diagram, Phasor notation, Angle
notation, Complex impedance, Impedance of the resistor, Impedance of the capacitor, Impedance
of the inductor, Reactance, KVL in phasor form, KCL in phasor form, Source transformation in
the frequency domain, Thévenin’s theorem for steady-state AC circuits, Norton theorem for
steady-state AC circuits, Superposition principle for multifrequency AC circuits

VIII-402
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Section 8.1 Harmonic Voltage and Current: Phasor

Whether you use a kitchen appliance or operate a piece of industrial machinery, they are
powered by alternating current (AC) power supplies. It may seem odd at first to create
power sources with alternating polarities that result in fluctuating voltage and current
directions through resistors, inductors, and capacitors. Consider the following situation:
current flows through a coil from left to right with increasing and then decreasing
magnitude, then changing direction, and again increasing and decreasing in magnitude
before the process repeats itself. A benefit of using periodic, sinusoidal waveforms is
related to the relative ease of stepping up or down AC voltages and currents with little,
ideally no, power losses via a transformer. The economic importance of AC circuits can
hardly be overstated; the bulk of the residential and industrial power demand is produced,
transformed, and distributed via AC circuits. To understand the basics of AC circuit
analysis, we start with the key characteristics of sinusoidal waveforms such as frequency,
phase, and amplitude. Special attention is paid to the phase. We define the meaning of
leading and lagging phase for steady-state AC voltages or currents. We note that in
electrical engineering, AC voltages are often called signals.

8.1.1 Harmonic Voltages and Currents

In steady-state AC circuits, all voltages and currents measured across or through the
elements are periodic and in the ideal case harmonic (i.e., sine or cosine) functions of
time. These voltages and currents have the same frequency but different phases and
amplitudes. Interestingly, the word harmonic originates from the reference to music
sounds of pure, single tones, pitches, or frequencies. The word steady-state means that
the circuit frequency, phases of all voltages and currents, and amplitudes of all voltages
and currents do not change over time. Transient effects are entirely excluded from our
consideration. Similarly, the AC voltage and currents are called steady-state AC voltage
and steady-state alternating current. It is common to use terms AC voltage and AC
power, but the term AC current does not make much sense, even though it might be a part
of the electrical engineering jargon. Figure 8.1 shows an AC harmonic signal. All
voltages and currents in an AC circuit will have exactly this form, regardless of whether
they are measured over a resistor, capacitor, or inductor. For this figure we consider the
voltage υ(t). The current i(t) could be treated in an identical manner. As a harmonic
function, the steady-state AC voltage can be written in the form

υðt Þ ¼ V m cos ðωt þ φÞ ð8:1aÞ

where
Vm is the voltage amplitude (maximum absolute voltage), with the unit of volts.
ω is the angular frequency, with the unit of rad/s.
φ is the phase, with the unit of radians.

VIII-403
Chapter 8 Steady-State AC Circuit Fundamentals

The use of a cosine function rather than a sine function, as a basis for any AC signal in
Eq. (8.1a), is common. The angular frequency relates to the frequency f and period T by

ω ¼ 2πf , T ¼ 1=f ð8:1bÞ

where f is measured in hertz or Hz (1 Hz ¼ 1 s1) and the period is recorded in seconds.

As the name implies, the frequency f defines the number of cycles (positive-to-negative
AC voltage transitions) per second. For example, the frequency of the AC wall plug in the
USA is 60 cycles per second and is recorded as
1
f ¼ 60 ¼ 60 Hz ð8:1cÞ
s
The angular frequency ω is essentially a replica of the frequency f; its use is primarily a
matter of convenience. Virtually all frequencies in electrical engineering are measured in
terms of f and not in terms of angular frequency ω. For example, we say that the AC
frequency in the USA is 60 Hz, or 50 Hz in Europe and other countries and not 377 rad/s
or 314 rad/s. The frequency unit hertz honors Heinrich Hertz (1857–1894), a German
scientist, who, at the age of 29, built the first radio-frequency transmitter based on a spark
gap circuit and a receiver (a loop antenna). He thus confirmed Maxwell’s theory of
electromagnetic wave propagation.

1 phase zero
0.8

0.4
amplitude
voltage, V

0 Pk-Pk

amplitude
-0.4

-0.8 period
-1
0 0.5 1 1.5 2 2.5 3
time, ms

Fig. 8.1. Harmonic voltage signal of Eq. (8.1a) and its parameters. The phase is zero.

Example 8.1: Determine the frequency in Hz, the angular frequency in rad/s, and the
amplitude of the harmonic voltage signal shown in Fig. 8.1.
Solution: The amplitude is the simplest parameter to find. By inspection, the maximum
voltage value is 0.6 V. It is exactly the amplitude value, V m ¼ 0:6 V, and therefore is the
absolute maximum voltage. The period is determined as the interval between two similar
zero

VIII-404
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Example 8.1 (cont.): crossings in Fig. 8.1; it is measured as one millisecond. Another
(sometimes more accurate) way to measure the period is to observe that there are exactly
three periods in the figure, recorded over the time interval from 0 to 3 ms. Dividing the time
interval over the number of periods, we obtain the duration of one period. The
frequency of the voltage is f ¼ 1=T ¼ 1000 Hz ¼ 1 kHz. The angular frequency results
in ω ¼ 2πf ¼ 6283:1 rad=s, and the voltage in Fig. 8.1 is υðtÞ ¼ 0:6 cos ð2π1000t Þ ½V.
In real circuits, you should avoid measuring the period by using two consecutive
maxima or minima of the sinusoidal signal. Although the same result is obtained on
paper, in reality the flat maximum plateau is frequently corrupted by noise. As a
result, rather inaccurate maxima positions are acquired.

Note that the amplitude is well defined for a pure harmonic in Fig. 8.1. It is symmetric
about the axis, i.e., does not have a DC offset. But what about other periodic signals that
have a significant DC offset? How can we define their amplitudes? The key is the peak-to-
peak (or in short: Pk-Pk) value shown in Fig. 8.1. For a harmonic AC voltage and current,
the peak-to-peak value is simply twice the amplitude value. For non-sinusoidal periodic
signals or sinusoidal signals with a DC offset, the peak-to-peak value is the only mean-
ingful measure of the alternating signal strength. Therefore, an oscilloscope always uses
the peak-to-peak value, instead of the amplitude.

8.1.2 Phase: Leading and Lagging

Perhaps the most confusing constant in Eq. (8.1a) is the phase φ. A nonzero phase means
that the cosine function is shifted with respect to a zero-phase cosine in Fig. 8.1 either to the
left or to the right. Let us ask the following question: if the phase in Eq. (8.1a) is positive, say
φ ¼ þπ=2, is the cosine function shifted to the left or to the right versus the base cosine
signal on the time axis? The answer might be somewhat unexpected: the positive phase
means a shift to the left. Why? Because the given value of the cosine function occurs earlier
in time than without the phase. Figure 8.2a shows a phase-shifted voltage signal υðt Þ ¼ 0:6
cos ð2π1000t þ π=2Þ ½V; see the dashed curve; the base signal of zero phase is shown by
the solid curve. At t ¼ 0, the argument is already ahead by þπ=2 and the cosine function is
zero, whereas with zero phase the zero occurs later at t ¼ 0:5T . A shift to the left means that
all events (e.g., the peaks or zero crossings) happen earlier in time than for the base cosine
with the phase zero. Therefore, the cosine with the positive phase always leads the cosine
with the zero phase. In the present case, it leads the base cosine by 90 . In contrast, the cosine
with the negative phase is shifted to the right. Figure 8.2b depicts the example of φ ¼ π=2.
A shift to the right means that all events (e.g., the peaks or zero crossings) happen later in
time than for the base cosine with zero phase. The voltage with the negative phase lags the
cosine with the zero phase (in fact, it lags by 90 or π/2 in Fig. 8.2b). The concept of leading
and lagging plays a prominent role in power electronics.

VIII-405
Chapter 8 Steady-State AC Circuit Fundamentals

Exercise 8.1: Determine frequency in Hz, angular frequency in rad/s, and the amplitude of
the harmonic voltage signals shown in Fig. 8.2.
Answer: All four signals in Fig. 8.2 have the same amplitude of 0.6 V, the same frequency
of 1 kHz, and the same radian frequency of ω ¼ 2πf ¼ 6283:1 rad=s.

a)
0.8
phase =+ /2
dashed voltage curve leads
solid voltage curve

0.4
voltage, V

-0.4

-0.8

b)
phase =- /2
0.8 dashed voltage curve lags
solid voltage curve

0.4
voltage, V

-0.4

-0.8

0 0.5 1 1.5 2 2.5 3

time, ms

Fig. 8.2. Phase-shifted AC voltages υðtÞ ¼ 0:6 cos ð2π1000t  π=2Þ ½V versus the zero-phase
voltage υðtÞ ¼ 0:6 cos ð2π1000t Þ ½V shown by a solid curve. In Fig. 8.2a, the dashed voltage
leads the base signal, whereas in Fig. 8.2b it lags the base signal.

The most useful facts related to phase measurements are:

1. The phase in electrical engineering applications ranges from π to +π radians; this
corresponds in degrees to 180 to +180 . The phase in degrees should be divided
by 180 and multiplied by π to obtain the phase in radians. The phase in radians
should be divided by π and then multiplied by 180 to obtain the phase in degrees.
2. The phase is a relative measure, with reference to a base signal. If the base signal is
not present, the phase cannot be defined uniquely. By default the base signal is the
cosine signal of the same frequency, that is, cos(ω t).
3. If a measured phase for some reason exceeds π, we need to subtract its value from
2π. If it is less than π, then we need to add 2π to its value. This will ensure that the
phase will stay in the range from π to +π radians.

VIII-406
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Example 8.2: Determine the frequency, amplitude, and phase of the harmonic voltage
signal shown in Fig. 8.3 versus the base cosine signal.
Solution: The amplitude is determined first: by inspection it is clear that V m ¼ 1:0V. Next
the frequency is determined: the entire interval from 0 to 3 ms contains three full periods;
hence T ¼ 1 ms, and f ¼ 1=T ¼ 1000 Hz ¼ 1 kHz. For the phase determination, we note
that the first maximum in Fig. 8.3 occurs later in time than for the base cosine, which already
peaks at t ¼ 0. Therefore, the phase must be negative, that is, φ < 0. The absolute value of the
phase in Fig. 8.3 is

ΔT
jφj ¼ 2π ð8:2Þ
T
which gives φ ¼ π=3 after measuring T and ΔT with a scale. A smarter way to obtain the
same result is to note that the cosine function is equal to 0.5Vm at t ¼ 0, so that
φ ¼  cos 1 ð0:5Þ ¼ π=3. And an even “smarter” way is to use a calculator to plot the
cosine function with all possible phases until it matches Fig. 8.3. This method only works
well during an exam if enough time is available.

1 T=T/6 = 2 * T/T= /3
0.8
0.6
0.4
0.2
voltage, V

0
-0.2
T T
-0.4
-0.6
-0.8
-1
0 0.5 1 1.5 2 2.5 3
time, ms

Fig. 8.3. A phase-shifted AC voltage υðtÞ ¼ 1 cos ð2π1000t  π=3Þ ½V.

Example 8.3: Assume that an AC signal is given by υðt Þ ¼ V m sin ðωt þ ψ Þ ½V. How do
we convert it to a cosine function?
Solution: The single most important trigonometric identity worth remembering is proba-
bly sin α ¼ cos ðα  π=2Þ. If you forget this identity, you may recall a visual picture: at
t ¼ 0 the sine with zero phase lags the cosine with zero phase by π/2 or by 90 . The result is
then υðt Þ ¼ V m cos ðωt þ ψ  π=2Þ ½V.

VIII-407
Chapter 8 Steady-State AC Circuit Fundamentals

8.1.3 Application Example: Measurements of Amplitude, Frequency,

and Phase
Different parameters of the steady-state AC voltage or of a general periodic voltage
studied in this section are readily measured with the oscilloscope. The oscilloscope always
measures the time-amplitude response of the voltage, not the current. Parameters include
frequency, amplitude (or peak-to-peak value), the rms voltage, the mean voltage (voltage
averaged over a period), etc. Figure 8.4 provides a practical measurement example.
Several voltage signals can be measured simultaneously with several oscilloscope chan-
nels. The phase is not measured for a single signal, only the phase difference between two
and more signals can be measured. The oscilloscope has two commonly used settings. In
the first setting called DC coupled, any voltage supplied to the channel will cause a
deflection of the trace. As a result, the actual value of the input voltage with respect to
oscilloscope ground can be measured. In the second setting called AC Coupled, any DC
offset to the periodic signal is eliminated via an internal coupling capacitor. The result,
after a short transient period, is that the trace will settle at 0 V regardless of the magnitude
of an applied DC voltage. This configuration is used whenever it is desired to ignore the
constant DC component of an applied voltage waveform and only observe the AC
component of the waveform with zero mean.

Fig. 8.4. Front panel of an inexpensive digital-storage oscilloscope. Note the measured signal
parameters.

8.1.4 Definition of a Phasor

This topic is critical for the steady-state AC circuit analysis. We are about to introduce the
method of solving AC circuits based on the use of so-called phasors or complex numbers.
Working with phasors allows us to “cancel” out the frequency dependence and the time
dependence. This is possible because in a linear system the harmonic behavior is the same
for all circuit components, we only need to keep the amplitude and phase information for
every sinusoid. The application of phasors will eventually allow us to reduce the AC
circuit to an equivalent “DC” circuit that is solved using standard tools. However, there is
no “free lunch”; the voltages and currents in the resulting “DC” circuit appear to be

VIII-408
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

complex numbers. They carry information of two parameters: the phase and the ampli-
tude of a sinusoid. The use of a complex number relies on two independent parameters,
real and imaginary parts (or magnitude and phase); it is ideally suited to represent AC
signals while entirely eliminating time-domain harmonics. The starting point of the
phasor concept is rooted in Euler’s formula in the form

e jα ¼ cos α þ j sin α ð8:3aÞ

The identity expresses the complex exponent in terms of the real-valued cosine and sine
pffiffiffiffiffiffiffi
functions. Here, j ¼ 1 is the imaginary unit and α is an arbitrary, real number. Note
pffiffiffiffiffiffiffi
that α can be equal to ωt or to ωt þ φ. In electrical engineering, the symbol j ¼ 1 is
preferred over the mathematical symbol i which may be confused with the electric
current. In terms of real and imaginary parts, we obtain
   
cos α ¼ Re e jα , sin α ¼ Im e jα ð8:3bÞ

Why do we need the complex exponent instead of cosine and sine? To answer this
question, let us study the following identity:

e jðωtþφÞ ¼ e jωt  e jφ ð8:3cÞ

We will write the current or the voltage in the form of Eq. (8.3c) and use the multiplicative
property of the exponent. Then, the factor ejωt can be exactly canceled out in every term in
the underlying differential equation. This is a major simplification, because the ODE
becomes an algebraic equation. The key is the function Re() or the real part of a complex
number:
 
υðt Þ ¼ V m cos ðωt þ φÞ ¼ Re V m e jðωtþφÞ
  ð8:4Þ
iðt Þ ¼ I m cos ðωt þ ψ Þ ¼ Re I m e jðωtþψ Þ

The remaining complex number available after cancellation of the time factor ejωt is
called a phasor: the phasor voltage V and the phasor current I:

V ¼ V m e jφ , I ¼ I m e jψ ð8:5Þ

Equation (8.5) as a definition tells us that the phasor is a complex number comprised of
two parameters: amplitude and phase. For related operations with complex numbers, you
can consult the chapter summary. You should notice that frequency is no longer present
since it remains the same for all circuit elements, and it is equal to the known frequency of
the voltage power supply. The phasor has the same units as the original quantity: the
phasor voltage has units of volts, and the phasor current has units of amperes. The theory

VIII-409
Chapter 8 Steady-State AC Circuit Fundamentals

and application of the symbolic method of alternating currents by means of complex

number algebra was started along with the famous book of C. P. Steinmetz, Theory and
Calculation of Alternating Current Phenomena, New York, McGraw Hill Publishing Co.,
1897. It happened only four years after the 1893 World’s Fair, the World’s Columbian
Exposition in Chicago, where Nikola Tesla and George Westinghouse introduced visitors
to AC power by using it to illuminate the exposition. Further, those methods received
wide attention in control theory, in communications and signal processing, and in RF
engineering.

8.1.5 From Real Signals to Phasors

Every AC voltage or current has its own phasor. The conversion from real-valued
voltages and currents to phasors is performed exactly according to the phasor definition
Eqs. (8.4) and (8.5).

Example 8.4: Determine the phasors for the real-valued AC voltages and currents:

υðt Þ ¼ 3 cos ðωt  π=3Þ ½V

ð8:6aÞ
iðtÞ ¼ 1 cos ðωt þ π=6Þ ½A

Solution: To construct the phasor, we only need the amplitude (Vm or Im) and the
phase (φ or ψ) for every signal in Eq. (8.6a). The result then uses the phasor
definition given by Eqs. (8.4) and (8.5):
 pffiffiffi
jπ=3 π π 3 3 3
V ¼ 3e ¼ 3 cos  j sin ¼ j ½V 
3 3 2 2
pffiffiffi ð8:6bÞ
 π π 3 1
I ¼ 1e jπ=6 ¼ 1 cos þ j sin ¼ þ j ½A
6 6 2 2
All results are a direct consequence of using Euler’s formula. The phasors are complex
numbers that can be plotted in the complex plane as dots or vectors; we will provide an
example of this later in this section. Moreover, we can add, subtract, multiply, and divide
phasors, and the results are again phasors. However, the addition and subtraction of real
signals is difficult, and the multiplication and division is practically impossible. Just
imagine how you would divide a sine by a cosine and then convert the result into a
cosine form.

To assure the uniqueness of phasor definition, the real signal must always be written in
the standard form of the cosine function with the positive amplitude. For example, voltage
υðt Þ ¼ 3 cos ðωt Þ is to be converted to υðt Þ ¼ 3 cos ðω t þ π Þ and then to the phasor
V ¼ 3e jπ .

VIII-410
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Exercise 8.2: Determine the phasor for the real-valued alternating current
iðt Þ ¼ 1 sin ðωt  π=2Þ ½A.
Answer: I ¼ 1e j0 ¼ 1 ½A.

8.1.6 From Phasors to Real Signals

If the phasors of the voltage and current are given, and the angular frequency of the AC
source is known, the real-valued AC voltages and currents can be restored using the
phasor definition given by Eqs. (8.4) and (8.5).

Example 8.5: The phasors of the AC voltage and current are given by

V ¼ 12∠π=3 ¼ 12e jπ=3 ½V


ð8:7aÞ
I ¼ 0:1∠60 ¼ 0:1e j60 ½A

An AC source has the angular frequency ω. Restore the corresponding real-valued voltages
and currents.
Solution: We construct the real signals in the form of Eq. (8.4) where the amplitude
(Vm or Im) and the phase (φ or ψ) of the corresponding sinusoidal function are extracted
from the phasors in Eqs. (8.7a):

υðt Þ ¼ 12 cos ðωt þ π=3Þ ½V

ð8:7bÞ
iðtÞ ¼ 0:1 cos ðωt þ π=3Þ ½A

Exercise 8.3: The phasor voltage is given by V ¼ 2∠π=3 ½V. Restore the
corresponding real-valued voltage signal.
Answer: υðtÞ ¼ 2 cos ðωt  2π=3Þ ½V:

8.1.7 Polar and Rectangular Forms: Phasor Magnitude

The phasor as a complex number has two forms: the polar form and the rectangular form.
Equation (8.8a) includes both forms:

V ¼ V m e jφ ¼ V m cos φ þ jV m sin φ ¼ x þ jy
|fflfflffl{zfflfflffl} |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl} |fflffl{zfflffl} ð8:8aÞ
polar form rectangular form rectangular form

In general, the conversion from polar form to rectangular form uses Euler’s identity. We
emphasize that V m > 0 is the magnitude of a complex number—the phasor. It is always

VIII-411
Chapter 8 Steady-State AC Circuit Fundamentals

positive and equals the amplitude of the corresponding real signal. As mentioned before,
φ is the phase; it changes from π to þπ and equals the phase of the corresponding real
signal. The phasor may be initially given in rectangular form, say as V ¼ x þ jy. In that
case, we may convert the phasor to polar form as follows:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi y 
V ¼ V m e jφ , V m ¼ x2 þ y2 , φ ¼ tan 1 ð8:8bÞ
x
Equation (8.8b) is only valid when x is positive. Otherwise, a factor of π must be added
as explained in Fig. 8.5. The conversion to the polar form and vice versa is routinely done
using a calculator, or MATLAB (function angle), or other software of your choice. For
other operations with complex numbers, please refer to Tables 1 and 2 at the end of this
section. Figure 8.5 plots a complex number (phasor) as given by Eqs. (8.8a,b) in the
complex plane. The x-axis of the complex plane is the real part of the complex number,
and the y-axis is the imaginary part. The complex number is either represented by a dot
with the coordinates x, y or by a vector drawn from the origin to that dot. The magnitude
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
of the complex number (phasor), jVj ¼ V m ¼ x2 þ y2 , is the length of this vector; the
phase φ is the angle with the positive x-axis. It is straightforward to add two phasors; this
operation corresponds to vector addition in Fig. 8.5.

quadrant II Im quadrant I
arctan(y/x)+ arctan(y/x)
x+jy
y

0 x Re

arctan(y/x)- arctan(y/x)

quadrant III quadrant IV

Fig. 8.5. A phasor in rectangular, or polar, form in the complex plane and the corresponding phase
conversion. The phase of φ is given for different quadrants.

Example 8.6: The phasors of AC voltage and current are given in rectangular form

V ¼ 5 þ j5 ½V 
ð8:9Þ
I ¼ 0:1  j0:2 ½A

VIII-412
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Example 8.6 (cont.): An AC source has the angular frequency ω. Restore the
corresponding real voltages and currents.
Solution: We convert the phasors to the polar form using Eq. (8.8b) and Fig. 8.5:

V ¼ 7:07e j 0:785 ½V

ð8:10Þ
I ¼ 0:224ej2:03 ½A

After that, we restore the real signals exactly following Example 8.5:

υðt Þ ¼ 7:07 cos ðω t þ 0:785Þ ½V

ð8:11Þ
iðtÞ ¼ 0:224 cos ðωt  2:03Þ ½A

The phase in Eqs. (8.11) is given in radians. Another equivalent form of the solution that
should please your engineering professor implies replacing 0.785 by π/4 or 45 and 2.03 by
π/1.54 or 116 .

8.1.8 Operations with Phasors and Phasor Diagram

Phasors allow us to perform a number of basic operations in AC circuits rather quickly.
Let us consider a part of an AC circuit with two arbitrary series circuit elements, shown in
Fig. 8.6. The AC voltages υ1(t) and υ2(t) across each element are known. How can we find
the AC voltage υ(t) for the series combinations?

a v1(t)
- - v2(t) b
+
+

A B
-
+

v(t)?

Fig. 8.6. A series combination of two AC circuit elements. The voltage across the series
combination (voltage between terminals a and b) is desired. You should note the voltage
polarities.

According to KVL we conclude υðt Þ ¼ υ1 ðt Þ  υ2 ðt Þ, where the second voltage has

opposite polarity. There are two ways to proceed from here. The first way in the time
domain is to add two cosine functions directly and convert them into another cosine
function. Unfortunately, if these cosine functions have different phases and amplitudes,
the operation is not straightforward. It requires a search for the corresponding trigonometric
formula and the accurate use of this formula, keeping in mind proper phase definition.

VIII-413
Chapter 8 Steady-State AC Circuit Fundamentals

The second way is to use phasors. We will see that this approach has a clear intuitive
background and is actually simpler to apply.

Example 8.7: Find the total voltage between terminals a and b in Fig. 8.6 if the element
voltages are given by:

υ1 ðtÞ ¼ 6:08 cos ðωt þ 80:5 Þ ½V


ð8:12aÞ
υ2 ðtÞ ¼ 5:00 cos ðωt þ 36:9 Þ ½V

Solution: We construct the phasors first:

V1 ¼ 6:08∠80:5 ½V

ð8:12bÞ
V2 ¼ 5:00∠36:9 ½V

We convert them into rectangular form using Euler’s formula next:

V1 ¼ 1:00 þ j6:00 ½V

ð8:12cÞ
V2 ¼ 4:00 þ j3:00 ½V

Both phasors are plotted in Fig. 8.7a. Such a plot is known as a phasor diagram. It is quite
useful since it provides us with a visual picture of magnitude and phase. We subtract the
phasors as complex numbers and obtain the resulting phasor for the desired voltage υ(t)

υðt Þ ¼ υ1 ðt Þ  υ2 ðtÞ ) V ¼ V1  V2 ¼ 1 þ j6  4  j3 ¼ 3 þ j3 ½V ð8:13aÞ

The same subtraction operation is done in the vector form in the phasor diagram of
Fig. 8.7a. The resulting phasor V is shown by a dashed arrow. We need to center it at the
origin to obtain agreement with Eq. (8.13a). Next, we convert the phasor V into polar form
using either Eq. (8.8b) or by just looking at the phasor diagram itself,
pffiffiffi  
V ¼ 3 þ j3 ¼ 3 2e j135 ¼ 4:23e j135 ¼ 4:23∠135 ½V ð8:13bÞ

Finally, we restore the voltage υ(t) from its phasor: υðt Þ ¼ 4:23 cos ðωt þ 135 Þ.
Using the same method, we could find the terminal voltage for a series combination of any
number of circuit elements with arbitrary polarities.

VIII-414
Chapter 8 Section 8.1: Harmonic Voltage and Current: Phasor

Im V1
5 V
V=V1-V2
V2

Re
phasor diagram a)
0 5

is equivalent to time-domain voltage subtraction b):

8 8 8
Voltage v1(t), V Voltage v2(t), V Voltage v(t), V
4 4 4

0
- =
-4 -4 -4

-8 -8 -8
0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3 0 0.5 1 1.5 2 2.5 3
time, ms time, ms time, ms

Fig. 8.7. (a) Finding series voltage for two circuit elements using phasors and (b) the time-domain
representation of the same result.

The phasor diagram introduced in Fig. 8.7a also allows us to find the relative lag or
lead for two or more AC voltages and/or currents, a topic you may recall from the
previous section. As an example, in Fig. 8.8 we show the phasor diagram for three
different time-domain voltages υ1(t), υ2(t), υ3(t). Irrespective of the specific values of
the amplitudes and phases, we may conclude that voltage υ2(t) leads voltage υ1(t) by 90 ,
but lags voltage υ3(t).

Im
V2

V3

Re
0
V1

Fig. 8.8. Phasor diagram for three time-domain voltages υ1(t), υ2(t), υ3(t). Voltage υ2(t) leads
voltage υ1(t) by 90 , but lags voltage υ3(t).

VIII-415
Chapter 8 Steady-State AC Circuit Fundamentals

Exercise 8.4: Does the voltage signal υðtÞ ¼ 6:08 cos ðωt þ 80:5 Þ lead or lag the voltage
signal υðt Þ ¼ 4:23 cos ðωt þ 135 Þ?
Answer: It lags by 54.5 .

At this point a legitimate question arises: is adding sinusoids the only application of
phasors? What will be the next step? Obviously, our goal now is to reduce an AC circuit
to the equivalent “DC” circuit. This step requires a new concept known as impedance. It
will be considered in the next section.

8.1.9 Shorthand Notation for the Complex Exponent

In electrical engineering and electronics, the shorthand notation (phasor notation or angle
notation) e jφ ¼ ∠φ is commonly used to simplify the notation of the complex exponent.
We will frequently use this shorthand notation in the following sections. All operations
with phasors remain the same. For example, the multiplicative operations are written in
the forms

V 1 e jφ V 2 e jψ ¼ V 1 V 2 e jðφþψ Þ ) V 1 ∠φ V 2 ∠ψ ¼ V 1 V 2 ∠ðφ þ ψ Þ ð8:14Þ

VIII-416
Chapter 8 Section 8.2: Impedance

Section 8.2 Impedance

8.2.1 The Concept of Impedance
We can avoid solving ODEs for the AC circuits entirely if we establish a relation between
the phasor voltage and the phasor current for the inductor and capacitor, similar to Ohm’s
law for the resistor. We consider three basic AC circuit elements shown in Fig. 8.9 in a
passive reference configuration and determine their phasor voltages and phasor currents.

resistance inductance capacitance

v(t) v(t) v(t)
+ - + - + -
i(t) i(t) i(t)

Fig. 8.9. Three AC circuit elements.

Resistance For the resistance, voltage will be given by a cosine function shown in
Fig. 8.10a. One has
Vm
υ ¼ Ri, υðt Þ ¼ V m cos ðωt þ φÞ ) iðt Þ ¼ cos ðωt þ φÞ
R ð8:15Þ
V m jφ
V ¼ V m e jφ I¼ e
R
Capacitance For the capacitance, voltage will again be given by a cosine function shown
in Fig. 8.10b. One has
dυ
i¼C , υðt Þ ¼ V m cos ðωt þ φÞ
dt
) iðt Þ ¼ ωCV m sin ðωt þ φÞ ¼ ωCV m cos ðωt þ φ þ π=2Þ ð8:16Þ
V ¼ V m e jφ I ¼ ωCV m e jφþjπ=2

Inductance For the inductance, current will be given by a cosine function shown in
Fig. 8.10c. One has
di
υ¼L , iðt Þ ¼ I m cos ðωt þ φÞ
dt
) υðt Þ ¼ ωLI m sin ðωt þ φÞ ¼ ωLI m cos ðωt þ φ þ π=2Þ ð8:17Þ
I ¼ I m e jφ V ¼ ωLI m ejφþjπ=2

Impedance Z of an arbitrary linear circuit element or of an arbitrary one-port network

comprised of such elements is defined by
V
Z ð8:18Þ
I

VIII-417
Chapter 8 Steady-State AC Circuit Fundamentals

Thus, the phasors for voltage and current are linked via a constant, which can be real or
complex. In order to emphasize that this constant is not exactly a resistance, the constant
is called the impedance. Substituting Eqs. (8.15), (8.16), and (8.17) into Eq. (8.18), we
obtain:
V
Resistance : ¼ R ½Ω 
ZR 
I
V 1 1 1
Capacitance : ZC  ¼ jπ=2 ¼ ∠  90 ¼ ½Ω ð8:19Þ
I e ωC ωC jωC
V
Inductance : ZL  ¼ e jπ=2 ωL ¼ ωL∠90 ¼ jωL ½Ω
I

Voltage(V) - dashed curve and current(A) - solid curve

2
a)

1
resistance

ZR R 0
-1
voltage and current are in phase

-2
2
b)

1
capacitance

0
1
ZC 90
C -1
voltage lags current by 90 deg

-2
2
c)

1
inductance

ZL L 90
voltage leads current by 90 deg -1

-2
0 2 3 4 t

Fig. 8.10. Voltage and current sinusoids for resistor, inductor, and capacitor (at phase zero).

VIII-418
Chapter 8 Section 8.2: Impedance

The impedance has units of ohms, exactly as the resistance does. The impedance is an
extension of the familiar DC resistance concept for dynamic AC circuit elements or any
combination of them. The term “impedance” (along with “inductance,” “permeability,”
etc.) belongs to Oliver Heaviside (1850–1925), a brilliant self-taught English electrical
engineer and mathematician. We note that the method of solving differential equations
with phasors (or using Laplace transforms) originates from him. Heaviside invented and
patented in England the first coaxial cable in 1880. His uncle was Sir Charles Wheatstone
(1802–1875); do you remember the Wheatstone bridge?

8.2.2 Physical Meaning of Impedance

The problem with the impedance definition is that in general it is a complex number.
What exactly is the physical meaning? For the resistor in Fig. 8.10a, we only need to
stretch the current sinusoid in order to obtain the voltage sinusoid. Therefore, we only
need one independent “stretching” parameter R to express voltage in terms of current.
However, for the capacitor in Fig. 8.10b, we need not only stretch, but also shift the
current sinusoid to the right in order to obtain the voltage sinusoid. To accomplish this,
we need two independent parameters. The impedance as a complex number has exactly
those two independent parameters. In polar form, it is the magnitude (responsible for
stretching) and the phase (responsible for shifting). Thus, only a complex number can be
used to express voltage in terms of current for the capacitor. Similarly, for the inductor in
Fig. 8.10c, we need not only stretch, but also shift to the left the current sinusoid in order
to obtain the voltage sinusoid. Therefore, we again need two independent parameters to
express the voltage in terms of current; we need the complex impedance with two
required independent parameters: the magnitude and the phase. In light of this reasoning,
it is useful to express the impedances given by Eq. (8.20) in polar form. For the imaginary
number, we use the equalities j ¼ ∠90 and 1=j ¼ ∠  90 and obtain
1
ZR ¼ R∠0 , ZC ¼ ∠  90 , ZL ¼ ω L∠90 ð8:20Þ
ωC

Example 8.8: Find impedances of resistor, capacitor, and inductor in Fig. 8.10.
Solution: In Fig. 8.10a,

iR ðt Þ ¼ 1 cos ðωtÞ ½A

) ZR ¼ R ¼ 2 ½Ω ð8:21aÞ
υR ðt Þ ¼ 2 cos ðωt Þ ½V

The resistor’s impedance is frequency independent. It is just the resistance R. In Fig. 8.10b,

VIII-419
Chapter 8 Steady-State AC Circuit Fundamentals

Example 8.8 (cont.):

iC ðtÞ ¼ 1 cos ðωt þ π=2Þ ½A

) ZC ¼ 2ejπ=2 ¼ j2 ½Ω ð8:21bÞ
υC ðt Þ ¼ 2 cos ðωtÞ ½V

The capacitor’s impedance is thus an imaginary negative number. This is always true. In
Fig. 8.10c,

iL ðt Þ ¼ 1 cos ðωt Þ ½A

) ZL ¼ 2eþjπ=2 ¼ þj2 ½Ω ð8:21cÞ
υL ðt Þ ¼ 2 cos ðωt þ π=2Þ ½V

The inductor’s impedance is thus an imaginary positive number. This is always true.

8.2.3 Magnitude and Phase of Complex Impedance

With reference to Fig. 8.10, the magnitude of the impedance is a factor by which we
should multiply (or stretch) the current sinusoid in order to obtain the voltage sinusoid.
The phase (or the polar angle) of the impedance is a factor by which we should shift to the
left the current sinusoid in order to obtain the voltage sinusoid.
1. For the resistor, we multiply the current by jZR j ¼ R ¼ 2 Ω and do not shift.
2. For the capacitor, we multiply the current by jZC j ¼ ωC1
¼ 2 Ω and shift it by 90
(or by a quarter of the period) to the right in Fig. 8.10b.
3. For the inductor, we multiply the current by jZL j ¼ ωL ¼ 2 Ω and shift by +90
(or by a quarter of the period) to the left in Fig. 8.10c.
The above operations hold for any resistor, inductor, and capacitor. Thus, the imped-
ance concept extends Ohm’s law to dynamic circuit elements in steady-state AC circuits.

Exercise 8.5: Establish phase relationships for voltages and currents in Fig. 8.10.
Answer: For the resistor, current and voltage are in phase. For the capacitor, current leads
voltage by 90 (or voltage lags current by 90 ). For the inductor, voltage leads current by
90 (or current lags voltage by 90 ).

Example 8.9: For two AC circuits shown in Fig. 8.11, find the impedance of the resistor,
capacitor, and inductor.
Solution: The resistor’s impedance is frequency independent. It is just the resistance R,

VIII-420
Chapter 8 Section 8.2: Impedance

Example 8.9 (cont.):

ZR ¼ R ¼ 2 ½Ω ð8:22aÞ

The capacitor’s impedance does depend on frequency. In Fig. 8.11a, ω ¼ 5000 rad/s.
Therefore,

1 1 1 2
ZC ¼ ¼ ¼ ¼ ¼ j2 ½Ω ð8:22bÞ
jωC j5000  104 j0:5 j

The inductor’s impedance also depends on frequency. In Fig. 8.11b, ω ¼ 20 rad/s.

Therefore,

ZL ¼ jωL ¼ j20  0:1 ¼ j2 ½Ω ð8:22cÞ

a) R=2 b) R=2

vS(t)=Vmcos(5000t) vS(t)=Vmcos(20t)
+ + L=100mH
- C=100 F -

Fig. 8.11. Two types of a series AC circuit with a voltage supply.

An equivalent representation of impedances in Eqs. (8.22a, b, c) is the impedance phasor

diagram shown in Fig. 8.12 where we plot the corresponding complex numbers.

Im( )

2 ZL

1
ZR
-2 -1 0 1 2 Re( )
-1

-2 ZC

Fig. 8.12. Complex impedances displayed in the phasor diagram.

VIII-421
Chapter 8 Steady-State AC Circuit Fundamentals

Note that the magnitudes of complex impedances in Fig. 8.12 are all equal to 2 Ω:

jZR j ¼ jZR j ¼ jZR j ¼ 2 Ω ð8:23Þ

8.2.4 Application Example: Impedance of a Human Body

The impedance is not only the characteristic of an electric circuit but also of an arbitrary
conducting object, which may be modeled as a combination of ideal resistance, induc-
tance, and capacitance. They can be connected in series, in parallel, or a combination of
series and parallel. As an example, we consider here the impedance of a human body in
the frequency range from 10 kHz to 3 MHz. The impedance magnitude is shown in
Fig. 8.13. It was measured for about 400 human subjects of ages between 18 and 70 years
and then averaged. The subject stood on a large aluminum sheet as a ground plane. The
electrode was a cylindrical brass rod for making a grasping contact with the hand. In its
simplest form, the concept of impedance measurement implies the simultaneous mea-
surements of harmonic voltage and current and then the extraction of the amplitude ratio
(magnitude of the impedance) and the phase difference (impedance phase).

Impedance magnitude,
660

600 Ten-year-old child

540

480

420
Female
360
Male
300
1 2 3 4
10 10 10 10
frequency, kHz

Fig. 8.13. Magnitude of the average impedance of the human body. From: I. Chatterjee et al.,
“Human Body Impedance and Threshold Currents for Perception and Pain for Contact Hazard
Analysis in the VLF-MF Band,” IEEE Trans Biomedical Eng., May 1986.

Figure 8.13 indicates that the impedance magnitude decreases with frequency.
Therefore, the human body impedance, at least at relatively low frequencies, behaves
similarly to the impedance of a capacitor, where the magnitude also decreases with
frequency. A purely resistive component is also present. The impedance measurements
have been used for the extraction of various biomedical data such as assessment of a
fat-free mass.

VIII-422
Chapter 8 Section 8.3: Principles of AC Circuit Analysis

Section 8.3 Principles of AC Circuit Analysis

8.3.1 AC Circuit Analysis: KVL, KCL, and Equivalent Impedances
The generic process of conducting an AC circuit analysis is outlined in Fig. 8.14 for a
steady-state circuit with a capacitor and a resistor. We replace the circuit components by
their impedances and replace the voltages and currents by their phasors. Hence, the “real”
time-domain AC circuit in Fig. 8.14a becomes an “imaginary” circuit with complex
phasor voltages VS, VR, VC and a complex phasor current I “flowing” through the
circuit; see Fig. 8.14b. However, the phasor VS is a constant and the resulting circuit is
the “DC” circuit.

a) R b) ZR

VR
vR
- -
+

+
+

+
vS(t) + vC C VS + VC ZC
- - - -
i(t) I

Fig. 8.14. Transformation of an AC circuit to phasor/impedance form.

KVL for the circuit in Fig. 8.14a in the time domain reads

υS ðt Þ þ υR ðt Þ þ υC ðt Þ ¼ 0 ð8:24Þ

Its phasor counterpart, Fig. 8.14b, in the frequency domain has exactly the same algebraic
form

VS þ VR þ VC ¼ 0 ð8:25Þ

This result can be proven using the method described in the Chapter’s summary. KCL
is formulated in terms of phasors exactly in the same manner. According to KCL,
the same phasor current I flows through all the elements in the circuit of Fig. 8.14b.
This results in

VS VS
VS þ ZR I þ ZC I ¼ 0 ) I ¼ ¼ ð8:26Þ
ZR þ ZC Zeq

where phasor voltages are related to phasor currents though the concept of impedances. Thus,
once the AC circuit has been transformed to phasors and impedances (frequency domain),

VIII-423
Chapter 8 Steady-State AC Circuit Fundamentals

the impedances may be combined as if they were simple resistors. The series/parallel
equivalents for the impedances are equally applicable. After plugging in numbers for
ω, R, C, VS, the phasor current and phasor voltages can be found and converted back
to time domain. This is the method of solving in AC circuits.

8.3.2 Complete Solution for an AC Circuit: KVL and KCL

on Phasor Diagram
Let us assume that the power supply voltage in Fig. 8.14a has the form υS ðt Þ ¼
V m cos ωt, and its amplitude and frequency are given by V m ¼ 5 V, ω ¼ 1000 rad=s.
Further, we know that C ¼ 1 μF, R ¼ 1 kΩ in Fig. 8.14a. The solution to this AC circuit
includes several steps as discussed above. First, we convert the circuit to the phasor/
impedance form as shown in Fig. 8.14b. All currents/voltages are replaced by their
phasors, and all circuit elements are replaced by their impedances. Next, we solve the
phasor circuit in Fig. 8.14b as if it were a “DC” circuit. The element impedances are
1 1
ZR ¼ 1000 ½Ω, ZC ¼ ¼ ¼ j1000 ½Ω ð8:27Þ
jωC j1000  1  106

Note that the impedances here are written in rectangular instead of polar form. We must
now find the equivalent impedance. From the series impedance combination:
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Zeq ¼ ZR þ ZC ¼ 1000  j1000 ¼ 10002 þ 10002 ∠  arctanð1Þ
1414∠  45 ½Ω ð8:28Þ

The phasor current (circuit current) has the form

VS 5
I¼ ¼ 
3:54∠45 ½mA ð8:29Þ
Zeq 1414∠  45

The phasor voltages across the resistor and the capacitor are found according to Ohm’s
law, that is:

VR ¼ ZR I ¼ 1000  0:00354∠45 ¼ 3:54∠45 ½V ð8:30Þ

VC ¼ ZC I ¼ j1000  0:00354∠45 ¼ 3:54∠  90 ∠45 ¼ 3:54∠  45 ½V ð8:31Þ

The phasor voltages are plotted in the phasor diagram as depicted in Fig. 8.15.

VIII-424
Chapter 8 Section 8.3: Principles of AC Circuit Analysis

Im(V)

4
VR
2
VS
-4 -2 0 2 4 Re(V)
-2
VC
-4

Fig. 8.15. Phasor diagram for the circuit in Fig. 8.14 and KVL in the vector form.

We may conclude that KVL in phasor form is equivalent to the addition of two
vectors VR, VC, which equals the supply phasor voltage VS. For an AC circuit in
the form of a current divider, KCL in phasor form will have exactly the same
representation.
As a final step to arrive at the solution, we convert the phasors in Eqs. (8.30) and (8.31)
to real-valued voltages:

υR ðt Þ ¼ 3:54 cos ðωt þ 45 Þ ½V

ð8:32Þ
υC ðt Þ ¼ 3:54 cos ðωt  45 Þ ½V

The AC circuit is thus solved. We note that the amplitudes of both voltages are the same;
in this particular case the voltage divider splits the power supply voltage “equally.” In
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
fact, 3:542 þ 3:542 ¼ 5V, which is KVL in terms of the voltage amplitudes.

Exercise 8.6: How does the phasor diagram in Fig. 8.15 change if the voltage source in
Fig. 8.14 is given by υS ðtÞ ¼ V m cos ðωt  45 Þ?
Answer: The entire phasor diagram rotates clockwise by 45 .

8.3.3 Source Transformation

After KVL and KCL in terms of phasors (in the frequency domain) have been
established, the circuit laws and principles from Chapters 3 and 4 are straightforwardly
extended to the steady-state AC circuits operating at a single frequency. The first example
is the source transformation in the frequency domain shown in Fig. 8.16.

VIII-425
Chapter 8 Steady-State AC Circuit Fundamentals

a a

ZT
VT + IN ZN
-

b b

Fig. 8.16. Source transformation in the frequency domain.

A voltage source with the phasor voltage VT and in series with a passive circuit element
having impedance ZT is equivalent to a current source with the phasor current IN and in
parallel with impedance ZN given that
VT
ZT ¼ ZN , VT ¼ ZT IN , IN ¼ ð8:33Þ
ZT

Equation (8.33) is the direct extension of the source transformation principle established
in Chapter 4 for DC circuits.

Example 8.10: Determine phasor voltage V1 in the AC circuit shown in Fig. 8.17 using
the method of source transformation. The impedance values are given at the frequency of
interest. Note that, for every impedance box in Fig. 8.17, its physical counterpart is shown
inside this box.
Solution: The source transformation follows Eq. (8.33) and leads to the AC circuit shown in
Fig. 8.18. This circuit is easier to solve.
 We can see that the circuit in Fig. 8.18 becomes the
current divider between the 10Ω10Ω ¼ 5Ω impedance and the j10Ω  j5Ω ¼ j5Ω
impedance, respectively. Therefore, the phasor current through the impedance of interest
is given by current division

5 5 1
I¼  1∠90 ¼ pffiffiffi  1∠90 ¼ pffiffiffi∠45 ½A ð8:34Þ
5 þ j5 5 2∠45  2

Thus, the resulting voltage is given by

1 5
V1 ¼ j5  I ¼ 5∠  90  pffiffiffi∠45 ¼ pffiffiffi∠  45 ½V ð8:35Þ
2 2

VIII-426
Chapter 8 Section 8.3: Principles of AC Circuit Analysis

10 j10

+
10 90 [V] + 10 -j5 V1
- -

Fig. 8.17. An AC circuit solved with the help of source transformation. Note that every impedance
box has a physical counterpart shown within this box.

Exercise 8.7: Three impedances j5Ω, 10Ω, 10Ω are combined in parallel. What is the
equivalent impedance?
Answer: 2:5 þ j2:5 Ω.

j10

I
+
90 [A] 10 10 -j5 V1
-

Fig. 8.18. Source transformation applied to the AC circuit from the previous figure.

Exercise 8.8: In Fig. 8.18, the impedance of the capacitor changes to j10Ω. What will be
the phasor current, I, through the capacitor?
Answer: I ¼ 1∠90 ½A

8.3.4 Thévenin and Norton Equivalent Circuits

The Thévenin’s theorem for steady-state AC circuits is formulated as follows. Any linear
AC network of resistors/capacitors/inductors and voltage/current power sources operat-
ing at the same frequency can be represented in the form of a Thévenin
equivalent network shown in Fig. 8.19b (phasor form). This result is a direct extension
of Thévenin’s theorem for DC circuits stated in Chapter 4. Thévenin’s theorem allows us
to reduce complicated AC circuits to a simple source and impedance configuration, with
the same power output to a load. The AC frequency remains the same. Phasor voltage VT
is known as Thévenin voltage or simply the source voltage; impedance ZT is called Thé
venin impedance or source impedance. The phasor voltage may have amplitude and

VIII-427
Chapter 8 Steady-State AC Circuit Fundamentals

phase that are different from the original AC sources. The Norton theorem for steady-
state AC circuits replaces the Thévenin equivalent circuit by the Norton equivalent circuit
from Fig. 8.16. As in resistive circuits, the Thévenin phasor voltage is equal to the open-
circuit phasor voltage of the original circuit in Fig. 8.19a:

VT ¼ Vab ¼ VOC ð8:36Þ

To find the Thévenin impedance ZT, we need to know the phasor, ISC, for the short-circuit
current of the network. The short-circuit current is obtained by shorting out the output
terminals a and b. This results in the source impedance:

a) a b) a

ZT
vS(t) + VT +
- -

b b

Fig. 8.19. Representation of a Thévenin equivalent circuit. The original AC circuit in (a) is
transformed into its Thévenin equivalent circuit (b) by providing the same voltage and current
to a load. Both circuits are indistinguishable when looking from the load.

VOC VT
ZT ¼ ¼ ð8:37Þ
ISC ISC

To find ZT we often use another, somewhat simpler, approach. We short out the voltage
source(s), i.e., replace them by wires. Then we zero the current sources (if present), i.e.,
replace them by an open circuit. This enables us to find the equivalent impedance of the
resulting purely passive circuit, which is equal to ZT.

Example 8.11: Find the Thévenin equivalent, i.e., VT and ZT, for the circuit shown in
Fig. 8.19a when ω ¼ 377 rad=s, L ¼ 26:5 mH, R ¼ 10 Ω, C ¼ 220 μF, and vS ðtÞ ¼
10 cos ωt ½V.
Solution: We convert the circuit in Fig. 8.19a into phasor form first. VT is the phasor voltage
between terminals a and b, i.e., the phasor voltage across the inductor. The resistor and the
inductor form a voltage divider with regard to the supply phasor voltage of 10 V. According
to the voltage division principle:

VIII-428
Chapter 8 Section 8.3: Principles of AC Circuit Analysis

Example 8.11 (cont.):

ZL j10 10∠90
VT ¼ VL ¼ 10 ¼ 10 ¼ 10 ¼ 7:07∠45 ½V  ð8:38Þ
ZL þ ZR 10 þ j10 14:14∠45

To find the Thévenin resistance, we short out the voltage source (by using the second
approach). This operation simultaneously shorts out the capacitor, with the result that the
resistor and the inductor are now in parallel. This gives
 j10  10 100∠90
ZT ¼ ZL ZR ¼ ¼ ¼ 7:07∠45 ½Ω ð8:39Þ
10 þ j10 14:14∠45

8.3.5 Summary of AC Circuit Analysis at a Single Frequency

As long as an AC circuit includes a single AC source or sources which all operate at the
same frequency, it can always be directly analyzed in terms of phasors/complex imped-
ances. All DC analysis techniques:
- Series/parallel equivalents and voltage/current dividers
- Superposition principle
- Source transformation; Thévenin and Norton equivalent circuits
- Nodal and mesh analyses
are equally applicable to steady-state AC circuits. The specific frequency value does not
matter. Indeed, we simply need to perform calculations with complex numbers instead of
real ones.

8.3.6 Multifrequency AC Circuit Analysis: Superposition Theorem

Unfortunately, when an AC circuit includes sources operating at different frequencies
and/or includes the DC sources, the situation is no longer straightforward. The key to the
circuit analysis in this case is the general superposition theorem, which is applicable to all
linear circuits with arbitrary independent and linear-dependent sources. The superposi-
tion theorem for multifrequency AC circuits is explained in Fig. 8.20. The original circuit
with two AC voltage sources of different frequencies and with one DC source is replaced
by three partial circuits, with two of three voltage sources shorted out at a time. Every
partial circuit is solved independently, either using the phasor/impedance method (for
single-frequency AC circuits) or the DC circuit analysis. Finally, the complete solution is
obtained as a sum of three real-valued contributions.

VIII-429
Chapter 8 Steady-State AC Circuit Fundamentals

vR
-

+
V1cos t + + V cos t
1
- - 2 2
V3=10 V DC
-

+
a) b) c)
vR1
- vR3
- vR2
-
+

+
V1cos t
1 V2cos t
2
+ +
- + V3=10 V DC
+ -
-

+
Fig. 8.20. Superposition theorem for a circuit with sources operating at different frequencies
(including a DC source).

Example 8.12: Find real-valued voltage υR(t) across the resistor for the circuit in Fig. 8.20
using the superposition principle. You are given R ¼ 10 Ω, V 1 ¼ 10 V, V 2 ¼ 1 V.
The impedance of the inductor is j2 Ω at frequency ω1 and j6 Ω at frequency ω2,
respectively.
Solution: Three partial equivalent circuits are shown in Fig. 8.20a,b,c. We solve the AC
circuits in Fig. 8.20a,c using the phasor/impedance method. Applying voltage division for
the circuits in frequency domain gives
10
VR1 ¼ þ 10 ¼ 9:81∠  11 V ð8:40Þ
10 þ j2
10
VR2 ¼  1 ¼ 0:86∠ þ 149 V ð8:41Þ
10 þ j6

The DC circuit in Fig. 8.20b gives V R3 ¼ 10 V since the inductor is the short circuit in the
DC steady state. Combining the solutions yields the resistor voltage,

υR ðt Þ ¼ 10 þ 9:81 cos ðω1 t  11 Þ þ 0:86 cos ðω2 t þ 149 Þ V ð8:42Þ

Exercise 8.9: How does the solution of the previous example change when a 20 Ω resistor
is placed in series with the DC source?
Answer: The solution becomes

υR ðtÞ ¼ 3:33 þ 3:33 cos ðω1 t  4 Þ þ 0:33 cos ðω2 t þ 169 Þ V:

VIII-430
Chapter 8 Summary

Summary
Term Meaning/Figure
υðt Þ ¼ Vm cos ðω t þ φÞ
Vm > 0 is the voltage
amplitude [V]
Steady-state AC ω ¼ 2π f > 0 is the angular
voltage (steady-state frequency [rad/s]
alternating current) f > 0 is the frequency [Hz]
T ¼ 1=f > 0 is the period [s]
π
φ
π is the phase
[rad] or [deg]

Leading/lagging

Euler’s identity e jα ¼ cos α þ j sin α, e jπ=2 ¼ j, ejπ=2 ¼ j

Time-domain
signal υ(t) versus
its phasor V; phasor
diagram

Complex phasors
and impedances
ZR ¼ R
1
ZC ¼
jωC
ZL ¼ jωL
(continued)

VIII-431
Chapter 8 Steady-State AC Circuit Fundamentals

Term Meaning/Figure

Meaning of
complex
impedance

Solution for
an AC circuit

Proof of the υS ðt Þ þ υR ðt Þ þ υC ðt Þ ¼ 0 )

conversion ReðVS e jω t Þ þ ReðVR e jω t Þ þ ReðVC e jω t Þ ¼ 0 )
from time
ReðVS e jω t þ VR e jω t þ VC e jω t Þ ¼ 0 )
domain to
frequency VS e jω t þ VR e jω t þ VC e jω t ¼ 0 )
domain VS þ VR þ VC ¼ 0
Analytical
solution
method
(calculator)
Numerical
solution
method
(MATLAB)
Source
transformation;
Thévenin and
Norton
equivalent
circuits ZT ¼ ZN , VT ¼ ZT IN , IN ¼ VZTT

VIII-432
Chapter 8 Summary

Table 1. Some basic operations with complex numbers.

e j0 ¼ 1, e jπ=2 ¼ j, ejπ=2 ¼ j, e jπ ¼ 1, ejπ ¼ 1, je jα j ¼ 1
j ¼ ∠90 , 1j ¼ ∠  90 , j2 ¼ 1, 1j ¼ j, jjj ¼ 1
V m e jφ ¼ V m ∠φ ¼ V m ð cos φ þ j sin φÞ, V m ejφ ¼ V m ∠  ϕ ¼ V m ð cos φ  j sin φÞ
jV m e jφ j ¼ jV m ejφ j ¼ V m
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi 
x þ jy ¼ V m e jφ , V m ¼ x2 þ y2 , φ ¼ arctan yx , x  0
¼ x2 þy
1
xþjy
x
2  jx2 þy2
y

pffiffiffiffiffiffiffiffiffiffiffiffiffiffi 
1
xþjy ¼ V1m ejφ , V m ¼ x2 þ y2 , φ ¼ arctan yx , x0
ðV m e jφ Þ ¼ V m ejφ ,
*
ðx þ jyÞ* ¼ x  jy, ðV m ∠φÞ* ¼ V m ∠  φ
V m ∠φ  I m ∠ψ ¼ ðV m e jφ ÞðI m e jψ Þ ¼ V m I m e jðφþψ Þ ¼ V m I m ∠ðφ þ ψ Þ
V m ∠φ
¼ VI mmee jψ ¼ VI mm e jðφψ Þ ¼ VI mm ∠ðφ  ψ Þ
jφ

I m ∠ψ

Table 2. Selected trigonometric identities.

sin α ¼ cos ðα  π=2Þ,  sin α ¼ cos ðα þ π=2Þ, cos 2α ¼ 2 cos 2 α  1, sin 2α ¼ 2 sin α cos α
cos ðα þ βÞ ¼ cos α cos β  sin α sin β cos α cos β ¼ 0:5ð cos ðα þ βÞ þ cos ðα  βÞÞ
sin ðα þ βÞ ¼ sin α cos β þ cos α sin β sin α sin β ¼ 0:5ð cos ðα  βÞ  cos ðα þ βÞÞ
sin α cos β ¼ 0:5ð sin ðα þ βÞ þ sin ðα  βÞÞ
αþβ αβ    
cos α þ cos β ¼ 2 cos 2 cos 2 sin α þ sin β ¼ 2 sin αþβ
2 cos αβ
2

arctanφ ¼ π2  arctanφ1 , φ > 0 arctanφ ¼ π2  arctanφ1 , φ < 0

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  
C 1 cos ωt þ C 2 sin ωt ¼ C 21 þ C 22 cos ðωt þ φÞ, φ ¼ arctan CC 21 C1, C2 > 0
C 1 cos ðω t þ φÞ  C 2 cos ðωt þ ψ Þ ¼ 0:5C 1 C 2 ð cos ðφ  ψ Þ þ cos ð2ωt þ φ þ ψ ÞÞ
qðffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
C 1 cos ωt þ φÞ þ C 2 cos ðω t þ ψ Þ ¼ C cos ðωt þ φÞ
C¼ C 21 þ C 22 þ 2C 1 C 2 cos ðφ  ψ Þ


C 2 sin ðφ  ψ Þ 0 C 1 þ C 2 cos ðφ  ψ Þ > 0
ϕ ¼ φ  arctan þ
C 1 þ C 2 cos ðφ  ψ Þ π C 1 þ C 2 cos ðφ  ψ Þ < 0

VIII-433
Chapter 8 Steady-State AC Circuit Fundamentals

Problems 0.8

8.1 Harmonic Voltage 0.4

and Current: Phasor

voltage, V
0
8.1.1 Harmonic Voltages and Currents
8.1.2 Phase: Leading and Lagging -0.4
Problem 8.1
A. Write a general expression for the AC
-0.8
harmonic voltage signal (steady-state AC
voltage) using the cosine function. 0 0.5 1 1.5 2 2.5 3
time, ms
B. Identify amplitude, angular frequency,
and phase.
Problem 8.4. Repeat problem 8.2 for a har-
C. Write relations between the angular fre-
monic voltage signal with a DC offset shown
quency, frequency, and the period.
in the figure below.
Problem 8.2
voltage, V
A. Determine frequency in Hz, angular fre-
quency in rad/s, and amplitude of the 10.0
harmonic voltage signal shown in the
figure below (show units for every 5.0
quantity).
B. Write the AC voltage in the form of a 0
cosine function with the corresponding
amplitude, frequency, and phase. -5.0

voltage, V -10.0

2
0 5 10 15 20 25 30
time, s

1
Problem 8.5
A. Determine frequency in Hz, angular fre-
0
quency in rad/s, amplitude, and phase
(versus the base cos ωt signal) of the
-1
harmonic voltage shown in the figure
below (show units for every quantity).
-2
B. Write the AC voltage in the form of a
0 15 30 45 60 75 90 cosine function with the corresponding
time, s amplitude, frequency, and phase.
Problem 8.3. Repeat problem 8.2 for the volt-
age signal shown in the figure below.

VIII-434
Chapter 8 Problems

voltage, V voltage, V
3
2
2

1
1

0 0

-1 -1

-2
-2
-3
0 15 30 45 60 75 90 0 5 10 15 20 25 30 35 40
time, s time, ms

Problem 8.6. Repeat problem 8.5 for the volt- Problem 8.9. An AC voltage in a circuit is
age signal shown in the figure below. given by υðt Þ ¼ 10 cos ð2π50tÞ ½V. Using soft-
ware of your choice, plot the voltage to scale
voltage, V over the time interval of two periods, i.e.,
0
t
2T . Label the axes.
2

Problem 8.10. An AC voltage in a

1
circuit is given by the voltage expression
υðt Þ ¼ 10 cos ð1000t  π=3Þ ½V. Using soft-
0 ware of your choice, plot the voltage to scale
over the time interval of four periods, i.e.,
-1 0
t
4T . Label the axes.

-2 Problem 8.11. The reference voltage is shown

by a solid curve in the figure; the AC voltage
0 15 30 45 60 75 90
time, s under study is shown by a dashed curve. Deter-
mine if the AC voltage under study leads or
Problem 8.7. Repeat problem 8.5 for the volt- lags the reference voltage, and, if so, by how
age signal shown in the figure below. many degrees.

voltage, V voltage, V
3 3

2 2

1 1

0 0

-1 -1

-2 -2

-3 -3
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
time, ms time, ms

Problem 8.8. Repeat problem 8.5 for the volt-

age signal shown in the figure below.

VIII-435
Chapter 8 Steady-State AC Circuit Fundamentals

Problem 8.12. Repeat problem 8.11 for the in V, and phase in radians (versus the base
voltage signal shown in the figure below. cosine signal) of the voltage signal in the form
υðt Þ ¼ 5 sin ð100 t þ 225 Þ ½V. Hint: Convert
3
voltage, V the signal to the base cosine form first.

2
Problem 8.17. The AC voltage is given by a
combination of two sinusoids:
1 A. υðt Þ ¼ 1sin ðω t þ π=2Þ  2sin ðωt π=2Þ
B. υðt Þ¼ 1 sin ðωt þ π=2Þ  2 sin ðωt π=3Þ
0
Convert this voltage to the basic cosine form
-1 υðt Þ ¼ V m cos ðωt þ φÞ and determine the
amplitude and the phase (versus the base cosine
-2 signal).
Hint: Trigonometric identities may be found in
-3
0 5 10 15 20 25 30 35 40 the summary to this chapter.
time, ms

Problem 8.13. Repeat problem 8.11 for the 8.1.4 Definition of a Phasor
voltage signal shown in the figure below. 8.1.5 From Real Signals to Phasors
voltage, V 8.1.6 From Phasors to Real Signals
3 Problem 8.18. Determine the phasors for the
real-valued AC voltages and currents. Show
2
units. Express all phase angles in radians.
1
υðt Þ ¼ 10 cos ðωt þ π=3Þ ½V
0 υðt Þ ¼ 3 cos ðωt  30 Þ ½V
-1
iðt Þ ¼ 12 cos ðωt þ π=6Þ ½A
iðt Þ ¼ 1 cos ðωt  π=2Þ ½A
-2
Problem 8.19. Determine the phasors for the
-3
0 5 10 15 20 25 30 35 40 real-valued AC voltages and currents. Use the
time, s shorthand notation ∠ for the complex exponent.
Show units. Express all phase angles in degrees.
Problem 8.14. Determine the frequency in
Hz, period in s, amplitude in V, and phase υðt Þ ¼ 10 sin ðωt þ π=3Þ ½V
in degrees (versus the base cosine signal) υðt Þ ¼ 3 cos ð100t  30 Þ ½V
of the voltage signal in the form iðt Þ ¼ 12 sin ðωt þ π=6Þ ½A
υðt Þ ¼ 15 sin ð100 t þ 45 Þ ½V. Hint: Convert iðt Þ ¼ cos ðω t Þ þ sin ðωt Þ ½A
the signal to the base cosine form first.
Problem 8.20. The phasors of the AC voltage
Problem 8.15. Determine the frequency in Hz, and current are given by
period in s, amplitude in V, and phase in
radians (versus the base cosine signal) of the V ¼ 5∠π=3 ½V
voltage signal in the form υðtÞ ¼ V ¼ 3∠π ½V
15 sin ð1000 t  35 Þ ½V. Hint: Convert the I ¼ 2:1∠45 ½A
signal to the base cosine form first. I ¼ 1∠  180 ½A
Problem 8.16. Determine the frequency in Hz,
period in s, amplitude in V, peak-to-peak value

VIII-436
Chapter 8 Problems

The AC source has the angular frequency ω. 6 Im

Restore the corresponding real-valued voltages
4
and currents. Show units; express all phase
angles in radians. 2
Re
Problem 8.21. The phasors of the AC voltage -6 -4 -2 0 2 4 6
and current are given by -2

V ¼ 10∠π=2 ½V -4

V ¼ 15∠  π=3 ½V -6

I ¼ 20∠  16 ½A
I ¼ j∠45 ½A
Problem 8.24. Repeat problem 8.23 for
The AC source has the angular frequency ω. V ¼ 4expðjπ=6Þ, but convert this number to
Restore the corresponding real-valued voltages the rectangular form.
and currents as functions of time. Show units.
Problem 8.25. Repeat problem 8.23 for
Express all phase angles in radians.
V ¼ 25=ð3 þ j4Þ.
Problem 8.22. The phasors of the AC voltage
Problem 8.26. Phasors of three AC voltage
and current are given by
signals are shown in Fig. 8.8. Every division in
the figure corresponds to 1 V. The AC source
V ¼ 1 ½V
has the angular frequency ω. Restore the corres-
V ¼ ∠π þ ∠  π ½V ponding real-valued voltages in time domain.
I ¼ 2∠45 ½A Show units. Express all phase angles in radians.
I ¼ ∠45 þ ∠  45 ½A
Problem 8.27. The phasors of the AC voltage
The AC source has the angular frequency ω. and current are given in the rectangular form:
Restore the corresponding real-valued voltages
and currents. Show units; express all phase V ¼ 3 þ j2 ½V
angles in degrees. I ¼ 2 þ j3 ½A

The AC source has the angular frequency ω.

8.1.7 Polar and Rectangular Forms: Restore the corresponding real-valued voltages
Phasor Magnitude and currents. Show units, express all phase angles
8.1.8 Operations with Phasors and Phasor in radians.
Diagram
Problem 8.23. A complex number V is given Problem 8.28*. Solve the previous problem
by V ¼ 4 þ j 2. using MATLAB. Present the corresponding
A. Convert it into polar form; express the MATLAB script.
phase angle in degrees.
Problem 8.29. The phasors of the AC voltage
B. Plot the number on the phasor diagram.
and current are given in the rectangular form:
C. If this number is a phasor of a voltage
signal with the units of volts, what is the
voltage signal in time domain? Assume V ¼ ð3 þ j2Þ2 ½V
the angular frequency ω. V ¼ ð2 þ j3Þð7 þ jÞ ½V
1
V¼ ½V
0:2 þ j0:1

VIII-437
Chapter 8 Steady-State AC Circuit Fundamentals

The AC source has the angular frequency ω. Problem 8.33. Phasors of three AC voltage
Restore the corresponding real-valued voltages signals are shown in Fig. 8.8. Every division in
and currents. Show units; express all phase the figure corresponds to 1 V. The AC source has
angles in radians. the angular frequency ω. A sum of three voltage
signals is desired. Restore the real-valued voltage
Problem 8.30. Solve the previous problem corresponding to the sum of three signals in time
using MATLAB. Present the corresponding domain. Express the phase angle in degrees.
MATLAB script.
Problem 8.34. Solve the previous problem
Problem 8.31. Two phasor voltages are shown using MATLAB. Present the corresponding
in the phasor diagram. The AC source has the MATLAB script.
angular frequency ω. Restore the corresponding
real-valued voltages. Show units. Express all Problem 8.35. Voltages of two series elements,
phase angles in radians. shown in the figure below,

a v1(t)
- v2(t) - b

+
6 Im(V)
A B
V1
4 -
V2

+
v(t)?
2 are given by
Re(V)
0 2 4 6 υ1 ðt Þ ¼ 5 cos ðωt þ 45 Þ ½V
υ2 ðt Þ ¼ 5 cos ðωt  45 Þ ½V

A. Draw the phasor diagram and

show phasors V1, V2 to scale as two vec-
tors in the complex plane.
Problem 8.32. Phasors of two AC current sig- B. Show the sum V ¼ V1 þ V2 as a vector
nals are shown in the following figure. The AC in the complex plane.
source has the angular frequency ω. A sum of A. Find voltage υ(t) between terminals a and
two current signals is desired. Restore the real- b using the phasor method. Express the
valued current corresponding to the sum of phase angle in degrees.
three signals in time domain. Express the C. Does voltage υ1(t) lag or lead voltage
phase angle in degrees. υ2(t)?

Problem 8.36. Repeat the previous problem for:

6 Im(A)

4
I1
υ1 ðt Þ ¼ 1 cos ðωt þ 60 Þ ½V
2
υ2 ðt Þ ¼ 2 cos ðωt  30 Þ ½V
Re(A)
0 2 4 6
Problem 8.37. Electric currents through two
I2 parallel circuit elements are shown in the fol-
lowing figure:

i1(t)
A
i(t)?
i2(t)
B

VIII-438
Chapter 8 Problems

The current expressions are given by that is described by the voltage-to-current rela-
3
tion υN ¼ N ddti3N where N is a constant:
i1 ðt Þ ¼ 4 cos ðωt þ 45 Þ ½A A. Obtain the impedance for this circuit
i2 ðt Þ ¼ 2 cos ðωt  60 Þ ½A element.
B. Do you think such a circuit element may
A. Draw the phasor diagram and exist? Why or why not? Hint: The real
show phasors I1, I2 to scale as two vec- part of the impedance is the element
tors in the complex plane. resistance. If the resistance is negative,
B. Show the difference I ¼ I1  I2 as a the element delivers power instead of
vector in the complex plane. absorbing it.
C. Find the net current i(t) of the parallel
combination using the phasor method. Problem 8.42. For three AC circuits shown in
Express the phase angle in degrees. the following figure, find the impedance of the
D. Does the net current i(t) lag or lead cur- resistor, inductor, and capacitor (when present).
rent i1(t)? Show units. Express the result in rectangular
form. Also express the result in polar form
8.2 Impedance using the shorthand notation ∠. Determine the
magnitude of the impedance. Does the strength
8.2.1 The Concept of Impedance of the power supply have an effect on the
8.2.2 Physical Meaning of the Impedance obtained impedance values?
Problem 8.38. Prove that the impedance of the
inductor, ZL ¼ jωL, has units of ohms. a)
Hint: The imaginary unit j is dimensionless. R=25 k

However, in the context of phasors, it may be +

assigned the units of rad1. - L=33 mH

10cos(1000t)=VS(t)
Problem 8.39. Prove that the impedance of the
capacitor, ZC ¼ jωC
1
has units of ohms. b)
R=10
Hint: The imaginary unit j is dimensionless.
However, in the context of phasors, it may be + C=47 F
-
assigned the units of rad1.
2cos(50000t)=VS(t)

Problem 8.40. An ECE student discovers a

c)
“new” dynamic passive circuit element R=10
N (in addition to the inductor and the capacitor)
that is described by the voltage-to-current rela- + C=68 F
2
-
tion υN ¼ N ddti2N where N is a positive constant: 2cos(50000t)=VS(t)
A. Obtain the impedance for this circuit
element. L=10 nH
B. Do you think such a circuit element may
exist? Why or why not? Hint: The real Problem 8.43. Repeat problem 8.42 if the AC
part of the impedance is the element voltage power supply in every figure is
resistance. If the resistance is negative, replaced by an alternating current power supply
the element delivers power instead of with the same frequency and amplitude of
absorbing it. 0.5 A.

Problem 8.41. Another ECE student discovers Problem 8.44. Voltages (dashed curves) and
a “truly new” dynamic passive circuit element the corresponding currents (solid curves) for
N (in addition to the inductor and the capacitor)
VIII-439
Chapter 8 Steady-State AC Circuit Fundamentals

three unknown circuit elements are shown in Problem 8.45. Voltages (dashed curves) and
the figure below. currents (solid curves) for three unknown cir-
cuit elements are shown in the figure below.
a) =1000 rad/sec
2
a) =1000 rad/sec
2
1.5
1.5
voltage(V) or current(A)

voltage(V) or current(A)
1
0.5
0.5
0
0
-0.5
-0.5
-1
-1
-1.5
-1.5
-2
0 2 3 4 t -2
0 2 3 4 t
b) 2
=100 rad/sec
b) =100 rad/sec
2
1.5
1.5
voltage(V) or current(A)

1
voltage(V) or current(A)

1
0.5
0.5
0
0
-0.5
-0.5
-1
-1
-1.5
-1.5
-2
0 2 3 4 t -2
0 2 3 4 t
c) =10000 rad/sec
2
c) =10000 rad/sec
2
1.5
1.5
voltage(V) or current(A)

1
voltage(V) or current(A)

1
0.5
0.5
0
0
-0.5
-0.5
-1
-1
-1.5
-1.5
-2
0 2 3 4 t -2
Determine: 0 2 3 4 t
A. The type of the element (resistor, capac- Determine:
itor, or inductor) A. The type of the element (resistor, capac-
B. The value of the corresponding resis- itor, or inductor)
tance, inductance, or capacitance B. The value of the corresponding resis-
tance, inductance, or capacitance
Note that the angular frequency is different in
every case.

VIII-440
Chapter 8 Problems

Problem 8.46. Phasor voltages and currents a) =1000 rad/sec 3 Im (V or A)

for three unknown circuit elements are shown V
in the figure below. Determine the type of 2
the element (R, L, or C) and the value of R, 1
L, or C. Re (V or A)
I 0 1 2
a) =1000 rad/sec 3 Im (V or A)
V
2

1 I

Re (V or A)
b) =3000 rad/sec 3 Im (V or A)
0 1 2
I 2

1
Re (V or A)
0 1 2
b) =3000 rad/sec 3 Im (V or A)
V V
2

I
1
Re (V or A)
c) =20000 rad/sec 3 Im (V or A)
0 1 2
2

1
Re (V or A)
I 0 1 2 V
c) =20000 rad/sec 3 Im (V or A)

1
Re (V or A)
0 1 2
I Problem 8.48*. The following MATLAB
script plots the real-valued signals in
time domain corresponding to the phasor
V
voltage V ¼ 5∠30 ½V and to the phasor
current I ¼ 2∠  60 ½A for an inductor.

Problem 8.47. Phasor voltages and phasor cur- clear all

rents for three unknown circuit elements are f ¼ 2e6; % frequency, Hz
shown in the figure below. Determine the type of T ¼ 1/f; % period, sec
the element (R, L, or C) and the value of R, L, or dt ¼ T/100; % sampling int.
C when appropriate. t ¼ [0:dt:2.5*T]; % time vector
vL ¼ 5*cos(2*pi*f*t+pi/6); % voltage
iL ¼ 2*cos(2*pi*f*t-pi/3); % current
t ¼ t/T; % time in periods

VIII-441
Chapter 8 Steady-State AC Circuit Fundamentals

plot(t, vL, ’b’); Problem 8.51. Find Zeq in the polar form for
grid on; hold on; the circuit element combination shown in the
plot(t, iL, ’r’); figure below when ω ¼ 100, 000 rad=s,
xlabel(’t/T’); C ¼ 100 nF, L ¼ 1 mH, R ¼ 100 Ω.
ylabel(’voltage/current’)

Modify the script and plot the real-valued volt- Z eq R C L

ages and currents corresponding to the phasors
shown in the phasor diagrams for Problem 8.46.

Problem 8.49. Repeat the previous problem for Problem 8.52. For three circuit element com-
the phasors shown in the phasor diagrams for binations shown in the figure below,
Problem 8.47. find Zeq given that ω ¼ 2000 rad=s, C ¼ 5 μF,
L ¼ 50 mH, R ¼ 1 kΩ.
8.3 Principles of AC Circuit a)

Analysis Z eq R C L
8.3.1 AC Circuit Analysis: KVL, KCL,
and Equivalent Impedances b)
8.3.2 Complete Solution for an AC R
Circuit: KVL and KCL on Phasor Z eq C L

Diagram
Problem 8.50. For the AC circuit element com- c) C
binations shown in the figure that follows,
Z eq R L
a) R b) R

Z eq C Z eq L
Problem 8.53. A complex impedance of any
circuit may be written in the form Z ¼ R þ jX
where R is called the resistance and X is called
c) R
the electrical reactance or simply the reac-
tance. An engineer measures a reactance of
Z eq L 2 Ω over an inductor at 60 Hz. What is the
inductance?
C Problem 8.54. The same engineer measures a
reactance of 1 kΩ over a capacitor at 60 Hz.
A. Find the equivalent impedance Zeq in What is the capacitance?
polar form given that
ω ¼ 10000 rad=s, C ¼ 0:1 μF, Problem 8.55. For the circuit shown
L ¼ 100 mH, R ¼ 1 kΩ. in the figure below, υS ðtÞ ¼ 10 cos ωt ½V,
B. Plot the result for the partial impedances ω ¼ 10, 000 rad=s, C ¼ 1 μF, R ¼ 100 Ω:
and for Zeq on the corresponding phasor A. Find phasor current I and phasor volt-
diagram. ages, VR, VC, and construct the voltage
phasor diagram for phasors VR, VC, VS.

VIII-442
Chapter 8 Problems

B. Find voltages across the resistor and L

capacitor, υR(t) and υC(t), as functions -
vL

+
of time.

+
vS(t) + vR R
- -
R i(t)

vR
-
+

+
vS(t) +
- vC
- C Problem 8.60. For the circuit shown in the fig-
i(t) ure, iS ðt Þ ¼ 1 cos ωt ½A and ω ¼ 10, 000 rad=s,
L ¼ 10 mH, R ¼ 100 Ω:
A. Find phasor voltages, VR, VL, V and
Problem 8.56. For the circuit shown in construct the voltage phasor diagram
the figure below, iS ðt Þ ¼ 1 cos ωt ½A and for phasors V, VR, VC.
ω ¼ 10, 000 rad=s, C ¼ 1 μF, R ¼ 100Ω. B. Find voltages across the resistor and
A. Find phasor voltages VR, VC, V and con- inductor, υR(t) and υL(t), as functions
struct the voltage phasor diagram for of time.
phasors V, VR, VC.
B. Find voltages across the resistor and L
capacitor, υR(t) and υC(t), as functions -
vL

+
of time.

+
iS(t) v(t)
- vR R
-
R
i(t)

vR
-
+
+

iS(t) v(t)
- vC
- C Problem 8.61. Repeat Problem 8.59 when
i(t) L ¼ 1:9 mH. Assume the other parameters to
be the same.
Problem 8.57. Repeat problem 8.55 when Problem 8.62. Repeat Problem 8.59 when
C ¼ 2:2 μF. The rest of the parameters are L ¼ 6:8 mH and f ¼ 1000 Hz. The rest of
the same. the parameters are the same.
Problem 8.58. Repeat problem 8.55 when C
¼ 2:2 μF and f ¼ 500 Hz. The rest of the 8.3.3 Source Transformation
parameters are the same. 8.3.4 Thévenin and Norton Equivalent
Problem 8.59. In the circuit shown in the figure Circuits
below, υS ðt Þ ¼ 12 cos ωt ½V, ω ¼ 10, 000 rad=s, 8.3.5 Summary of AC Circuit Analysis at
L ¼ 1 mH, R ¼ 10 Ω: a Single Frequency
A. Find phasor current I and phasor volt- 8.3.6 Multi-frequency AC Circuit Anal-
ages, VR, VL, and construct the voltage ysis: Superposition Principle
phasor diagram for phasors VR, VC, VS. Problem 8.64. A current source with the
B. Find voltages across the resistor and phasor current IN and in parallel with the
inductor, υR(t) and υL(t), as functions impedance ZN shown in the following figure
of time. is equivalent to a voltage source with the phasor

VIII-443
Chapter 8 Steady-State AC Circuit Fundamentals

voltage VT and in series with the impedance Problem 8.67. Find the Thévenin equivalent
ZT. Determine VT and ZT given that circuit, i.e., VT and ZT, for the circuit shown
IN ¼ 2 þ j3 A, ZN ¼ 2  j3 Ω. Express your in the figure below when ω ¼ 377 rad=s, L ¼
result in the polar form. 26:5 mH, R ¼ 10 Ω, C ¼ 500 μF, and V S ðt Þ ¼
10 cos ωt ½V:
a
a

IN ZN
vS(t) +
-
b
b
Problem 8.65. Determine phasor voltage V1 in Problem 8.68. Describe the meaning of the
the AC circuit shown in the following figure superposition principle for multifrequency AC
using the method of source transformation. circuits in your own words.
The impedance values are given at the fre-
quency of interest. Problem 8.69. Find real-valued voltage υR(t)
across the resistor for the circuit in the follow-
V1 ing figure using the superposition principle.
-
+

You are given V 1 ¼ 10 V, V 2 ¼ 1 V. The

5 90 A 2 j10
AC frequencies are ω1 ¼ 377 rad=s and
3 -j5 ω2 ¼ 3ω1 , respectively.

vR
-
+
L=10 mH

10
Problem 8.66. Find the Thévenin equivalent
+ V cos t +
circuit, i.e., VT and ZT, for the circuit shown in - 1 1 V2cos t
2 -
the figure below when ω ¼ 377 rad=s, L ¼ 26:5 -
mH, R ¼ 10 Ω, C ¼ 220 μF, υS ðt Þ ¼
+

10 10
10 cos ωt ½V: V3=10 V DC

vS(t) +
-

VIII-444
Chapter 9

Chapter 9: Filter Circuits: Frequency

Response, Bode Plots, and Fourier
Transform

Overview
Prerequisites:
- Knowledge of complex arithmetic
- Knowledge of superposition principle for linear circuits (Chapter 3)
- Knowledge of harmonic voltage and current behavior (Chapter 8)
- Knowledge of phasor/impedance method for AC circuit analysis (Chapter 8)
- Knowledge of an operational amplifier with negative feedback (Chapter 5)

Objectives of Section 9.1:

- Establish the concept of a first-order analog filter as a two-port network
- Understand the difference between high-pass and low-pass filters
- Understand the effect of filter termination
- Become familiar with the fundamental filter characteristics including transfer
function, break frequency, roll-off, and high-/low-frequency asymptotes
- Understand the construction of the Bode plot including decibels; become familiar
with some of the jargon used by electrical engineers
- Establish the close agreement between first-order RC and RL filters; become
familiar with the concept of cascaded filter networks

Objectives of Section 9.2:

- Establish the model for the open-loop gain of an operational amplifier as a function
of frequency
- Understand the meaning of datasheet parameters such as unity-gain bandwidth and
gain-bandwidth product
- Establish the model for the closed-loop gain of an operational amplifier as a function
of frequency from first principles
- Find the frequency bandwidth for any practical operational amplifier circuit using
the datasheet

© Springer Nature Switzerland AG 2019 IX-445

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_9
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Objectives of Section 9.3:

- Obtain an introductory exposure to the continuous Fourier transform and be able to
compute the transform for simple examples including the meaning of a sinc function
- Be able to relate continuous and discrete Fourier transform via the Riemann sum
approximation
- Be able to define sampling points of the DFT in both time and frequency domain
- Understand the structure and ordering of the DFT frequency spectrum including its
relation to negative frequencies
- Apply the DFT to a filter with a given transfer function and generate the discrete
frequency spectrum of the output signal
- Apply the DFT to filter operation with input pulse or nonperiodic signals
- Apply the DFT (FFT and IFFT) in MATLAB

Application examples:
- Effect of a load connected to the filter
- Effect of next-stage filter load
- Finding bandwidth of an amplifier circuit using the datasheet
- Selection of an amplifier IC for proper frequency bandwidth
- Numerical differentiation via the FFT
- Filter operation for an input pulse signal
- Converting computational electromagnetic solution from frequency domain to time
domain

Keywords:
Analog filter, RC filter, RL filter, Port, Two-port network, First-order high-pass filter, First-order
low-pass filter, Filter termination, Amplitude transfer function, Phase transfer function, Power
transfer function, Complex transfer function, Frequency response, Break frequency, Half-power
frequency, 3-dB frequency, Corner frequency, Bode plot, Decibel, Roll-off, High-frequency
asymptote, Low-frequency asymptote, Frequency band, Passband, Stopband, Decade, Octave,
Power gain, Open-loop amplifier gain, Unity-gain bandwidth, Gain-bandwidth product, Internal
compensation, Open-loop AC gain, Closed-loop AC gain, Amplifier circuit bandwidth, Fourier
transform continuous (direct inverse Fourier spectrum, direct inverse Fourier spectrum,
bandlimited spectrum, reversal property, sinc function, mathematical properties, amplitude-
modulated signal, Parseval’s theorem, energy spectral density), Fourier transform discrete
(Fast digital signal processing (DSP), sampling points, sampling interval, sampling frequency,
sampling theorem, Riemann sum approximation, rectangle rule, fundamental frequency, direct
(DFT), inverse (IDFT), standard form, reversal property, structure of discrete spectrum, numerical
differentiation, filter operation for pulse signals)

IX-446
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

Section 9.1 First-Order Filter Circuits and Their

Combinations
AC voltage divider circuits (either RC or RL) generally operate as analog filters. They
pass certain voltage signals but stop or cut out other signals, depending on the signal’s
frequency content. The analog filters studied in this section are first-order filters since
they may be described by first-order differential equations—we discussed them in
Chapter 7. The phasor/impedance method is applied to solve the AC circuit, both in
analytical and in numerical form. Although over the years the value of the numerical
analysis has greatly increased in engineering, the analytical method remains important if
we are interested in a parametric study such as a rigorous filter analysis. The analytical
method involves (multiple) conversions of complex numbers or expressions from the
rectangular to polar form and vice versa. Generally, division and multiplication are better
carried out in polar form, whereas addition and subtraction require a rectangular form.

9.1.1 RC Voltage Divider as an Analog Filter

The RC voltage divider circuit shown in Fig. 9.1a is perhaps the oldest and best-known
version of an analog filter. In order to understand its operation, we must obtain a general
solution to the RC circuit in Fig. 9.1a. Even though the solution has to work at any
frequency f or angular frequency ω of interest, it is not difficult to find.

a) R b) ZR

VR
vR
- -
+

+
+

+ VS +
vin(t)=vS(t)
- vC
- C - VC
- ZC
i(t) I

Fig. 9.1. RC voltage divider circuit and its solution by the phasor method. We note the phasors for
the voltages VS, VR, VC and the phasor for the circuit current I.

What is an analog filter? The goal of the filter is to accept an AC voltage signal at its input
and either pass the signal to the next circuit block or to stop (or “cut out”) the signal,
depending on its particular frequency. Imagine a human voice that is mixed with white noise
with a spectrum extended over all frequencies. If the noise level is high, we will probably
only hear screaming at high frequencies. However, if we only pass the voltage signals with
frequencies below 3 kHz, where most of the voice power is concentrated, the resulting total
signal will be much clearer for listening. The analog filter is an electric circuit which, in its
simplest form, is identical to the circuit in Fig. 9.1.

IX-447
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

General Solution
Let us first convert the circuit in Fig. 9.1a to a phasor form as shown in Fig. 9.1b. We
assume that υS ðt Þ ¼ V m cos ωt; therefore VS ¼ V m . Next, we solve the resulting “DC
circuit” in the complex domain. The voltage division yields
1
ZC jωC 1 1
VC ¼ Vm ¼ Vm ¼ Vm ¼ V m ½V ð9:1aÞ
ZR þ ZC R þ jωC
1 1 þ jωRC 1 þ jω τ
ZR R jωRC jωτ
VR ¼ Vm ¼ Vm ¼ Vm ¼ V m ½V ð9:1bÞ
ZR þ ZC R þ jωC
1 1 þ jωRC 1 þ jω τ

where τ ¼ RC is exactly the same time constant that appears for transient circuits in
Chapter 7. Converting Eq. (9.1a) and (9.1b) into polar form gives
1
VC ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ∠φC , φC ¼  tan 1 ðωτÞ ð9:1cÞ
1 þ ðωτÞ2
ωτ π
VR ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ∠φR , φR ¼  tan 1 ðω τÞ ð9:1dÞ
2
1 þ ðωτÞ2

After the polar form has been obtained, the real-valued voltages are found in the form
1
υC ðt Þ ¼ V mC cos ðωt þ φC Þ ½V, V mC ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiV m ½V
1 þ ðωτÞ2
ωτ ð9:1eÞ
υR ðt Þ ¼ V mR cos ðωt þ φR Þ ½V, V mR ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ½V
1 þ ðωτÞ2

The general solution of the RC circuit in Fig. 9.1a is now complete. The key observations
are that the amplitudes of the resistor voltage and the capacitor voltage now become
functions of frequency.

Qualitative Analysis
The circuit in Fig. 9.1 is a voltage divider. The supply voltage (or the input voltage to the
filter) is divided between the capacitor and the resistor. Which voltage dominates at low
frequencies and which at high frequencies? To answer those questions, we consider
Eq. (9.1e). When ω ! 0,

IX-448
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

1
V mC ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiV m ! V m
1 þ ðωτÞ2
ωτ ð9:2aÞ
V mR ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ! 0
1 þ ðωτÞ2

Therefore, at low frequencies, the capacitor voltage dominates; it is approximately equal

to the supply voltage. This fact is quite clear because the capacitor acts like an open
circuit for DC, implying that the capacitor voltage “sees” nearly all the supply voltage.
On the other hand, when ω ! 1,
1
V mC ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiV m ! 0
1 þ ðωτÞ2
ωτ ð9:2bÞ
V mR ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ! V m
1 þ ðωτÞ2

Therefore, at high frequencies, the resistor voltage dominates; it approximately equals the
supply voltage. This fact is also easy to understand because the capacitor acts like a short
circuit for a high-frequency AC, jZC j ¼ 1=ðωC Þ ! 0, so that the capacitor voltage is
nearly zero and all the supply voltage is “seen” by the resistor.

Filter Concept: Two-Port Network

Now, we will explore the concept of an analog low-pass RC filter. We consider the power
supply AC voltage as the input voltage υin(t) into the filter. We consider the capacitor voltage
as the output voltage υout(t) of the filter. According to Eq. (9.2a, 9.2b),

υout ðt Þ  υin ðt Þ at low frequencies

υout ðt Þ  0 at high frequencies

The circuit so constructed passes voltage signals with lower frequencies (like the human
voice) but cuts out voltage signals with higher frequencies (like noise). Figure 9.2 on the
left depicts the corresponding circuit transformation. This transformation implies that the
input voltage is acquired from another circuit block and the output voltage is passed to
another circuit block. The qualitative filter description is complete. You should note that
both circuits on the right of Fig. 9.2 are called two-port networks. A port is nothing else
but a pair of voltage terminals, either related to the input voltage or to the output voltage,
respectively. Can we construct an RC filter that passes high frequencies but cuts out low
frequencies? In other words, can we create a so-called high-pass filter? The solution is
simple and elegant; the output voltage is now the resistor voltage, not the capacitor
voltage. Figure 9.2b shows the corresponding circuit transformation.

IX-449
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

low-pass analog RC filter a)

R R

+
+
vin(t)=vS(t) + vin(t) vout(t)
- vC
- C C
- -

high-pass analog RC filter b)

C C

+
+
vin(t)=vS(t) + vR R vin(t) R vout(t)
- - - -

Fig. 9.2. (a) Transformation of the series RC circuit into the low-pass analog RC filter.
(b) A similar transformation into the high-pass RC filter.

Example 9.1: The input voltage to the low-pass filter in Fig. 9.2a is a combination of three
harmonics: υin ðt Þ ¼ 10 cos ω1 t þ 10 cos ω2 t þ 10 cos ω3 t ½V, each with an amplitude of
10 V. The filter has the following parameters: C ¼ 530 nF, R ¼ 100 Ω. Determine the
output voltage υout(t) given that:
1. f 1 ¼ 20 Hz (lower frequency of the acoustic range)
2. f 2 ¼ 3000 Hz (frequency below which most of the acoustic power is present)
3. f 3 ¼ 20, 000 Hz (higher frequency of the acoustic range)
Solution: The key is the superposition principle, which is based on circuit linearity. Using
the superposition principle, we apply Eq. (9.1e) (and Eq. (9.1c) reporting the phases) for
the capacitor voltage (the output voltage to the filter) to each harmonic separately and
then find the sum of three partial solutions. This will be the filter output voltage, which is
given by

υout ðt Þ ¼ 10:00 cos ðω1 t  0:4 Þ þ 7:07 cos ðω2 t  45:0 Þ

þ 1:48 cos ðω3 t  81:5 Þ ½V ð9:3Þ

The filter reduces the amplitudes of higher-frequency harmonics and simultaneously

creates a certain phase shift. The high-pass filter operates in an opposite manner. The
phase shift becomes positive.

Example 9.2: Solve Example 9.1 using MATLAB.

Solution: The text of the corresponding MATLAB is listed below. It is vectorized in the
sense that any number of input harmonics may be taken into consideration:

IX-450
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

Example 9.2 (cont.):

Vm = [10 10 10]; % input voltage amplitudes, V

f = [20 3000 20000]; % input voltage frequencies, Hz
omega = 2*pi*f; % angular frequencies, rad/sec
R = 100; % resistance, Ohm
C = 530e-9; % capacitance, F
tau = R*C;
VmC = 1./sqrt(1+(omega*tau).^2).*Vm % output voltage ampl., V
phiC = - atan(omega*tau)*180/pi % output phases in deg

Exercise 9.1: The input voltage to a low-pass filter circuit is a combination of two
harmonics, υin ðt Þ ¼ 2 cos ω1 t þ 2 cos ω2 t, with the amplitude of 2 V each. The filter has
the following parameters: R ¼ 100 kΩ and C ¼ 1:59 nF. Determine the output voltage
υout(t) to the filter given that f 1 ¼ 100 Hz and f 2 ¼ 100 kHz.
Answer: υout ðt Þ ¼ 1:99 cos ðω1 t  5:7 Þ þ 0:02 cos ðω2 t  89:4 Þ ½V.

Application Example: Effect of a Load Connected to the Filter

The initial excitement about the simplicity of the theoretical filter model often fades
quickly once we try to construct the filter circuit of Fig. 9.2a or Fig. 9.2b in the laboratory.
And the circuit does not work. The major reason for this is the effect of a load connected
to the filter. We consider the low-pass filter in Fig. 9.3.

R
+

vin (t) C vout (t) Load

- -

Fig. 9.3. A generic load connected to the low-pass RC filter.

To solve the circuit with the load, we need to apply the phasor method. The input
voltage is now divided between the resistance R and the parallel combination of the
capacitor impedance and the load resistance, RL. Instead of Eq. (9.1c), we will have
 
1 1 ωτ
VC ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi V m ∠φC ½V, φC ¼  tan ð9:4Þ
2 2 1 þ R=RL
ð1 þ R=RL Þ þ ðωτÞ

The proof of this result is suggested in Problems 9.5 and 9.6. The necessary condition for
proper filter operation (both high pass or low pass) is that the filter termination resistance

IX-451
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

should be much greater than the filter’s resistance R. Put in approximate mathematical
terms: R=RL  1. The low-resistance load (e.g., a loudspeaker) would simply short out
the capacitor! To avoid this effect, a buffer amplifier may have to be inserted between the
load and the filter.

9.1.2 Half-Power Frequency and Amplitude Transfer Function

Low-Pass Filter
We are going to show how to construct a low-pass RC filter for a particular application.
The design engineer needs to know at approximately which frequency the signal should
be cut out. It is a common agreement to choose this frequency so that the amplitude of the
pffiffiffi
output voltage is exactly 1= 2  0:707 of the input voltage amplitude Vm. In other
words, the output filter power, which is proportional to the square of the output voltage,
becomes exactly half of the input power. The corresponding frequency is called the break
frequency or half-power frequency of the low-pass filter. According to Eq. (9.1e), the
break frequency f b is found using the amplitude of the output (capacitor) voltage in the
following way:
1 1 1 ω 1 1
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ¼ pffiffiffi ) ωb τ ¼ 1 ) ωb ¼ ) f b ¼ b ¼ ¼ ½Hz ð9:5Þ
2 τ 2π 2πτ 2πRC
1 þ ðω b τ Þ2

Expressed in terms of the break frequency, the amplitude of the output voltage or the
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
voltage across the capacitor in Eq. (9.1e), has the form V m = 1 þ ð f =f b Þ2 since
ωτ ¼ f =f b . With the input voltage amplitude to the filter being Vm, the ratio of the two
amplitudes is the amplitude transfer function of the low-pass filter Hm. This transfer
function is given by
1
H m ðf Þ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  1 ð9:6aÞ
1 þ ð f =f b Þ2

We note that the transfer function is dimensionless (or has the units of V/V). For a
given input voltage, the amplitude transfer function allows us to determine the output
voltage amplitude. The behavior of Eq. (9.6a) is such that the amplitude transfer function
is always less than one: the output voltage cannot exceed the input voltage.

High-Pass Filter
The break frequency, ωb or fb, of the high-pass filter has the meaning of reducing the
pffiffiffi
voltage amplitude by a factor of 1= 2 and reducing the signal power by the factor of ½.
According to Eq. (9.1e) for the resistor voltage, it is found using the equality
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi pffiffiffi
ωb τ= 1 þ ðωb τÞ2 ¼ 1= 2, which gives us exactly the same value as the break

IX-452
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

frequency for the low-pass filter; see Eq. (9.5). In other words, the definitions of the break
frequency or the half-power frequency coincide for the first-order low-pass filter and the
first-order high-pass filter, respectively. In terms of the break frequency, the amplitude of
the output voltage to the high-pass filter, the voltage across the resistor in Eq. (9.1e), has
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
the form V m f =f b = 1 þ ð f =f b Þ2 , whereas the input voltage amplitude to the filter is still
Vm. The ratio of the two amplitudes is the amplitude transfer function of the high-pass
filter, denoted here by the same letter Hm. This transfer function is given by

f =f b
H m ð f Þ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi1 ð9:6bÞ
1 þ ð f =f b Þ2

We note again that Hm cannot exceed 1. The implication is that the output voltage is
always less than or equal to the input voltage; the filter cannot amplify the input.

Example 9.3: With the values of C ¼ 530 nF, R ¼ 100 Ω, determine the break frequency
of both the low-pass RC filter and the high-pass RC filter, respectively.
Solution: We utilize the definition of Eq. (9.5), f b ¼ 2πRC1
½Hz, and obtain f b ¼ 3:00 kHz
in either case. This is exactly why the particular signal at 3 kHz in Eq. (9.3) of Example 9.1
(the example uses the same parameters) was reduced by a factor of 0.707 at the output of
the low-pass filter. If a high-pass filter were used, the corresponding output signal would
have exactly the same form but with the phase shift of +45 instead of 45 .

Exercise 9.2: The input signal to a high-pass RC filter includes a 180-Hz component. Its
amplitude is to be reduced by a factor of 10. What break frequency should the filter have?
Answer: 1791 Hz.

9.1.3 Bode Plot, Decibel, and Roll-Off

The Bode plot displays the amplitude transfer function defined by Eq. (9.6a, 9.6b) as a
function of frequency on a logarithmic scale. It was first suggested by an electrical
engineer and mathematician Hendrik Wade Bode (1905–1982), Bell Labs, NJ, USA.
The advantage of the logarithmic scale is the ability to simultaneously observe the (very
large) function variations at small and large frequencies. Furthermore, you can more
clearly see the roll-off of the transfer function as a straight line (asymptote). This is
impossible to see when using the linear scale. As the x-variable, we will always choose
frequency f (and avoid the angular frequency ω). As the y-variable, we plot the logarith-
mic function

IX-453
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

H m ð f ÞdB ¼ 20log10 H m ð f Þ ½dB ð9:7Þ

The dimensionless units for the amplitude transfer function in Eq. (9.7) are decibels or
dB. Figure 9.4 shows the Bode plot for transfer function Eq. (9.6a) with f b ¼ 100 Hz.
The selected values of the transfer function are given in Table 9.1 and where the last row is
given in dB. The particular values of the resistance and capacitance are yet to be found;
only their combination τ ¼ RC ¼ 1=ð2πf b Þ ¼ 1:6 ms is really important for the Bode plot.
Despite the apparent simplicity of this operation, the Bode plot for an RC filter is a very
likely question on the entrance exam for an industrial position in electrical engineering.

Table 9.1. Values of amplitude transfer function for a low-pass filter with f b ¼ 100 Hz.
f, Hz 1 10 100 1000 104 105 106 107
Hm( f ) 1.000 0.995 0.707 1.0  101 1.0  102 1.0  103 1.0  104 1.0  105
20 log10Hm( f ) 0.0004 0.0432 3.0103 20.043 40.000 60.000 80.000 100.00

Historical: The decibel is named in honor of Alexander Graham Bell (1847–1922), a

Scottish scientist and inventor who later became a professor at Boston University,
MA. Bell invented the first practical telephone at the age of 28 (US Patent 174,465) and
very quickly became a millionaire. His father-in-law Gardiner Greene Hubbard founded
the Bell Telephone Company in 1878, which subsequently transformed into American
Telephone & Telegraph Company (AT&T).

Hm, dB Bode plot

0
3dB or 70.7% amplitude
attenuation
-20

-40
roll-off of 20dB
per decade
-60

passband
-80

decade
-100

fb
100 101 102 103 104 105 106 107
Frequency of input voltage, Hz

Fig. 9.4. Construction of a Bode plot for the amplitude transfer function of a low-pass RC filter
with break frequency of f b ¼ 100 Hz.

A legitimate question to ask is what is the meaning of the factor 20 in Eq. (9.7)? The
answer is based on the equality 20log10 H m ð f Þ ¼ 10log10 H 2m ð f Þ where H2m ( f ) is not the

IX-454
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

amplitude transfer function but rather attempts to represent power, which is proportional
to voltage squared for a resistor. Therefore, Eq. (9.7) in fact attempts to plot the power
transfer function even though the capacitor in Fig. 9.2a does not consume any power in the
average sense, see Chapter 11. Also note that, when f ¼ f b , the transfer function in Table 9.1
is approximately 3 dB. Therefore, the break frequency is also called the 3-dB frequency
for obvious reasons. Another name, the corner frequency, will be explained shortly.
The interval on the Bode plot for which the frequencies differ by a factor of 10 is called
a decade. Every division on the x-axis in Fig. 9.4 is one decade. The transfer function for
any first-order low-pass filter decreases by 20 dB per decade or has the 20-dB-per-decade
roll-off as seen in Fig. 9.4. This not only occurs away from the break frequency, i.e., when
f f b , but it is also approximately valid in the interval from fb to 10fb; see Table 9.1.
Note that an interval of frequencies is called the frequency band. The roll-off of 20 dB per
decade (or equivalently, the slope of -20 dB per decade) means that the output amplitude
of the filter decreases by a factor of 10 per decade (see Table 9.1), whereas the output
power decreases by the factor of 100. Figure 9.4 shows a frequency band from 0 to fb,
which is the passband of the low-pass filter. The passband is the range of frequencies that
are passed through a filter without being (significantly) attenuated. The opposite of the
passband is the stopband. The required attenuation within the stopband may be specified
between 20 and 120 dB as compared to the value of 0 dB, which means no attenuation.
Besides the decade, the relative frequency interval of one octave is sometimes used. In
this interval, the frequencies differ by the factor of 2, not 10. For example, a TV antenna
that has the bandwidth of one octave (400–800 MHz) may be used to receive most of the
(digital) commercial TV channels in the USA. It can be shown that the RC filter has a
6-dB-per-octave roll-off, away from the break frequency.

Historical: The career of Hendrik Wade Bode (1905–1982), a pioneer of modern control
theory and electronic telecommunications, gives us an example of how important it is to
have a comprehensive education in calculus and a solid background in electrical engineer-
ing. A graduate of the Ohio State University (BS in mathematics at the age of 19 and then
MS in mathematics two years later), Hendrik Bode started his job at Bell Labs as a designer
of electronic filters and invented the asymptotic plots we now call them Bode plots in 1938.
These plots have proven to be extremely useful in feedback control theory. Today, any
electrical engineer who works with amplifiers and their frequency responses is relying on
Bode plots. Some consider Bode a pioneer of robotics as well, based on his invention of
robotic antiaircraft artillery during WWII.

Exercise 9.3: The following values of the amplitude transfer function are given:
H m ð f Þ ¼ 0:707, 0:0707, and 0:00707. Find the corresponding values of Hm( f )dB.
Answer: 3.01 dB, 23.01 dB, and 43.01 dB.

IX-455
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Example 9.4: Design a medium-frequency-range RC low-pass filter (LPF) that has a break
frequency of 1 kHz. The filter load has the resistance of R ¼ 100 kΩ. Create the amplitude
Bode plot in the range from 10 Hz to 100 kHz. Label the filter passband. Repeat the same task
for the high-pass filter (HPF).
Solution: The condition f b ¼ 1=ð2πRC Þ yields C ¼ 1=ð2πRf b Þ ¼ 1:6 nF. The Bode plot
may be generated by finding transfer function values for (at least) every decade and filling
out a table similar to Table 9.1. The result is shown in Fig. 9.5a. The passband is the
frequency band from 0 to fb. For the high-pass filter, we repeat the same steps but replace
the transfer function given by Eq. (9.6a) by the transfer function given by Eq. (9.6b).
The result is shown in Fig. 9.5b. The passband extends from fb to infinity and is only
limited by the upper frequency of the Bode plot. Note that the Bode plot for the high-pass
filter has the same form, but it is mirror reflected about the break frequency. This is another
advantage of the logarithmic scale.
Figure 9.5 indicates that the amplitude response of both the low-pass filter and the high-
pass filter follows two straight lines, which are known as high-frequency and low-frequency
asymptotes. The corner between them is the break frequency, also called the corner fre-
quency. Note that, for the high-pass filter, the meaning of high-frequency and low-frequency
asymptotes is interchanged in Fig. 9.5b.

a) b)
Bode plot - low-pass filter Bode plot - high-pass filter
Hm, dB Hm, dB
0
low-frequency high-frequency
-5 asymptote high-frequency low-frequency asymptote
asymptote asymptote
-10

-15
passband passband
-20

-25

-30

-35

-40
fb fb
-45
101 102 103 104 105 101 102 103 104 105
frequency, Hz frequency, Hz

Fig. 9.5. (a) Bode plot for the amplitude transfer function of the low-pass RC filter with the break
frequency f b ¼ 1 kHz. (b) The same Bode plot but for the high-pass RC filter. Note high-
frequency and low-frequency asymptotes.

9.1.4 Phase Transfer Function and Its Bode Plot

According to Eq. (9.1e), it is not only the amplitude but also the phase of the input signal
that undergoes a transformation when the signal is passed through the filter. The phase
transformation is important since different frequencies (or harmonics) of the input signal
may have a certain phase relation that is distorted by the filter. The phase transfer function is

IX-456
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

given by the phase variation of the filter’s output voltage, which is either the capacitor
voltage for the low-pass filter or the resistor voltage for the high-pass filter. From Eq. (9.1c)
for the low-pass filter, the phase transfer function has the form
 
1 1 f
φH ð f Þ ¼  tan ðωτÞ ¼  tan low-pass RC filter ð9:8aÞ
fb

From Eq. (9.1d) for the high-pass filter, the phase transfer function has the form
 
π 1 π 1 f
φH ð f Þ ¼  tan ðωτÞ ¼  tan high-pass RC filter ð9:8bÞ
2 2 fb

where the break frequency is given by Eq. (9.5).

Example 9.5: Generate the phase Bode plots for the low-pass filter and the high-pass
filter, respectively, with the same break frequency f b ¼ 1 kHz. The frequency band is from
10 Hz to 100 kHz.
Solution: The phase Bode plots in Fig. 9.6 may be generated by calculating the phase
transfer function according to Eq. (9.8a, 9.8b) for (at least) every decade. The result is
shown in Fig. 9.6. You can see that the Bode plots only differ by a phase shift of 90 .
Alternatively, a MATLAB script may be used:

f = logspace(1, 5); % frequency vector, Hz (from 10^1 to 10^5 Hz)

fb = 1000; % break frequency, Hz
phiH1 = -atan(f/fb); % low-pass filter phase transfer function
phiH2 = pi/2-atan(f/fb); % high-pass filter phase transfer function
semilogx(f, phiH1/pi*180); grid on;
title('Bode plot'); ylabel('phase transfer function, deg'); xlabel('f, Hz')

9.1.5 Complex Transfer Function: Cascading Filter Circuits

The complex transfer function of the filter, H( f ), is often called the frequency response of the
filter. It describes not only the amplitude transformation but also the phase transformation.
The transfer function now becomes a complex expression. It is equal to the ratio of two
phasors; specifically, it denotes the ratio of the output phasor voltage to the input phasor
voltage. The low-pass filter has the form of Fig. 9.1 with the input voltage equal to the supply
voltage and the output voltage equal to the capacitor voltage. Its complex transfer function is
given by Eq. (9.1c) divided by Vm. The high-pass filter also has the form of Fig. 9.1 with the
input voltage equal to the supply voltage and the output voltage equal to the resistor voltage.
Its complex transfer function is given by Eq. (9.1d) divided by Vm. Thus, we obtain

IX-457
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

a) b)
φH, deg Bode plot - low-pass filter φH, deg Bode plot - high-pass filter
0 90

-10 80

-20 70

-30 60

-40 50

-50 40

-60 30

-70 20

-80 10
fb fb
-90 0
1 2 3 4 5 1 2 3 4 5
10 10 10 10 10 10 10 10 10 10
frequency, Hz frequency, Hz

Fig. 9.6. Comparison of the phase Bode plots for (a) the low-pass-filter and (b) for the high-pass filter
with the same break frequency f b ¼ 1 kHz. Both plots are identical to within a phase shift.
8  
>
> 1 1 1 f
>
> ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ∠  tan ; low-pass RC filter
>
< 1 þ jð f =f b Þ fb
1 þ ð f =f b Þ2
Hð f Þ
H m ð f Þ∠φH ¼   ð9:9aÞ
>
> ð f =f b Þ f =f b π 1 f
>
> ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ∠  tan ; high-pass RC filter
>
: 1 þ jð f =f b Þ 2 fb
1 þ ð f =f b Þ2

This is consistent with Eqs. (9.6a, b) and (9.8a, b), respectively. Given the phasor of input
voltage Vin, the phasor of the output voltage is simply expressed by

Vout ¼ Hð f ÞVin ð9:9bÞ

Equation (9.9b), which is valid for any linear electronic filter and other linear systems,
fully describes the filter operation and has great practical value.

Example 9.6: For a low-pass RC filter with the values C ¼ 530 nF, R ¼ 100 Ω,
determine the output voltage in time domain when the input voltage is given by
υin ðtÞ ¼ 1 cos ðωt þ 30 Þ ½V where ω ¼ 2π  3000 rad=s.
Solution: The break frequency of the low-pass filter is f b ¼ 3:00 kHz, which coincides
with the signal frequency in this particular case. According to first Eq. (9.9a) and
Fig. 9.6a, at that frequency, Hð f Þ ¼ p1ffiffi2 ∠  45 ; therefore, the output voltage has the
form Vout ¼ p1ffiffi ∠  15 or υout ðt Þ ¼ 0:71 cos ðω t  15 Þ ½V. The same analysis may be
2
applied at any frequency and phase of the input harmonic voltage signal.

Another advantage of the complex transfer function lies in the fact that the series or
cascade combination of any number of filters (or two-port networks) shown in Fig. 9.7
has a transfer function that is simply the product of the corresponding transfer functions:

IX-458
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

Hð f Þ ¼ H1 ð f ÞH2 ð f Þ ) H m ð f Þ ¼ H m1 ð f ÞH m2 ð f Þ ð9:10Þ
In this manner, a more advanced filter may be constructed from the individual filter
blocks. To prove Eq. (9.10), we state that the phasor for the intermediate output voltage
υout1(t) in Fig. 9.7 is given by Vout1 ¼ H1 ð f ÞVin. Hence, the phasor for the output voltage
υout(t) in Fig. 9.7 becomes Vout ¼ H1 ð f ÞVout1 ¼ H1 ð f ÞH2 ð f ÞVin which is equivalent to
Eq. (9.10). Due to the logarithmic scale of the Bode plot, the product in Eq. (9.10) is
replaced by the sum of two contributions when the decibel scale is used:
H m ð f ÞdB ¼ H m1 ð f ÞdB þ H m1 ð f ÞdB ð9:11Þ
Thus, we simply add up two magnitude transfer functions in dB and obtain the resulting
magnitude transfer function also in dB.

C1 H1(f) R2 H2(f)
+

+
vin(t) R1 vout1(t) C2 vout(t)
- - -

Fig. 9.7. Cascading a high-pass and a low-pass filter into a more complex filter structure.

Application Example: Effect of Next-Stage Filter Load

Equation (9.10) requires great care. For example, the equivalent impedance seen by the
leftmost high-pass filter stage in Fig. 9.7 should be much greater than R1; otherwise this
stage will not operate as expected, and Eq. (9.10) will be inaccurate. In other words, a
following filter stage should not appreciably load the previous one.

Example 9.7: For the combined circuit in Fig. 9.7, create the Bode plot for the transfer
function of the cascade connection in the frequency band from 1 Hz to 1 MHz. You are
given R1 ¼ 159:1 Ω, C 1 ¼ 10 μF and R2 ¼ 159:1 Ω, C 2 ¼ 0:1 μF.
Solution: The break frequency of the high-pass filter is calculated as 100.0 Hz, and the
break frequency of the low-pass filter is found to be 10.0 kHz. The combined Bode plot is
generated using Eqs. (9.6a, 9.6b) and (9.10). Alternatively, the transfer functions in dB,
specified by Eq. (9.7), may be added. The result is a band-pass filter as shown in Fig. 9.8
by the solid curve. This result is expected to be accurate only if jR2 þ Z C 2 j R1 . Though
valid at low frequencies below 1 kHz, this inequality is violated above 1 kHz. The exact
transfer function is obtained by solving the complete AC circuit in Fig. 9.7 with the open-
circuited capacitor C2. It is plotted in Fig. 9.8 by a dashed curve. There is clearly a
significant deviation from the solution given by Eq. (9.10) at higher frequencies above
1 kHz. To avoid the loading effect seen in Fig. 9.8, a buffer amplifier may be inserted
between the filter stages shown in Fig. 9.7.

IX-459
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Hm, dB Amplitude Bode plot - combined filter

5

-5

-10

-15

-20

-25

-30

-35

-40

-45
100 101 102 103 104 105 106
frequency, Hz

Fig. 9.8. Solid curve: Eq. (9.10) for the cascaded filters. Dashed curve: the exact solution with the
open-circuited capacitor C2.

In Fig. 9.8, the exact transfer function may exceed 0 dB. In other words, the voltage
gain of the combined (still passive) filter may be greater than one. How is this possible?
The answer is that, in contrast to the circuits in Fig. 9.2, the circuit in Fig. 9.7 is in fact
already a second-order circuit. Second-order circuits may experience a resonance behav-
ior where the circuit voltages across individual elements may (very considerably) exceed
the original supply voltage. This effect, called voltage multiplication, is of great practical
importance and will be considered in detail in Chapter 10 devoted to second-order AC
circuits. Note that that the true power gain of a passive filter of any order and any
topology is always less than one (less than 0 dB). Only electronic amplifiers may have a
positive, and often high, power gain; this is discussed in the next section.

9.1.6 RL Filter Circuits

The RL circuits are used for the same filtering purposes as the RC circuits. Figure 9.9
depicts the concept. It may be demonstrated that the corresponding circuit theory and
Eqs. (9.6a, 9.6b) for the transfer functions become equivalent to first-order RC filter
circuits under the following conditions:
1. The time constant τ ¼ RC is replaced by the time constant τ ¼ L=R, similar to the
corresponding operation for the first-order transient circuits. The break frequency
f b ¼ 1=ð2πτÞ remains the same.
2. The role of the capacitor and inductor are interchanged. For example, the RL circuit
in Fig. 9.9a is a first-order high-pass filter because the inductor voltage, which is
the output filter voltage, is exactly zero for a DC signal. However, it becomes a
first-order low-pass filter if the inductor and resistor positions are interchanged, as
shown in Fig. 9.9b.

IX-460
Chapter 9 Section 9.1: First-Order Filter Circuits and Their Combinations

3. Similarly, the RL circuit in Fig. 9.9b is a first-order low-pass filter simply because
the inductor becomes a short circuit at DC and the DC signal will pass through.
However, it becomes a first-order high-pass filter if the inductor and resistor
positions are interchanged, as shown in Fig. 9.9a.
Furthermore, the filter specifications might require large inductance values, which lead
to physically large inductor sizes.

a) high-pass analog RL filter

R R

+
vin(t)=vS(t) + + vin(t) vout(t)
- vC
- L L
- -

b) low-pass analog RL filter L

L

+
+

vin(t)=vS(t) + vin(t) R vout(t)

- vR
- R
- -

Fig. 9.9. (a) Transformation of a series RL circuit into a high-pass analog RL filter. (b) Similar
transformation into the low-pass RL filter.

Example 9.8: For the two filter circuits in Fig. 9.9, create the amplitude Bode plots in the
frequency band from 10 Hz to 100 kHz. You are given R ¼ 31:4 Ω, L ¼ 5 mH.
Solution: The break frequencies of the high-pass filter and the low-pass filter in Fig. 9.9
coincide. In either case, we obtain f b ¼ 1=ð2πτÞ, τ ¼ L=R ¼ 1:59 104 s. Thus,
f b ¼ 1:00 kHz. The Bode plots may be generated by finding transfer function values
based on Eqs. (9.6) for (at least) every decade and filling out a table similar to Table 9.1.
The result is shown in Fig. 9.10 along with high- and low-frequency asymptotes. We
again observe the 20-dB roll-off per decade. The Bode plots given in Fig. 9.10 coincide
with the Bode plots for RC filters having the same break frequency, see Fig. 9.5.
However, given an identical component topology, the filter function is interchanged.

IX-461
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

a) b)
Bode plot - high-pass RL filter Bode plot - low-pass RL filter
Hm, dB Hm, dB
0
high-frequency low-frequency
-5 low-frequency asymptote asymptote high-frequency
asymptote asymptote
-10

-15
passband passband
-20

-25

-30

-35

-40
fb fb
-45
1
101 102 103 104 105 10 102 103 104 105
frequency, Hz frequency, Hz

Fig. 9.10. (a) Bode plot for the amplitude transfer function of the high-pass RL filter with break
frequency f b ¼ 1 kHz. (b) The same Bode plot but for the low-pass RL filter.

Exercise 9.4: An RL filter circuit in Fig. 9.9a has R ¼ 100 Ω and L ¼ 1 mH. Establish
the capacitance value of an equivalent RC filter, given that the resistances are the same in
both cases.
Answer: 100 nF.

IX-462
Chapter 9 Section 9.2: Bandwidth of an Operational Amplifier

Section 9.2 Bandwidth of an Operational Amplifier

The operational amplifier circuits introduced earlier are implicitly assumed to operate
equally well for any frequency of the input signal. In reality this is not true. An
operational amplifier may operate only over a certain frequency band, and the associated
frequency bandwidth is perhaps the most critical device parameter. Frequently we do not
realize how severe this limitation can be and how difficult it is to build a high-frequency
or radio-frequency amplifier. As an example, we should point out that none of the
common amplifier ICs studied in introductory ECE classes can be used as a front-end
amplifier for an AM radio receiver (520–1610 kHz), even if the noise levels were low.
Indeed, high-frequency amplifiers with larger frequency bandwidths exist. A case in point
is the accessible LM7171 chip. Key to understanding the amplifier frequency behavior is
the theory of the first-order RC filters developed in the previous section.

9.2.1 Bode Plot of the Open-Loop Amplifier Gain

Open-Loop Amplifier Gain and Its Relation to the Previous Results
The (amplitude) frequency response of an operational amplifier is simply a plot of its gain
magnitude versus frequency of the input AC voltage signal. This response is usually a
Bode plot. The problem is that the gain of the amplifier (both open loop and closed loop)
generally decreases with increasing frequency. We consider the open-loop gain (gain
without the feedback loop) first. The open-loop gain magnitude will be denoted here by
AOL ¼ AOL ð f Þ. Note that in Chapter 5 we have already introduced the open-circuit gain,
A, of an amplifier at DC without the feedback loop. What is the relation between AOL and
A introduced previously? The answer is given by the equality A ¼ AOL ð f ¼ 0Þ as long as
the amplifier is open circuited.

Open-Loop Gain Behavior

The open-loop gain decreases with increasing frequency of the input signal. Figure 9.11
shows the frequency response of an open-loop amplifier on a log-log scale. You may
recall that the log-log scale used in this figure is simply the Bode plot introduced in the
previous section. This figure is typical for the LM741 amplifier IC and similar general-
purpose devices. Comparing the Bode plot in Fig. 9.11 with the Bode plot of the RC filter
in Fig. 9.4 of the previous section, we discover that the amplifier’s gain as a function of
frequency is virtually identical to the transfer function of the RC filter for the same break
frequency of 10 Hz, as seen in Fig. 9.11! In both cases, we have a roll-off of 20 dB per
decade. Obviously, the scale is different. Why is this so? This occurs because the
amplifier ICs are usually internally compensated, which means incorporating a simple
RC filter network (in practice, it may be a single capacitor C) into the IC chip itself. This
process is called internal compensation of the amplifier. The goal of such a modification
is to ensure that the amplifier circuit will be stable. Stability refers to the amplifier’s
immunity to spontaneous oscillations. These undesired oscillations occur when the input

IX-463
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

frequency excites internal resonances, similar to a mechanical mass-spring system, that

might continue ad infinitum.

106 120
3dB or 0.707A OL (0)

Open-loop gain [dB], 20log10 (AOL)

105 100

Open-loop gain, AOL

20 dB per decade
104 roll-off 80

103 60

102 40

101 20

fb BW
1 0
1 2
1 10 10 103 104 105 106 107
Frequency of input voltage, Hz

Fig. 9.11. Bode plot of the open-loop gain magnitude for the LM741-type amplifier IC. Note the
logarithmic scale on the left and the corresponding scale in dB on the right. The frequency
bandwidth given by the break frequency fb is only 10 Hz.

9.2.2 Unity-Gain Bandwidth Versus Gain-Bandwidth Product

The amplifier gain in Fig. 9.11 decreases by a factor of 0.1 (or 20 dB) gain roll-off per
frequency decade. The decay already starts at a relatively low break frequency of 10 Hz
pffiffiffi
where the DC open-loop gain drops by the factor of 0.707 or 1= 2. The corresponding
pffiffiffi
value in dB is 20log10 1= 2 ¼ 3 dB. The gain continues to decrease further and
reaches unity at the frequency of 1 MHz. This frequency is equal to the unity-gain
bandwidth (BW) of the amplifier, i.e., for the amplifier IC depicted in Fig. 9.11:

BW ¼ 1 MHz ð9:12Þ

A remarkable observation from Fig. 9.11 is that the gain-bandwidth product (sometimes
denoted by GBW or GB in datasheets) remains constant over the band for every particular
gain value. The gain-bandwidth product is equal to the length of every single arrow
(in Hz) in Fig. 9.11 times the corresponding gain value (dimensionless), that is,

f ¼ 102 Hz ) GBW ¼ 102  104 ¼ 106 Hz ¼ BW,

f ¼ 103 Hz ) GBW ¼ 103  103 ¼ 106 Hz ¼ BW, ð9:13Þ
f ¼ 104 Hz ) GBW ¼ 104  102 ¼ 106 Hz ¼ BW;

etc. Thus, the gain-bandwidth product is exactly equal to the unity-gain bandwidth BW;
it is frequently specified in the manufacturer datasheet. In what follows, we will use the
unity-gain bandwidth as the major parameter of interest. Note that instead of, or along

IX-464
Chapter 9 Section 9.2: Bandwidth of an Operational Amplifier

with, the unity-gain bandwidth, the rise time of an amplifier may be specified in the
datasheet. Approximately, we can state that BW ¼ 0:35=rise time ½Hz.

9.2.3 Model of the Open-Loop AC Gain

The open-loop gain dependence on the frequency has the form of a low-pass filter. We
could therefore describe the open-loop gain in a complex form that is identical to the
complex transfer function of the low-pass filter given, for example, by Eqs. (9.9a, b) of
the previous section. The open-loop AC gain in complex phasor form states

AOL ð0Þ
AOL ð f Þ ¼ , AOL ð0Þ is the open-loop DC gain ð9:14Þ
1 þ jð f =f b Þ

For example, AOL ð0Þ ¼ 105 in Fig. 9.11. According to Eq. (9.14), the open-loop AC gain
is a complex-valued frequency-dependent transfer function. This circumstance is
reflected in a phase difference between the output and input voltages. To be consistent
with Fig. 9.11 and with the previous DC amplifier analysis, the magnitude of the complex
gain function in Eq. (9.14) is denoted by the same symbol, AOL, i.e.,

AOL ð0Þ
jAOL j ¼ AOL ð f Þ ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð9:15Þ
1 þ ð f =f b Þ2

The Bode plot applied to Eq. (9.15) will give us exactly the dependence shown in
Fig. 9.11. According to Eq. (9.15), the unity-gain bandwidth satisfies the equality

AOL ð0Þ
1 ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð9:16Þ
1 þ ðBW=f b Þ2
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Since BW =f b 1, one has 1 þ ðBW =f b Þ2  BW=f b with a high degree of accuracy.
Therefore, according to Eq. (9.16),

BW ¼ AOL ð0Þf b ð9:17Þ

Looking at Fig. 9.11, we observe a very significant decrease of the open-loop gain, even
in the audio frequency range. For example, the open-loop gain decreases by a factor of
1000 in the audio range from 10 Hz to 10 kHz. Does it mean that the LM741 or any
general-purpose amplifier cannot be used in this range? The general answer is that the
operational amplifier is mostly used with a negative feedback loop. When the closed-loop
DC gain is not very high (say 10), the corresponding closed-loop AC gain appears to be
nearly constant over a much wider bandwidth (say up to 100 kHz). This critical result will
be proved mathematically shortly.

IX-465
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Example 9.9: The internally compensated LM148-series amplifiers (LM148/248/348)

have a unity-gain bandwidth BW of 1 MHz. The typical large-signal voltage gain at
room temperature reported in the datasheet is 160 V/mV.
A. Find the open-loop DC gain in dB and the open-loop break frequency fb.
B. Find the open-loop gain at 100 Hz, 1 kHz, and 10 kHz.
Solution: The open-loop DC gain is AOL ð0Þ ¼ 160, 000 or 20log10 ð160; 000Þ ¼ 104 dB.
The break frequency may be found from Eq. (9.17):

BW
fb ¼ ¼ 6:25 Hz ð9:18Þ
160, 000
According to Eq. (9.15), the open-loop gain at 100 Hz, 1 kHz, and 10 kHz becomes 104,
103, and 100, which corresponds to 80 dB, 60 dB, and 40 dB.

Exercise 9.5: For an internally compensated amplifier IC, the open-loop DC gain is
120 dB. The break frequency is 100 Hz. Determine the unity-gain bandwidth.
Answer: BW ¼ 100 MHz.

9.2.4 Model of the Closed-Loop AC Gain

Consider a negative feedback amplifier in an inverting configuration, as shown in
Fig. 9.12. Since the open-loop gain significantly decreases with frequency, we can no
longer apply the second summing-point constraint (the differential input voltage is zero),
which was justified based on the condition of the very high (ideally infinite) open-loop
gain. However, the first summing-point constraint of no current into the amplifier is still
valid. Therefore, a direct theoretical derivation of the closed-loop gain can be performed.

R2

Vin R1 Vout
+ V* - I2
Vx + +
I1 - -

Fig. 9.12. Circuit configuration for deriving the frequency-dependent closed-loop gain.

We use the complex open-loop gain given by Eq. (9.14) and employ phasor voltages.
 
Looking at Fig. 9.12, we conclude that AOL 0V  V* ¼ Vout , based on the amplifier
definition. This definition is valid for either real (time-dependent) voltages or complex
phasors. By KCL at the node associated with V*, we can develop

IX-466
Chapter 9 Section 9.2: Bandwidth of an Operational Amplifier

Vin  V* V*  Vout Vin þ Vout =AOL Vout =AOL  Vout

¼ ) ¼ )
R1 R2 R1 R2
  ð9:19Þ
Vin 1 1 1
¼    Vout
R1 AOL R1 AOL R2 R2

It follows from Eq. (9.19) that the output phasor voltage to the amplifier and the closed-
loop amplifier phasor gain ACL become
R2 Vin V R 1
Vout ¼    ) ACL
out ¼  2   ð9:20Þ
R1 1 þ V R
AOL 1 þ R1 1 þ AOL 1 þ R1
1 R2 in 1 1 R2

Next, we substitute Eq. (9.15) into Eq. (9.20) and rearrange terms to obtain the form
R2 1
ACL ð f Þ ¼  h  i   ð9:21Þ
R1 1 þ 1
1 þ RR21 þ AOL1ð0Þ 1 þ RR21 jff
AOL ð0Þ b

The first term in the denominator on the right-hand side of Eq. (9.20) is one with a high
degree of accuracy since AOL ð0Þ  105  108 . This approximation is valid for any
realistic resistor values. Therefore, we again arrive at the first-order low-pass filter
response:

ACL ð0Þ
ACL ð f Þ ¼  ,
closed loop
1 þ j f =f b
ð9:22Þ
R2 AOL ð0Þf b BW
ACL ð0Þ ¼  , f closed loop
¼ ¼
R1 b
1 þ R2 =R1 1 þ R2 =R1

but with a very different break frequency fbclosed loop. A similar treatment holds for the
non-inverting amplifier configuration. The result is identical to Eq. (9.22); however, the
closed-loop DC gain ACL(0) is now given by
R2
ACL ð0Þ ¼ 1 þ ð9:23Þ
R1
9.2.5 Application Example: Finding Bandwidth of an Amplifier Circuit
The relation reported in Eq. (9.22) is perhaps the single most important result with regard
to the AC behavior of operational amplifiers. It reveals that the closed-loop AC gain has
conceptually the same RC filter response as the open-loop gain; see Eq. (9.15). However,

IX-467
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

the corresponding break frequency fbclosed loop is much larger, namely, by a factor of
AOL ð0Þ=ð1 þ R2 =R1 Þ. This implies that the frequency response remains flat up to a very
high frequency. The amplifier bandwidth in the closed-loop configuration coincides with
the break frequency fbclosed loop determined by Eq. (9.22). Therefore, the bandwidth is
directly proportional to the unity-gain bandwidth BW reported in the datasheet and
inversely proportional to the factor 1 þ R2 =R1, which is straightforwardly calculated
using the known values of the feedback resistances.

Example 9.10: An amplifier with the open-loop gain of Fig. 9.11 (AOL ð0Þ ¼ 105 , f b ¼ 10Hz)
is used in the closed-loop inverting configuration with R2 =R1 ¼ 9 (the DC inverting gain
is 9). Create the Bode plot for the gain magnitude ACL( f ), compare this result with
the open-loop gain, and determine the bandwidth of the amplifier.
Solution: According to Eq. (9.22), the gain magnitude is given by

R2 1 105
ACL ð f Þ ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2 , f bclosed loop ¼  10 Hz ¼ 100 kHz ð9:24Þ
R1 closed loop 10
1 þ f =f b

In Fig. 9.13, we plot the closed-loop gain versus the open-loop gain given by Eq. (9.15).
The amplifier bandwidth in the closed-loop configuration is now a respectable 100 kHz.

Exercise 9.6: The unity-gain bandwidth of an amplifier IC is 1 MHz. Determine the

bandwidth of the non-inverting amplifier circuit with a gain of 200.
Answer: 5 kHz.

9.2.6 Application Example: Selection of an Amplifier IC for Proper

Frequency Bandwidth
The required bandwidth and closed-loop gain usually are known to the circuit designer.
Using Eq. (9.22), we can estimate whether or not a specific amplifier IC will meet those
requirements. There is clearly a trade-off between the closed-loop gain and bandwidth
according to Fig. 9.13 and Eq. (9.22). For a given amplifier IC, the lower the closed-loop
gain requirement, the wider the achievable bandwidth.

IX-468
Chapter 9 Section 9.2: Bandwidth of an Operational Amplifier

a)
6
10 120

Open-loop gain vs. closed-loop gain [dB]

open-loop gain 3dB or 0.707A OL (0)

Open-loop gain vs. closed-loop gain

5
10 100
20 dB per decade
4
10 roll-off 80

3
10 60

2
10 40
3dB or 0.707G
1
closed-loop gain
10 20
closed loop
fb fb
1 0
1 2 3 4 5 6 7
1 10 10 10 10 10 10 10
Frequency of input voltage, Hz
b)

closed-loop bandwidth

Fig. 9.13. Closed-loop AC gain ACL( f ) (lower curve) versus open-loop AC gain AOL( f ) (upper
curve) for an inverting amplifier with AOL ð0Þ ¼ 105 and 1 þ R2 =R1 ¼ 10 (the amplifier DC gain
is 9).

Example 9.11: An inverting amplifier with a gain of 20 and bandwidth of at least
20 kHz is needed. Is the LM348 chip appropriate for this purpose?
Solution: From the LM348 datasheet, we obtain BW ¼ 1 MHz. Because the inverting
gain is 20, we should use a ratio of R2 =R1 ¼ 20. According to Eq. (9.22), this gives
f closed
b
loop
¼ 47:6 kHz. The closed-loop 3-dB bandwidth of the amplifier coincides with
this value. Therefore, the LM348 chip is sufficient for our purposes. However, if its gain is
forced to a higher value, say to 100, then the useful bandwidth reduces to 10 kHz.

Exercise 9.7: A non-inverting amplifier with a gain of 31 and a bandwidth of at least

90 kHz is needed. Is an LM741-based amplifier IC appropriate for this circuit?
Answer: No.

IX-469
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Section 9.3 Introduction to Continuous

and Discrete Fourier Transform
9.3.1 Meaning and Definition of Fourier Transform
Consider a phasor as introduced in the previous chapter. This phasor is in fact a
transform. It converts a harmonic sinusoidal time-domain signal into a complex number
for easier, algebraic computation of circuit values. After determining the phasor value of a
voltage or current signal, we transform it back to the time-domain expression. What if we
do not have a pure sinusoidal tone, but an arbitrary voltage pulse υ(t) in the time domain?
Another important example is a bit stream of arbitrary data, which can also be described
by a certain voltage function f(t). Could we still introduce a “phasor” for an arbitrary
signal f(t) in the time domain? The answer is yes; however, instead of a single complex
number, we will have an entire complex function F(ω) of angular frequency ω. This
function essentially consists of individual phasors, corresponding to all possible har-
monic signals, which form the time-domain signal f(t). Mathematically, the direct Fourier
transform (from time domain to frequency domain) is given by
ð
1

F ðωÞ
f ðt Þejωt dt; ð9:25aÞ
1

whereas the inverse Fourier transform (from frequency domain to time domain) is given
by a similar integral
ð
1
1
f ðt Þ
F ðωÞejωt dω ð9:25bÞ
2π
1

The pair of integrals in Eqs. (9.25a, 9.25b) completely describes the Fourier transform.
Function F(ω) is called the Fourier spectrum (or simply the spectrum) of the signal f(t).
This function is generally complex; however, in contrast to the previous convention, we
will not use boldface here in order to preserve the most common mathematical notations.

Exercise 9.8: Establish a relation between F ðωÞ and F(ω) for a real signal f(t), which is
called a reversal property of the Fourier transform.
Answer:

F ðωÞ ¼ F * ðωÞ ð9:26Þ

where the star denotes complex conjugate.

IX-470
Chapter 9 Section 9.3: Introduction to Continuous and Discrete. . .

The spectrum is said to be bandlimited if F(ω) is zero above a certain angular

frequency ωmax. According to Eq. (9.26), this simultaneously means that F(ω) is zero
below ωmax. Many useful signals are approximately bandlimited.

Example 9.12: Derive the Fourier transform of a rectangular pulse in the form of one bit
of data shown in Fig. 9.14a.
Solution: The integral in Eq. (9.25a) is reduced to
Tð=2
V m jωt T =2 sin ωT =2
F ðωÞ ¼ V m ejωt dt ¼ e ¼ V mT ð9:27Þ
jω T=2 ωT =2
T =2

The function sincðxÞ

sin π x=ðπ xÞ is called a sinc function. Using its definition, the final
result for the spectrum has the form
 
ωT
F ðωÞ ¼ V m T sinc ð9:28Þ
2π
and is plotted in Fig. 9.14b using a few lines of MATLAB code:

Vm = 1; % input voltage amplitude, V

T = 1e-6; % pulse duration, s
omega = linspace(-12*pi/T, 12*pi/T); % angular frequency, rad/s
F = Vm*T*sinc(omega*T/(2*pi)); % inverse Fourier transform
plot(omega, F); grid on;

In contrast to the original signal, the pulse spectrum is not bounded and extends to infinity.
This is due to the fact that the original pulse has sharp edges, which are described by
higher-frequency harmonics. The pulse spectrum in the form of a sinc function is famous in
communications theory. Figure 9.15 shows the shape of the sinc function depicted above
the entrance of an electronics store in Silicon Valley, CA.

a) b) F(w)/(VmT)
1

f(t)
Vm

-0.4
0 t -4 -2 0 2 4
-T/2 T/2 w, 10 rad/s
7

Fig. 9.14. (a) Rectangular pulse f(t) of duration T ¼ 1 μs and (b) its Fourier spectrum in the form
of a sinc function.

IX-471
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Fig. 9.15. Fry’s Electronics store in Sunnyvale, Silicon Valley, with an emblem depicting the sinc
function.

9.3.2 Mathematical Properties of Fourier Transform

Major mathematical properties of the Fourier transform follow from its definition and are
listed in Table 9.2.

Table 9.2. Major mathematical properties of Fourier transform.

f(t) F(ω)
Kf(t) KF(ω)
f 1 ðt Þ  f 2 ðt Þ þ f 3 ðt Þ F 1 ðωÞ  F 2 ðωÞ þ F 3 ðωÞ
dnf(t)/dtn ( jω)nF(ω)
ðt 1ð
1
f ðτÞdτ, f ðt Þdt ¼ 0 F ðωÞ
jω
1 1
f(at) ð1=aÞF ðω=aÞ, a > 0
f ð t  aÞ ejωa F ðωÞ

The first two properties follow from Fourier transform linearity. Multiplication of f(t)
by a constant corresponds to multiplying F(ω) by the same constant. Also, addition
(subtraction) in the time domain corresponds to addition (subtraction) in the frequency
domain. The next two properties (differentiation and integration) make the Fourier
transform useful for solving ODEs since the time-domain derivatives and integrals will
correspond to multiplication and division by jω in the frequency domain. The two last
properties (scaling and translation) directly follow from Eq. (9.25a).

Exercise 9.9: The Fourier transform of f(t) is F(ω). What is the Fourier transform of
df ðt Þ=dt þ 5f ðt Þ?
Answer: ð5 þ jωÞF ðωÞ.

IX-472
Chapter 9 Section 9.3: Introduction to Continuous and Discrete. . .

We emphasize that the properties listed in Table 9.2 also apply to the discrete Fourier
transform studied below, but the corresponding indexing of discrete frequencies has to be
carefully arranged.

9.3.3 Discrete Fourier Transform and Its Implementation

Direct Discrete Fourier Transform
Present and future demands are such that we must process continuous signals by discrete
methods. Perhaps the most important method is the discrete Fourier transform (DFT) and
its fast versions: fast Fourier transform (FFT) and inverse fast Fourier transform (IFFT).
Let f(t) be a continuous pulse signal which is the source of the data. We assume that f(t) is
zero outside of the interval 0  t < T . Let f ðt n Þ, n ¼ 0, . . . , N  1 be its values at
N uniformly distributed sampling points t n ¼ ΔT n, n ¼ 0, . . . , N  1 within the
interval of interest. Here
T
ΔT ¼ ð9:29Þ
N
is the sampling interval. Then, the integral of the direct Fourier transform in Eq. (9.25a)
may be found using the rectangle rule (or the Riemann sum approximation)

X
N 1
F ðωÞ ¼ ΔT ejωnΔT f ðt n Þ ð9:30Þ
n¼0

We could in principle evaluate this expression at any value of ω. However, with only N data
points to start with, only N final outputs will be significant. We choose those N uniformly
distributed frequency sampling points as ωm ¼ ω0 m, m ¼ 0, . . . , N  1, where
2π
ω0 ¼ ð9:31Þ
T
is the fundamental frequency (with one period over the interval T). Let F ðωm Þ, m ¼ 0,
. . . , N  1 be the values of F(ω) at the frequency sampling points. Then, Eq. (9.30) gives

X
N 1 2π
F ðωm Þ ¼ ΔT ej N mn f ðt n Þ, m ¼ 0, . . . , N  1 ð9:32Þ
n¼0

Inverse Discrete Fourier Transform (IDFT)

A very similar operation is applied to the integral of the inverse Fourier transform given
by Eq. (9.25b). We first assume that F(ω) is zero outside of the interval 0  ω < N ω0; in
other words, it is bandlimited. Then, the corresponding integral in Eq. (9.25b) is again
approximated using the rectangle rule so that the final result has the form

IX-473
 Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

1 XN 1 2π
f ðt n Þ ¼ e j N mn F ðωm Þ, n ¼ 0, . . . , N  1 ð9:33Þ
N ΔT m¼0

Definition of Discrete Fourier Transform

It is rather inconvenient to keep the factor ΔT in both Eqs. (9.32) and (9.33), respectively.
Therefore, we may introduce the notation

f ½n
ΔT f ðt n Þ, F ½m
F ðωm Þ ð9:34Þ

and obtain the standard form of the discrete Fourier transform

X
N 1 2π
F ½m ¼ ej N mn f ½n, m ¼ 0, . . . , N  1 ð9:35aÞ
n¼0

1XN 1 2π
f ½n ¼ e j N mn F ½m, n ¼ 0, . . . , N  1 ð9:35bÞ
N m¼0

Here, f [n] may be treated as an impulse having the area of ΔTf(tn).

Exercise 9.10: Establish a relation between F ½N  m and F[m] for a real signal f(t), which
is called a reversal property of the discrete Fourier transform.
Answer:
F * ½N  m ¼ F ½m ð9:36Þ
where the star again denotes complex conjugate.

Example 9.13: It is possible to very significantly minimize the actual number of multi-
plications necessary to compute a given DFT in Eqs. (9.35a, b). The DFT so constructed
is the fast Fourier transform (FFT) and inverse fast Fourier
 transform  (IFFT). It works
best when N is a power of two. For a pulse f ðt Þ ¼ exp 2ðt  5Þ2 , 0 s  t < 10 s,
compute its FFT and then the IFFT and finally compare the end result with the original
pulse form given that N ¼ 64.
Solution: The solution is conveniently programmed using a few lines of a self-explanatory
MATLAB code, which uses Eq. (9.29) and plots two final curves:

IX-474
Chapter 9 Section 9.3: Introduction to Continuous and Discrete. . .

Example 9.13 (cont.):

T = 10; N = 64;
dT = T/N; t = dT*(0:N-1);
f0 = exp(-2*(t-5).^2);
F = fft(f0); f = ifft(F);
plot(t, f, t, f0, '*');

Both curves are virtually identical: the relative integral error (integral of signal difference
magnitude over the integral of signal magnitude) does not exceed 1016.

Structure of Discrete Fourier Spectrum

The set of spectrum values F[m], m ¼ 0, . . . , N  1, of the DFT has an important
redundancy property illustrated in the following example.

Example 9.14: Express all discrete Fourier spectrum values F[m] present in Eq. (9.35a)
through N/2 first values of F[m] only. Hint: Use Eq. (9.36).
Solution:




N N N
F ½0, F ½1, . . . , F  1 ,F ,F þ 1 , . . . , F ½N  1 ¼
2 2 2



 ð9:37Þ
N N N
F ½0, F ½1, . . . , F  1 ,F , F*  1 , . . . , F * ½1
2 2 2
Equation (9.37) demonstrates how the output of the DFT (and of the FFT, in particular in
MATLAB) is arranged in reality. It is a symmetric conjugate about m ¼ N =2. Equation (9.37)
is a key to finding derivatives and arbitrary filter transformations of the input signal with the
FFT. Only a frequency with m  N =2 is considered to be valid; its mirror reflection about
m ¼ N =2 is a higher “aliasing frequency.” We emphasize that, according to Eq. (9.26), the
complex conjugates may be replaced by spectrum values at a negative frequency, i.e.,
F * ½1 ¼ F ½1. Thus, the spectrum above m ¼ N =2 corresponds to negative frequencies
with m > N =2.

9.3.4 Sampling Theorem

It follows from Example 9.14 that only frequency samples with ωm  N2 ω0 are really
needed. This fact is a consequence of the sampling theorem, which states that any signal
bandlimited to ωmax can be reproduced exactly using the discrete Fourier transform if
N
ωmax  ω0 ð9:38aÞ
2

IX-475
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Accordingly, the maximum possible sampling interval may be found from inequality
1 1 ωmax
ΔT  , f max ¼ ð9:38bÞ
2 f max 2π

Exercise 9.11: Examples of the maximum frequency of interest for some biomedical
signals are:
1. Electrocardiogram (ECG) where f max  250 Hz
2. Blood flow where f max  25 Hz
3. Respiratory rate where f max  10 Hz
4. Electromyogram where f max  10 kHz
Establish the maximum possible sampling interval of the DFT and the minimum possible
sampling frequency, which is equal to 1/ΔT.
Answer: (1) 2 ms and 500 Hz; (2) 20 ms and 50 Hz; (3) 50 ms and 20 Hz; (4) 50 μs and
20 kHz.

9.3.5 Applications of Discrete Fourier Transform

The DFT is one of the most important tools in digital signal processing (DSP). In
particular, the DFT can calculate a signal’s frequency spectrum. This is a direct exami-
nation of information encoded in the frequency, phase, and amplitude of the component
sinusoids. For example, human speech and hearing use signals with this type of encoding.
Second, the DFT or rather its variation, the discrete cosine transform, is used in sound
compression; the MP3 format is one such example. The DFT is also an important image
processing tool which is used to decompose an image into its sine and cosine compo-
nents. The output of the transformation represents the image in the Fourier or frequency
domain, while the input image is the spatial domain equivalent. In the Fourier domain
image, each point represents a particular frequency contained in the spatial domain
image. In particular, the JPEG format is using a modification of the DFT for image
compression; the DFT is also used for image filtering and reconstruction. Along with this,
the DFT is used widely in bioinformatics/computational biology to analyze DNA
sequences. Last but not least, many computational modeling tools, such as antenna and
high-speed circuit simulators, typically operate at one particular signal frequency (in the
frequency domain). Collecting the solutions at many such frequencies makes it possible
to establish evolution of an arbitrary signal or wave field in time.

9.3.6 Application Example: Numerical Differentiation via the FFT

We have established that a filter is characterized by its transfer function H( f ) or H(ω) and
found this transfer function for simple cases. Given the input sinusoidal signal, we have
also shown how to evaluate the filter’s output when its transfer function is known.
But what if the input signal is an arbitrary pulse? How could the solution for the output

IX-476
Chapter 9 Section 9.3: Introduction to Continuous and Discrete. . .

pulse be obtained? The answer relies upon an observation that the transfer function given
by Eq. (9.9b) may be applied to every harmonic component of the input signal fin(t)
separately. Those harmonics are all described by the Fourier spectrum of the pulse, F(ω).
Therefore, the output Fourier pulse spectrum is given by

F out ðωÞ ¼ HðωÞF in ðωÞ ð9:39Þ

The remaining part is to find the output pulse itself, which is clearly the inverse Fourier
transform:
ð
1
1
f out ðt Þ
F out ðωÞejωt dω ð9:40Þ
2π
1

When moving from continuous toward discrete Fourier transform and toward digital
signal processing, Eq. (9.39) becomes a somewhat tricky operation. According to
Eq. (9.37), the discrete version of Eq. (9.39) must have the form





N N N N
HF ! H½0F ½0, H½1F ½1, . . . , H 1 F 1 , H F ,
2 2 2 2


 ð9:41Þ
* N N
H 1 F þ 1 , . . . H ½1F ½N  1
*
2 2

This version corresponds to the full list of monotonic frequency data

ωm ¼ ω0 m, m ¼ 0, . . . , N  1. Also note that, in all realistic linear systems,

H* ðmÞ ¼ HðmÞ ð9:42Þ

Therefore, Eq. (9.41) simultaneously describes a set  of data for the following
non-monotonic frequency list 0, ωm , . . . , N2 ω0 , 1  N2 ω0 , 2  N2 , . . . ,  ω0 , which
also includes the negative frequencies.

Example 9.15 (numerical differentiation via the FFT): Prove Eq. (9.41) for a pulse
2
f ðt Þ ¼ exp 2ðt  5Þ , 0 s  t < 10 s and for HðωÞ ¼ jω. Such a transfer function
corresponds to numerical differentiation via the FFT. Use the FFT and IFFT with N ¼ 64.
Solution: The solution is conveniently programmed in a self-explanatory MATLAB code,
which uses Eq. (9.41) and plots two final results in Fig. 9.16, the numerical pulse derivative
and the analytical derivative, respectively:

IX-477
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

Example 9.15 (numerical differentiation via the FFT) (cont.):

T = 10; N = 64;
dT = T/N; t = dT*(0:N-1);
f = exp(-2*(t-5).^2); % input pulse
omega = (2*pi/T)*[0:N/2]; % non-aliasing frequencies
H = j*omega; % H at non-aliasing frequencies
F = fft(f); % FFT spectrum
HF = F.*[H, conj(H(end-1:-1:2))]; % HF according to Eq. (9.40)
fder = real(ifft(HF)); % numerical derivative
fder0 = -4*(t-5).*f; % analytical derivative
plot(t, fder0, t, fder, 'd'); % compare both derivatives

Both curves are virtually identical: the relative integral error (integral of signal difference
magnitude over the integral of analytical signal magnitude) does not exceed 1.3  1015.

df(t)/dt
1.5

0.5

-0.5

-1

-1.5
0 5 t, s 10

Fig. 9.16. Analytical (solid curve) and numerical (diamonds) differentiation of the original
Gaussian pulse.

9.3.7 Application Example: Filter Operation for an Input Pulse Signal

The filter operation for an input pulse signal exactly follows Example 9.15 but with a
different transfer function H(ω).

 
Example 9.16: A pulse f ðt Þ ¼ exp 2ðt  5Þ2 , 0 s  t < 10 s is an input to a first-
order high-pass filter. Find the filter output when its (angular) break frequency is given by
a) ω0 ¼ 1 rad=s and b) ω0 ¼ 10 rad=s. Use the FFT and IFFT with N ¼ 64.
Solution: The solution is performed and programmed exactly as described in the previous
example, but the transfer function is now given by Eq. (9.1b):
H ¼ j*omega/omega0./(1+j*omega/omega0);

IX-478
Chapter 9 Section 9.3: Introduction to Continuous and Discrete. . .

Example 9.16 (cont.):

Figure 9.17 plots two output pulse forms corresponding to two different values of the break
frequency. Note that the value ω0 ¼ 10 rad=s approximately corresponds to a mean value for
non-aliasing frequencies of the FFT. Also note that when the break frequency becomes
sufficiently high, the HPF behaves as an ideal differentiator but with a significant amplitude
decay.

a) b)
1.5 1.5

1 1

0.5 0.5

0 0

-0.5 -0.5

-1 -1

-1.5 -1.5
0 5 t, s 10 0 5 t, s 10

Fig. 9.17. Diamonds: HPF output for (a) ω0 ¼ 1 rad=s and (b) ω0 ¼ 10 rad=s, respectively. Solid
curve: analytical derivative of the input Gaussian pulse.

9.3.8 Application Example: Converting Computational Electromagnetic

Solution from Frequency Domain to Time Domain
Many computational modeling tools operate at one particular signal frequency or at a set of
those (in frequency domain). To obtain the solution for an arbitrary pulse at an arbitrary
point in space, we can again use the method of the transfer function and the FFT described
previously. As an example, we consider a TMS (transcranial magnetic stimulation) coil
above the head of a computational human phantom in Fig. 9.18a. Once a current pulse is
applied to the coil, an electric field will be excited in the brain according to Faraday’s law of
induction. This field may help to reestablish some neuron connections lost, for example, in
Parkinson’s disease. For safety considerations, the field at arbitrary locations within the
body needs to be evaluated, let’s say at node 2 in Fig. 9.18a. In order to do so, the problem
is first solved for about 40 single-frequency excitations, which will presumably cover the
spectrum content of the desired TMS pulse in Fig. 9.18b well. The ratio of the electric field
phasor at the observation point to the coil current phasor is the transfer function value at a
desired frequency, H(ω). This ratio does not depend on particular amplitude of the coil
current. Next, we introduce the DFT of size N for the original pulse shown in Fig. 9.18b,
interpolate the transfer function over N =2 þ 1 required frequency points, and apply the
method of Examples 91.5 and 9.16 with this new transfer function. The result is an electric
field pulse at node 2 shown in Fig. 9.18c (the dominant z-component has been plotted),
which is excited by the coil current shown in Fig. 9.18b.
IX-479
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

a) b)
Coil current, kA
1

0.5

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time, ms
c)
Electric field (Ez), mV/m
3.5

2.5

1.5

0.5

-0.5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
time, ms

Fig. 9.18. Time-domain computational solution for the induced electric field within a human body
obtained from the frequency-domain data via the FFT.

IX-480
Chapter 9 Summary

Summary
Property First-order low-pass filter First-order high-pass filter

Circuit schematic

Transmission at 1 0
f ¼ 0 (DC) (DC path through the resistor) (No DC path)
Transmission at 0 1
f !1 (Inductor is an open circuit (DC path through the resistor)
at f ! 1)
Transfer function 1 ð f =f b Þ
H( f ) 1 þ jð f =f b Þ 1 þ jð f =f b Þ
Decibels of H ¼ jHj 20 log10H [dB] 20 log10H [dB]
Decibels of 1 and 0.1 0 dB and 20 dB 0 dB and 20 dB
Transfer function 1 f =f b
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
magnitude Hm( f )
1 þ ð f =f b Þ2 1 þ ð f =f b Þ 2
   
Transfer function f π f
∠  tan 1  tan 1
phase ∠φH fb 2 fb
Break frequency, 1 1
fb ¼ ½Hz fb ¼ ½Hz
(half-power fre- 2πτ 2πτ
quency, 3-dB L L
frequency, corner τ ¼ RC or ½ s τ ¼ RC or ½ s
R R
frequency)
Passband (3 dB From 0 to fb From fb to 1
bandwidth), Hz

Filter with a resistive

load RL

(continued)

IX-481
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

1 ð f =f b Þ
Hð f Þ ¼ Hð f Þ ¼
Transfer function 1 þ R=RL þ jð f =f b Þ 1 þ jð f =f b Þ
with the load H( f ) 1 1  
fb ¼ , τ ¼ RC fb ¼ , τ ¼ R RL C
2πτ 2πτ

Amplitude Bode
plots

Phase Bode plots

Meaning of the
Vout ¼ Hð f ÞVin
transfer function for
harmonic signals where Vin is the input voltage phasor and Vout is the output voltage phasor
represented by
phasors
Cascading filters
Hð f Þ ¼ H1 ð f ÞH2 ð f Þ
and linear systems
(series combination) only if the loading effect of individual blocks is minimized
3-dB bandwidth
of an operational
amplifier circuit
(inverting or non-
inverting amplifier
configuration)
From 0 to f bclosed loop (closed-loop 3-dB frequency) where
BW
f bclosed loop ¼
1 þ R2 =R1
BW is the unity-gain bandwidth reported in the datasheet
(continued)

IX-482
Chapter 9 Summary

Property Continuous and discrete Fourier transform

ð
1 ð
1
jωt 1
Fourier transform F ðωÞ
f ðt Þe dt f ðt Þ
F ðωÞejωt dω
2π
definition 1 1
F ðωÞ ¼ F * ðωÞ if f(t) is real

Fourier transform
of a rectangular pulse

 
ωT
F ðωÞ ¼ V m T sinc
2π
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Fourier transform π ω2 =ð4aÞ
f ðt Þ ¼ eat
2

of a Gaussian pulse , F ð ωÞ ¼ e
a

Major properties of
Fourier transform

Definition of sam- t n ¼ ΔT n, n ¼ 0, . . . , N  1, T ¼ N ΔT
pling points: discrete 2π
ωm ¼ ω0 m, m ¼ 0, . . . , N  1, ω0 ¼
Fourier transform T
Definition of f ½n
ΔT f ðt n Þ, n ¼ 0, . . . , N  1
samples: discrete
Fourier transform F ½m
F ðωm Þ, m ¼ 0, . . . , N  1

2π
X
N 1 j mn
F ½m  ¼ e N f ½n, m ¼ 0, . . . , N  1
Discrete/fast Fourier n¼0
transform
N 1 j 2π mn
X
f ½n ¼ 1
Ne N F ½m, n ¼ 0, . . . , N  1
m¼0

(continued)

IX-483
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

1. Any signal bandlimited to ωmax can be reproduced exactly using the

N
discrete Fourier transform if ωmax  ω0
2
Sampling theorem 2. Alternatively, the sampling interval must satisfy inequality
1 1 ωmax
ΔT  , f max ¼
2 f max 2π

F * ½N  m ¼ F ½m
+




Structure of discrete N N N
Fourier spectrum F ½0, F ½1, . . . , F  1 ,F ,F þ 1 , . . . , F ½ N  1 ¼
2 2 2




N N N
F ½0, F ½1, . . . , F  1 ,F , F*  1 , . . . , F * ½1
2 2 2
Equivalent frequency    
N N N
samples for negative 0, ωm , . . . , ω0 , 1  ω0 , 2  , . . . ,  ω0
2 2 2
frequencies





N N N N
HF ! H½0F ½0, H½1F ½1, . . . , H 1 F 1 , H F ,
Transfer function 2 2 2 2
multiplication


* N N
H 1 F þ 1 , . . . , H* ½1F ½N  1
2 2

IX-484
Chapter 9 Problems

following figure. All other parameters remain

Problems the same.
9.1 First-Order Filter
C
Circuits and Their
Combinations

+
vin(t) R vout(t)
9.1.1 RC Voltage Divider as an Analog - -
Filter
Problem 9.1
A. Explain the function of an analog RC Problem 9.5. The input voltage to the RC
filter. filter circuit shown in the figure is
B. Write the capacitor and resistor voltages V in ðtÞ ¼ 5 cos ω t ½V. The filter has the fol-
υR(t) and υC(t) of the series RC circuits lowing parameters: C ¼ 1 μF and
in the general form, as functions of the R ¼ 100 Ω. The filter operates in the fre-
AC angular frequency. quency band from 100 Hz to 50 kHz. The filter
C. Which circuit element (or which voltage) is connected to a load with the load resistance
dominates at low frequencies? At high of 1 MΩ. By solving the corresponding AC
frequencies? circuit, determine the output voltage amplitude
across the load (and its percentage versus the
Problem 9.2 input voltage amplitude) with and without the
A. Draw a schematic of the low-pass analog load at f ¼ 100 Hz, f ¼ 1592 Hz, and
RC filter. Show the input and output f ¼ 50 kHz.
ports.
B. Repeat the same task for the high-pass R
analog RC filter.
+

+
Problem 9.3. The input voltage to the filter vin(t)=5cos(wt) [V] C vout(t) Load
circuit shown in the following figure is a com- - -
bination of two harmonics,
υin ðtÞ ¼ 1 cos ω1 t þ 1 cos ω2 t, with the ampli-
tude of 1 V each. The filter has the following Problem 9.6. Repeat the previous problem
parameters: R ¼ 100 kΩ and C ¼ 1:59 nF. when the load resistance changes from 1 MΩ
Determine the output voltage υout(t) to the filter to 100 Ω (decreases).
given that f 1 ¼ 100 Hz and f 2 ¼ 100 kHz.
Express all phase angles in degrees. Problem 9.7. Repeat Problem 9.5 for the filter
circuit shown in the following figure. Assume
R the load resistance of 100 Ω.

C
+

vin(t) C vout(t)
+

- -
vin(t)=5cos(wt) [V] R vout(t) Load
- -
Problem 9.4. Repeat the previous problem
for the filter circuit shown in the

IX-485
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

9.1.2 Half-Power Frequency and Ampli- between the two corresponding decibel
tude Transfer Function measures in dB?
9.1.3 Bode Plot, Decibel, and Roll-off B. When the ratio
pffiffiofffi the amplitudes of two
Problem 9.8. signals is 1/ 2, what is the difference
A. Describe the physical meaning of the between the two corresponding decibel
(half-power) break frequency in your measures in dB?
own words. C. When the ratio
pffiffiffiffiffiof the amplitudes of two
B. Give the expression for the break fre- signals is 20, what is the difference
quency in terms of circuit parameters of between the two corresponding decibel
an RC filter. Is it different for low-pass measures in dB?
and high-pass filters? D. When the ratio of the powers of two
signals is 1000, what is the difference
Problem 9.9. Given R ¼ 100 kΩ and between the two corresponding decibel
C ¼ 1:59 nF, determine the break frequency measures in dB?
of the low-pass RC filter and of the high-pass
RC filter, respectively. Problem 9.18. What do engineers mean by one
decade? One octave?
Problem 9.10. List all possible alternative
names for the break frequency. Problem 9.19. For the filter circuit shown in
the following figure, given that R ¼ 100 kΩ
Problem 9.11. Write the amplitude transfer and C ¼ 159 pF:
function for the low-pass RC filter. Repeat for A. Create the amplitude Bode plot by find-
the high-pass RC filter. Indicate units (if any). ing transfer function values for (at least)
every decade.
Problem 9.12. The input signal to a high-pass B. Label the break frequency.
RC filter includes a 60-Hz component. Its C. Label the filter passband.
amplitude is to be reduced by a factor of 10.
What break frequency should the filter have? C
+

+
Problem 9.13. The input signal to a low-pass
RC filter includes a 10-kHz component. Its vin(t) R vout(t)
amplitude is to be reduced by a factor of - -
5. What break frequency should the filter have? Hm, dB
0
Problem 9.14. Describe the meaning of the -5
Bode plot in your own words. -10

-15
Problem 9.15. It is known that
H m ð f ÞdB ¼ 0,  6,  20 ½dB. Find the -20

-25
corresponding values of Hm( f ).
-30
Problem 9.16. The following values are given -35
H m ð f Þ ¼ 1, 0:707, 0:1, and 100. Find the -40
corresponding values of Hm( f )dB. -45
2 3 4 5 6
10 10 10 10 10
Problem 9.17 frequency, Hz

A. When the ratio

pffiffiffi of the amplitudes of two Problem 9.20. Repeat the previous problem
signals is 2, what is the difference
with R ¼ 100 kΩ and C ¼ 53 pF.

IX-486
Chapter 9 Problems

Problem 9.21. For the filter circuit Problem 9.24. An amplitude Bode plot for a
shown in the following figure, assume the certain RC filter is shown in the figure below.
values R ¼ 10 kΩ and C ¼ 1:59 nF. A. Approximately determine the filter’s
A. Create the amplitude Bode plot by finding capacitance, C, for a given
the transfer function values for (at least) R ¼ 100 kΩ. Describe each step of
every decade. your approach.
B. Label the break frequency. B. Suggest a way to verify your solution.
C. Label the filter passband.
R
+

+
vin(t) C vout(t)
- -
Hm, dB
0

-5

-10

-15
2 3 4 5
-20 10 10 10 10
frequency, Hz
-25

-30 Problem 9.25. Prove analytically that the ampli-

-35 tude transfer functions of the low-pass filter and
-40 the high-pass filter are the mirror reflections of
-45 each other about the break frequency in the
2 3 4 5 6
10 10 10 10 10 Bode plot.
frequency, Hz

Problem 9.22. Repeat the previous problem

with R ¼ 100 kΩ and C ¼ 53 pF. 9.1.4 Phase Transfer Function and Its
Bode Plot
Problem 9.23. An amplitude Bode plot for a Problem 9.26. Write the phase transfer function
certain RC filter is shown in the figure below. for the low-pass RC filter. Repeat for the high-
A. Approximately determine the filter’s resis- pass RC filter. Show units.
tance R if it is known that C ¼ 265 pF.
Describe each step of your approach. Problem 9.27. The input voltage to a low-pass
B. Suggest a way to verify your solution. RC filter has a zero phase. At what frequency in
Hm, dB
terms of the break frequency fb is the phase shift
at the output equal to 1 , 45 , and 89 ?

Problem 9.28. The input voltage to a high-pass

RC filter has a zero phase. At what frequency in
terms of the break frequency fb is the phase shift
at the output equal to 5 , 45 , and 85 ?

Problem 9.29. A low-pass RC filter has the

break frequency of 10 kHz. Create the phase
Bode plot by finding the transfer function values
2
10
3
10 10
4
10
5
for (at least) every decade.
frequency, Hz

IX-487
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

φH, deg
0
Problem 9.32
-10
A. Repeat the previous problem for the filter
circuit shown in the figure below.
-20
B. Analytically determine the roll-off per
-30
decade in dB.
-40

-50 C1 C1

-60

+
-70
vin(t) R1 vout1(t) R1 vout(t)
-80 - - -
fb
-90
2 3 4 5 6
10 10 10 10 10
frequency, Hz
Hm, dB
0
Problem 9.30. Repeat the previous problem for
a high-pass RC filter with the same break fre-
-20
quency.
-40

9.1.5 Complex Transfer Function: Cas-

cading Filter Circuits -60

9.1.6 RL Filter Circuits

Problem 9.31. For the filter circuit shown in
-80

the figure below, create the amplitude response

of the Bode plot by finding transfer function
-100
1 2 3 4 5
10 10 10 10 10
values for (at least) every decade. The two frequency, Hz

individual filter blocks both have the break

frequency of 1 kHz. Assume that the loading Problem 9.33. For two RL filter circuits with
effect of the filter stages is negligibly small; in R ¼ 31:4 Ω and L ¼ 1 mH shown in the figure
practice, a buffer amplifier stage could be used. below:
A. Determine the break frequency.
C1 R2 B. Draw the corresponding RC counterpart.
C. Establish the capacitance values of the
RC filters, which assure the equivalent
+

vin(t) R1 vout1(t) C2 vout(t) transfer functions, given that the resis-

tances of the RC filters are 100 kΩ in
- - -

both cases.
Hm, dB
0
a) R

-10
+

vin(t) L vout(t)
- -
-20

-30 b) L
+

-40
vin(t) R vout(t)
- -
-50 1 2 3 4 5
10 10 10 10 10
frequency, Hz

IX-488
Chapter 9 Problems

Problem 9.34. For the filter circuit shown in Problem 9.37. Using a manufacturing
the figure below, assume the values R1 ¼ 628 company’s website (usually it is a more accu-
Ω and L1 ¼ 10 mH. rate frequently updated source) or the
A. Create the amplitude Bode plot by find- corresponding datasheet, find the unity-gain
ing transfer function values for (at least) bandwidth for the following amplifier ICs:
every decade. A. TL082
B. Determine the roll-off per decade in dB. B. LM741
C. LM7171
Assume that the loading effect of the filter
stages is negligibly small (e.g., a buffer ampli-
fier stage is used). Problem 9.38. Frequency response of an
amplifier is characterized by the open-loop DC
L1 L1 gain AOL ð0Þ ¼ 1:41  106 and the break fre-
quency of f b ¼ 20 Hz. Numerically calculate
the gain-bandwidth product for the amplifier at:
+

vin(t) R1 vout1(t) R1 vout(t)

- - - A. 20 Hz,
B. 2 kHz,
C. 2 MHz.
Hm, dB
0
9.2.3 Model of the Open-Loop AC Gain
Problem 9.39. Frequency response of an
-20
amplifier is characterized by the open-loop
DC gain AOL ð0Þ ¼ 106 and the break frequency
-40
of f b ¼ 20 Hz. Plot the open-loop gain magni-
tude in dB over the range of frequencies (the
-60
frequency band) from 1 Hz to 10 MHz on the
log-log scale (the Bode plot) and label the axes.
-80

-100
Problem 9.40. In the previous problem, find
10
2 3
10 10
4 5
10 10
6
the unity-gain bandwidth BW of the amplifier.
frequency, Hz

Problem 9.35. The transfer function of a filter Problem 9.41. Internally compensated
1þjð f =1000Þ LM358-series amplifiers have the unity-gain
circuit is given by H ð f Þ ¼ 1þ ð f =1000Þ2
. Create its
bandwidth (BW) of 1 MHz. The typical large-
amplitude and phase Bode plots in the frequency signal DC voltage gain at room temperature is
band from 10 Hz to 100 kHz by finding transfer 100 V/mV.
function values for (at least) every decade. A. Find the open-loop DC gain in dB and the
open-loop break frequency fb.
9.2 Bandwidth of an Opera- B. Find the open-loop gain at 100 Hz,
1 kHz, and 10 kHz.
tional Amplifier
9.2.1 Bode Plot of the Open-loop Ampli- Problem 9.42. The open-loop gain magnitude
fier Gain of an internally compensated high-frequency
9.2.2 Unity-gain Bandwidth Versus amplifier has been given as
Gain-Bandwidth Product AOL ð100 HzÞ ¼ 0:9  106 ,
Problem 9.36. An amplifier has the unity-gain
bandwidth BW of 5 MHz. What exactly does AOL ð1 MHzÞ ¼ 1:0  102
this mean? Explain and provide equations.

IX-489
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

at room temperature. Determine: A. An inverting amplifier with the gain of

A. 3-dB break frequency, 1,
B. DC open-loop gain, B. An inverting amplifier with the gain of
C. Unity-gain bandwidth BWof the amplifier. 10,
C. A non-inverting amplifier with the gain
Problem 9.43. Repeat the previous problem for of 100,
D. A voltage follower (buffer amplifier).
AOL ð100 HzÞ ¼ 0:5  106 ,
constructed using the same IC.
AOL ð1 MHzÞ ¼ 1:0  102 :
9.2.6 Application Example: Selection of
9.2.4 Model of the Closed-loop AC Gain an Amplifier IC for Proper Frequency
9.2.5 Application Example: Finding Bandwidth
Bandwidth of an Amplifier Circuit Problem 9.46. An inverting amplifier with a
Problem 9.44. An amplifier with the open-loop gain of 20 and a bandwidth of at least
gain described by the first-order RC circuit 200 kHz is needed. Which amplifier chip is
response with AOL ð0Þ ¼ 105 and f b ¼ 20 Hz appropriate for this circuit (and which is not)?
is used in a closed-loop inverting configuration A. LM358
with R2 =R1 ¼ 9 and R2 =R1 ¼ 99, respectively. B. TL082
A. Using the template that follows, create C. LM741
the Bode plots for the corresponding D. LM7171
frequency response (closed-loop gain), E. LM8272
G( f ), in the band from 10 Hz to
10 MHz on the same graph. Plot the Problem 9.47. A non-inverting amplifier with a
gain values for (at least) every decade. gain of 31 and a bandwidth of at least 90 kHz is
B. Also on the same graph, plot the open- needed. Which amplifier chip is appropriate for
loop gain as a function of frequency. this circuit (and which is not)?
C. Determine the bandwidth of the closed- A. TL082
loop amplifier so constructed in B. LM7171
every case. C. LM8272

Bode plot
100 9.3 Introduction
80 to Continuous and Discrete
60 Fourier Transform
gain, dB

40 9.3.1 Meaning and Definition of Fourier

Transform
20 Problem 9.48. Establish all values of the angu-
0
lar frequency ω in Fig. 9.14b at which the
Fourier spectrum F(ω) of a rectangular pulse
-20 crosses the frequency axis (becomes zero).
1 2 3 4 5 6 7
10 10 10 10 10 10 10 Express your result in terms of pulse duration T.
frequency, Hz

Problem 9.45. The unity-gain bandwidth of an Problem 9.49

amplifier IC is 1 MHz. Determine the band- A. Establish the value of the Fourier trans-
width of the following amplifier circuits: form F(ω) for the pulse shown in the
following figure at ω ¼ 0.

IX-490
Chapter 9 Problems

B. Establish the complete pulse spectrum Problem 9.53. The Fourier transform of f(t) is
F(ω) at all values of angular frequency ω. F(ω). What is the Fourier transform of f ðt Þ?

f(t)
Vm Problem 9.54. The function f(t)cos ω0t is an
amplitude-modulated signal: a high-frequency
carrier cos ω0t, which is transmitted wirelessly,
has a low-frequency envelope f(t), which
t
carries information and is being demodulated
-T/2 0 T/2
at the receiver. If the Fourier transform of f(t)
is F(ω), what is the Fourier transform of
f(t)cos ω0t?

Problem 9.55. If f(t) represents the voltage

Problem 9.50 across a 1-Ω load, then f2(t) is the power deliv-
Establish the Fourier transform F(ω) for the ð
1

following voltage signals

π  in time domain: ered to the load and f 2 ðtÞdt is the total
f ðtÞ ¼ A sin t , 2  t < 2 1
A. 2 energy delivered to the load. Prove Parseval’s
f ðt Þ ¼ 0,   otherwise theorem,
π
f ðt Þ ¼ A cos t , 2  t < 2
B. 2 ð
1 ð
1
f ðt Þ ¼ 0, otherwise 1
f ðt Þdt ¼
2
jF ðωÞj2 dω;
2π
1 1
Problem 9.51
Show that for an arbitrary real voltage signal which relates the total energy to an inte-
f(t): gral of the energy spectral density,
A. The real part of F(ω) is an even function jF ðωÞj2 ¼ F ðωÞF * ðωÞ, of the signal.
of angular frequency ω. Hint: Use the reversal property of the Fourier
B. The imaginary part of F(ω) is an odd transform given by Eq. (9.26).
function of angular frequency ω.
C. The magnitude of F(ω) is an even func- Problem 9.56. Based on Parseval’s theorem
tion of angular frequency ω. established in the previous problem, find the
D. Replacing ω by ω generates the com- ð
1
plex conjugate of F(ω); in other words, value of the integral sinc2 ðt Þdt.
F ðωÞ ¼ F * ðωÞ. 1

9.3.3 Discrete Fourier Transform and Its

9.3.2 Mathematical Properties of Fou- Implementation
rier Transform Problem 9.57. You are using the discrete Fou-
Problem 9.52. The Fourier transform of rier transform of length 8 (N ¼ 8) for a signal
f(t) is F(ω). What is the Fourier transform of f ðt Þ ¼ sin t over a time interval from 0 to 2π s.
ðt A. Compute all sampling points in the time
d f ðt Þ=dt  2 f ðτÞdτ?
2 2
domain.
1 B. Compute all sampling points in the fre-
quency domain.

IX-491
Chapter 9 Filter Circuits: Frequency Response, Bode Plots. . .

C. Compute equivalent frequency samples Problem 9.61*. Present the text of a MATLAB
using negative frequencies. script that numerically differentiates the input
D. Compute all discrete samples f [n]. signal f ðt Þ ¼ sin t over the time interval from
E. Compute all discrete samples F[m] using 0 to 4π s using the FFT with 4096 sampling
the definition of the discrete Fourier points and plot the resulting signal derivative.
transform. Explain the physical meaning
of their values. Problem 9.62. Repeatthe previousproblem for
F. Repeat the previous step using function
the signal f ðt Þ ¼ exp ðt  2π Þ2 . All other
fft of MATLAB. Compare both sets of
F[m]. parameters remain the same.
G. Restore all discrete samples f [n] using
the definition of the inverse discrete Fou- Problem  9.63. A  monopolar pulse
rier transform. Compare them with the
f ðt Þ ¼ exp 2ðt  5Þ2 , 0  t < 10 s is
exact function values.
H. Repeat the previous step using function an input to a series combination of two identical
ifft of MATLAB. Compare both sets of first-order high-pass filters. Find the output of
f [n]. the filter combination when the (angular) break
frequency is given by:
Problem 9.58. Repeat the previous problem for A. ω0 ¼ 0:5 rad=s
the signal f ðt Þ ¼ cos t. All other parameters B. ω0 ¼ 10 rad=s
remain the same.
Use the FFT and IFFT with N ¼ 64. Plot the
filter output and explain the output signal
Problem 9.59. For Problem 9.57, establish and behavior in every case.
prove a discrete version of Parseval’s theorem
formulated in Problem 9.55.
Problem 9.64.
 A bipolar pulse
2
Problem 9.60. An input signal to a filter has a f ðt Þ ¼ ð5  t Þexp 2ðt  5Þ , 0  t < 10 s is
discrete frequency spectrum an input to a first-order low-pass filter. Find the
F ½m, m ¼ 0, . . . , N  1 computed via the filter output when its (angular) break frequency
FFT. You are given filter transfer function is given by:
H computed at N2 þ 1 frequency points of the A. ω0 ¼ 0:5 rad=s
FFT, H½m, m ¼ 0, . . . , N =2. Compute the B. ω0 ¼ 5 rad=s
discrete spectrum of the filter’s output to be Use the FFT and IFFT with N ¼ 64. Plot the
fed into the IFFT. filter output along with the input signal on the
same graph and explain the output signal
9.3.6 Application Example: Numerical behavior in both cases.
Differentiation via the FFT
9.3.7 Application Example: Filter Oper-
ation for an Input Pulse Signal

IX-492
Chapter 10

Chapter 10: Second-Order RLC Circuits

Overview
Prerequisites:
- Knowledge of complex arithmetic
- Knowledge of phasor/impedance method for AC circuit analysis (Chapter 8)

Objectives of Section 10.1:

- Learn the concept of a resonant circuit and its relation to other engineering
disciplines
- Understand the internal dynamics of the series/parallel RLC resonator including
voltage and current behavior near the resonant frequency
- Establish the meaning and be able to calculate resonant frequency, quality factor, and
bandwidth of the second-order resonant circuits
- Establish and quantify the duality between series and parallel RLC resonators

Objectives of Section 10.2:

- Construct four major types of the second-order RLC filters
- Relate all filter concepts to the corresponding circuit diagrams
- Specify two filter design parameters: the undamped resonant frequency and the
quality factor
- Realize the advantages of the second-order filters versus the first-order filters

Objectives of Section 10.3:

- Become familiar with the concept of the near-field wireless link
- Apply the theory of the series resonant RLC circuit to the basic design of the near-
field wireless transmitter and receiver
- Understand the operation of proximity sensors based on resonant RLC circuits

Application examples:
- Near-field wireless link in undergraduate laboratory
- Proximity sensors

© Springer Nature Switzerland AG 2019 X-493

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_10
Chapter 10 Second-Order RLC Circuits

Keywords:
Self-oscillating LC circuit, Series resonant RLC circuit, Parallel resonant RLC circuit, Series RLC
tank circuit, Parallel RLC tank circuit, Undamped resonant frequency, Resonant frequency,
Quality factor of the series resonant RLC circuit, Quality factor of the parallel resonant RLC
circuit, Quality factor (general definition, interpretation, mechanical analogy, trade-off between
Q-factor and inductance value), Bandwidth of the series resonant RLC circuit, Bandwidth of the
parallel resonant RLC circuit, Half-power bandwidth, Upper half-power frequency, Lower
half-power frequency, Duality of series/parallel RLC circuits, Ideal filter, Cutoff frequency,
Second-order band-pass RLC filter, Second-order low-pass RLC filter, Second-order band-
reject (or band-stop or notch) RLC filter, Second-order high-pass RLC filter, Quality factor of
the filter circuit, Center frequency of the band-pass filter, Lower and upper half-power
frequencies, Butterworth response, Quality factor of the nonideal inductor, Voltage multiplier
circuit, Voltage multiplication, Near-field wireless link, Horseshoe coil

X-494
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

Section 10.1 Theory of Second-Order Resonant

RLC Circuits
In this section, we study the last group of standard AC circuits—the resonators. They are
second-order AC circuits in LC or RLC configuration. The term second order means that
the circuits will be described by second-order differential equations if we work in the time
domain. The value of a resonator circuit in electronics cannot be overstated. In order to
proceed with any type of wireless communication, we first need to create a high-
frequency AC signal as part of a resonator circuit. Beyond high-frequency circuits,
resonators are often used in power electronics and as sensors. In this section, we apply
the phasor/impedance method to analyze resonator circuits. We will discover that the
most important characteristic is the resonant frequency. Another important parameter is
the quality factor, which also determines the resonator bandwidth.

10.1.1 Self-Oscillating Ideal LC Circuit

The circuit shown in Fig. 10.1a includes an inductor and a capacitor and there is no power
source connected to the circuit. The circuit is also ideal, which means that there is no
resistance. In other words, the parasitic resistance of the inductor, parasitic resistance of
the capacitor, and the wire resistance are all neglected. We assume that the power supply
(voltage or current) was disconnected at t ¼ 0, after the resonator was excited. The
steady-state alternating current and the AC voltages across the circuit elements are sought
once the oscillation process has been stabilized, i.e., at t ! 1.

Fig. 10.1. Self-oscillating ideal LC circuit and its phasor representation.

When we apply the phasor/impedance method to the circuit in Fig. 10.1a, we obtain the
circuit shown in Fig. 10.1b. KVL in phasor form yields (note the passive reference
configuration)
VL þ VC ¼ 0 ) ZL I þ ZC I ¼ 0 ) ðZL þ ZC ÞI ¼ 0 ð10:1Þ

Generally, Eq. (10.1) requires the phasor current I to be zero. Obviously, if the phasor
current is zero, then the real current is also zero and so are the voltages across the inductor
and the capacitor. The circuit is not functioning. However, you should note that, if

X-495
Chapter 10 Second-Order RLC Circuits

ZL þ ZC ¼ 0 ð10:2Þ

in Eq. (10.1), the phasor current does not have to be zero and may have any value
depending on the initial excitation. Equation (10.2) is satisfied at only one single
frequency f0:
1 1
jω0 L þ ¼ 0 ) ðmultiply by jÞ ) ω0 L þ ¼0)
jω0 C ω0 C
ð10:3Þ
1 1
ω0 ¼ pffiffiffiffiffiffiffi ) f 0 ¼ pffiffiffiffiffiffiffi
LC 2π LC

which is the undamped resonant frequency of the LC circuit. Equation (10.3) is perhaps
the most important result of resonator theory. Once Eq. (10.3) is satisfied, the solution for
the circuit current is obtained in the form:
I ¼ I0 ) iðt Þ ¼ I m cos ω0 t ð10:4aÞ

The current amplitude Im may be arbitrary; it is determined by the initial excitation. The
voltages are found accordingly:
VL ¼ ZL I ) υL ðt Þ ¼ ω0 LI m cos ðω0 t þ 90 Þ
ð10:4bÞ
VC ¼ ZC I ) υC ðt Þ ¼ 1=ðω0 C ÞI m cos ðω0 t  90 Þ

The ideal self-oscillating LC circuit in Fig. 10.1 can oscillate indefinitely long. What is the
physical basis of self-oscillations in an LC circuit? To answer this question, let us take a
closer look at Eqs. (10.4). When the circuit current is at its maximum, the magnetic field
energy stored in the inductor also has reached its maximum. Since the voltages are shifted
by  π/2 versus the current, they are exactly zero at that time instance. The zero capacitor
voltage means that no energy of the electric field is stored in the capacitor. All of the energy
stored in the circuit is concentrated in the inductor. When the circuit current reaches zero,
the situation becomes the opposite: the capacitor stores the entire circuit energy, and the
inductor does not have any stored energy. As time progresses, the process continues so that
the current flows back and forth in the circuit charging and discharging the capacitor (and in
certain sense the inductor) periodically. Figure 10.2 shows the ideal mechanical counterpart
of the circuit in Fig. 10.1. A massive wheel with a rotational inertia represents the inductor
and the flexible membrane, the capacitor. The fluid flows back and forth either rotating the
wheel (increasing its rotational energy) or bending the membrane (increasing its release
energy).

X-496
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

massive wheel

flexible membrane

Fig. 10.2. Self-oscillating mechanical counterpart of the LC circuit shown in Fig. 10.1.

Example 10.1: An LC circuit in Fig. 10.1 has the circuit parameters L ¼ 1 μH,
C ¼ 1 μF. Determine its resonant frequency, also known as the self-oscillation frequency.
Solution: Equation (10.3) is applied, which gives ω0 ¼ 106 rad=s ) f 0 ¼ 159 kHz. For
practical reasons, the resonant frequency is most often measured and reported in Hz,
instead of rad/s.

10.1.2 Series Resonant Ideal LC Circuit

What if an alternating pressure pump is connected to the oscillating mechanical system in
Fig. 10.2, which will add a small pressure “push” at every period of oscillation? Since there
is no friction, the oscillations may grow up indefinitely. This phenomenon is known as
resonance. The corresponding electrical counterpart is the circuit shown in Fig. 10.3a. The
circuit in Fig 10.3a is solved using the phasor method, see Fig. 10.3b. We assume
υS ðt Þ ¼ V m cos ωt. The equivalent impedance is given by,
 
1 1 1  LCω2
Zeq ¼ ZL þ ZC ¼ jω L þ ¼ j ωL  ¼ j ½Ω ð10:5aÞ
jωC ωC ωC

Fig. 10.3. Series resonant ideal LC circuit and its phasor representation.

X-497
Chapter 10 Second-Order RLC Circuits

The phasor voltages and phasor current become

V m ωC V m LCω2
I ¼ V m =Zeq ¼ ∠90 , VL ¼ ZL I ¼ ∠180 ,
1  LCω 2 1  LCω 2
ð10:5bÞ
Vm
V C ¼ ZC I ¼ ∠0
1  LCω 2

The real-valued circuit parameters are given by

V m ωC V m LCω2
iðt Þ ¼ cos ðω t þ 90 Þ, υ L ðt Þ ¼ cos ðωt þ 180 Þ,
1  LCω2 1  LCω2 ð10:5cÞ
Vm
υC ðt Þ ¼ cos ðωt Þ
1  LCω2
The solution remains finite at any source frequency except the undamped resonant
pffiffiffiffiffiffiffi
frequency ω0 ¼ 1= LC or f 0 ¼ ω0 =ð2π Þ. The closer the source frequency approaches
the undamped resonant frequency, the higher the circuit current, capacitor voltage, and
the inductor voltage become. Eventually, at the exact undamped resonant frequency, they
all become infinitely high! The denominator in Eq. (10.5c) approaches zero and the
circuit starts to “resonate.” At the undamped resonant frequency, the impedances of the
inductor and capacitor cancel out and their combination is a short circuit: an ideal wire of
zero resistance. Moreover, the voltage source is shorted out. Note that the resonant
frequency of an LC circuit was first derived by James Clerk Maxwell in 1868. A young
man at this point, he spent a night working over this problem, which arose from an
experiment of Sir William Grove, and wrote a report to him the next morning.

Example 10.2: Find the sum of the real-valued voltages υL(t), υC(t) in Fig. 10.3a.
Solution: The capacitor and inductor voltages are in antiphase (the phases differ by 180 ).
Therefore, they largely cancel out. According to Eq. (10.5c), the sum of the voltages is
exactly the supply voltage υS(t), irrespective of how high the individual voltages are.

Exercise 10.1: For the ideal series resonant ideal LC circuit in Fig. 10.3, determine the
phasor voltages across the inductor and capacitor given that V m ¼ 1 V and ω2 ¼ 0:9ω20 .
Answer: VL ¼ 9∠180 ½V, VC ¼ 10∠0 ½V

10.1.3 Series Resonant RLC Circuit: Resonance Condition

The ideal LC circuit shown in Fig. 10.1a, or the series LC resonator shown in Fig. 10.3a,
never exists in practice. Internal power supply resistance, wire resistance, or parasitic
resistances of realistic capacitors and inductors lead to the realistic resonant RLC circuit

X-498
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

model shown in Fig. 10.4a. Here, the resistance R models the combined resistances
present in the circuit. The resistance reduces the resonant effect and leads to finite
voltages/currents at the resonance.

a) i(t) L b) I ZL
vL
-

+
VL
-

+
+

+
vS(t) + vR R VS + VR ZR
- - - -
VC
- vC
-

+
C ZC

Fig. 10.4. Realistic series resonant RLC circuit and its phasor representation.

The circuit in Fig 10.4a is solved using the phasor method, see Fig. 10.4b. We again
assume υS ðt Þ ¼ V m cos ωt. The equivalent impedance is given by
 
1 1
Zeq ¼ ZR þ ZL þ ZC ¼ R þ jωL þ ¼ R þ j ωL  ½Ω ð10:6aÞ
jωC ωC

The resonance condition for any AC circuit, and not necessarily the circuit shown in
Fig. 10.4, states that the imaginary part (the reactance X) of the equivalent circuit
impedance seen by the power source must be equal to zero:
 
Im Zeq ¼ X ¼ 0 ½Ω ð10:6bÞ

When applied to Eq. (10.6a), this condition defines the circuit’s resonant frequency in the
pffiffiffiffiffiffiffi  pffiffiffiffiffiffiffi
form ω0 ¼ 1= LC , f 0 ¼ 1= 2π LC . For the series RLC circuit, the resonant fre-
quency clearly coincides with the undamped resonant frequency of the ideal LC circuit.
Unfortunately, this is not always the case for general RLC circuits. This important
question is addressed in the homework problems.
At the resonant frequency, the impedances of the inductor and capacitor in Eq. (10.6a)
cancel out; their combination is a short circuit since only the resistance R remains. The
real-valued circuit current and the real-valued voltages are given by
Vm Vm Vm 1
i ðt Þ ¼ cos ðω0 tÞ, υL ðt Þ ¼ ω0 L cos ðω0 t þ 90 Þ, υC ðt Þ ¼ cos ðω0 t  90 Þ
R R R ω0 C
ð10:6cÞ

at resonance. Those are the highest amplitudes of the circuit current and the individual
voltages that could be achieved in the series RLC circuit. If the frequency deviates from
the resonant frequency, smaller amplitude values are obtained. When the circuit

X-499
Chapter 10 Second-Order RLC Circuits

resistance is small, large circuit current and large capacitor and inductor voltages may be
achieved at the resonance. You have to be aware of the fact that it is not uncommon to
measure voltage amplitudes of 50–500 V across the individual elements in the laboratory,
whereas the driving source voltage may only have an amplitude of 10 V. The circuit in
Fig. 10.4 is also called the series RLC tank circuit.

Exercise 10.2: In the series resonant RLC circuit shown in Fig. 10.4, V m ¼ 10 V,
L ¼ 50 μH, C ¼ 0:5 nF, R ¼ 50 Ω. Determine the real-valued circuit current and the
inductor/capacitor voltages at the resonance.
Answer:

iðt Þ ¼ 0:2 cos ðω tÞ ½A, υL ðt Þ ¼ 63:3 cos ðωt þ 90 Þ ½V,

ð10:7Þ
υC ðt Þ ¼ 63:3 cos ðωt  90 Þ ½V:

Could we increase the resonant voltage amplitudes of the series RLC circuit in
Fig. 10.4a [see Eq. (10.6c)] while keeping the voltage source and the circuit resistance
unaltered? Yes we can. However, one more concept is required for this and similar
problems: the concept of the quality factor of a resonator.

10.1.4 Quality Factor Q of the Series Resonant RLC Circuit

Multiple factors in front of resonant voltages and currents expressions can be reduced to
one single factor. Using the definition of the resonant frequency ω0, Eq. (10.6c) at the
resonance may be rewritten in the simple form
Vm
i ðt Þ ¼ cos ðω0 t Þ, υL ðt Þ ¼ QV m cos ðω0 t þ 90 Þ, υC ðt Þ ¼ QV m cos ðω0 t  90 Þ
R
ð10:8Þ

where the dimensionless constant

pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffi
L=R L=C
Q ¼ pffiffiffiffiffiffiffi ¼ ð10:9Þ
RC R

is called the quality factor of the series resonant RLC circuit. The equivalent forms are
1 L
Q¼ ¼ ω0 ð10:10Þ
ω0 RC R

Thus, in order to increase the resonant voltage amplitudes in Eq. (10.8), we should simply
increase the quality factor of the resonator. Even if the circuit resistance remains the same,
we can still improve Q by increasing the ratio of L/C in Eq. (10.9). This observation

X-500
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

provides one physical interpretation of the quality factor: it determines the maximum
amplitude of the resonant oscillations. A higher Q-factor results in larger amplitudes. Yet
another, perhaps even more important, interpretation is related to the “sharpness” of the
resonance at frequencies close to ω0. What is the physical meaning of Eq. (10.9)? Why
does the Q-factor increase with increasing the inductance but not the capacitance? To
answer these questions, consider the fluid mechanics analogy in Fig. 10.2. The high
Q implies a massive wheel (note: high inductance is equivalent to high wheel mass).
Simultaneously, it implies a large membrane stiffness (the small capacitance, which is
inversely proportional to the stiffness). The mechanical resonator so constructed will be
less susceptible to losses at resonance but will not resonate at all if the driving force has a
frequency far away from the resonance.
A general definition of the quality factor also applicable to mechanical engineering and
physics is as follows. The quality factor is 2π times the ratio per cycle of the energy stored
in the resonator to the energy supplied by a source, while keeping the signal amplitudes
constant at the resonant frequency. According to Eq. (10.6c), the instantaneous energies
stored in the inductor and capacitor are given by
 2
1 2 Vm L
EL ðt Þ ¼ Li ðt Þ ¼ cos 2 ðω0 t Þ,
2 R 2
 2 ð10:11Þ
1 2 Vm 1
EC ðt Þ ¼ CυC ðt Þ ¼ sin 2 ðω0 t Þ
2 R 2ω20 C
 
Since 1= ω20 C ¼ L, the coefficients in front of the cosine squared and sine squared terms
are equal. It means that even though both energies vary over time, their sum is a constant:
 2
Vm L
E L ðt Þ þ E C ðt Þ ¼ ð10:12aÞ
R 2

The energy dissipated in the resistance is the integral of the instantaneous absorbed power
over the period; this integral is equal to

ð 0
2π=ω ð 0
2π=ω
V 2m V2 πV 2m
Ediss ¼ cos 2 ðω0 t Þdt ¼ m ð1 þ cos ð2ω0 t ÞÞdt ¼ ð10:12bÞ
R 2R ω0 R
0 0

The ratio of the two energies times 2π precisely equals Eq. (10.9).

X-501
Chapter 10 Second-Order RLC Circuits

Example 10.3: A series resonant LC circuit is driven by a laboratory AC voltage source

with amplitude V m ¼ 10 V and an internal resistance of 50 Ω (a function generator).
Which value should the ratio L/C have to obtain the amplitude of the capacitor voltage to
be equal to 50 V at resonance?
Solution: We replace the realistic voltage source by its Thévenin equivalent: the ideal
voltage source with the amplitude of 10 V and the series resistance of 50 Ω. Here, we again
arrive at the standard RLC circuit shown in Fig. 10.4a. According to Eq. (10.8), the
Q-factor of the RLC circuit should be equal to 5. From the definition of the Q-factor
given by Eq. (10.9), one has

pffiffiffiffiffiffiffiffiffi
L=C ¼ RQ ¼ 250 ) L=C ¼ 62, 500 Ω2 ð10:13Þ

This result is valid for any resonant frequency. For example, the set L ¼ 1 mH, C ¼ 16
nF will give us the desired amplitude value.

Note that large Q-factors usually imply large inductances which increase the
series resistance of the inductor coil and thus increase the net circuit resistance (increase
circuit loss). Therefore, there is a trade-off between the circuit Q and the inductance
value.

Example 10.4: A series resonant RLC circuit is needed with the resonant frequency of
100 kHz and a Q-factor of 50. The circuit resistance is 10 Ω. Determine the necessary
values of L and C.
Solution: From the definition of the resonant frequency and the Q-factor, we obtain
pffiffiffiffiffiffiffi pffiffiffiffiffiffiffiffiffi
1= LC ¼ 2π  105 , L=C ¼ RQ ¼ 500 )
1 ð10:14Þ
¼ 2π  105  500 ) C ¼ 3:2 nF
C
Consequently, L ¼ 0:80 mH.

10.1.5 Bandwidth of the Series Resonant RLC Circuit

The bandwidth of the series resonant RLC circuit may be obtained by analyzing the
behavior of the amplitude of the circuit current as a function of source frequency. An
alternative definition implies analyzing the behavior of the amplitude of the resistor
voltage; however, both quantities are equal to within a constant R. The phasor current
at any frequency is obtained from Eq. (10.6a). It will be written here in terms of quality
factor Q, frequency f, and resonant frequency f0 in the form:

X-502
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

Vm V 1
I¼ ¼  m  ¼ Im   ð10:15aÞ
Zeq R þ j ωL  ωC
1
1 þ jQ ff  ff0
0

Here, I m ¼ V m =R is the maximum (resonant) current amplitude in the series RLC circuit;
 pffiffiffiffiffiffiffi
f 0 ¼ 1= 2π LC is the resonant frequency in Hz. The real-valued circuit current and the
real-valued resistor voltage are both found from the phasor current given by Eq. (10.15a):

iðt Þ ¼ I m H cos ðωt þ φÞ ½A 1
, H ðf Þ ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2ffi ð10:15bÞ
υR ðt Þ ¼ V m H cos ðωt þ φÞ ½V f0
1þQ f  f 2 f
0

Here, H( f ) is the dimensionless function of frequency that peaks at the resonant frequency
f ¼ f 0 , H ðf 0 Þ ¼ 1. One may treat H( f ) as an amplitude transfer function of an associated
RLC filter with the input voltage being the source voltage and the output voltage being the
resistor voltage. In this case, H( f ) is equal to the amplitude ratio of the two voltages.
Simultaneously, H( f ) characterizes how fast the circuit current amplitude decays when
the circuit frequency deviates from the resonant frequency f0. To be specific, we assume
f 0 ¼ 10 kHz and Q ¼ 1, 2, 5 in Eq. (10.15b). Figure 10.5 plots the function H( f ) in
decibels, H ðf ÞdB ¼ 20log10 ðH ðf ÞÞ ½dB, using a log-log scale, i.e., creates its Bode plot.
The bandwidth B of the series resonant RLC circuit is defined as the interval of
pffiffiffi
frequencies over which the function H( f ) is greater than or equal to 1= 2 ¼ 0:707.
In other words, the signal power at the resistor (which is proportional to the square of
H( f )) is no less than 50 % of the maximum power at the exact resonance. We call
the bandwidth so defined the half-power bandwidth. In terms of the transfer function
H( f )dB in decibels, this condition corresponds to the inequality H ðf ÞdB  3 dB.

H(f), dB Bode plot

0
-3 dB -3 dB
-5 Q=1

-10 Q=2

-15
Q=5
-20
B(Q=5)
-25 B(Q=2)
B(Q=1)
f0
-30
3 4 5
10 fL 10 fU frequency, Hz 10

Fig. 10.5. Amplitude of the circuit current (or amplitude of the resistor voltage) normalized by its
peak value at resonance. The resulting graph is the Bode plot.

X-503
Chapter 10 Second-Order RLC Circuits

The first resonant curve in the form of Fig. 10.5 was published by Heinrich Hertz in
1887 although he used a linear, not a logarithmic, frequency scale so that the curve did not
look quite symmetric. Figure 10.5 indicates that the bandwidth increases when the quality
factor decreases and vice versa. In other words, the low-Q resonant circuit has a large
bandwidth; we may say it resonates “equally bad” over a wider band of frequencies.
However, the high-Q circuit has a small bandwidth; it resonates well but only over a small
band of frequencies. The lower and upper half-power frequencies, fL, fU, are obtained by
pffiffiffi
setting H ðf Þ ¼ 1= 2 in Eq. (10.15b) and solving for f. The resulting quadratic equation
has two roots:
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 
2
fL ¼ f0 1 þ 1=ð2QÞ  1=ð2QÞ ,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  ð10:16Þ
2
fU ¼ f0 1 þ 1=ð2QÞ þ 1=ð2QÞ ½Hz

Despite the complexity of those expressions, the final result for the half-power bandwidth
is surprisingly simple and understandable:
f0 R
B  fU fL ¼ ¼ ½Hz ð10:17Þ
Q 2πL

Exercise 10.3: Determine the bandwidth B of the series resonant RLC circuit with the
resonant frequency of 1 MHz and a Q-factor of 10.
Answer: B ¼ 100 kHz.

Example 10.5: A series resonant RLC circuit is needed with a resonant frequency of
500 kHz and a bandwidth of 20 kHz. Given the circuit resistance of 15 Ω, determine
L and C.
Solution: From the bandwidth definition, the required Q-factor is equal to 500=20 ¼ 25.
Further, we may follow the solution developed in Example 10.4. From the definition of
thepffiffiffiffiffiffiresonant
ffi frequency
pffiffiffiffiffiffiffiffiffi and the Q-factor, we subsequently obtain
1= LC ¼ 2π  5  105 , L=C ¼ RQ ¼ 375 ) C 0:85 nF. Next, we determine
L ¼ C ðRQÞ2 0:12 mH. Alternatively, one could find the inductance L directly from
Eq. (10.17), that is, L ¼ R=ð2π BÞ ¼ 0:12 mH.

X-504
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

Example 10.6: Create Bode plots as seen in Fig. 10.5 using MATLAB.
Solution: We create only one bandwidth curve, for Q ¼ 2. Other curves are obtained
similarly, using the command hold on.

10.1.6 Parallel Resonant RLC Circuit: Duality

The parallel resonant RLC circuit is shown in Fig. 10.6. It is driven by an alternating
current source iS ðt Þ ¼ I m cos ωt. The parallel RLC resonator is a current divider circuit,
which is the dual to the series RLC resonator (which is a voltage divider) in Fig. 10.4.
While the series RLC resonator is capable of creating large voltages (or “amplifying” the
supply voltage), the parallel RLC resonator circuit is capable of producing large currents.
The amplitudes of the currents through the inductor and the capacitor may be large, much
larger than the supply current itself. The circuit in Fig. 10.6 is also called the parallel RLC
tank circuit.

a) iS (t) b) IS

R L C IS ZR ZL ZC
+ + + +
+

iS (t) v v v v V
- V
- V
-
- - - -

Fig. 10.6. Parallel resonant RLC circuit and its phasor representation.

The circuit in Fig 10.6a is solved by using the phasor method; see Fig. 10.6b. The
equivalent impedance is given by

1 1 1 1 1 1 1 1  LCω2
¼ þ þ ¼ þ þ jωC ¼  j ½Ω ð10:18Þ
Zeq ZR ZL ZC R jωL R ωL

The resonance condition for any AC circuit states that the impedance Zeq must be a
purely real number at resonance. If the impedance is real, its reciprocal, the admittance, is
also real and vice versa. Therefore, from Eq. (10.18), we obtain the resonant frequency

X-505
Chapter 10 Second-Order RLC Circuits

1 1
ω0 ¼ pffiffiffiffiffiffiffi , f 0 ¼ pffiffiffiffiffiffiffi ð10:19Þ
LC 2π LC

which coincides with the resonant frequency of the series RLC tank circuit and with the
undamped resonant frequency of the LC circuit. Thus, at resonance, Zeq ¼ R and one has
V ¼ RI m for the phasor voltage in Fig. 10.6b. Knowing the phasor voltage, we can
establish the phasor currents. The corresponding real-valued voltage and currents at
resonance take on the forms
RI m
υðt Þ ¼ RI m cos ðω0 t Þ, i L ðt Þ ¼ cos ðω0 t  90 Þ,
ω0 L ð10:20Þ
iC ðt Þ ¼ RI m ðω0 C Þ cos ðω0 t þ 90 Þ

The amplitude of the circuit voltage, along with the amplitudes of inductor and capacitor
currents in Eq. (10.20), reaches a maximum at resonance. Next, we wish to introduce the
Q-factor of the circuit, similar to Eq. (10.8) for the series resonator, that is,

υðt Þ ¼ RI m cos ðω0 t Þ, iL ðt Þ ¼ I m Q cos ðω0 t  90 Þ,

ð10:21Þ
iC ðt Þ ¼ I m Q cos ðω0 t þ 90 Þ

Comparing Eq. (10.20) with Eq. (10.21), we obtain a different expression:

pffiffiffiffiffiffiffi
RC R
Q ¼ pffiffiffiffiffiffiffiffi ¼ ω0 RC ¼ ð10:22Þ
L=R ω0 L

which is exactly the reciprocal of the Q-factor of the series RLC circuit. This means that a
high-Q parallel resonant circuit will require higher capacitances than inductances.
Fortunately, all the results related to the series resonant RLC circuit can directly be
converted to the parallel RLC resonant circuit using the substitutions:
υðt Þ ! Riðt Þ, iL ðt Þ ! υC ðt Þ=R, iC ! υL ðt Þ=R ð10:23aÞ

Here, the left-hand side corresponds to the parallel RLC circuit, whereas the right-hand
side corresponds to the series RLC circuit. Furthermore, we need to replace Vm by RIm
and interchange the role of two partial time constants:
L
$ RC ð10:23bÞ
R
in the original solution for the series RLC circuit; see Eq. (10.22). The solution so
constructed will match exactly the solution of the parallel RLC circuit depicted in

X-506
Chapter 10 Section 10.1: Theory of Second-Order Resonant RLC Circuits

Fig. 10.6. This fact is proved by direct substitution. Thus, Eqs. (10.23) reflect the duality
of the series/parallel RLC AC steady-state electric circuits driven by voltage and current
sources, respectively. It means that the results established for one circuit may be applied
to the other circuit and vice versa. A similar duality is established for the transient RLC
circuits. Consequently, we can concentrate our study on the series RLC circuit.

Exercise 10.4: For the parallel resonant RLC circuit in Fig. 10.6, determine resonant
phasor currents IR, IL, and IC if I m ¼ 100 mA, L ¼ 30 μH, C ¼ 1 μF, and R ¼ 100 Ω.
Answer: IR ¼ IS ¼ 0:1∠0 , IL ¼ 1:83∠  90 , IC ¼ 1:83∠ þ 90 ½A.

Example 10.7: The circuit voltage for the parallel RLC circuit in Fig. 10.6 at any
frequency may be written in the form υðtÞ ¼ RI m H ðf Þ cos ð2πf t þ φÞ where H( f ) is a
dimensionless amplitude transfer function, which peaks at the resonant frequency,
H ðf 0 Þ ¼ 1. Create the Bode plot of H( f ) at f 0 ¼ 10 kHz and for Q ¼ 1, 2, 5.
Solution: The amplitude transfer function H( f ) for the parallel RLC circuit coincides with
the expression (10.15b) for the series RLC circuit. The Bode plot also coincides with the
corresponding result for the series RLC circuit shown in Fig. 10.5. However, the Q-factor
is now given by Eq. (10.22). The bandwidth of the parallel resonant circuit is still given
by the expression B ¼ f 0 =Q ½Hz but with the modified Q-factor. This results in
B ¼ 1=ð2πRC Þ ½Hz.

X-507
Chapter 10 Second-Order RLC Circuits

Section 10.2 Construction of Second-Order RLC Filters

An immediate application of the RLC resonant circuits relates to the concept of filter
design. We have already studied the first-order low-pass and high-pass filters on the basis
of RC and RL steady-state AC circuits. In fact, the RLC circuits could also be used as
low-pass and high-pass filters. They perform even better, i.e., more closely linked to the
frequency response of the ideal filter, which implies passing all the frequency compo-
nents below (or above) a certain cutoff frequency and rejecting all other frequency
components. In other words, the frequency responses of the RLC low-pass and high-
pass filters are “steeper” than first-order filters. Not only that, resonant RLC circuits can
form band-pass and band-reject (or band-stop or notch) filters, a task which is impossible
with first-order RC or RL circuits.

10.2.1 Second-Order Band-Pass RLC Filter

A series RLC circuit is shown in Fig. 10.7a. We consider the supply voltage as the input
voltage υin(t) into the filter and the resistor voltage υR(t) as the output voltage υout(t).
Figure 10.7b depicts the corresponding circuit transformation. This transformation
implies that the input voltage is provided by another circuit block and the output voltage
is passed to another circuit block. The circuit in Fig. 10.7b is thus a two-port network.
Qualitatively then, when the frequency of the input voltage is the resonant frequency of
the RLC circuit, the LC block in Fig. 10.7 is replaced by a short circuit (a wire). The
input voltage passes through unchanged. However, if the frequency differs from the
resonant one, the LC block exhibits a large finite impedance that is added to the
resistance R. As a result, the circuit current decreases in amplitude, as does the output
voltage (voltage across the resistor), which is proportional to the current. Those frequen-
cies are thus rejected. The filter so constructed is known as a second-order band-pass
RLC filter.

second-order band-pass RLC filter

a) L C b) L C
+

+
+

vin(t)=vS(t) + R vin(t) R vout(t)

- vR
- - -

Fig. 10.7. Transformation of the series RLC circuit into the band-pass analog RLC filter.

We assume the source voltage (the input filter voltage) is given in the form
υS ðt Þ ¼ V m cos ωt. The phasor current for the corresponding series RLC circuit was
found in the previous section (see Eq. (10.15a)):

X-508
Chapter 10 Section 10.2: Construction of Second-Order RLC Filters

V m =R
I¼   ð10:24aÞ
1 þ jQ ff  ff0
0

pffiffiffiffiffiffiffiffiffi
Here, Q ¼ L=C =R is the corresponding quality factor of the series RLC circuit (quality
pffiffiffiffiffiffiffi
factor of the filter circuit), and f 0 ¼ 1= LC is its resonant frequency. The complex filter
transfer function is defined by the ratio of the corresponding phasors:
VR RI
H ðf Þ  ¼ ð10:24bÞ
VS V m

Substitution of Eq. (10.24a) into Eq. (10.24b) gives the transfer function in the form:
  
1 1 f f0
Hðf Þ ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2ffi ∠  tan Q  ð10:24cÞ
f0
f0 f
1þQ f  f2 f
0

which is equivalent to Eq. (10.15b) of the previous section. Therefore, the results
derived for the series resonant RLC circuit are also valid for the band-pass RLC filter
in Fig. 10.7. In particular, the center frequency of the band-pass filter is the circuit
resonant frequency. The half-power bandwidth of the filter is given by Eq. (10.17),
i.e., B ¼ f 0 =Q ¼ R=ð2πLÞ, and the lower and upper half-power frequencies are known
from Eq. (10.16) of the previous section. Second-order filter circuits are designed
by choosing the values of R, L, C in such a way as to obtain the required values of
Q and f0 (the filter center frequency and the required bandwidth). Thus, we have two
equations for three unknowns. The remaining degree of freedom is used to match the filter
impedances.

Example 10.8: A band-pass RLC filter is required with the center (resonant) frequency of
1 MHz and a half-power bandwidth B of 100 kHz. Create amplitude and phase Bode plots
for the filter in the frequency band from 100 kHz to 10 MHz.
Solution: Clearly, f 0 ¼ 1 MHz. The quality factor of the RLC circuit is found to be
Q ¼ f 0 =B ¼ 10. We plot the magnitude of the transfer function, H( f ), in decibels and its
phase in degrees according to Eq. (10.24c). The result is shown in Fig. 10.8. You should
note that far away from the passband, the filter follows a 20-dB-per-decade roll-off.

X-509
Chapter 10 Second-Order RLC Circuits

H(f), dB Bode plot

0
-3 dB

-10

half-power bandwidth
-20 (passband)

-30

roll-of f of 20 dB roll-of f of 20 dB
per decade per decade
-40
80
Phase transfer function, degrees

60

40

20

-20

-40

-60

-80
5 6 7
10 10 f, Hz 10

Fig. 10.8. Amplitude and phase Bode plot of a band-pass series RLC filter with Q ¼ 10.

Exercise 10.5: In the band-pass filter circuit of Fig.10.7, L ¼ 100 μH, C ¼ 15:9 μF, and
R ¼ 10 Ω. What is the filter bandwidth?
Answer: 15.9 kHz.

Example 10.9: In the previous example, determine the necessary values of L and C given
R ¼ 20 Ω.

pffiffiffiffiffiffiffi From the definition

Solution: pffiffiffiffiffiffiffiffiffi of the resonant frequency and the Q-factor, we obtain
1= LC ¼ 2π  106 , L=C ¼ RQ ¼ 200 ) C 796 pF. Then, we find the required
2
inductance, L ¼ C ðRQÞ 31:8 μH. Alternatively, one could find the inductance
L directly from the definition of the bandwidth for the series RLC resonator, that is,
L ¼ R=ð2π BÞ ¼ 31:8 μH.

X-510
Chapter 10 Section 10.2: Construction of Second-Order RLC Filters

10.2.2 Second-Order Low-Pass RLC Filter

A series RLC circuit is again shown in Fig. 10.9a. We consider the power supply AC
voltage as the input voltage υin(t) into the filter. We monitor the capacitor voltage υC(t) as
the output voltage υout(t) of the filter. Figure 10.9b depicts the corresponding circuit.

second-order low-pass RLC filter

a) R L b) R L

+
+
vin(t)=vS(t) + C vin(t) C vout(t)
- vC
- - -

Fig. 10.9. Transformation of the series RLC circuit into the low-pass analog RLC filter.

The filter so constructed is a second-order low-pass RLC filter. Qualitatively then,

when the frequency of the input voltage is low, the inductor behaves as a short circuit and
the capacitor as an open circuit. The input voltage passes through unchanged. However, if
the frequency is higher than the resonant frequency, both the inductor and the capacitor
prevent transmission: the capacitor shorts out the output voltage, whereas the inductor
reduces the circuit current. The complex filter transfer function is defined by the ratio of
the corresponding phasors:
VC I
H ðf Þ  ¼ ð10:25aÞ
VS jωCV m

We substitute the expression for the phasor current of the series RLC circuit from
Eq. (10.24a) and obtain the transfer function in the form:
  
f0 1 π 1 f f0
Hðf Þ ¼ Q rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2ffi ∠  2  tan Q  ð10:25bÞ
f f
f 0 f
1 þ Q2 f  f0
f
0

Example 10.10: A low-pass RLC filter is required with a passband from 0 to 1 MHz. In
other words, the low-pass filter bandwidth, which extends from zero hertz to the half-
power frequency, should be 1 MHz. Create amplitude and phase Bode plots for the filter
in the frequency band from 100 kHz to 10 MHz.
Solution: The critical point for the low-pass RLC filter design is the proper selection of
the quality factor. The amplitude transfer function in Eq. (10.25b) can exhibit a sharp
peak in the passband with its value higher than one. p Such
ffiffiffi a peak (further investigated
pffiffiffi
in the homework problems) occurs only for Q ¼ 1= 2. The value Q ¼ 1= 2 corre-
sponds to the maximally flat but still steep transfer function (maximally flat Butterworth

X-511
Chapter 10 Second-Order RLC Circuits

Example 10.10 (cont.): response). We will use this value in Eq. (10.25b). Then, the half-
power or 3-dB frequency of the filter will be exactly the resonant frequency f0.
The resulting Bode plots are shown in Fig. 10.10. The transfer function of the filter has
40-dB-per-decade roll-off.

Figure 10.10 shows three amplitude responses: for an ideal filter with the cutoff
frequency of 1 MHz, for a second-order RLC filter with the 3-dB frequency which
coincides with f 0 ¼ 1 MHz, and for a first-order RC (or RL) filter with the break
(half-power) frequency f b ¼ f 0 ¼ 1 MHz. Clearly, the second-order filter better
approaches the desired ideal response. This observation encourages us to consider filters
of higher order.

H(f), dB
0
first-order RC
-10 -3 dB roll-off of 20 dB
per decade
-20
passband
-30

-40
second-order RLC
-50 roll-off of 40 dB
per decade
-60

-70 ideal filter

fb=f0
-80
0
Phase transfer function, degrees

-20
-40
-60
-80
-100
-120
-140
-160
-180
4 5 6 7 8
10 10 10 10 f, Hz 10

Fig. 10.10. Amplitude and phase Bode plots for the low-pass RLC filter (solid curves) compared
with a first-order RC filter (dotted curve).

10.2.3 Second-Order High-Pass RLC Filter

A series RLC circuit is again shown in Fig. 10.11a. We consider the power supply AC
voltage as the input voltage υin(t) into the filter. We next consider the inductor voltage
υL(t) as the output voltage υout(t) of the filter, see Fig. 10.11b. The constructed circuit is a

X-512
Chapter 10 Section 10.2: Construction of Second-Order RLC Filters

second-order high-pass RLC filter. Qualitatively, when the frequency of the input voltage
is low, the capacitor behaves like an open circuit, while the inductor behaves like a short
circuit. Both the inductor and the capacitor prevent transmission. However, if the
frequency is higher than the resonant frequency, the capacitor becomes a short circuit
and the inductor becomes an open circuit. The input voltage is passed through the filter
nearly unchanged. The complex filter transfer function is defined by the ratio of the
corresponding phasors:

second-order high-pass RLC filter

a) R C b) R C

+
+
vin(t)=vS(t) + L vin(t) L vout(t)
- vL
- - -

Fig. 10.11. Transformation of the series RLC circuit into the high-pass analog RLC filter.

VL jωLI
H ðf Þ  ¼ ð10:26aÞ
VS Vm
We substitute the expression for the phasor current of the series RLC circuit from
Eq. (10.24a) and obtain the transfer function in the form:
  
f 1 π 1 f f0
H ðf Þ ¼ Q rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
2 ∠2  tan Q  ð10:26bÞ
f0 f0
f0 f
1þQ f  f 2 f
0

The amplitude transfer function of the high-pass filter in Eq. (10.26b) is the mirror
reflection of the amplitude transfer function for the low-pass filter in Eq. (10.25b) if a
logarithmic frequency scale is used. This fact is seen by substituting f $ 1=f , f 0 $ 1=f 0 ,
which makes both expressions identical.

Example 10.11: A high-pass RLC filter is required with the passband from 0 to 1 MHz.
The high-pass filter half-power frequency should be 1 MHz. Create amplitude and phase
Bode plots for the filter in the band from 100 kHz to 10 MHz.
Solution: The important point for the high-pass RLC filter design is again the proper
ffi the quality factor. Similar to the low-pass filter, we choose the value
selectionpffiffiof
Q ¼ 1= 2, which corresponds to the maximally flat transfer function (Butterworth
response). Then, the half-power or 3-dB frequency of the filter will be exactly the resonant
frequency f0. The resulting Bode plots are shown in Fig. 10.12 in comparison with
the transfer function of the first-order high-pass filter. The amplitude transfer function of
the filter again has the 40-dB-per-decade roll-off.

X-513
Chapter 10 Second-Order RLC Circuits

Exercise 10.6: A band-pass filter is as a series combination of the second-order low-pass

RLC filter and the second-order high-pass RLC filter, respectively. Both filters have the
same half-power frequency. What is the transfer function roll-off far away from the
passband per one octave?
Answer: 12 dB.

H(f), dB
0
first-order RC -3 dB
-10 roll-off of 20 dB
per decade
-20
passband
-30

-40
second-order RLC
roll-off of 40 dB
-50
per decade
-60
ideal filter
-70
fb=f0
-80
Phase transfer function, degrees

160
140
120
100
80
60
40
20
0
4 5 6 7 8
10 10 10 10 f, Hz 10

Fig. 10.12. Amplitude and phase Bode plots for the high-pass RLC filter (solid curves) and
amplitude comparison with a first-order RC filter (dotted curve).

10.2.4 Second-Order Band-Reject RLC Filter

A series RLC circuit is again shown in Fig. 10.13a. We consider the power supply
AC voltage as the input voltage υin(t) into the filter, and the voltage υLC(t) across the LC
block is recorded as the output voltage υout(t). Figure 10.13b depicts the corresponding
filter circuit.

X-514
Chapter 10 Section 10.2: Construction of Second-Order RLC Filters

second-order band-reject RLC filter

a) R b) R

+
+ L L
vin(t)=vS(t) vLC vin(t) vout(t)
- - -
- C C

Fig. 10.13. Transformation of the series RLC circuit into the band-reject analog RLC filter.

The filter so constructed is a second-order band-reject (known as band-stop or notch)

RLC filter. At resonance, the LC block forms a short circuit: the output filter voltage is
thus shorted out. All other frequencies pass through. This filter is useful when a single
tone (e.g., 60 Hz) needs to be rejected. By KVL, its transfer function is equal to one minus
the transfer function of the band-pass filter in Eq. (10.24c), i.e.,


  
1
f f 0
1 π f f0
Hð f Þ ¼ 1   

¼ Q

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1
f0 f  2 ∠ 2  tan Q 
f0 f
1 þ jQ ff  ff0 1þQ f  f2 f f0
0
0

ð10:27Þ
Therefore, the filter behavior is the opposite of the band-pass filter previously analyzed.

Example 10.12: A band-reject RLC filter is required with the center frequency of 1 MHz
and the half-power bandwidth, B, of 100 kHz. Create amplitude and phase Bode plots for
the filter in the frequency band from 100 kHz to 10 MHz.
Solution: The quality factor of the RLC circuit is found to be Q ¼ f 0 =B ¼ 10. We plot the
magnitude of the transfer function H( f ) in decibels and its phase in degrees according to
Eq. (10.27). The result is shown in Fig. 10.14. The filter response is very steep over the
specified frequency range. We can lower the Q-factor, which will lead to a wider band-
width. Note that the amplitude transfer function formally equals zero at the exact resonant
frequency. This result is physically unrealizable since real inductors have a small parasitic
series resistance.

X-515
Chapter 10 Second-Order RLC Circuits

H(f), dB Bode plot

0
-1
-2
-3 dB
-3
-4
-5
-6
-7 half-power bandwidth
-8 (stopband)
-9
-10
Phase transfer function, degrees

160
140
120
100
80
60
40
20
0
5 6 7
10 10 f, Hz 10

Fig. 10.14. Amplitude and phase Bode plots for the band-reject series RLC filter.

10.2.5 Second-Order RLC Filters Derived from the Parallel RLC Circuit
All second-order filters considered so far are derived from the series RLC circuit, with the
same quality factor given by Q ¼ ω0 L=R. The natural structure after shorting out the
input voltage source is shown in Fig. 10.15a. A complementary group of these filter
circuits exists; after shorting out the input voltage source, its natural structure is that of the
parallel RLC circuit seen in Fig. 10.15b. These circuits operate quite similarly, but all of
them have the quality factor of the parallel RLC resonator, that is, Q ¼ ω0 RC.

a) L b)

R L C

Fig. 10.15. (a) Series RLC circuit with no excitation and (b) parallel RLC circuit with no
excitation.

X-516
Chapter 10 Section 10.2: Construction of Second-Order RLC Filters

For the filter circuits derived from the parallel RLC circuit, the resonant frequency
still has to satisfy the condition that a real circuit impedance is “seen” by the voltage
source. The resonant frequency found this way either does not equal the undamped
 pffiffiffiffiffiffiffi
resonant frequency f 0 ¼ 1= 2π LC or does not exist at all. However, the structure of
the filter equations is not affected by this result. Only the undamped resonant frequency
f0 appears to be important for the voltage transfer function.

X-517
Chapter 10 Second-Order RLC Circuits

Section 10.3 RLC Circuits for Near-Field

Communications and Proximity Sensors
10.3.1 Near-Field Wireless Link
Near-field wireless communication can transfer data, power, or both of them simulta-
neously. Common data-related applications include radio-frequency identification (RFID)
systems of 125/134 kHz and 13.56 MHz, electronic article surveillance (EAS) for
electronic anti-theft devices in shops, and mobile and other portable device near-field
communication (NFC). Promising applications in biomedical engineering have also been
explored. Figure 10.16 shows the key concept of a near-field wireless link. The transmitter
and the receiver inductor coils share a common magnetic flux density ~ B in the near field.
The transmitter/receiver system in Fig. 10.16 is known as an inductively coupled system.

z
Inductive
Power, data coupling
Transmitter Receiver
R
B
+ Processing
vS(t) L
- device

Feedback, data

Fig. 10.16. The concept of the near-field wireless link; the magnetic flux density is shared between
receiver and transmitter coils.

In contrast to the radio-frequency radiating fields, the near field ~

B is very strong in the
vicinity of the coil antenna. However, this field very rapidly decays at larger distances
from the transmitter. For example, consider a transmitter coil with N loops of area A each.
The corresponding near field of the transmitter coil in Fig. 10.16 with current i(t) on the
coil axis at the axial distances z much greater than the coil length (and the loop radius)
may be found in the form:
 
μ0 A iðt Þ
B¼N 3
½T, μ0 ¼ 4π  107 H=m ð10:28Þ
2π z

where B is recorded in tesla T. The expression in square brackets is the contribution of a

single loop. Thus, the near-field decay is inversely proportional to the third degree of the
separation distance. This observation (obtained via an asymptotic analysis of the related
magnetostatic expressions) is also valid for any quasi-static magnetic (and electric)
dipole. Such a short-range wireless communication is both safe and effective.

X-518
Chapter 10 Section 10.3: RLC Circuits for Near-Field Communications. . .

10.3.2 Transmitter Circuit

In a transmitter circuit shown in Fig. 10.17a, the function generator is modeled by an ideal
voltage source υS ðt Þ ¼ V m cos ω t in series with resistance R. The function generator is
connected to the transmitter coil modeled by the inductance L. The goal is to increase the
magnetic flux density ~ B of the transmitter. According to Eq. (10.28), the obvious choice is
to increase the inductance of the transmitter. However, this operation would decrease the
circuit current i(t) due to an increase of the impedance magnitude. We will attempt to
solve this problem with the series resonant RLC circuit shown in Fig. 10.17b.

a) b)
R R C
B B
vS(t) + L vS(t) + L
- -

Fig. 10.17. Using a series capacitor in order to increase the circuit current.

The original and the modified circuits in Fig. 10.2 are both solved by using the phasor
method. We denote the desired operating frequency by f0. For the original circuit in
Fig. 10.17a, the phasor current may then be written in the form:
 
Vm Vm Vm 1 1 f
I¼ ¼ ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2ffi ∠  tan Q , Q ¼ ω0 ðL=RÞ ð10:29Þ
Zeq R þ jωL R f0
1þQ f 2 f
0

For the series RLC circuit in Fig. 10.17b, the capacitance is chosen in such a way that the
resonant frequency of the circuit coincides with the operation frequency f0. The phasor
current for the RLC circuit has been derived in the previous sections. It has the form:
  
Vm Vm Vm 1 1 f f0
I¼ ¼   ¼ r ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 ffi
2 ∠  tan Q  ð10:30Þ
Zeq R þ j ωL  ωC 1 R f0
f0 f
1þQ f  f 2 f
0

Note the presence of the quality factor, Q ¼ ω0 ðL=RÞ, for the series RLC circuit in both
Eqs. (10.29) and (10.30). At exactly the operation frequency, f ¼ f 0 , the ratio of current
magnitudes (both phasors ∠
have the magnitude of one) becomes
qffiffiffiffiffiffiffiffiffiffiffiffiffiffi
I m circuit with series capacitor
¼ 1 þ Q2 Q f or Q  1 ð10:31Þ
I m original circuit
This ratio may significantly exceed one for high Q values. Thus, the series RLC circuit
may considerably increase the circuit current and the associated magnetic field.

X-519
Chapter 10 Second-Order RLC Circuits

Exercise 10.7: It is suggested to increase the magnetic field for the circuit without the
capacitor in Fig. 10.7a by simply doubling the number of coil turns and increasing the coil
length by the factor of two. Given that:
1. Q ¼ ω0 ðL=RÞ  1 at the operation frequency
2. Q ¼ ω0 ðL=RÞ  1 at the operation frequency
how does the magnetic field B change?
Answer: (i) B remains the same. (ii) B doubles.

Example 10.13: Given the operation frequency (center band frequency) of f 0 ¼ 1 MHz
and V m ¼ 10 V, L ¼ 50 μH, R ¼ 50 Ω, plot the amplitude of the circuit current as a
function of source frequency for the original (RL) and modified (resonant RLC) circuits in
in Fig. 10.17 over the frequency band from 0.5 to 1.5 MHz.
Solution: The quality factor is found to be Q ¼ 6:283. Next, we plot both current
amplitudes based on Eqs. (10.29) and (10.30) using a linear scale. The result is shown in
Fig. 10.18. The amplitude of the circuit current increases from 31.4 to 200 mA at the
operation frequency f0 (resonant frequency of the RLC circuit).

Amplitude of the circuit current, mA

200

160 ~QI0

with series capacitor C

120

without capacitor C
80

f0
40

I0
0
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
source frequency, MHz

Fig. 10.18. Amplitudes of the circuit current for the original and modified circuits in Fig. 10.17.

10.3.3 Receiver Circuit

Consider the receiver coil in Fig. 10.19a. Its equivalent circuit includes the ideal inductor
L in series with R, which is the resistance of the coil winding. It also includes an induced
emf (electromotive force) voltage source υemf(t), which follows Faraday’s law of induction:

X-520
Chapter 10 Section 10.3: RLC Circuits for Near-Field Communications. . .

dBðt Þ
υemf ðt Þ  AN ¼ V m cos ðωt Þ, Bðt Þ ¼ Bm sin ðωt Þ ð10:32Þ
dt
Here, B(t) is the coaxial component of the external, time-varying magnetic flux
density of the transmitter at the receiver location. The source voltage amplitude is
given by V m ¼ AN ωBm where A is the area of the receiver coil and N is the number
of coil turns. The major parameter of interest is the (small) open-circuit voltage of
the receiver coil, υout(t). It is desired to increase this voltage. For the circuit shown
in Fig. 10.19a, υout(t) is always equal to υemf(t). However, the situation will change if
we create a series RLC circuit as shown in Fig. 10.19b. The output voltage becomes the
capacitor voltage. We will attempt to increase υout(t) by using the resonance condition.

a) R L
B
+

+
vemf(t)
vout(t) + vout(t)
- - -

b) R L
B
+

+
vemf(t)
C vout(t) + C vout(t)
- - -

Fig. 10.19. (a) Receiver coil and (b) receiver coil with capacitance to increase the open-circuit
voltage.

The circuit in Fig. 10.19b is solved using the phasor method. The desired resonant
frequency (operating frequency) is f0. The phasor for the output voltage has the form:
  
V m =ðjωC Þ f0 Vm π 1 f f0
Vout ¼ VC ¼ ¼ Q rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
 2ffi ∠  2  tan Q 
R þ jωL þ jωC
1 f f0
f0 f
1þQ f  f
2 f
0

ð10:33Þ
where Q ¼ ω0 ðL=RÞ is again the quality factor of the series RLC resonant circuit (and
simultaneously the quality factor of the nonideal inductor with series resistance R). At the
exact resonant frequency, the output voltage amplitude becomes
V out ¼ QV m ð10:34Þ

X-521
Chapter 10 Second-Order RLC Circuits

This value may significantly exceed Vm given a high value of Q. Thus, the series RLC
circuit formed with the help of the shunt capacitor C in Fig. 10.19b may considerably
increase the received voltage.

Example 10.14: Given the operating frequency (center band frequency) of f 0 ¼ 1 MHz
and V m ¼ 10 mV, L ¼ 78 μH, R ¼ 10 Ω, plot the amplitude of the output voltage for
the original and modified circuits in Fig. 10.19 as a function of source frequency over the
band from 0.5 to 1.5 MHz.
Solution:
 We  find the required capacitance value first. Specifically,
C ¼ 1= Lð2πf 0 Þ2 325 pF. The quality factor is given by Q 49:0. Next, we plot
both voltage amplitudes. The first one is simply Vm. The second plot is based on
Eq. (10.33). The results are shown in Fig. 10.20. The amplitude of the output voltage
increases from 10 to 490 mV at the operating frequency f0 (resonant frequency of the series
RLC circuit).

Amplitude of the open-circuit voltage, mV

500

400
~QV0

300
with shunt capacitor C

200

100 without capacitor C

f0
0
0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 1.4 1.5
V0 source frequency, MHz

Fig. 10.20. Amplitudes of the output voltage for the original and modified circuits in Fig. 10.19.

The circuit in Fig. 10.19b is the low-pass second-order RLC filter studied in the
previous section, right? Why is it boosting the source voltage instead of just passing it
through? The key is the Q-factor. The present circuit operates as a filter at relatively small
values of the quality factors, i.e., Q 1. At higher Q values, the circuit generates a
voltage spike close to the resonant frequency and operates as a voltage multiplier. This
operation is similar to the operation of an electric transformer.

10.3.4 Application Example: Near-Field Wireless Link in Laboratory

Figure 10.21 shows a prototype of the near-field link implemented in an undergraduate
laboratory. The operating frequency of the transmitter is tunable; it ranges from 400 kHz
to 1.2 MHz. Despite this relatively high frequencies, the circuitry can still be implemented

X-522
Chapter 10 Section 10.3: RLC Circuits for Near-Field Communications. . .

on standard protoboards. The key is the tunability of both the transmitter and the receiver,
which simultaneously accounts for the parasitic capacitance of the board.

function transmitter coil

generator

trimmer capacitor LM386 audio amplifier chip

receiver coil and associated circuitry

tuning capacitance made of fixed

capacitors

Fig. 10.21. Prototype of the near-field link implemented in laboratory.

The transmitter is driven by a function generator; the tuning is made by a bank of fixed
capacitors from a laboratory kit. The receiver uses a single-ended magnetic-core coil
(a loopstick antenna) with an inductance of approximately 1 mH and a series resistance of
about 10 Ω. The RLC circuit at the receiver uses a trimmer capacitor of 10–180 pF range for
tuning. When the transmitter is sending an amplitude-modulated signal from the function
generator, the receiver operates as an AM radio given the subsequent rectifying circuit with a
germanium diode and an audio amplifier IC. Frequency modulation is also possible; how-
ever, the receiver circuit has to be modified accordingly. The system operating range is up to
two feet on average. When an external modulation input of the function generator is used, an
audio clip may be transmitted.

10.3.5 Application Example: Proximity Sensors

The idea of the resonant RLC proximity sensor is quite simple. Assume that the
inductance is a large coil or simply a loop of wire. When a metal object to be detected
is brought in proximity to the loop, its (self) inductance changes. This causes a detectable
change in the resonant frequency f0 of an RLC tank circuit. After encoding, information
may be extracted about the presence of the object and sometimes of its size. This is the
well-known principle of a metal detector. A large variety of metal-detecting circuits
already exist, and more are still awaiting discovery. Another idea is to change the
capacitance by putting a dielectric object (such as a medical pill) inside the capacitor.
A similar change in the resonant frequency may be observed and detected. Such a device
may be used, for example, as an automatic pill counter.
The detector circuit itself can operate based on three different principles. First, a simple
method is to use the series RLC tank circuit with the external AC power supply.
The measured parameter is the amplitude of the circuit current (resistor voltage) at the

X-523
Chapter 10 Second-Order RLC Circuits

frequency of the AC source. When the resonant frequency is close to the AC frequency,
the circuit voltage is large. However, when the resonant frequency deviates from the
source frequency, the circuit voltage becomes smaller. The change in the circuit voltage is
detected. A second method is to make the tank circuit self-resonant, by using an amplifier
with a positive feedback. A resonant circuit so built does not need an AC power supply. It
oscillates exactly at f0 when there is no object to be detected. When the object is present
the oscillation frequency changes. The change in the AC frequency is encoded by another
electronic circuit. Using self-resonant tank circuits is perhaps the most common method
in practice. A third method is based on the effect of the resistance in the tank circuit.
When a metal object is placed close to the coil, the coil’s series resistance significantly
increases, due to the so-called eddy current losses (for all metals) and, possibly, hysteresis
losses (for magnetic metals such as iron, nickel, steel alloys, etc.). The increase in the
resistance leads to smaller voltage oscillations in the self-resonant circuit. The circuit may
be tuned in such a way as to stop oscillating at a given value of the extra resistance. Great
sensitivity may be achieved with this method.
Figure 10.22 shows the inductor assembly in a resonant sensor for an automatic traffic
light. The inductor now is a single-turn (or multi-turn) pavement loop. When a vehicle is
located above the loop, its (self) inductance L decreases. This leads to an increase in the
resonance frequency. The change in frequency, not the change in the amplitude, is
typically detected and encoded. The latter is used to indicate the presence of a vehicle
and to adjust the traffic light control. Most vehicle detectors based on loop inductors
operate with frequencies from 10 to 100 kHz. A (simplified) equivalent tank circuit for
the traffic light control is shown in Fig. 10.23b. We note the series resistance R, which is
the parasitic resistance of the loop. The parasitic resistance includes both the effect of the
passing vehicle and of the ground.

Fig. 10.22. Multiple vehicle detection loops after installation at an intersection. Courtesy of the US
Traffic Corporation, Loop Application Note of 3/10/03.

X-524
Chapter 10 Section 10.3: RLC Circuits for Near-Field Communications. . .

a)
Parallel fixed
capacitor C Vehicle detecting
a loop (L+R)

Detector
b)
a
L
C
R
b

Fig. 10.23. (a) Simplified schematic of the vehicle detecting system and (b) equivalent resonant
circuit.

The circuit in Fig. 10.23b may be analyzed exactly in the same way as the series/
parallel RLC tank circuits in Section 10.1. The tricky part for the tank circuit block in
pffiffiffiffiffiffiffi
Fig. 10.23b is that its resonant angular frequency is no longer 1= LC. However, it can
still be found from the condition of a purely real equivalent impedance Zeq, see the
summary of this chapter.
Single coils of special shapes—the horseshoe shape—may be used to detect the level
and the presence of molten metals through the walls of (large) casting molds and for other
purposes. The equivalent circuit is the parallel RLC tank circuit. When properly tuned,
the self-resonating circuit quantitatively detects variations in molten metal level through
4–500 thick walls, see Fig. 10.24.

Fig. 10.24. A half-toroidal (horseshoe) coil used to concentrate the magnetic field between its tips
in a molten metal detector (Foley, Biederman, Ludwig, and Makarov, US Patent 7,828,043 Nov.
9th 2010).

X-525
Chapter 10 Second-Order RLC Circuits

Summary
TERM Series RLC circuit Parallel RLC circuit
Series and
parallel RLC
resonators

Resonant pffiffiffiffiffiffiffi  pffiffiffiffiffiffiffi pffiffiffiffiffiffiffi  pffiffiffiffiffiffiffi

ω0 ¼ 1= LC , f 0 ¼ 1= 2π LC ω0 ¼ 1= LC , f 0 ¼ 1= 2π LC
frequency
Coincides with the undamped resonant Coincides with the undamped resonant
frequency frequency
Quality factor pffiffiffiffiffiffiffiffi pffiffiffiffiffiffiffi
L=R 1 RC 1
of the Q ¼ pffiffiffiffiffiffiffi ¼ ¼ ω0 ðL=RÞ Q ¼ pffiffiffiffiffiffiffiffi ¼ ω0 RC ¼
RC ω0 RC L=R ω0 ðL=RÞ
resonant
circuit dimensionless dimensionless
Bandwidth of f0 1 f0 1
B fU fL ¼ ¼ ½Hz B  fU fL ¼ ¼ ½Hz
the resonant Q 2π ð L=RÞ Q 2π RC
circuit
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ! sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !
Half-power 1 1 1 1
lower and f L, U ¼ f 0 1þ  ½Hz f L, U ¼ f 0 1þ  ½Hz
ð2QÞ2 2Q ð2QÞ2 2Q
upper
frequencies
Other RLC Circuit diagram Resonant frequency
resonators
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
1 RC
ω0 ¼ 1
ðRC Þ L=R
L+R||C Different from the undamped resonant
frequency

1 1
ω0 ¼ rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ðL=RÞ RC
1
C+L||R L=R
Different from the undamped resonant
frequency
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
(R+L)||C 1 L=R
ω0 ¼ 1
ðL=RÞ RC
Different from the undamped resonant
frequency

(continued)

X-526
Chapter 10 Summary

 pffiffiffiffiffiffiffi
RLC filter circuits derived from the series RLC circuit: f 0 ¼ 1= 2π LC , Q ¼ 1=ðω0 RC Þ
Band-pass
filter H0 ð f Þ ¼ 
1

f f
1 þ jQ  0
f0 f

Low-pass
filter f0
Hð f Þ ¼ Q H0 ð f Þ
f

High-pass
filter f
Hð f Þ ¼ Q H0 ð f Þ
f0

Band-reject
filter  
f f0
Hð f Þ ¼ jQ  H0 ð f Þ
f0 f

 pffiffiffiffiffiffiffi
RLC filter circuits derived from the parallel RLC circuit: f 0 ¼ 1= 2π LC , Q ¼ ω0 RC
Band-pass
1
filter based H0 ð f Þ ¼  
f f
on parallel 1 þ jQ  0
RLC circuit f0 f

Low-pass
filter based f0
on parallel Hð f Þ ¼ Q H0 ð f Þ
f
RLC circuit

High-pass
filter based f
Hð f Þ ¼ Q H0 ð f Þ
on parallel f0
RLC circuit

Band-reject
filter based  
on parallel f f0
Hð f Þ ¼ jQ  H0 ð f Þ
RLC circuit f0 f

(continued)

X-527
Chapter 10 Second-Order RLC Circuits

Near-field wireless transmitter/receiver

Resonant
circuit at the
transmitter
(TX)
The series capacitor forms the series RLC circuit and increases the amplitude of the
pffiffiffiffiffiffiffiffiffiffiffiffiffiffi
magnetic flux density anywhere in space by the factor 1 þ Q2, Q ¼ ω0 ðL=RÞ

Resonant
circuit at the
receiver (RX)
The shunt capacitor again forms the series RLC circuit and increases the amplitude
of the output voltage by the factor Q, Q ¼ ω0 ðL=RÞ

X-528
Chapter 10 Problems

Problems A. Determine the undamped resonant fre-

quency f0 of the circuit.
10.1. Theory of the Second- B. Construct the phasor diagram for phasor
Order RLC Resonator voltages VS, VL, and VC when the source
frequency is 90 % of the resonant
10.1.1 Self-Oscillating Ideal LC Circuit frequency.
10.1.2 Series Resonant Ideal LC Circuit C. Describe how your phasor diagram
Problem 10.1. Give an example of a self- would change if the inductance becomes
oscillating (resonant) mechanical system differ- 1 mH instead of 0.5 mH.
ent from that in Fig. 10.2 of this section.
i(t) L
-
Problem 10.2. For the self-oscillating circuit vL

+
shown in the figure below, the circuit current +
vS(t)
is specified by iðtÞ ¼ I m cos ω t. Given -
I m ¼ 200 mA, L ¼ 0:63 mH, C ¼ 1 μF:
- vC

+
A. Determine the undamped resonant fre- C
quency f0. 30 Im
B. Construct the phasor diagram for phasor
voltages VL and VC and phasor 20

current I on the same plot. Assume that 10

every plot division corresponds to 1 V or Re
to 100 mA. -30 -20 -10 0 10 20 30
-10
L
vL
-
+

-20
i(t) i(t)
-30
- vC
+

C
Problem 10.4. For a series ideal LC circuit
6 Im shown in the figure below, the voltage
source has the form υS ðt Þ ¼ V m cos ωt. Given
4 V m ¼ 5 V, L ¼ 1 mH, C ¼ 0:5μF:
2 A. Determine the undamped resonant fre-
Re
quency f0 of the circuit.
-6 -4 -2 0 2 4 6 B. Construct the phasor diagram for phasor
-2
voltages VS, VL, and VC when the source
frequency is 111 % of the resonant
-4 frequency.
-6
C. Describe how your phasor diagram
would change if the capacitance
becomes 1 μF instead of 0.5 μF.
Problem 10.3. For a series ideal LC circuit
shown in the figure below, the voltage source
has the form υS ðtÞ ¼ V m cos ωt. Given
V m ¼ 5 V, L ¼ 0:5 mH, C ¼ 1 μF:

X-529
Chapter 10 Second-Order RLC Circuits

i(t) C C. Determine the real-valued circuit current

vC
- i(t) and the inductor/capacitor voltages

+
+ υL(t), υC(t) at resonance.
v S (t)
-
- vL i(t)

+
L
vL
-

+
L

+
30 Im
vS(t) + vR R
- -
20
- vC

+
10 C
Im 600
Re
-30 -20 -10 0 10 20 30
400
-10
200
-20
Re
-30 -600 -400 -200 0 200 400 600

-200

-400
10.1.3 Series Resonant RLC Circuit:
-600
Resonance Condition
10.1.4 Quality Factor Q of the Series
Resonant RLC Circuit Problem 10.8. In the series resonant RLC
circuit shown in the figure below, given V ¼
10.1.5 Bandwidth of the Series Resonant 5 V, L ¼ 30 μH, C ¼ 0:48 nF, R ¼ 50m Ω:
RLC Circuit A. Determine resonant frequency and the
Problem 10.5. For a generic series resonant RLC Q-factor.
circuit with the supply voltage υS ðt Þ ¼ B. Determine resonant phasor voltages VR,
V m cos ω t, resistance R, inductance L, and capac- VL, and VC; construct the phasor diagram.
itance C, give the expressions (show units) for: C. Determine the real-valued resistor
A. Equivalent circuit impedance at the voltage υR(t) and the inductor/capacitor
resonance voltages υL(t), υC(t) at the resonance.
B. Resonant frequency
C. Quality factor of the resonant circuit i(t) L
vL
-
+

Problem 10.6. Describe the physical meaning

+

+
of the quality factor of the series RLC resonator v S (t)
-
vR
- R

circuit in your own words. - vC

+

Problem 10.7. In the series resonant RLC C

circuit shown in the figure that follows, given 30 Im

V m ¼ 1 V, L ¼ 1 mH, C ¼ 80 pF, R ¼ 10 Ω: 20
A. Determine resonant frequency and the
Q-factor. 10

B. Determine resonant phasor current and Re

phasor voltages VR, VL, and VC; con- -30 -20 -10 0 10 20 30

struct the phasor diagram. Assume volt- -10

age scale in volts and current scale in -20

milliamperes.
-30

X-530
Chapter 10 Problems

Problem 10.9. A series resonant LC circuit is Hint: The resonance is defined by the con-
driven by a laboratory AC voltage source with dition of the purely real equivalent impedance
an amplitude V m ¼ 12 V and an internal resis- between terminals a and b. In other words, Im
 
tance of 50 Ω (a function generator). Which Zeq ¼ 0 at the resonance.
value should the ratio L/C have to obtain the
amplitude of the capacitor voltage equal to a L
200 V at the resonance?

Problem 10.10. A series resonant RLC circuit R C

is needed with the resonant frequency of 1 MHz
and a Q-factor of 100. The circuit resistance is
10 Ω. Determine the necessary values of b

L and C.
Problem 10.16. Repeat the previous problem
Problem 10.11. Describe the physical meaning for the circuit shown in the figure that follows.
of the resonance bandwidth of the series reso-
nant RLC circuit in your own words. C
a

Problem 10.12. A series resonant RLC circuit

has the resonant frequency of 1 MHz and the R L
quality factor of 10. Create the Bode plot for the
amplitude of the circuit current normalized by
b
its maximum value at the resonance over fre-
quency band from 0.5 to 2 MHz.
H(f), dB Bode plot
10.1.6 Parallel Resonant RLC Circuit:
0
Duality
-5
Problem 10.17. For a generic parallel RLC
-10
resonant circuit with the supply current
iS ðt Þ ¼ I m cos ωt, resistance R, inductance L,
-15

-20
and capacitance C, give the expressions (show
-25
units) for:
0.5MHz 1MHz 2MHz
A. Equivalent circuit impedance at the
resonance
Problem 10.13. Determine the bandwidth, B, B. Resonant frequency
of the series resonant RLC circuit with the C. Quality factor of the resonant circuit
resonant frequency of 1 MHz and a Q-factor
of 100. Problem 10.18. In the parallel resonant RLC
circuit shown in the figure that follows, given
Problem 10.14. A series resonant RLC circuit I m ¼ 0:5 A, L ¼ 30 μH,C ¼ 0:43 μF, R ¼ 50 Ω:
is needed with the resonant frequency of 1 MHz A. Determine the resonant frequency and
and the bandwidth of 10 kHz. Given the circuit the Q-factor.
resistance of 10 Ω, determine L and C. B. Determine resonant phasor currents IR,
IL, and IC; construct the phasor diagram.
Problem 10.15. For the RLC circuit block
C. Determine the real-valued resistor cur-
shown in the figure, establish the resonant fre-
rent iR(t) and the inductor/capacitor cur-
quency in terms of component values.
rents iL(t), iC(t) at the resonance.

X-531
Chapter 10 Second-Order RLC Circuits

Problem 10.22. A band-pass RLC filter is

R L C required with the center (resonant) frequency
+ + + +
iS(t) v v v v of 100 kHz and the half-power bandwidth, B, of
- - - -
20 kHz.
A. Create its amplitude Bode plot in the
3 Im
frequency band from 10 kHz to 1 MHz.
B. Label the filter passband.
2 C. Determine the necessary values of L and
1
C given R ¼ 20 Ω.
Re H(f), dB Bode plot
0
-3 -2 -1 0 1 2 3 -3 dB

-1 -10

-2 -20

-3 -30

-40
4 5 6
10 10 f, Hz 10
Problem 10.19. Determine the bandwidth, B,
of the parallel resonant RLC circuit with the Problem 10.23
resonant frequency of 0.5 MHz and a Q-factor A. Draw the circuit diagram of the second-
of 50. order RLC low-pass filter. Label R, L,
and C.
Problem 10.20. A parallel resonant RLC cir-
B. Show the input and output ports (input
cuit is needed with the resonant frequency of
and output voltages)
1 MHz and the bandwidth of 10 kHz. Given the
C. Define the resonant frequency and the
circuit resistance of 100 Ω, determine L and C.
Q-factor of the filter circuit.
D. Which Q-factor is required for the max-
10.2: Construction of imally flat response?
E. What is the filter’s half-power frequency
Second-Order RLC Filters for the maximally flat response?
10.2.1 Second-Order Band-Pass RLC
Filter Problem 10.24. A low-pass RLC filter is required
with the passband from 0 to 1 MHz. Create ampli-
10.2.2 Second-Order Low-Pass RLC tude Bode plots for the filter in the frequency band
Filter from 100 kHz to 10 MHz given the resonant
10.2.3 Second-Order High-Pass RLC frequency of the filter circuit of 1 MHz and
Filter A. Q ¼ 5 pffiffiffi
10.2.4 Second-Order Band-Reject RLC B. Q ¼ 1= 2
Filter C. Q ¼ 0:2
Problem 10.21 H(f), dB
20
A. Draw the circuit diagram of the second- 10
order RLC band-pass filter. Label R, L, 0

and C. -10
-20
B. Show the input and output ports (input -30
and output voltages) -40

C. Define the resonant frequency and the Q- -50

f0

factor of the filter circuit.

-60
104 105 106 107 f, Hz 108

X-532
Chapter 10 Problems

Problem 10.25*. Generate Fig. 10.4 of this 0

H(f), dB Bode plot

section, the Bode plots for the low-pass filter -1

-2
-3 dB
using MATLAB. -3
-4
-5
Problem 10.26 -6
-7
A. Draw the circuit diagram of the second- -8

order RLC high-pass filter. Label R, L, -9

-10
104 105 f, Hz 106
and C.
B. Show the input and output ports (input
and output voltages) 10.2.6. Second-Order RLC Filters
C. Define the resonant frequency and the Q- Derived from the Parallel RLC Circuit
factor of the filter circuit. Problem 10.30. Second-order RLC filters may
D. Which Q-factor is required for the max- be constructed either on the basis of the series
imally flat response? RLC circuit or on the basis of the parallel RLC
E. What is the filter’s half-power frequency circuit. Thepffiffiffiffiffiffi

undamped
ffi resonant frequency,
for the maximally flat response? f 0 ¼ 1= 2π LC , which is present in the filter
equations, remains the same in either case. How-
Problem 10.27. A high-pass RLC filter is ever, the quality factor does not. Three unknown
required with the passband from 0 to 1 MHz. second-order RLC filter circuits are shown in the
Create amplitude Bode plots for the filter in the figure that follows.
frequency band from 100 kHz to 10 MHz given A. Determine the filter function (band-pass,
the resonant frequency of the filter circuit of low-pass, high-pass, or band-reject).
1 MHz and B. By analyzing filter’s natural structure (after
A. Q ¼ 10pffiffiffi shorting out the input voltage source),
B. Q ¼ 1= 2 determine the expression for the filter qual-
C. Q ¼ 0:1 ity factor.

Problem 10.28 L C a)
A. Draw the circuit diagram of the second-
order RLC band-reject filter. Label R, L,
+

and C. vin(t) R vout(t)

- -
B. Show the input and output ports (input
and output voltages)
C. Define the resonant frequency and the L b)
Q-factor of the filter circuit.
+

Problem 10.29. A band-reject RLC filter is vin(t)

- C R vout(t)
-
required with the center (resonant) frequency
of 100 kHz and the half-power bandwidth, B,
of 20 kHz. C c)
A. Create its amplitude Bode plot in the
frequency band from 10 kHz to 1 MHz.
+

B. Label the filter passband. vin(t) L R vout(t)

C. Determine the necessary values of L and - -
C given R ¼ 20 Ω.

X-533
Chapter 10 Second-Order RLC Circuits

10.3. RLC Circuits for Near- Problem 10.34. In the circuit shown in the
figure that follows, a receiver coil antenna is
Field Communications subject to an external magnetic field oscillating
and Proximity Sensors at 1 MHz. A capacitor C is introduced in parallel
with the coil having inductance L and series
10.3.1 Near-Field Wireless Link resistance R in order to set up a series resonant
10.3.2 Transmitter Circuit RLC circuit and increase the amplitude of the
10.3.3 Receiver Circuit output voltage υout(t). Determine the ratio of the
10.3.4 Application Example: Near-Field output voltage amplitudes with and without the
capacitor given that L ¼ 1000 μH, R ¼ 10 Ω.
Wireless Link in Laboratory
R L
10.3.5 Application Example: Proximity

+
Sensors vemf(t)
vout(t) + vout(t)
Problem 10.31. Describe the concept and pur- C
- C

pose of the near-field wireless link in your own - -

words. Think of an example where the link may
be used solely for the power transfer.
Problem 10.35. Given the operation frequency
(center band frequency) of f 0 ¼ 1 MHz and
Problem 10.32. In the circuit shown in the υemf ðtÞ ¼ V m cos ðωtÞ, V m ¼ 1 mV, L ¼
following figure, a capacitor C is introduced in 500 μH, R ¼ 50 Ω, plot to scale the ampli-
series with an ideal coil having inductance L in tude of the output voltage as a function of
order to set up a series resonant RLC circuit and source frequency for the original (RL) and
increase the amplitude of the magnetic flux den- modified (resonant series RLC) circuits in the
sity ~
B oscillating at 1 MHz. Determine the ratio figure to the previous problem over the fre-
of the magnetic flux amplitudes quency band from 0.5 to 1.5 MHz. Label the
amplitude values at the operation frequency.
Bm circuit with series capacitor
Bm original circuit Problem 10.36. In the circuit shown in the
figure below, a receive coil antenna is subject
with and without the capacitor anywhere in to an external magnetic field oscillating at
space given that L ¼ 100 μH, R ¼ 25 Ω. 1 MHz. A capacitor C is introduced in parallel
with the coil having inductance L and series
R C resistance R in order to set up a series resonant
B RLC circuit and increase the amplitude of
+
the output voltage υout(t). Determine the output
vS(t) L
- voltage amplitudes with and without the capac-
itor given that υemf ðt Þ ¼ 1 cos ðωt Þ ½mV and
that L ¼ 1000 μH, R ¼ 10 Ω, Rf ¼ 100 Ω.
Rf
receiving coil antenna
Problem 10.33. Given the operation frequency L R C a
(center band frequency) of f 0 ¼ 1 MHz and -
+
V m ¼ 1 V, L ¼ 100 μH, R ¼ 20 Ω, plot to i(t) +
+ v vout
scale the amplitude of the circuit current as a - emf
-
function of source frequency for the original b
(RL) and modified (resonant RLC) circuits in
the figure to the previous problem over the fre-
0V
quency band from 0.5 to 1.5 MHz. Label the
amplitude values at the operation frequency.
X-534
Chapter 11

Chapter 11: AC Power and Power

Distribution

Overview
Prerequisites:
- Knowledge of complex arithmetic
- Knowledge of basic circuit analysis (Chapters 3 and 4)
- Knowledge of phasor/impedance method for AC circuit analysis (Chapter 8)

Objectives of Section 11.1:

- Find average AC power for a resistive load and understand the rms values
- Express average power for any AC load in terms of power angle and power factor
- Express average power in terms of phasors/impedances
- Define major AC power types: average power, reactive power, complex power, and
apparent power
- Be able to construct the power triangle and classify the load power factor

Objectives of Section 11.2:

- Be able to perform power factor correction of an inductive load (AC motor)
- Learn about maximum power efficiency technique in general
- Derive and test a simple condition for maximum power transfer to a load from an
arbitrary AC source

Objectives of Section 11.3:

- Learn the structure of power distribution systems
- Establish the concept of the three-phase power transmission system
- Understand the meaning and realization of three-phase source and three-phase load
- Solve for phase and line voltages and line currents in the three-phase balanced
wye-wye system
- Establish the meaning and the role of the neutral conductor in the wye-wye power
distribution system
Objectives of Section 11.4:
- Establish that the instantaneous power in balanced three-phase systems is constant
- Extend the concepts of reactive power, complex power, and apparent power to the
three-phase systems
- Compare conductor material consumption in single-phase and three-phase systems
- Become familiar with delta-connected three-phase sources and loads
- Establish equivalency between delta and wye topologies with no ground

© Springer Nature Switzerland AG 2019 XI-535

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_11
Chapter 11 AC Power and Power Distribution

Application Examples:
- rms voltages and AC frequencies around the world
- Wattmeter
- Automatic power factor correction system
- Examples of three-phase source and the load
- Conductor material consumption in three-phase systems

Keywords:
Time averaging, Average power, rms voltage, rms current, AC fuse, Root mean square, Sawtooth
wave, Triangular wave, Noise signals, Power angle, Power factor, Reactance, Capacitive
reactance, Inductive reactance, Active power, True power, Reactive power, Complex power,
Apparent power, VAR (volt-amperes reactive), VA (volt-amperes), Power triangle, Lagging
power factor, Leading power factor, Wattmeter, Wattmeter current coil, Wattmeter potential
coil, AC power conservation laws, Power factor correction, Power factor correction capacitor,
PFC capacitor, Principle of maximum power efficiency for AC circuits, Principle of maximum
power transfer for AC circuits, Impedance matching, Single-phase two-wire power distribution
system, Single-phase three-wire power distribution system, Neutral conductor, Neutral wire,
Split-phase distribution system, Polyphase distribution systems, Three-phase four-wire power
distribution system, Phase voltages, Line-to-neutral voltages, abc phase sequence, Positive phase
sequence, acb phase sequence, Negative phase sequence, Balanced phase voltages, Wye (or Y)
configuration, Balanced three-phase source, Wye-connected source, Wye-connected load, Wye-
wye distribution system, Phase impedances, Load impedances per phase, Balanced three-phase
load, Synchronous three-phase AC generator, Alternator, Rotor, Stator, Synchronous AC motor,
Rotating magnetic field, Line-to-line voltages, Line voltages, Line currents, Superposition
principle for three-phase circuits, Per-phase solution, Total instantaneous load power of the
three-phase system, Average load power of the balanced three-phase system, Reactive load
power of the balanced three-phase system, Complex load power of the balanced three-phase
system, Balanced delta-connected load, Balanced delta-connected source, Delta-delta distribution
system

XI-536
Chapter 11 Section 11.1: AC Power Types and Their Meaning

Section 11.1 AC Power Types and Their Meaning

The present section studies the basics of AC power. We begin with the root-mean-square
(rms) representation of AC voltages and currents. The rms concept enables us to develop a
DC equivalent representation, which compares AC to DC conditions in terms of power
delivered to the load. It is important to understand that the rms concept is a general power
concept; it applies not only to periodic AC circuits but virtually to any circuit, even with
nonperiodic power sources, like noise power sources. Further results presented in this
section are primarily intended for power electronic circuits; they have an equal applica-
bility to radio-frequency communication circuits.

11.1.1 Instantaneous AC Power

We consider an arbitrary load with resistance R, load current i(t), and load voltage υ(t), in
the passive reference configuration. The instantaneous power delivered into the load is
given by

υ 2 ðt Þ
pðt Þ ¼ υðt Þiðt Þ ¼ ð11:1Þ
R
according to Ohm’s law. If we use the load voltage in the form υðt Þ ¼ V m cos ωt ½V,
then

V 2m V2
pðt Þ ¼ υðt Þiðt Þ ¼ cos 2 ω t ¼ m ð1 þ cos 2ωt Þ ð11:2Þ
R 2R
where we applied the trigonometric identity cos 2 ωt ¼ 0:5ð1 þ cos 2ωt Þ. Interestingly,
the load power is not constant; it varies in time, and the behavior is shown in Fig. 11.1 for
a load voltage amplitude V m ¼ 3V, frequency f ¼ 50 Hz, and load resistance R ¼ 5 Ω.

6 6

T
4 4
instantaneous
AC power
2 2
load power, W
voltage, V

0 0

-2 -2

AC voltage
-4 -4

-6 -6
0 5 10 15 20 25 30 35 40
time, ms

Fig. 11.1. Power (solid line) for a load voltage υðt Þ ¼ 3 cos ð2π50t Þ ½V (dotted line).

XI-537
Chapter 11 AC Power and Power Distribution

11.1.2 Time-averaged AC Power

An interesting question arises when we have to determine the bill for the variable AC
power in Fig. 11.1. As far as the utility power is concerned, a consumer would prefer to
pay for the minimum amount of power. The power minima in Fig. 11.1 occur at t ¼ 0:25T ,
0:75T , 1:25T , etc:; here T is the period of the AC voltage signal. Since the power is exactly
zero at its minima, we would pay nothing. On the other hand, a utility would prefer to charge
for the maxima of the power, which occur at t ¼ 0, 0:5T , 1:0T , 1:5T , etc: A fair solution
is clearly somewhere in the middle. It is based on time averaging the load power and then
charging the consumer for the average (or mean) power as indicated in Fig. 11.1 by the
shaded rectangle. Thus, we are interested in the averaged instantaneous power of Eq. (11.2).
The time averaging is always done over a full period T of the AC voltage signal. The
notation for the time-average value is often denoted by an overbar. Thus, the definition
reads:

ðT
1
P ¼ pðt Þ  pðt Þdt ð11:3Þ
T
0

where P is now the average power delivered to the load. We note that the average power
times the period T gives us the energy E (in J or more often in Wh, 1 Wh ¼ 3600 J)
delivered to the load per period, i.e., E ¼ T P:

rms Voltage and rms Current

Using Eq. (11.2) we obtain from Eq. (11.3)
0T 1
ðT ð ðT
1 V 2m 1 V 2m @ 1  dt þ cos 2ω tdt A
P¼ ð1 þ cos 2ω t Þ dt ¼ ð11:4Þ
T 2R T 2R
0 0 0

The first integral yields a nonzero contribution, whereas the second integral is exactly
equal to zero, due to fact that the average of the sine or cosine functions over a period, or
multiple periods, is zero. Thus,

ðT ðT
1 1
1  dt ¼ T , cos 2ωtdt ¼ sin 2ωtj0T ¼ sin ð4π t=T Þj0T ¼ 0 ð11:5Þ
2ω 2ω
0 0

Inserting these values into Eq. (11.4) results in

V 2m V2 Vm
P¼ ¼ rms , V rms ¼ pffiffiffi ð11:6Þ
2R R 2

XI-538
Chapter 11 Section 11.1: AC Power Types and Their Meaning

where Vrms is the rms (root-mean-square) value for the load voltage υðt Þ ¼ V m cos ωt or
simply the rms voltage. According to Eq. (11.6), the rms voltage is the equivalent DC
voltage that provides the same power into the load. Once we know the rms voltage, the
average power is given by the “DC” formula Vrms 2
/R. The rms voltage is always less than
the voltage amplitude by a factor of 0.707 (or 71 %). You should notice that Fig. 11.2 is a
replica of Fig. 11.1; additionally, it shows the rms voltage and the averaged power for the
load voltage υðt Þ ¼ 3 cos ð2π50t Þ ½V. If a nonzero phase is present in this expression, the
result will not change. The signal will be shifted but all averages over the period will
remain the same. The corresponding mathematical proof is suggested as one of the
homework problems.

a) 6 b) 6
rms voltage instantaneous average
4 4 AC power p(t) AC power P

2 2
load power, W
voltage, V

71%
0 0

-2 -2

AC voltage
-4 -4

-6 -6
0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40
time, ms time, ms

Fig. 11.2. (a) Load voltage υðtÞ ¼ 3 cos ð2π50t Þ ½V (dotted line) and its rms DC voltage, which
delivers the same power into the load and (b) instantaneous power and the average power for the
load voltage υðt Þ ¼ 3 cos ð2π50t Þ ½V.

A similar expression is obtained for the alternating current iðt Þ ¼ I m cos ωt across the
load. The rms current is given by

Im I 2m
I rms ¼ pffiffiffi, V rms ¼ RI rms , P¼R ¼ RI 2rms ð11:7Þ
2 2

Example 11.1: Determine average power delivered to a 10-Ω load when the applied AC
voltage is given by υðt Þ ¼ 170 cos ð2π60tÞ ½V (US).
Solution: We find the rms voltage first:
pffiffiffi
V rms ¼ 170= 2  120:21 V ð11:8Þ

XI-539
Chapter 11 AC Power and Power Distribution

V2
Example 11.1 (cont.): The average power is then given by P ¼ Rrms ¼ 1:445 kW. Note
that the rms current is equal to 12.02 A. Therefore, a 15-A or a 20-A AC fuse should be
used. The fuse rating is based on the rms electric current value.

Exercise 11.1: Determine average power delivered to a 10-Ω load when the alternating
load current is given by iðt Þ ¼ 5 cos ð2π60t Þ ½A.
Answer: 125 W.

11.1.3 Application Example: rms Voltages and AC Frequencies

Around the World
The AC voltage reported for the wall plug is an rms voltage. In the USA, the rms voltage
typically ranges from 110 V to 127 V. Variations are caused by a specific type of a three-
phase secondary distribution system studied later in this chapter. In this text we assume an
average nominal value of Vrms ¼ 120 V, perhaps a safe estimate. Using this value we
pffiffiffi
obtain the voltage amplitude (the peak voltage value) of V m ¼ 2  120 V  170 V. This
number is considerably greater than the reported 120 V AC. In other countries, the
nominal rms wall plug voltage ranges from 220 V to 240 V, depending on the country.
Using the nominal rms value of 220 V (People’s Republic of China, Russia, France,
pffiffiffi
Argentina, etc.), we obtain a voltage amplitude of V m ¼ 2  220 V ¼ 311 V. This
number is again greater than the reported 220 V AC. India uses the nominal rms value
of 230 V at 50 Hz, as do the European Union and Great Britain. This yields a voltage
pffiffiffi
amplitude of V m ¼ 2  230 V ¼ 325 V. Depending on the country you live in, the AC
frequency is either 50 Hz or 60 Hz; for instance, it is 60 Hz in the USA.

Historical: From the IEEE Historical FAQ’s and other sources: The person responsible
for adopting 60 Hz was probably Nikola Tesla who figured that for the Westinghouse-
designed central stations for incandescent lamps, the efficient distribution was 59 Hz, and it
was then rounded to 60 Hz. The German company AEG (Allgemeine Elektrizitäts-
Gesellschaft), originally influenced by Thomas Edison, started using 50 Hz as a more
“metric” number. Their standard spread to the rest of Europe and to other countries.

Figure 11.3 shows the rms voltage (and frequency) world map. Some countries have a
dual distribution system that operates at 120 V and 220 V simultaneously. With the help
of an electric transformer, studied in the following text, we can convert the voltages to
higher or lower values. However, we cannot convert frequency with a linear circuit or a
transformer. Some transformers are designed for both 50 Hz and 60 Hz but unfortunately
not all. As time progresses, the frequency difference between the load and the source may

XI-540
Chapter 11 Section 11.1: AC Power Types and Their Meaning

have a severe effect on motorized applications and their transformers (power loss,
overheating, and even eventual burnout).

Fig. 11.3. The rms voltage world map, courtesy of Mr. Conrad H. McGregor, UK, and reproduced
with the author’s permission.

Example 11.2: Determine average power delivered to a 100-Ω load when the applied AC
voltage has an rms value of 220 V (People’s Republic of China).
Solution: The average power is given by

V 2rms
P¼ ¼ 484 W ð11:9Þ
R
Note that the rms current is equal to 2.2 A. Therefore, a 5-A AC fuse (but not the 2-A fuse)
is sufficient in the present case. The fuse is an overcurrent protective device; a soldered
joint within the fuse is melted when the rms current exceeds a threshold.

Exercise 11.2: The load from Example 11.2 is connected to a wall plug in the USA. How
would the average load power change?
Answer: The average load power will be 144 W.

11.1.4 rms Voltages for Arbitrary Periodic AC Signals

Analytical expressions given by Eqs. (11.6) and (11.7) are quite sufficient for finding the
mean power but only for single-frequency AC signals. In certain cases, the signal may
still be periodic with a period of T but may contain multiple frequency components. One

XI-541
Chapter 11 AC Power and Power Distribution

such example is the clock signal for which we estimate the average electric power. In this
case, we return to the definition Eq. (11.3) and rewrite it in the form, which literally
explains the meaning of the root-mean-square value:
0vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi12
u ðT
u
Bu 1 2 C
@tT υ ðt Þ dt A
ðT ðT ð11:10aÞ
1 1 1 2 0
P ¼ pðt Þdt ¼ υ ðt Þdt ¼
T R T R
0 0

We again wish to define the rms voltage as the DC voltage that gives the same power into
the load resistance R. Therefore, it should be
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u ðT
u
u1 V2
V rms ¼ t υ2 ðt Þ dt ) P ¼ rms ð11:10bÞ
T R
0

For single-frequency voltage signals, this result reduces to Eq. (11.6). For more compli-
cated voltages or for voltages measured directly, the calculation of the integral in
Eq. (11.10b) may constitute some difficulties. At the end of this chapter, we provide a
few homework problems tasking you to calculate the integral in Eq. (11.10b) directly.
Once the rms voltage is found, the rms current through the resistive load is expressed by
I rms ¼ V rms =R, irrespective of the particular signal type.

Example 11.3: Determine the average power delivered to a 100-Ω load when the applied
periodic voltage has the form υðtÞ ¼ 10 t=T ½V over a period T ¼ 10 ms. This signal is
known as a sawtooth or a triangular wave.
Solution: We find the rms voltage first:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u ðT u ðT
u u pffiffiffiffiffiffiffiffiffiffiffiffi
u1 2 u1
V rms ¼t υ ðt Þ dt ¼ t 100t 2 =T 2 dt ¼ 100=3  5:77 V ð11:10cÞ
T T
0 0

The average power is thus given by P ¼ V 2rms =R  333 mW.

Frequently encountered voltage signals in microelectronic circuits are noise signals,

which are neither sinusoidal nor periodic. In this case, Eq. (11.10b) applies again, but
only in the limit as T ! 1. Advanced analog electronics deals with certain electronic
circuits where such noise sources become important and even critical.

XI-542
Chapter 11 Section 11.1: AC Power Types and Their Meaning

11.1.5 Average AC Power in Terms of Phasors: Power Angle

For arbitrary dynamic circuit elements, the power analysis is carried out in terms of
phasors. Consider element A in Fig. 11.4 which has real-valued voltage and current
given by

υðt Þ ¼ V m cos ðωt þ φÞ

ð11:11Þ
iðt Þ ¼ I m cos ðωt þ ψ Þ

P?
v(t)
-

+
a b
A
i(t)

Fig. 11.4. Arbitrary circuit element in the passive reference configuration.

When element A is a resistor, the phases in Eq. (11.11) are the same, and finding the
average power is straightforward. However, when element A is an inductor, capacitor, or
a combination of resistor and inductor/capacitor, the situation becomes different. In this
case, the phases of voltage and current in Eqs. (11.11) do not necessarily coincide. By
definition:

ðT ðT
1 1
P ¼ pðt Þ ¼ υðt Þiðt Þdt ¼ V m I m cos ðωt þ φÞ cos ðωt þ ψ Þdt ð11:12Þ
T T
0 0

To manipulate the cosine expression in Eq. (11.12), we can use the trigonometric identity
cos ðωt þ φÞ cos ðωt þ ψ Þ ¼ 0:5 cos ðφ  ψ Þ þ 0:5 cos ð2ωt þ φ þ ψ Þ. The integral of
the second term in Eq. (11.12) will be equal to zero since it is the integral of the plain
cosine function over two periods. The result then has the form:

ðT
1 V mI m V mI m
P ¼ pðt Þ ¼ υðt Þiðt Þdt ¼ cos ðφ  ψ Þ ¼ cos θ ¼ V rms I rms cos θ
T 2 2
0
ð11:13Þ

Equation (11.13) is of great importance for power electronics since it introduces the
so-called power angle θ

θ ¼ φ  ψ,  90o  θ  þ90o ð11:14Þ

and the power factor

PF ¼ cos ðφ  ψ Þ ¼ cos θ ð11:15Þ

XI-543
Chapter 11 AC Power and Power Distribution

Both of these expressions determine the average power delivered to the circuit element.
For any passive load, the power angle must be between 90 and +90 ; i.e., the average
power delivered to the element must be nonnegative! However, for an active load
(an amplifier), it is possible that the power angle is no longer within those limits. Then,
the element actually delivers power to the circuit rather than absorbing it. Equation
(11.13) can now be expressed in terms of phasor voltage V ¼ V m ∠φ and phasor current
I ¼ I m ∠ψ. The result is simple and elegant:
   
Re V  I* Re V*  I
P¼ ¼ ð11:16Þ
2 2

where the star denotes the complex conjugate, ðe jα Þ ¼ ejα , and Re is the real part of a
*

complex number. The proof is based on the phasor substitution, that is,
 
Re V  I* ReðV m ∠φ  I m ∠  ψ Þ V m I m V mI m
¼ ¼ Reð∠φ  ψ Þ ¼ cos ðφ  ψ Þ
2 2 2 2
ð11:17Þ

Example 11.4: The phasor voltage across a purely resistive load with a resistance
R ¼ 10 Ω is given by V ¼ 3 þ j3 ½V. Find the average power delivered to the load.
Solution: According to Eq. (11.16),
   
jVj2 ¼ 3 þ 3 ¼ 18 ¼ 900 mW
2 2
Re V  I* Re V  V*
P¼ ¼ ¼ 20 20
2 2R 2R

Exercise 11.3: The phasor voltage and phasor current for an AC load are given by
V ¼ 3 þ j3 ½V, I ¼ þ 2 þ j3 ½A. Find the average power delivered to the load.
Answer: P ¼ 1:5 W.

11.1.6 Average Power for Resistor, Capacitor, and Inductor

For arbitrary passive circuit elements, the phasor voltage V and the phasor current I are
related by Ohm’s law in the impedance form, V ¼ Z I, where Z is the element impedance
(or the equivalent impedance of a circuit block). We can substitute this result into
Eq. (11.16) and obtain

XI-544
Chapter 11 Section 11.1: AC Power Types and Their Meaning

 
    2
Re V  I *
Re Z  I  I * Re Z  j I j ReðZÞjIj2
P¼ ¼ ¼ ¼ ð11:18Þ
2 2 2 2

since for any complex number I, the following equality holds: I  I* ¼ I m ∠ψ  I m ∠  ψ ¼

I 2m ¼ jIj2 > 0. Equation (11.18) is a remarkable result: if the impedance of an element is
purely imaginary, the average power delivered to the circuit element must be zero.
Indeed, so are the impedances for the inductor and the capacitor. Therefore, the average
power delivered to either the inductor or the capacitor must be zero! The same result
may be explained using Eq. (11.13). We put the phase of the current, ψ, equal to zero for
simplicity. The voltage phase φ will then be +90 for the inductor and 90 for the
capacitor. As the cosine of 90 is zero, Eq. (11.13) will also give zero average power.
An additional explanation is related to the phasor diagrams for voltages and currents
shown in Fig. 11.5. The average power is half of the dot products of two vectors (phasor
voltage and phasor current) in the complex plane. The dot product of two perpendicular
vectors is exactly zero.

Im VR Im Im
VL
resistor inductor capacitor
IC
IR
o o
Re 90 Re 90 Re
0 0 0
IL VC

Fig. 11.5. Average power for a resistor, inductor, and capacitor and the related phasor diagrams.

11.1.7 Average Power, Reactive Power, and Apparent Power

We have just seen that the impedance of a load, Z, is most important for the average AC
power delivered to the load. If the impedance is a pure resistance, there is no problem.
Otherwise, almost no power may be delivered to the load even though large currents can
flow in the circuit and large AC voltages are observed. For example, if the load is a pure
inductance or capacitance, then no average power will be delivered to the load, no matter
which voltages and currents we use. Instead, we will only heat up wires and other circuit
components. We write the impedance for an arbitrary load both in rectangular and in polar
form:

Z ¼ R þ jX ½Ω, Z ¼ jZj∠θ ½Ω, R ¼ jZj cos θ, X ¼ jZj sin θ ð11:19aÞ

XI-545
Chapter 11 AC Power and Power Distribution

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  
1 X
jZj ¼ R2 þ X 2 ½Ω, θ ¼ tan ð11:19bÞ
R

The real part of the impedance, R, is the resistance of the load, and the imaginary part, X,
is the load reactance. For example, the inductor and the capacitor are purely reactive
loads (have only X but not R), whereas the resistor is purely “resistive”. The angle θ is the
power angle of the load; the power angle has already been introduced in Eqs. (11.13),
(11.14), and (11.15). Thus, the power factor PF is simply the cosine of the angle of the
load impedance.

Example 11.5: Determine the resistance and the reactance of an RLC series load shown in
Fig. 11.6. The AC angular frequency is 1000 rad/s.
Solution: The three impedances are combined in series (added to each other),

1
Z ¼ 100 þ jωL  j ¼ 100 þ j1  j100 ¼ 100  j99 ½Ω ð11:20Þ
ωC

The resistance is 100 Ω and the reactance is equal to 99 Ω. The reactance is
negative, i.e., capacitive. In other words, the capacitive reactance dominates.

a
1 mH

100

10 F
b

Fig. 11.6. A RLC series load.

Using resistance R and reactance X of the load, we now introduce three different AC
power types for that load. The first type is the average or active power P studied before in
this section. The active power is expressed by
 
Re V  I* RjIj2 jZjjIj2 jVjjIj
P¼ ¼ ¼ cos θ ¼ cos θ ¼ V rms I rms cos θ ½W 
2 2 2 2
ð11:21aÞ
pffiffiffi pffiffiffi
since jVj ¼ V m ¼ 2V rms , jIj ¼ I m ¼ 2I rms . This is the true or useful power deliv-
ered to the load, with the units of watts.
Note the operations with complex magnitudes:
jabj ¼ jajjbj, ja=bj ¼ jaj=jbj, a* ¼ jaj, which directly follow from the complex
number definition in polar form. The second power type is the reactive power Q that is

XI-546
Chapter 11 Section 11.1: AC Power Types and Their Meaning

 
Im V  I* X jIj2 jZjjIj2 jVjjIj
Q¼ ¼ ¼ sin θ ¼ sin θ ¼ V rms I rms sin θ ½VAR
2 2 2 2
ð11:21bÞ

The physical units of the reactive power are also watts. However, to underscore the fact
that this power is not an active useful power, the units of VAR (volt-amperes reactive) are
used. The reactive power flows back and forth from the source to the load, through an
electric line but does not do real work. The last power type is the complex power S that is
simply

V  I*
S¼ ¼ P þ jQ ½VA ð11:21cÞ
2

The complex power is measured in volt-amperes (VA). The magnitude of the complex
power S ¼ jSj is called the apparent power. We can see that the apparent power is given by

jZjjIj2 jVjjIj
S ¼ jSj ¼ ¼ ¼ V rms I rms ½VA ð11:21dÞ
2 2

The apparent power is the “best possible” load power that can be obtained if one
measures current and voltage and ignores the phase shift between them. Equations
(11.21) and (11.22) raise the obvious question: why do we need so many AC power
types? The answer is that a purely resistive load (the power angle θ equals zero) is merely
a dream and not realistic. Any AC load generally has a significant reactive impedance
part. So does an electric motor, a small antenna in your cellphone, and even a household
electric heater whose heating spiral is a series combination of a resistance and a small, but
often visible, inductance. Therefore, we always deal with active and reactive power; the
sum of their squares is the square of the magnitude of the apparent power. The reactive
power increases the electric current flowing in the circuit and thus increases the
unrecoverable losses in (sometimes very long) power lines, which have a finite resistance.
Therefore, our goal is to decrease the percentage of the reactive power and thus decrease the
net power loss. The three power definitions show us how to accomplish this task. In the next
section, we will need to decrease the power angle θ by modifying the load through adding
other circuit components; in other words, we are attempting to load match the circuit.

11.1.8 Power Triangle

Since cos 2 θ þ sin 2 θ ¼ 1, the three powers (average, reactive, and apparent) are
interconnected by the relation

S 2 ¼ P2avg þ Q2 ½W ð11:22Þ

XI-547
Chapter 11 AC Power and Power Distribution

This relation is also called the power triangle.

Example 11.6: Determine the average (or true) power and the reactive power for the
inductive load shown in Fig 11.7a. Construct the corresponding power triangle. The circuit
parameters are as follows:
V m ¼ 170 V, ω ¼ 377 rad=s, L ¼ 25:7 mH, R ¼ 9:7 Ω.
Solution: We need to find active and reactive powers according to Eq. (11.21). To do so,
we need the load impedance and the load or circuit current. We convert the circuit to the
phasor/impedance form. The equivalent impedance is given by
pffiffiffi
Z ¼ ZR þ ZL ¼ R þ jωL ¼ 9:7 þ j9:7 ¼ 9:7 2∠45 ½Ω ð11:23Þ

The power angle (the phase of the complex impedance) is 45 . The phasor voltage V across
is ffiffiffiequal to V m ¼ 170V. The load phasor current is given by I ¼ V m =Z ¼
the load p
 
170= 9:7 2∠45 ¼ 12:39∠  45 ½A. According to Eq. (11.21),

P ¼ 0:5
170
12:39
cos 45 ¼ 745 ½W
ð11:24Þ
Q ¼ 0:5
170
12:39
sin 45 ¼ 745 ½VAR

The corresponding power triangle is plotted in Fig. 11.8a.

a) load I
+

L VL
- ZL
vS(t) + VS +
- -
+

R VR ZR
-

b) load I
+

C VC
- ZC
vS(t) + VS +
- -
+

R VR ZR
-

Fig. 11.7. Two circuits for power calculation of (a) inductive load and (b) capacitive load.

The power angle (the phase of the load impedance) of the power triangle in Fig. 11.8a
is +45 . When the power angle θ is positive, as in the present case, the corresponding
power factor is said to be lagging. The lagging power factor means that the load current
lags the load voltage. Thus, the power factor in Fig. 11.8a is 0.707 lagging or, which is the
same, is 70.7 % lagging. Similarly, the power factor in Fig. 11.8b is 44.8 % lagging.
However, the power angle (the phase of the load impedance) in Fig. 11.8c is 45 .

XI-548
Chapter 11 Section 11.1: AC Power Types and Their Meaning

When the power angle θ is negative, as is the case here, the corresponding power factor is
said to be leading. The leading power factor means that the load current leads the load
voltage. The power factor in Fig. 11.8c is 70.7 % leading. The lagging occurs for a
predominantly inductive load, whereas the leading occurs for a predominantly capacitive
load; see Fig. 11.7.
a) 70.7% lagging b) 44.8% lagging c) 70.7% leading
Im (VAR) Im (VAR) Im (VAR)
1000 1000 1000

S
Q
S Q
o o o
θ=45 Re (W) θ =63.4 Re (W) θ =-45 P Re (W)
0 P 1000 0 P 1000 0 1000

S Q

Fig. 11.8. Power triangles for inductive loads (a, b), and for a capacitive load (c). The real axis
corresponds to the average (true) power; the imaginary axis is the reactive power.

Exercise 11.4: Determine the average (or true) power and the reactive power for the inductive
load shown in Fig 11.7a. Construct the corresponding power triangle. You are given V m ¼
170 V, ω ¼ 377 rad=s, L ¼ 26:5 mH, R ¼ 5 Ω.
Solution: P ¼ 578:9 ½W, Q ¼ 1156:7 ½VAR, θ ¼ 63:4 . The power triangle is
plotted in Fig. 11.8b. The power factor is 44.8 % lagging.

Exercise 11.5: Determine the average (or true) power and the reactive power for the
capacitive load shown in Fig 11.7b. Construct the corresponding power triangle. The
circuit parameters are V m ¼ 170 V, ω ¼ 377 rad=s, C ¼ 265 μF, R ¼ 10 Ω.
Answer: P ¼ 722:5 ½W, Q ¼ 722:5 ½VAR, θ ¼ 45 . Note that the reactive
power becomes negative for the capacitive load. The corresponding power triangle is
plotted in Fig. 11.8c. The power factor is 70.7 % leading.

It might be interesting to mention that the circuit in Fig. 11.7b is an equivalent circuit
model of a short dipole or monopole antenna, the so-called whip antenna; it represents
predominantly a capacitive load. Indeed, much higher frequencies are employed, but the
concept remains the same. Whip antennas are common on ships, trucks, and other
vehicles. Many radio amateurs use whip antennas as well.

XI-549
Chapter 11 AC Power and Power Distribution

11.1.9 Application Example: Wattmeter

AC power is measured with a wattmeter. The idea of an analog wattmeter operation is
schematically illustrated in Fig. 11.9a. The wattmeter includes at least two coils: the massive
immovable current coil and the lighter suspended, or pivoted, voltage (or potential) coil. The
current coil has a very low impedance; it is connected in series with the load in Fig. 11.9b.
The voltage coil has a very high impedance; it is connected in parallel with the load in
Fig. 11.9b. The voltage coil typically has a high-value resistor connected in series to increase
the impedance. The voltage coil is constructed of a fine wire, whereas the fixed (current) coil
uses a thicker wire to carry the load current. When the current and voltage are in phase, the
magnetic fluxes in both coils attempt to align with each other so that the arrow in Fig. 11.9a
will move to the right. When the current and voltage are out of phase (phase difference of
180 ; the load is in fact an AC source), the arrow in Fig. 11.9a would move to the left. When
the phase difference is 90 , the arrow stays at the center. Thus, the power angle could be
measured, and the average and reactive powers could be reported in an analog or digital way.

a) b)

voltage coil
i(t) wattmeter
+

v(t)
+
- voltage
B coil v(t)
R
-

current
i(t) i(t) coil
current coil

Fig. 11.9. (a) Wattmeter concept and (b) wattmeter coil connection to the load.

XI-550
Chapter 11 Section 11.2: Power Factor Correction: Maximum Power. . .

Section 11.2 Power Factor Correction: Maximum Power

Efficiency and Maximum Power Transfer
11.2.1 Power Factor Correction
We are about to proceed with the correction of the power factor PF ¼ cos θ for an
arbitrary AC load. The correction has several equivalent approaches. One can make:
1. the power angle exactly equal to zero.
2. the power factor equal to one.
3. the reactive power equal to zero.
4. the load impedance purely resistive.
5. the imaginary part of the load impedance (reactance) equal to zero.
The last approach is perhaps most useful from a practical point of view. We attempt to
modify the load by adding extra circuit elements so that the impedance of the modified
load becomes purely resistive (the imaginary part of the impedance is zero). It is worth
noting that the condition of zero reactance is simultaneously the resonance condition for
the various RLC tank circuits studied in the previous chapter. It means that we need to
make the load “resonant” in order to correct the power factor! This is often achieved by
converting the load to an RLC circuit: adding a capacitor to the inductive load or an
inductor to a capacitive load. Most residential loads (washer, dryer, air conditioner,
refrigerator, etc.) and industrial loads are powered by an induction motor. A simplified
equivalent circuit of it is an inductive load shown in Fig. 11.10a. We intend to add a
capacitor in parallel with the load as in Fig. 11.10b. We attempt to choose the capacitance
value in such a fashion as to make the power factor of the modified load equal to one. The
capacitor in Fig. 11.10b is called the power factor correction capacitor or the PFC
capacitor.

a) inductive load b) inductive load with a power factor

correction capacitor

L L
vS(t) + vS(t) + C
- -
R R

Fig. 11.10. Power correction for an inductive load with the shunt capacitor (capacitor in parallel).

We solve both circuits in Fig. 11.10 in the phasor form. The equivalent impedance
(or better its reciprocal, the admittance) for the modified load in Fig. 11.10b is found first,
that is,

XI-551
Chapter 11 AC Power and Power Distribution

1 1 1 R  jωL
¼ þ ¼ jωC þ 2 ð11:25Þ
Z ZC ZL þ ZR R þ ðω LÞ2

If the impedance is a real number, then the admittance is a real number and vice versa.
Therefore, the condition of a real impedance is equivalent to the condition of a real
admittance. From Eq. (11.25), one has

jωL L L
jω C þ ¼ 0 ) R2 þ ðωLÞ2 ¼ )C¼ 2 2 ð11:26Þ
R þ ðωLÞ
2 2 C ω L þ R2

The capacitance value is thus found from the equation

L
C¼ ð11:27Þ
ω2 L2 þ R2

Equation (11.27) is a mathematical statement for the power factor correction capacitor. Its
practical value will become apparent from the example that follows. The equivalent
impedance of the load with the matching capacitor is then found using the real part of
Eq. (11.25), i.e.,

1 R ðωLÞ2
¼ 2 ) Z ¼ R þ ð11:28Þ
Z R þ ðωLÞ2 R

We are interested in the phasor circuit current I with and without the PFC capacitor. Given
the voltage source υS ðt Þ ¼ V m cos ωt, one obtains for the circuits in Fig. 11.10:

without capacitor
zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{
RV m jωLV m
I¼ 2 2
 2 ð11:29Þ
R þ ðωLÞ R þ ðωLÞ2
|fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflffl}
with capacitor

As you can see, the two terms of the expression without capacitor are reduced to the first
term when the power correction capacitor is included. This completes the analysis of the
circuits in Fig. 11.10.

Example 11.7: For the circuit in Fig. 11.10, find the average (or true) load power and the
reactive load power with and without the power correction factor capacitor. You are given
V m ¼ 170 V, ω ¼ 377 rad=s, L ¼ 25:7 mH, R ¼ 9:7 Ω.

XI-552
Chapter 11 Section 11.2: Power Factor Correction: Maximum Power. . .

Example 11.7 (cont.):

Solution: To find the power expressions, we need the phasor voltage across the load,
which is simply Vm. We also need the phasor current, which is given by Eq. (11.29) in
either case. Plugging in the numbers we obtain,

I ¼ 8:77 A or I ¼ 8:77  j8:77 A ð11:30Þ

with and without the PFC capacitor, respectively. Now, we use the power
definitions:
 
Re V  I*
P¼ ) P ¼ 745 W or P ¼ 745 W ð11:31Þ
2
 
Im V  I*
Q¼ ) Q ¼ 0 VAR or Q ¼ 745 VAR ð11:32Þ
2
with and without the PFC capacitor, respectively.

In summary, we have found the following information from this example:

No power factor correction:
- Average (active or true) power: P ¼ 745 W
- Reactive power: Q ¼ 745 VAR
- Amplitude of the circuit current: 12.41 A
Power factor correction:
- Average (active or true) power: P ¼ 745 W
- Reactive power: 0
- Amplitude of the circuit current: 8.77 A
By correcting the load power factor with the capacitor in parallel, we did not change
the average power (power delivered to the load), but we eliminated the reactive power and
pffiffiffi
decreased the amplitude of the circuit current by 70.7 % (1= 2). This means that the ohmic
losses in the electric line connecting load and generator will decrease by 50 %, since these
losses are proportional to the square of the current amplitude. In other examples, the loss
reduction factor may be even more significant. Is it worth doing a power factor correction?
Well, if the electric power line is long enough or the initial power factor is not high enough,
it is definitely a very useful and professional task. To support this conclusion, we mention a
citation from IEEE Transactions on Power Electronics: “Everyone knows that correcting
power factor is the easiest and fastest way to save energy dollars.”

XI-553
Chapter 11 AC Power and Power Distribution

Exercise 11.6: Find the value of the power factor correction capacitor in Example 11.7.
Answer: C ¼ 136:73 μF:

Exercise 11.7: Find the value of the load impedance in Example 11.7 with and without the
power factor correction capacitor. Express your result in polar form.
Answer: Z ¼ 13:7∠45 Ω and Z ¼ 19:4∠0 Ω, respectively.

11.2.2 Application Example: Automatic Power Factor Correction System

The power factor correction capacitors are frequently seen on residential power poles in
the form of pole-mounted capacitor banks. Figure 11.11 shows an automatically switched
power factor correction system that measures all three power types (active, reactive, and
apparent power) using the same wattmeter principle described in the previous section.
Based on the recorded measurements, the required capacitor value is selected, which
assures the targeted power factor.

Fig. 11.11. Automatically switched power factor correction systems for low-voltage applications. Six
capacitor cells are seen on the bottom. Technical Data TD02607001E Cutler-Hammer.

11.2.3 Principle of Maximum Power Efficiency for AC Circuits

Why is the power correction capacitor placed in parallel, not in series with the load? To
answer this question, we should establish and understand the principle of maximum
power efficiency for AC circuits. Consider a generic source-load AC circuit depicted in
Fig. 11.12a in phasor form. It is based on a Thévenin equivalent circuit for an AC source
with the source impedance ZT connected to the load impedance ZL. The source imped-
ance will also include the loss resistance of power lines. Figure 11.12b shows the

XI-554
Chapter 11 Section 11.2: Power Factor Correction: Maximum Power. . .

corresponding DC counterpart, which is useful for the subsequent analysis. The source
resistance RT will include the loss resistance of power lines as well.

a) b)

ZT I RT I
VT + ZL VT + RL
- -

Fig. 11.12. Thévenin equivalent circuit for an AC source with the load impedance ZL and its DC
counterpart with the load resistance RL.

In the DC case, the source-load configuration of Fig. 11.12b is maximally efficient

when the load resistance is much greater than the source (loss) resistance. This fact has
been established in Chapter 4. When RL  RT , the useful power delivered to the load
resistance RL is much larger than the power loss in RT. We emphasize that the load power
itself could be relatively small in this case as compared to the maximum available load
power at RL ¼ RT . The same situation occurs for the AC system shown in Fig. 11.12a. Do
we wish to deliver the maximum available power of a remote megawatt AC source to the
household? No, this is not our goal. We would rather deliver a reasonable amount of
power but in a most efficient way. It means that not only do we need to make ZL real but
also as high as possible. This operation would further reduce the circuit current and the
associated loss. Exactly this goal is accomplished by the shunt PFC capacitor in
Fig. 11.10. If we consider the series-connected PFC capacitor as an alternative, we will
obtain
ðω LÞ2
ZL ¼ R f or series connection versus ZL ¼ R þ f or shunt connection:
R
ð11:33Þ

Both impedances in Eq. (11.33) are real; there is no reactive power in either case.
However, the second impedance is considerably greater than the first one for poor power
factors, i.e., for ωL  R! Hence, considerably smaller circuit currents and considerably
better efficiencies are achieved. Moreover, the parallel connection is easier to accomplish
in practice—we remember how easy it is to connect a voltmeter as opposed to an ammeter

Exercise 11.8: Find the amplitude of the circuit current in Example 11.7 if the power
correction capacitor were in series.
Answer: I m ¼ 17:53 A.

XI-555
Chapter 11 AC Power and Power Distribution

11.2.4 Principle of Maximum Power Transfer for AC Circuits

The principle of maximum power transfer is perhaps less important for residential power
distribution systems where efficiency counts. However, it is critical for radio-frequency
and communication circuits, which are conceptually the same AC circuits but operating
at much higher frequencies. With reference to Fig. 11.12a, the following question should
now be asked: at which value of the load impedance ZL ¼ RL þ jX L is the average
(true) power delivered to the load maximized? The phasor current in Fig. 11.12a is
given by

VT
I¼ ½A ð11:34Þ
ZL þ ZT

so that the average power delivered to the load becomes

R L j Ij 2 RL jVT j2 0:5RL jVT j2

P¼ ¼ ¼ ½W ð11:35Þ
2 2jZL þ ZT j2 ðRL þ RT Þ2 þ ðX L þ X T Þ2

Let us take a closer look at Eq. (11.35); in order to reach the maximum true power, the
load reactance XL should be equal to the generator reactance XT taken with the opposite
sign so that X L þ X T ¼ 0. This yields for the average load power

0:5RL jVT j2
P¼ ½W ð11:36Þ
ðRL þ RT Þ2

Consequently, the problem reduces to the maximum power transfer of a DC circuit as

studied in Chapter 2. The corresponding condition for the maximum load power is

RL ¼ RT ð11:37Þ

This condition, augmented by the equality for the reactances

X L ¼ X T ð11:38Þ

leads to a simple, yet very useful result for the maximum power transfer to the load:

1 jVT j2
ZL ¼ Z*T ) Pmax ¼ ½W  ð11:39Þ
8 RT

We note that the load impedance should be the complex conjugate of the generator
impedance. Along with the maximum power transfer, Eq. (11.39) assures that there is no
reflection of radio-frequency waves propagating along the circuit transmission lines from

XI-556
Chapter 11 Section 11.2: Power Factor Correction: Maximum Power. . .

the source to the load, which may even be a more important factor. A process of
modifying the load impedance in order to satisfy Eq. (11.39) is called impedance
matching.

Example 11.8: A generator impedance is 50 Ω. The load impedance is 10 + j100 Ω. What

percentage of maximum available power (at a load impedance of 50 Ω) is transferred to the
load?
Solution: According to Eq. (11.36), when the load impedance is exactly 50 Ω,

0:5
50jVT j2 jV T j2
P¼ ¼ ¼ 0:0025jVT j2 ½ W ð11:40aÞ
ð50 þ 50Þ2 8
50

When the load impedance is 10 + j100Ω, Eq. (11.35) gives

0:5
10jVT j2 jV T j2
P¼ ¼ ¼ 0:00037jVT j2 ½ W ð11:40bÞ
ð50 þ 10Þ2 þ 1002 2720

The ratio of the two power expressions is 0.147, or 14.7%. In other words, 85.3% of
available power is lost!

Exercise 11.9: A generator’s impedance is 50  j100 Ω. What should the load impedance
be for maximum power transfer?
Answer: 50 + j100 Ω.

Exercise 11.10: Solve Example 11.8 when the load impedance is 10  j100 Ω.
Answer: The same result of 14.7 % is obtained.

XI-557
Chapter 11 AC Power and Power Distribution

Section 11.3 AC Power Distribution: Balanced Three-Phase

Power Distribution System
11.3.1 AC Power Distribution Systems
Representative AC power distribution systems are shown in Fig. 11.13. A single-phase
two-wire power distribution system is depicted in Fig. 11.13a. It consists of a generator
pffiffiffi
with a voltage amplitude of Vm, an rms value of V rms ¼ V m = 2, and a phase φ connected
through two conductors to a load with impedance Z. The previous analysis of AC power
was solely restricted to this configuration. An extension is the single-phase three-wire
power distribution system shown in Fig. 11.13b. Such a system contains two identical AC
sources of the same amplitude and phase connected to two (Z1, Z2) loads or to one (Z)
load through two outer conductors and the neutral conductor (or neutral wire). This
system is the common household distribution system. It allows us to connect both 120-V
and 240-V appliances as shown in Fig. 11.13b; we sometimes called it the split-phase
distribution system. The neutral wire is usually physically grounded. In contrast to those
two cases, the power distribution systems shown in Fig. 11.13c, d are the polyphase
distribution systems in the sense that they use AC sources with different phases. For
example, Fig. 11.13c illustrates a two-phase three-wire distribution system with two
voltage sources; the second one lags the former by 90 . Finally, Fig. 11.13d shows the
most important and practical three-phase four-wire power distribution system with three
sources and three load impedances Z1, Z2, and Z3. Generally, the three-phase system also
uses a (grounded) neutral wire. We will show that this wire may be omitted for balanced
power distribution circuits, with the earth itself acting as the neutral conductor. This is
important for long-distance, high-power transmissions. Power systems designed in this
way are grounded at critical points to ensure safety.
Today, a vast majority of electric power is generated and distributed via the three-phase
power systems. Why is this so? You will soon learn that in contrast to the single-phase
systems, the instantaneous power in balanced three-phase systems is constant or inde-
pendent of time rather than pulsating. This circumstance results in more uniform power
transmission and less vibration of electric machines. Furthermore, three-phase AC motors
have a nonzero starting torque in contrast to the single-phase motors. Last but not least, it
will be shown that the three-phase system surprisingly requires a lesser amount of wire
compared to the single-phase system.

XI-558
Chapter 11 Section 11.3: AC Power Distribution: Balanced Three-Phase. . .

a) single-phase two-wire system b) single-phase three-wire system

a A

+ V Z + V Z1
- m - m
n neutral N
Z

+ V Z2
- m
b B

c) d) three-phase four-wire system

a A
-

+
two-phase three-wire system
a A Vm 0 Z1

b B
+ Z1 -
- Vm 0 +
n N Vm 120 Z2

cb C
+ Vm 90 Z2 -
+

-
b B Vm 120 Z3

n neutral N

Fig. 11.13. Various AC power distribution systems. N or n indicates the neutral line.

11.3.2 Phase Voltages: Phase Sequence

The voltage sources in the three-phase system in Fig. 11.13d are set between lines a, b,
c and the neutral line n. Those voltages are called phase voltages or line-to-neutral
voltages. The phase voltages are 120 out of phase. One possible scenario for the real-
valued phase voltages is

υan ðt Þ ¼ V m cos ðωt Þ, υbn ðt Þ ¼ V m cos ðωt  120 Þ,

ð11:41aÞ
υcn ðt Þ ¼ V m cos ðωt þ 120 Þ
Van ¼ V m , Vbn ¼ V m ∠  120 , Vcn ¼ V m ∠ þ 120 ð11:41bÞ

Phase voltage υan leads phase voltage υbn, which in turn leads υcn. This set of voltages is
shown in Fig. 11.14. It has a positive or abc phase sequence since the voltages reach their
peak values in the order abc as seen in Fig. 11.14. Simultaneously, the phasor voltages are
obtained from each other by clockwise rotation in the phasor diagram. This is shown in
Fig. 11.15a.

XI-559
Chapter 11 AC Power and Power Distribution

Vm van(t) vbn(t) vcn(t)

0 240
120 t

Fig. 11.14. Balanced phase voltages in positive phase sequence.

An alternative is the negative or acb phase sequence, which corresponds to

0, þ120 , 120 phases in Eqs. (11.41a, b). In this case, the phasor voltages are
obtained from each other by counterclockwise rotation in the phasor diagram of
Fig. 11.15a. Thus, the balanced phase voltages are those which have equal amplitudes
and are out of phase with each other by 120 (either in positive 0, 120 , þ120 or in
negative 0, þ120 , 120 phase sequence). An example of balanced voltages is given
by Eqs. (11.41a, b). The concept of balanced phase voltages is critical for the subsequent
analysis.

Example 11.9: Determine whether the phase voltages

υan ðt Þ ¼ 3cos ðωt  90 Þ, υbn ðt Þ ¼ 3cos ðωt þ 150 Þ, υcn ðt Þ ¼ 3cos ðωt þ 30 Þ ð11:42Þ

of a three-phase system are balanced or not. If yes, determine the corresponding

phase sequence.
Solution: The amplitudes of the phase voltages are equal, which is the first necessary
condition of the balanced sources. To analyze the phases, we plot the voltages in the phasor
diagram and obtain Fig. 11.15b. Despite the common phase shift of 90 as compared to
Eqs. (11.41a, b), the phase voltages are still out of phase with each other by 120 and form
the same positive phase sequence; see Fig. 11.15b.

Exercise 11.11: The phase voltage Vbn is given by V m ∠ þ 45. Determine the remaining
phase voltages Van, Vcn of the balanced three-phase system for the positive phase
sequence. Express your result in phasor form.
Answer: Van ¼ V m ∠165 , Vcn ¼ V m ∠75 .

XI-560
Chapter 11 Section 11.3: AC Power Distribution: Balanced Three-Phase. . .

a) b)
Vcn Im Im
Vbn Vcn
2 2

1 1
Van
Re
0 1 2 3 0 1 2 3
Re

Vbn Van

Fig. 11.15. (a) Phasor diagram for the phase sequence 0, þ120 , 120 and (b) phasor diagram
for individual phase voltages from Eq. (11.42). Both three-phase sources are equivalent.

11.3.3 Wye (Y) Source and Load Configurations for Three-Phase Circuits
The voltage sources in the three-phase system in Fig. 11.13d are now rearranged as
shown in Fig. 11.16a. This configuration is indeed equivalent to the original one; it is
known as the wye (or Y) configuration. Accordingly, the balanced three-phase source in
Fig. 11.16a is the wye-connected source, and the load in Fig. 11.16b is the wye-connected
load.

a) a A b)

Van Vbn b B Z Z

transmission line
+

+
-

n N
-
Vcn +
Z
c C

Fig. 11.16. Wye configuration for the three-phase source connected to a three-phase load.

The AC voltage source in Fig. 11.16a has four terminals. The corresponding load
should also have four terminals. The concept is shown in Fig. 11.16b. This load assembly
is also identical with the topology of Fig. 11.13d. The load includes three impedance
elements (phase impedances or load impedances per phase) Z ¼ Z∠θ with impedance
magnitude Z and phasor angle (power angle) θ each. The load so assembled is the
balanced three-phase load. In the balanced load, the phase impedances are equal in
magnitude and phase. The source and the load are typically connected by (long) wire
transmission lines. When necessary, the wire resistance may be added to each individual
load impedance.

XI-561
Chapter 11 AC Power and Power Distribution

11.3.4 Application: Examples of Three-Phase Source and the Load

Synchronous Three-Phase AC Generator
Despite the apparent complexity, the three-phase source and the three-phase loads are
relatively simple to realize in practice. Figure 11.17 shows a synchronous three-phase AC
generator (or alternator), which is equivalent to the three-phase source in Fig. 11.16a.
a) a b) a a’

a b N
b’ c’
N
S

c S b
a’
n

Fig. 11.17. Structure of the three-phase AC generator. (a) Cross-section view. (b) Simplified
outline of one of the three windings.

Consider first the generator cross section shown in Fig. 11.17a. The generator’s rotor is
a permanent magnet (small scale) or an electromagnet (industrial scale) rotated by a
mechanical torque (a turbine). Three individual coil windings aa0 , bb0 , and cc0 in the
stator are spaced exactly 120 apart around the stator. When the rotor moves, an induced
emf (induced voltage) will be created in every individual winding according to Faraday’s
law of induction—see Fig. 11.17b. From the geometry considerations, the induced
voltages are equal in magnitude and out of phase by 120 . When the coil terminals a0 ,
b0 , and c0 are all connected to the neutral wire, see Fig. 11.17a, we obtain exactly the
three-phase source with the neutral wire in Fig. 11.16a.

Automotive Alternator
The automotive alternator operates based on the same principle. However, the resulting
three-phase voltage is further converted to the DC voltage (rectified, see Chapter 16).

Synchronous Three-Phase AC Motor

The counterpart of the three-phase generator is the three-phase AC motor (the synchro-
nous AC motor). The stator, which is subject to the three-phase voltage source, creates a
rotating magnetic field; the rotor magnet is aligned with this field at every time moment
and rotates accordingly. The stator’s circuit model is similar to the three-phase load model
in Fig. 11.16b where each load impedance Z includes resistance and inductance of the
individual (identical) coil windings. The induced emf should be included into our
consideration as well. Why is the phase sequence important for power distribution?
There is a simple answer to this question. Assume that the machine in Fig. 11.17 operates

XI-562
Chapter 11 Section 11.3: AC Power Distribution: Balanced Three-Phase. . .

in the motor mode. Changing the phase sequence from abc to acb will reverse the
direction of the magnetic field rotation and thus reverse the direction of the motor
rotation! This method is used in practice since it requires interchanging only two
connections.

Residential Household
Another example of the load impedance per phase is related to a typical residential
household in the USA. A single phase of a three-phase residential distribution system
is normally used to power them up; see Fig. 11.18. This single phase still has a high rms
voltage (4800 V or 7200 V). A step-down center-tap transformer is used to decrease this
voltage level to the desired level of 120–240 V and provide the neutral contact necessary
for the three-wire single-phase residential system shown in Fig. 11.13b. This transformer
case is also seen in Fig. 11.18. In the USA, a pole-mounted transformer in a suburban
setting may supply one to three houses.

Fig. 11.18. Three-phase to three-wire residential system connected via a step-down transformer.
From the pole transformer, the residential power system serving two houses is run down the pole
underground. Cape Cod, MA.

11.3.5 Solution for the Balanced Three-Phase Wye-Wye Circuit

Phase Voltages and Line Voltages
A three-phase balanced circuit (wye-wye configuration) which includes the source and the
load is shown in Fig. 11.19. We place the nodes n and N at the originally anticipated
center positions. The positive phase sequence of 0, 120 , þ120 is assumed. The
sum of phase voltages is to be found first. In the phasor form,

Van þ Vbn þ Vcn ¼ V m ð1 þ 1∠120 þ 1∠þ120 Þ ¼ V m ð1 þ 2 cos 120 Þ

ð11:43Þ
¼ V m ð1  1Þ ¼ 0

XI-563
Chapter 11 AC Power and Power Distribution

a Ia

Van Vbn b Z Z
Ib

+
-

-
n N
- In
Vcn +
Z
c
Ic

Fig. 11.19. Three-phase, four-wire balanced wye-wye distribution system. Ground connection is
implied for the neutral wire.

Thus, the sum of the balanced phase voltages is exactly zero, either in the phasor form or
in the time domain. Now, along with the phase (line-to-neutral) voltages, we define line-
to-line voltages (or just line voltages) Vab, Vbc, Vca between nodes a–b, b–c, and c–a, as
indicated in Fig. 11.19. These voltages are expressed through the phase voltages using
pffiffiffi
KVL. Using the trigonometric identity 1  1∠120 ¼ 3∠30 three times, it can be
shown that
pffiffiffi 
Vab ¼ Van  Vbn ¼ p3
ffiffiffiVan ∠30 ,
Vbc ¼ Vbn  Vcn ¼ pffiffi3ffi Vbn ∠30 , ð11:44Þ
Vca ¼ Vcn  Van ¼ 3Vcn ∠30
It is seen that the line voltages are higher in amplitude than the phasor voltages by a factor
pffiffiffi
of 3  1:73. Furthermore, they lead their corresponding phase voltages by 30 .
According to Eqs. (11.43) and (11.44), the sum of the line voltages is also equal to
zero. Both the phase voltages and the line voltages may be used in the three-phase system.

Exercise 11.12: Is Eq. (11.44) also valid for the negative phase sequence?
Answer: Not exactly. A substitution 30 ! 30 has to be made.

Example 11.10: The electric service for commercial buildings (university campus build-
ings) in the USA is a three-phase, four-wire wye system schematically shown in Fig. 11.19.
Determine rms phase voltages if the line voltages are all equal to 208 V rms.
pffiffiffi
Solution: According to Eq. (11.44), we should divide the line voltage of 208 V rms by 3.
This gives us exactly 120 V rms voltage per phase. Thus, the present wye system is also
powering common 120 V wall plugs with any of the line-to-neutral voltages. Note that the
source in Fig. 11.19 typically models an output of a three-phase transformer.

XI-564
Chapter 11 Section 11.3: AC Power Distribution: Balanced Three-Phase. . .

Line Currents: Per-Phase Solution

The currents Ia,b,c in Fig. 11.19 are called line currents. To find the line currents, the circuit
may be solved separately for every phase using the superposition principle. The superpo-
sition principle implies shorting out two of the three voltage sources at a time. This method
applies to both balanced and unbalanced circuits. Shorting out voltage sources Vbn and Vcn
leads to a single-phase equivalent circuit, shown in Fig. 11.20, since the two remaining
source impedances will be shorted out by the neutral wire. As long as the system is balanced,
the same equivalent circuit will be derived for every other phase.

a
Ia
Van + ZL
-
n

Fig. 11.20. A single-phase equivalent circuit by shorting out Vbn and Vcn.

Applying this method to every phase, we obtain

Ia ¼ Van =Z ¼ I m ∠  θ, Ib ¼ Vbn =Z ¼ I m ∠120  θ,

ð11:45Þ
Ic ¼ Vcn =Z ¼ I m ∠120 θ

where I m ¼ V m =Z. The sum of the line currents is given by

Ia þ Ib þ Ic ¼ ðVan þ Vbn þ Vcn Þ=Z ¼ 0 ð11:46Þ

according to Eq. (11.43). Thus, the sum of the balanced line currents is also exactly zero,
either in phasor form or in the time domain. Equations (11.43), (11.44), (11.45), and
(11.46) hold for any phase sequence, with or without the common phase shift.

11.3.6 Removing the Neutral Wire in Long-Distance Power Transmission

Equation (11.46) for the line currents has an important implication. By taking into
account Eq. (11.46), KCL for node n in Fig. 11.19 yields

In ¼  ðIa þ Ib þ Ic Þ ¼ 0 ð11:47Þ

Equation (11.47) states that the neutral conductor in the balanced circuit carries no
current. Such a wire could in principle be removed from the balanced circuit without
affecting the rest of it. Removing the neutral conductor is economically beneficial in long-
distance high-voltage power transmission, which utilizes the balanced circuits. In high-
voltage power lines, the conductors in multiples of three are used; see Fig. 11.21.

XI-565
Chapter 11 AC Power and Power Distribution

a) b) x

Fig. 11.21. (a) Three-phase single-circuit high-voltage overhead power transmission line and (b)
three-phase double-circuit, high-voltage overhead line. Both lines include (thinner) shielding
wires on top of it to protect against lightning strikes (F. Kiessling, et al., “Overhead Power
Lines: Planning, Design, Construction,” Springer 2003).

In fact, the neutral wire is not removed entirely since the earth ground itself plays the
role of the neutral conductor. We will not draw the neutral wire in the three-phase
balanced wye-wye circuit; in Fig. 11.22 only three wires are drawn. However, the
meaning of the phase voltage or the phase-to-neutral voltage still remains unchanged—
this voltage is simply defined with regard to the reference node n in Fig. 11.22.
If a three-phase circuit is unbalanced, like having different individual loads in
Fig. 11.19, then a significant current may flow in the neutral wire. The neutral wire is
thus meant to carry unbalanced currents in the electrical system. It should be kept in place
for potentially unbalanced systems.

Example 11.11: Determine line currents in the balanced three-phase wye-wye circuit
shown in Fig. 11.22 given the acb sequence of phase voltages Van ¼ 325∠0 , Vbn
¼ 325∠120 , Vcn ¼ 325∠120 ½V and load impedance per phase Z ¼ 8:333 þ
j14:434 Ω.
Solution: The three-phase circuit in Fig. 11.22 is balanced; hence, the single-phase circuit
in Fig. 11.20 applies to every phase (to visualize the per-phase method, we can still imagine
the neural wire present). We convert the load impedance to polar form first, i.e.,
Z ¼ 16:667∠60 Ω. Then, we solve the circuit in Fig. 11.20 for every phase and obtain
Ia ¼ 19:5∠60 , Ib ¼ 19:5∠60 , Ic ¼ 19:5∠180 ½A. The solution is shown in the
phasor diagram in Fig. 11.23. Note that the phasor voltages/currents are obtained from each
other by counter clockwise rotation in the phasor diagram, which corresponds to the
negative or acb phase sequence. Also note that the rms values for the phase voltages in
this example are 230 V, which corresponds to the European residential power distribution
system.

XI-566
Chapter 11 Section 11.3: AC Power Distribution: Balanced Three-Phase. . .

Ia

Van Vbn b Z Z
a
Ib

+
-

-
n N
-
Vcn +
Z
c Ic

Fig. 11.22. Three-phase balanced wye-wye system with the neutral conductor removed. The
neutral conductor may still be implied for the solution using the per-phase method.

The wye-wye circuit in Fig. 11.22, along with the similar circuits in Figs. 11.13 and
11.19, may contain extra impedances. Those are line impedance, which characterizes
transmission line loss and inductance, and source impedances, which are present for
nonideal voltage sources. Fortunately, all those (equal) impedances are combined in
series along the line into one impedance Z which is called the total load impedance
per phase. In this sense, Fig. 11.22 represents this general case as well.

Vbn Im
325 V

Ib
Van
Ic

Re
19.5 A

Ia
Vcn

Fig. 11.23. Phasor diagram for the three-phase circuit of Example 11.11. Note the separate scales
for the phasors of voltages and phasors of currents.

XI-567
Chapter 11 AC Power and Power Distribution

Section 11.4 Power in Balanced Three-Phase Systems:

Delta-connected Three-Phase Circuits
11.4.1 Instantaneous Power
The analysis of the instantaneous power requires a source-load circuit in terms of real-
value expressions of voltages and currents. This is shown in Fig. 11.24 for the wye-wye
configuration. We assume a balanced source and a balanced load. This means that
individual loads a, b, and c in Fig. 11.24 are identical. Each of them can be a mixed
RLC load, with an arbitrary impedance Z ¼ Z∠θ.

Fig. 11.24. Three-phase balanced wye-wye circuit in the time domain. Three individual loads are
identical. Each of them is characterized by the impedance Z ¼ Z∠θ in the frequency domain.

We consider the positive phase sequence. According to Eqs. (11.41) and (11.45) of the
previous section, the phase voltages and line currents in Fig. 11.24 are given by

υan ðt Þ ¼ V m cos ðωtÞ, υbn ðt Þ ¼ V m cos ðωt  120 Þ, υcn ðt Þ ¼ V m cos ðωt þ 120 Þ
ð11:48aÞ
ia ðt Þ ¼ I m cos ðωt  θÞ, ib ðt Þ ¼ I m cos ðωt  120  θÞ,
ð11:48bÞ
ic ðt Þ ¼ I m cos ðω t þ 120  θÞ

The total instantaneous load power of the three-phase system is the sum of the three
power contributions for each phase voltage, that is,

pðt Þ ¼ υan ðt Þia ðt Þ þ υbn ðt Þib ðt Þ þ υcn ðt Þic ðt Þ ð11:49Þ

Every summand on the right-hand side of Eq. (11.49) is the product of two cosines. To
transform this product back to cosines, we use the trigonometric identity cos α cos β
¼ 12 ð cos ðα þ βÞ þ cos ðα  βÞÞ and obtain
 
3 1  
pðt Þ ¼ V m I m cos θ þ V m I m cos ð2ωt  θÞ þ cos ð2ωt  θ þ 120 Þ þ cos ð2ωt  θ  120 Þ
2 2 |fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}
ð11:50Þ

XI-568
Chapter 11 Section 11.4: Power in Balanced Three-Phase Systems. . .

We then use the above trigonometric identity again to convert the underlined term to
2 cos ð2ωt  θÞ cos ð120 Þ ¼  cos ð2ωt  θÞ. Consequently, the entire term in the
square brackets in Eq. (11.50) is equal to zero, and the final result for the instantaneous
power is

3
pðt Þ ¼ V m I m cos θ ¼ 3V rms I rms cos θ ¼ const ! ð11:51Þ
2

where the rms values of phase voltages and line currents are indeed related to the
pffiffiffi pffiffiffi
amplitudes by V m ¼ 2V rms , I m ¼ 2I rms.

Example 11.12: A balanced wye-wye three-phase system in Fig. 11.24 operates at 60 Hz.
The line-to-neutral voltages have the amplitudes of 170 V, i.e. V m ¼ 170 V. Every phase
impedance is a 77.2-mH inductance in series with a 29.1-Ω resistance. Find the instanta-
neous load power.
Solution: The first step is to find the impedance for every phase of the load. We have
pffiffiffi
Z ¼ R þ jωL ¼ 29:1 þ j29:1 Ω ¼ 29:1 2∠45 Ω ð11:52Þ

Next, we find the line currents. Since the circuit is balanced, the per-phase solution applies,
with the equivalent circuit shown in Fig. 11.20 of the previous section. It yields

Vm Vm 170
Im ¼ ¼ ¼ ¼ 4:1309 A ð11:53Þ
j Zj Z 41:154

The instantaneous load power follows Eq. (11.51) with the power angle, θ ¼ 45 .
Therefore, we obtain

3
pðt Þ ¼ V m I m cos θ ¼ 745 W ¼ const ð11:54Þ
2

Equation (11.51) is critical for three-phase systems. It tells us that the total instanta-
neous power delivered to the load remains constant at any instance in time. This is in
contrast to the instantaneous power of every individual single phase, which is still
pulsating in time. Equation (11.51) implies that a three-phase load (e.g., an induction
motor) as well as the three-phase generator introduced in the previous section should
generate or require a constant torque. Thus, they undergo less vibration since the net
power transfer is uniform.

XI-569
Chapter 11 AC Power and Power Distribution

11.4.2 Average Power, Reactive Power, and Apparent Power

The AC power types defined for the single-phase power distribution in Section 11.1 of
this chapter also apply for the three-phase circuits. The average or active load power P,
the reactive load power Q, and the complex load power S of the balanced three-phase
system are given by

P ¼ 3V rms I rms cos θ ½W, Q ¼ 3V rms I rms sin θ ½VAR, S ¼ 3V rms I rms ∠θ ½VA
ð11:55aÞ

While the instantaneous powers per phase are pulsating, their average values labeled with
indexes a, b, c are exactly one third of the load powers. One has

Pa, b, c ¼ V rms I rms cos θ ½W, Qa, b, c ¼ V rms I rms sin θ ½VAR,
ð11:55bÞ
Sa, b, c ¼ V rms I rms ∠θ ½VA

per phase. Equation (11.55) uses the rms values of phase voltages and line currents.

Example 11.13: For the previous example, determine the load average power, reactive
power, and the apparent power. Do these powers coincide with the corresponding source
measures?
Solution: The average power is simply the load instantaneous power, P ¼ 745 W. The
reactive power is Q ¼ 32 V m I m sin θ ¼ 745 VAR. The apparent power is
S ¼ jSj ¼ 32 V m I m ¼ 1053 VA. And the apparent power can be also found from the
power triangle. All load powers coincide with the corresponding source powers since the
transmission lines in Fig. 11.24 are assumed to be ideal conductors.

Exercise 11.13: A three-phase induction motor is modeled by a balanced wye load in

Fig. 11.24. The motor (active) power is 6 kW; the line current is 20 A rms, and the line
voltage is 208 V rms. Determine the power factor of the motor, which is the ratio of the
active load power P to the magnitude of the total apparent power, |S|.
Answer: PF  5=6 ¼ 0:833.

11.4.3 Application Example: Material Consumption in Three-Phase

Systems
A comparison is made between the conductor material consumption in a single-phase
two-wire transmission system (shown in Fig. 11.25a) and the three-phase, three-wire
transmission system shown in Fig. 11.25b. Both systems have the identical distance from
the source to the load, the same average power P distributed to a purely resistive load, and

XI-570
Chapter 11 Section 11.4: Power in Balanced Three-Phase Systems. . .

the same rms line voltages close to the load. They all use the same conductor material.
The distributed resistance per wire is modeled by a lumped resistor R for the single-phase
line and by a lumped resistor R0 for the three-phase line. Given the same load power and
line voltage, the rms line currents are expressed as

P P
I single phase ¼ , I three phase ¼ pffiffiffi ð11:56Þ
V 3V

Equating power loss in the wire conductors, we obtain

 2  2
0 P 0 P 0
2RI 2single phase ¼ 3R I 2three phase ) 2R ¼ 3R pffiffiffi ) R ¼ 2R ð11:57Þ
V 3V

a)
R

source V load
R
b)
R’

V V
source R’ load

R’ V

Fig. 11.25. Comparison of single-phase and three-phase transmission systems.

Since resistance R is twice as large as resistance R0 , the cross section of the

corresponding cylindrical conductor is smaller by a factor of two in the three-phase
pffiffiffi
configuration. Hence, its radius is 1= 2 times less than the radius, r, of the single-phase
line. Figure 11.26 depicts the radii of the equivalent conductors.

a) single-wire transmission

r r

b)
three-phase transmission

r/ 2

Fig. 11.26. Radii of equivalent conductors for the two systems in Fig. 11.25.

For the single-phase line, the total conductor cross section is 2πr2; for the three-phase
line, the total cross section is 1.5πr2. Given the same length, the ratio of conductor
material required is exactly the cross-section ratio, that is, 1:5=2 ¼ 0:75. In other words,

XI-571
Chapter 11 AC Power and Power Distribution

the three-phase system consumes 25 % less conductor material compared to the single-
phase system. The key is the absence of the neutral wire (or, possibly, using a much
thinner neutral wire). Other examples for particular loads might result in even more
dramatic savings.

11.4.4 Balanced Delta-Connected Load

Along with the wye-connected load, an important example of the three-phase load is the
delta-connected load, which is shown in Fig. 11.27a in the balanced configuration. The
balanced delta-connected load is common, along with a balanced wye-connected load.
The delta-connected load inherently does not have a neutral port. This load may be
converted to the wye-connected load shown in Fig. 11.27b by using the Y-Δ transforma-
tion algorithm established in Chapter 3. This algorithm equally applies to the impedance
circuits. The algorithm considerably simplifies when the loads are balanced (load resis-
tances or impedances are equal). With reference to Fig. 11.27, one has

1
ΖY ¼ ΖΔ $ ΖΔ ¼ 3ΖY ð11:58Þ
3

for phase impedance transformation. Here, indexes Y and Δ refer to the wye-connected
and delta-connected loads, respectively.

a) Delta-connected load b) Wye-connected load

a a

b Z b
1 1
Z Z Z Z
3 1 3
Z
c c 3

Fig. 11.27. Delta-connected load versus ad.

11.4.5 Balanced Delta-Connected Source

The balanced delta-connected source is shown in Fig. 11.28b. In its original configura-
tion, it is not using the ground terminal or a neutral conductor. The delta-connected source
so wound is generally less common and less safe than the wye-connected source. It may
be created by the three-phase generator shown in Fig. 11.17 of the previous section,
assuming the three individual coil windings aa0 , bb0 , and cc0 are interconnected in a
closed loop.

XI-572
Chapter 11 Section 11.4: Power in Balanced Three-Phase Systems. . .

a) Wye-connected source b) Delta-connected source

a Vab a
Van Vbn -

+
b b

+
-

-
- +
Vca Vbc
+
- -
Vcn +
c c

Fig. 11.28. Wye-connected source versus delta-connected source.

The balanced wye-connected source without a neutral or ground conductor can be

easily converted to the balanced delta source and vice versa. The concept is shown in
Fig. 11.28. The line voltages Vab, Vbc, Vca between nodes a–b, b–c, and c–a of the wye
source become the phase voltages of the delta source. The relation between two voltage
types is given by Eq. (11.44) of the previous section, that is (positive phase sequence),
pffiffiffi pffiffiffi pffiffiffi
Vab ¼ 3Van ∠30 , Vbc ¼ 3Vbn ∠30 , Vca ¼ 3Vcn ∠30 ð11:59Þ

Thus, according to Eq. (11.59) and Fig. 11.28, the phase voltages of the equivalent
pffiffiffi
delta-connected source Vab, Vbc, Vca are greater in amplitude by a factor of 3  1:73
as compared to the phase voltages Van, Vbn, Vcn of the equivalent wye-connected source
in Fig. 11.28. The line voltages of the delta-connected source coincide with its phase
voltages given lossless conductors and coincide with the line voltages of the wye-wye
source; all of them are simply Vab, Vbc, Vca. Indeed, the sum of the phase voltages for
the delta-connected source is still equal to zero according to Eq. (11.43) of the previous
section. Hence, there is no current circulation in the (ideal) delta loop in Fig. 11.28b.
Transformations given by Eqs. (11.58) and (11.59) allow us to consider four distinct
source-load configurations in the three-phase systems: wye-wye, wye-delta, delta-wye,
and delta-delta. All of them may be reduced to the wye-wye circuit or solved indepen-
dently. Figure 11.29 shows one such configuration: a balanced delta-delta distribution
system. In the delta-delta system, the line voltages coincide with the phase voltages,
whereas the line currents Ia, Ib, and Ic are different from the load (or phase) currents IAB,
IBC, and ICA. This is in contrast to the wye-wye system where the line and phase voltages
are different, but the line and load currents remain the same.

XI-573
Chapter 11 AC Power and Power Distribution

Ia
- Vab I AB Z

+
a A
b B I BC I CA
Ib
- Z Z
+
Vca Vbc
+
-
Ic
c C

Fig. 11.29. Three-phase balanced delta-delta distribution system. Note the load (phase) currents
circulating in the delta loop.

Example 11.14: A balanced delta-delta system in Fig. 11.29 operates at 60 Hz. The phase
voltages of the delta source, Vab, Vbc, Vca, have amplitudes of V m ¼ 294:5 V each.
Moreover, each phase impedance is a 0.2315 H inductance in series with a 87.3 Ω
resistance. Find the average load power.
Solution: We find the impedance for each phase of the load first. One has
pffiffiffi
Z ¼ R þ jωL ¼ 87:3 þ j87:3 Ω ¼ 87:3 2∠45 Ω ð11:60Þ

The power angle is thus given by θ ¼ 45. Next, we find the load (phase) currents
IAB, IBC, ICA circulating in the delta-connected load. Since the individual voltage
sources in Fig. 11.6 are now directly connected to the individual load phases, one
has for the amplitude of the phase current IAB:

Vab V m 294:5 294:5

IAB ¼ ) Im ¼ ¼ ¼ ¼ 2:385 A ð11:61Þ
Z jZj Z 123:46

The remaining phases have the same amplitudes: the per-phase method is used again. Both
the average load power and the instantaneous load powers are the sum of three individual
contributions, that is,
 
1 3
P ¼ pðtÞ ¼ 3
V m I m cos θ ¼ 294:5
2:385
0:707 ¼ 745 W ð11:62Þ
2 2

The instantaneous power may be calculated; it is constant and equals 745 W. Note that the
rms line voltages in this example are 208 V.

XI-574
Chapter 11 Section 11.4: Power in Balanced Three-Phase Systems. . .

Example 11.15: Solve the previous example by converting the delta-delta system to the
equivalent wye-wye system.
Solution: First, the phase impedance of the wye load pffiffiffishould be three times less than the

pffiffiffiis, Z ¼ 29:1 2∠45 Ω. Second, the amplitude of
phase impedance of the delta load, that
the wye phase source should be 1= 3 times less than the amplitude of the delta phase
source, that is, V m ¼ 170 V. These numbers have been used in Examples 11.12 and 11.13
for the wye-wye system, which gave us exactly the same value of 745 W (one horsepower)
for the average and instantaneous powers, respectively.

Apart from the circuit equivalence, one may look at Fig. 11.28 from a slightly different
perspective. What if the voltage sources in Fig. 11.28 are all the same (the same windings
of the three-phase generator just connected differently)? In this case, the wye connection
pffiffiffi
gives us a line voltage 3 times greater than the delta connection. Hence, the line current,
pffiffiffi
which is required for the same power transfer, will be 1= 3 times less. Reducing line
currents reduces line losses. This explains why the wye source connection is preferable
for long-distance power transmission.

XI-575
Chapter 11 AC Power and Power Distribution

Summary
rms Voltages and currents in terms of sine/cosine amplitudes and in the general case
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u ðT
u
Vm Im u1 2
For sinusoidal signals: V rms ¼ pffiffiffi, I rms ¼ pffiffiffi General periodic case: V rms ¼ t υ ðt Þ dt
2 2 T
0

1
Average power for resistive load: P ¼ V m I m ¼ V rms I rms , V rms ¼ RI rms
2
Power angle θ and power factor PF
υðt Þ ¼ V m cos ðωt þ φÞ
) θ ¼ φ  ψ,  90  θ  þ90 PF ¼ cos ðφ  ψ Þ ¼ cos θ
iðt Þ ¼ I m cos ðω t þ ψ Þ
V mI m
Average power for arbitrary load: P ¼ cos θ ¼ V rms I rms cos θ (zero for L and C)
2
 
Re V  I*
Average power and power angle in terms of phasors: P ¼ , Z ¼ jZj∠θ
2
Average power P, reactive power Q, complex power S, and apparent power S
 
Re V  I* RjIj2 jZjjIj2 jVjjIj
P¼ ¼ ¼ cos θ ¼ cos θ ¼ V rms I rms cos θ ½W
2 2 2 2
 
Im V  I* X jIj2 jZjjIj2 jVjjIj
Q¼ ¼ ¼ sin θ ¼ sin θ ¼ V rms I rms sin θ ½VAR
2 2 2 2
V  I*
S¼ ¼ P þ jQ ½VA
2
jZjjIj2 jVjjIj
S ¼ jSj ¼ ¼ ¼ V rms I rms ½VA
2 2
Power triangle (lagging/leading power factor)

AC power conservation laws

For any network of N loads connected in series, parallel, or in general:
S ¼ S1 þ S2 þ . . . SN , P ¼ P1 þ P2 þ . . . PN , Q ¼ Q1 þ Q2 þ . . . QN
(continued)

XI-576
Chapter 11 Summary

Power factor correction

L ðω LÞ2 1 þ ω2 R 2 C 2 1
C¼ ) Z ¼ R þ ) L¼ )Z¼Rþ )
ω2 L2 þ R2 R ω C
2
RðωC Þ2
without capacitor without inductor
zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{ zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{
2
I ¼ RðωC Þ V m2 þ
jωLV m jωCV m
I ¼ 2 RV m 2  2
R þ ðω LÞ R þ ðωLÞ2 1 þ ðωR C Þ 1 þ ðωR C Þ2
|fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflffl} |fflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflffl}
with capacitor with inductor

P remains exactly the same, Q becomes zero, PF becomes 100 %

Maximum power transfer
0:5RL jVT j2
P¼ ½ W
ðRL þ RT Þ2 þ ðX L þ X T Þ2
1 jVT j2
Pmax ¼ ½ W
8 RT
at ZL ¼ Z*T
Some equivalent drawings of the same balanced three-phase four-wire wye–wye power
distribution system

(continued)

XI-577
Chapter 11 AC Power and Power Distribution

Major parameters of the balanced three-phase four-wire wye–wye power distribution system
Positive phase sequence Van ¼ V m ∠0 , Vbn ¼ V m ∠ 120 , Vcn ¼ V m ∠ þ120
Negative phase sequence Van ¼ V m ∠0 , Vbn ¼ V m ∠ þ120 , Vcn ¼ V m ∠ 120
Current in the neutral wire: In ¼ 0
Per phase solution: Ia ¼ I m ∠  θ, Ib ¼ I m ∠ 120  θ, Ic ¼ I m ∠ 120  θ
I m ¼ V m =Z, Z ¼ Z∠θ
pffiffiffi
Vab ¼ Van  Vbn ¼3Van ∠30 ,
pffiffiffi
Line voltages (positive phase sequence): Vbc ¼ Vbn  Vcn ¼ 3Vbn ∠30 ,
pffiffiffi
Vca ¼ Vcn  Van ¼ 3Vcn ∠30
3
Instantaneous/average load power: pðt Þ ¼ P ¼ V m I m cos θ ¼ 3V rms I rms cos θ ¼ const
2
3
Apparent load power: S ¼ V m I m ¼ 3V rms I rms
2
Some common wye distribution systems
3-Phase, 4-Wire 208Y=120 V ðUSÞ Line : V rms ¼ 208 V, V m ¼ 294 V
Phase : V rms ¼ 120 V, V m ¼ 170 V
3-Phase, 4-Wire 400Y=230 V ðEU, OthersÞ Line : V rms ¼ 400 V, V m ¼ 566 V
Phase : V rms ¼ 230 V, V m ¼ 325 V
Wye load to delta load conversion

Wye source to delta source conversion

pffiffiffi pffiffiffi pffiffiffi

Vab ¼ 3Van ∠30 , Vbc ¼ 3Vbn ∠30 , Vca ¼ 3Vcn ∠30 (positive phase sequence)

XI-578
Chapter 11 Problems

Problems Problem 11.6. Present a mathematical proof of

the fact that the expression for the average
11.1 AC Power Types V2
power, P ¼ 2Rm , holds for an AC voltage signal
and Their Meaning given by υðt Þ ¼ V m cos ðωt þ φÞ ½V where φ
11.1.1 Instantaneous AC Power is an arbitrary phase.
11.1.2 Time-Averaged AC Power
Problem 11.1. An AC voltage signal across a 11.1.3 Application Example: rms Volt-
resistive load with R ¼ 10 Ω is given by: ages and AC Frequencies Around the
A. υðt Þ ¼ V m cos 1000t ½V World
B. υðt Þ ¼ V m sin 60t ½V Problem 11.7. A 100 Ω resistive load is
C. υðt Þ ¼ V m cos ð60t þ 45 Þ ½V connected to an AC wall plug in:
where V m ¼ 10 V. Determine the average AC A. Peoples Republic of China
power into the load in every case. B. India
C. USA
Problem 11.2. An alternating current through a D. Germany
resistive load with R ¼ 100 Ω is given by: Determine the average power delivered to the
A. iðtÞ ¼ I m cos 106 t ½V load in every case. Also determine the rms load
B. iðt Þ ¼ I m cos 37t ½V current in every case.
C. iðt Þ ¼ I m sin ð2011t þ 45 Þ ½V
where I m ¼ 1 A. Determine the average AC Problem 11.8. What do you think is a major
power into the load in every case. A. Advantage
B. Disadvantage
Problem 11.3. An rms voltage across a resis- of having a higher AC voltage?
tive load with R ¼ 100 Ω is given by:
A. V rms ¼ 5 V
B. V rms ¼ 100 V 11.1.4 rms Voltages for Arbitrary Peri-
C. V rms ¼ 0 V odic AC Signals
Determine the average power into the load in Problem 11.9. Determine the average power
every case. delivered to a 100 Ω resistive load when the
applied periodic voltage signal has the form
Problem 11.4. An rms current through a resis- υðt Þ ¼ ð5t þ 0:01Þ=T ½V over one period
tive load with R ¼ 1 kΩ is given by: T ¼ 0.01 s. This signal is known as the saw-
A. I rms ¼ 1 A tooth or the triangular wave:
B. I rms ¼ 100 μA A. Use the analytical calculation of the rms
C. I rms ¼ 0 A voltage.
Determine the average power into the load in B. Use the rms voltage found numerically,
every case. based on a MATLAB script or any soft-
ware of your choice.
Problem 11.5. An AC voltage signal is given
by: Problem 11.10. Determine the average
A. υðt Þ ¼ V m cos ðω t þ φÞ ½V power delivered to a 100 Ω resistive load
B. υðt Þ ¼ 1V þ V m cos ðωt þ φÞ ½V when thepffiapplied periodic voltage has the form
C. υðt Þ ¼ 1 V  V m sin ðωt þ φÞ ½V υðt Þ ¼ t =T ½V over one period T ¼ 0.01 s:
where V m ¼ 1 V, ω ¼ 100 rad=s, and A. Use the analytical calculation of the rms
φ ¼ π=2 rad. Find the time-average voltage voltage.
υðt Þ in every case. B. Use the rms voltage found numerically,
based on a MATLAB script or any soft-
ware of your choice.

XI-579
Chapter 11 AC Power and Power Distribution

Problem 11.11. Of the two periodic voltage sig- The periodic voltage on the top graph is
nals shown in the figures below, the cosine function. The periodic voltage
on the bottom graph is given by υðt Þ ¼ 3:2

a) 6 105 ðt  0:005Þ2  4 ½V over the time interval

voltage, V
4 from 0 to T.

2 a) 6
voltage, V
0 4

-2 2

-4 0

-6 -2
0 5 10 15 20 25 30 35 40
time, ms
-4
b) 6 time, ms
voltage, V
-6
4 0 5 10 15 20 25 30 35 40
time, ms
2 b) 6 time, ms
voltage, V
0 4

-2 2

-4 0

-6 -2
0 5 10 15 20 25 30 35 40
time, ms -4
which signal delivers more average power into
-6
a resistive load? The periodic voltage on the top 0 5 10 15 20 25 30 35 40
graph is the cosine function. Explain your time, ms
answer and provide an analytical proof (find
the rms voltages and the average power in Problem 11.13. Of the two periodic signals
every case). shown in the figure that follows, which signal
delivers more average power into a resistive
Problem 11.12. Of the two periodic signals
load? The periodic voltage on the top graph is
shown in the figures that follow, which signal
the cosine function. Explain your answer and
delivers more average power into a resistive
provide:
load? Explain your answer and provide:
A. An analytical proof—find the rms volt-
A. An analytical proof—find the rms volt-
age and the average power in every case
age and the average power in every case
B. A numerical proof (use MATLAB or any
B. A numerical proof (use MATLAB or any
software of your choice).
software of your choice).

XI-580
Chapter 11 Problems

a) 6 A. Find the average power delivered to the

voltage, V load analytically.
4
B. Find the average power delivered to the
2 load numerically using MATLAB or any
software of your choice.
0
Problem 11.17. Repeat the previous problem
-2
for phasor voltage and phasor current in the
-4 form:
-6
0 5 10 15 20 25 30 35 40 V ¼ 2 þ j2 ½V
b) 6
time, ms
time, ms
I ¼ 1  j1 ½A
voltage, V
4
Problem 11.18. Express the average power
ReðVI* Þ
2 given by P ¼ 2 in terms of the following
0 three quantities: magnitude of the phasor volt-
age, |V|; the impedance magnitude, |Z|; and the
-2 real part of the impedance, Re(Z).
-4
Problem 11.19. Determine the average power
-6 delivered to the load circuit between terminals
0 5 10 15 20 25 30 35 40
a and b shown in the figure that follows. The
time, ms
AC angular frequency is 100 rad/s.

A) a 1 mH
11.1.5 Average AC Power in Terms
of Phasors: Power Angle
10cos t [A]
11.1.6 Average Power for the Resistor, 100 F
Capacitor, and Inductor
Problem 11.14. The phasor voltage across a
purely resistive load with the resistance b 100
R ¼ 100 Ω is given by V ¼ 2  j1:5 ½V. B) a
Find the average power delivered to the load.
10cos t [A]
Problem 11.15. The phasor current through a 1 mH 100

purely resistive load with the resistance

R ¼ 100 Ω is given by I ¼ 1  j0:5 ½A.
b
Find the average power delivered to the load.

Problem 11.16. The phasor voltage across an Problem 11.20. Determine the average power
AC load and the phasor current through the delivered to the load circuit between terminals
same AC load are given by: a and b shown in the figure that follows. The
AC angular frequency is 1000 rad/s.
V ¼ 3 þ j3 ½V
I ¼ þ j0:1 ½A

XI-581
Chapter 11 AC Power and Power Distribution

A) a 1 mH Problem 11.23. Determine the resistance and

the reactance of the circuit block (the load)

+
shown in the figure in terms of R, L, and C in
100cos t [V] 100 F a general form. The AC angular frequency is ω.
-
a
b 100

B) a
R L C
+

1 mH
1 F
100cos t [V] 100 b
-
Problem 11.24. Write the expressions (and
b show units) for the average power P and the
reactive power Q in terms of:
Problem 11.21. Determine the average power A. Phasor current I through the load and the
delivered to the load circuit shown in the figure load resistance R and the reactance X
below. The AC signal frequency is 106 Hz. B. Phasor voltage V across the load, the
load impedance magnitude |Z|, and the
a
impedance phase (or the power angle) θ
10cos t [mA] 1 F
1 mH 100 Problem 11.25. For the circuit shown in the
figure with the parameters V m ¼ 170 V, ω ¼
377 rad=s, L ¼ 26:5 mH, R ¼ 25 Ω:
b A. Determine the power angle and the
power factor.
B. Determine the average (or true) power and
11.1.7. Average Power, Reactive Power, the reactive power for the inductive load
and Apparent Power shown in the figure.
11.1.8. Power Triangle C. Construct the corresponding power
Problem 11.22. Determine the resistance and triangle.
the reactance of the circuit blocks (the load)
shown in the figure. The AC angular frequency load
is 1000 rad/s.
L
A) a 10 mH vS(t) +
-
R

100 F

Im
b 100 500 VAR

B) a

10 mH 100 0 500 W
Re

XI-582
Chapter 11 Problems

Problem 11.26. For the circuit shown in the Im

figure with the parameters V m ¼ 170 V, ω ¼ 1 kVAR
377 rad=s, C ¼ 500 μF, R ¼ 10 Ω:
A. Determine the power angle and the
power factor.
0 1 kW Re
B. Determine the average (or true) power
and the reactive power for the capacitive
load shown in the figure.
C. Construct the corresponding power
triangle.

load

C 11.2 Power Factor Correc-

+
vS(t)
-
R
tion: Maximum Power Effi-
ciency and Maximum Power
Transfer
Im
1 kVAR 11.2.1 Power Factor Correction
11.2.3 Principle of Maximum Power
Efficiency for AC Circuits
1 kW Problem 11.28. Correct the power factor for
0 Re the inductive load shown in the figure below.
The circuit parameters are V m ¼ 170 V, ω ¼
377 rad=s and L ¼ 53 mH, R ¼ 10 Ω:
A. Present the circuit diagram of the modi-
fied load and determine the required
capacitance.
Problem 11.27. For the circuit shown in the B. Determine average (true) power, reactive
figure with the parameters power, power factor, and amplitude of
V m ¼ 170 V, ω ¼ 377 rad=s, L ¼ 14:07 mH, the circuit current before the power fac-
C ¼ 500 μF, R ¼ 10 Ω: tor correction.
A. Determine the power angle and the C. Determine average (true) power, reactive
power factor. power, power factor, and amplitude of
B. Determine the average (or true) power and the circuit current after the power factor
the reactive power for the complex load correction.
shown in the figure.
C. Construct the corresponding power
triangle. L
vS(t) +
load
-
R

L
C
VS(t) +
- Problem 11.29. Correct the power factor for
R
the capacitive load shown in the figure that

XI-583
Chapter 11 AC Power and Power Distribution

follows. The circuit parameters are V m ¼ 170 B. Repeat the same task for the load imped-
V, ω ¼ 377 rad=s and C ¼ 265 μF, R ¼ ance of 1  j50 [Ω].
10 Ω: C. Repeat the same task for the load imped-
A. Present the circuit diagram of the modi- ance of 5 + j50 [Ω].
fied load and determine the required Hint: Derive the general expression for the
inductance. power ratio first and then plug in the numbers.
B. Determine average (true) power, reactive
power, power factor, and amplitude of
the circuit current before the power fac- 11.3 AC Power Distribution:
tor correction.
Balanced Three-Phase
Power Distribution System
C 11.3.1 AC Power Distribution Systems
vS(t) +
-
R
11.3.2 Phase Voltages: Phase Sequence
Problem 11.34. Draw generic circuits for the
following representative AC power distribution
systems:
Problem 11.30. A whip monopole antenna A. Single-phase two-wire system
used in US Coast Guard ships has an equivalent B. Single-phase three-wire system
electric circuit shown in the figure of the previ- C. Two-phase three-wire system
ous problem. Its (radiation) resistance is 1 Ω, D. Three-phase four-wire system
and the reactance is j1000 Ω. By modifying Show loads and phasor voltages with the
the antenna circuit with a lumped inductor, it is corresponding phases.
required to make the antenna impedance real
and as large as possible: Problem 11.35. Determine the phase sequence
A. Present the circuit diagram of the modi- for the phase voltages given by:
fied load
B. Determine the required impedance of the υan ðt Þ ¼ 240 cos ð314t þ 75 Þ ½V,
inductor. υbn ðt Þ ¼ 240 cos ð314t  165 Þ ½V,
υcn ðt Þ ¼ 240 cos ð314t  45 Þ ½V:
11.2.4 Principle of Maximum Power
Transfer for AC Circuits To simplify the solution, construct the
Problem 11.31. Describe in your own words corresponding phasor diagram in the figure
the difference between the concepts of maxi- below:
mum power efficiency and maximum power
transfer for AC circuits.
Im

Problem 11.32. A generator’s impedance is

50∠30 ½Ω. What should the load imped-
ance be to allow the maximum power transfer
to the load? Re

Problem 11.33
A. A generator’s impedance is 50 Ω. The
load impedance is 1 + j50 [Ω]. What per-
centage of the maximum available
power (at the load impedance of 50 Ω)
is transferred to the load?

XI-584
Chapter 11 Problems

Problem 11.36. Given Vbn ¼ 120 ∠45 ½V, Problem 11.39. Repeat the previous problem
find Van and Vcn assuming: for the circuit shown in the figure below:
A. The positive abc phase sequence
B. The negative acb phase sequence
Express your result in phasor form. Make sure Vm 0 + Z Z
-
that the phase ranges from 180 to +180 . To
check the solution, you may want to use the Vm 120
n N

-
-

+
+
corresponding phasor diagram shown in the
figure for the previous problem. Vm 120 Z

11.3.3 Wye (Y) Source and Load

Configurations for Three-phase Circuits Problem 11.40. Repeat Problem 11.38 for the
circuits shown in the figure that follows:
11.3.4 Application: Examples of
Three-phase Source and the Load Z2
11.3.5 Solution for the Balanced
Three-Phase Wye-Wye Circuit Vm 0 + Z1 Z1
-
11.3.6 Removing the Neutral Wire in Z2

+
-
Long-Distance Balanced High-power Vm 120
-
+

Transmission Z
Vm 120 1
Problem 11.37
A. Draw the circuit diagram for a generic
Z2
three-phase four-wire balanced wye-wye
power distribution system. Problem 11.41. Prove that Eq. (11.28) of this
B. Labelphasevoltages and phaseimpedances chapter for line voltages also holds for the
(load impedances per phase). negative phase sequence to within the substitu-
C. Label line currents. tion 30 ! 30 .

Problem 11.38. A three-phase circuit is shown Problem 11.42. The local electric service in
in the figure that follows: the European Union is provided by a three-
A. Is it a balanced wye-wye circuit? phase four-wire abcn wye system with the line
B. If not, show your corrections on the voltages equal to 400 V rms each (so-called
figure. Niederspannungsnetz):
A. Determine the rms phase voltages.
B. By connecting terminals abcn in any
Vm 0 sequence of your choice, could you in
+
- Z1
2Z1 principle obtain the rms voltages higher
Vm 120 than 400 V?
-
+

-
+ Z1 Problem 11.43. Determine line currents in the
Vm 120 balanced three-phase wye-wye circuit shown in
the figure that follows. You are given the acb
sequence of phase voltages Van ¼ 170∠0 ½V,

XI-585
Chapter 11 AC Power and Power Distribution

Vbn ¼ 170∠120 ½V, Vbn ¼ 170∠120 ½V,

and load impedance per phase, Z ¼ 8 þ 11.4 Power in Balanced
j30 Ω. Three-Phase Systems:
Plot phasor currents on the phasor diagram that
follows. Delta-connected
Three-Phase Circuits
Ia Z Z
Van Vbn b 11.4.1 Intantaneous Power
a
11.4.2 Average Power, Reactive
+

Ib
-

n N Power, and Apparent Power

Vcn
-
+ Ic Problem 11.45. In a three-phase balanced
Z wye-wye system, the rms phase voltages are
c 120 V, and the rms line currents are 10 A. The
impedance has the power angle of θ ¼ 75 .
Find:
Im 1. The instantaneous load power
2. The average load power

Problem 11.46. In a three-phase balanced

wye-wye system, the rms line voltages are
Re 400 V, and the rms line currents are 10 A. The
impedance has the power angle of θ ¼ 60 .
Find:
1. The instantaneous load power
2. The average load power

Problem 11.47. In the three-phase system

Problem 11.44. In the balanced three-phase
shown in the figure that follows, Z ¼ 40∠60 .
wye-wye circuit shown in the figure that fol-
The sources have the relative phases
lows, the power line resistance and inductance
0,  120 , þ 120 . The rms line voltages
are additionally included into consideration. The
are 208 V. Determine:
three-phase source operates at 60 Hz; R ¼ 2 Ω,
A. The type of the three-phase system
L ¼ 9:6 mH. You are given the abc sequence
B. Instantaneous power delivered to the
of phase voltages Van ¼ 170∠0 ½V, Vbn ¼
three-phase load
170∠  120 ½V, Vbn ¼ 170∠120 ½V, and
C. Average power delivered to the three-
load impedance per phase, Z ¼ 7 þ j30 Ω.
phase load
A. Determine line currents.
B. Plot phasor line currents on the phasor
diagram to the previous problem. a
+ Z
- Z
R L
Ia b
+
-

-
Van Vbn R L Z Z +
Z
a c
+

+
-

Ib
Z
-
Vcn + Problem 11.48. A balanced wye-wye three-
R L phase system in the figure that follows uses loss-
c
less transmission lines and operates at 60 Hz.
Ic
XI-586
Chapter 11 Problems

The line-to-neutral voltages have the amplitudes a)

of 170 V,V m ¼ 170 V. Every phase impedance is 120 0 120 120 30 60 30 60
a 92-mH inductance in series with a 20 Ω resis- a b

+
tance. Find the instantaneous load power.

-
n N
-
+ 120 120 30 60
Van Vbn b Z Z
a c
+

+
-

b)
n N
- -
Vcn Z

+
+
a b B A
c -
+ +
-
Problem 11.49. In the previous problem:
A. Determine the load average power, reac- c C
tive power, and the apparent power.
B. Do these powers coincide with the Problem 11.53. A three-phase balanced delta-
corresponding source measures? delta system is shown in the figure that follows.
Its wye-wye equivalent is sought, which is shown
Problem 11.50. A three-phase induction motor
in the same figure. For the wye-wye system, write
is modeled by a balanced wye load. The motor
the corresponding voltage and impedance values
(active) power is 2.5 kW; the line current is
10 A rms, and the phase voltage of a three- in the phasor form close to every circuit element.
phase wye source is 120 V rms. Determine the
a)
power factor of the motor. - 400 30
+

a b B A
Problem 11.51. In the previous problem, the 90 60
motor (active) power is 9 kW; the line current -
+ +
-
is 15 A rms, and the line voltage of a three-
phase wye source is 400 V rms. Determine the 400 150 400 90 90 60 90 60
power factor of the motor. c C
b)
11.4.4 Balanced Delta-Connected Load
11.4.5 Balanced Delta-Connected Source a b
+

Problem 11.52. A three-phase balanced

-

wye-wye system is shown in the figure that n N

-
follows. Its delta-delta equivalent is sought, +
which is shown in the same figure. For the
c
delta-delta system, write the corresponding
voltage and impedance values in the phasor
form close to every circuit element.

XI-587
Chapter 11 AC Power and Power Distribution

Problem 11.54. A balanced delta-delta system Problem 11.55. A balanced wye-delta system
shown in the figure below operates at 50 Hz. shown in the figure below operates at 60 Hz.
The phase voltages of the delta source, Vab, The phase voltages of the wye source, Van,
Vbc, Vca, have the amplitudes of V m ¼ 563:4 Vbn, Vcn, have the amplitudes of V m ¼ 170 V
V each. Every phase impedance is a 0.21 H each. Each phase impedance is a 0.18 H induc-
inductance in series with a 38-Ω resistance: tance in series with a 90 Ω resistance:
A. Find the average load power, A. Find the average load power,
B. Find the instantaneous load power, B. Find the instantaneous load power,
C. Find the apparent load power. C. Find the apparent load power.

Vab B I AB A
- Ia
+

a Van Vbn b
b I BC I CA a
Z

+
- + Ib

-
Vca Vbc
+
-
- Z Z
Vcn +
Ic
c C
c

XI-588
Chapter 12

Chapter 12: Electric Transformer

and Coupled Inductors

Overview
Prerequisites:
- Knowledge of complex arithmetic
- Knowledge of basic circuit analysis (Chapters 3 and 4)
- Knowledge of self- and mutual inductances (Chapter 6)
- Knowledge of phasor/impedance method for AC circuit analysis (Chapter 8) and of
basic AC power analysis (Chapter 11)

Objectives of Section 12.1:

- Derive the ideal transformer model from the first principles
- Understand the role of the ideal magnetic core
- Understand the role of Faraday’s law and Ampere’s law
- Prepare the background for introducing magnetic circuits
- Understand and apply the dot convention
- Relate ideal transformer model to a model with dependent sources

Objectives of Section 12.2:

- Be able to analyze electric circuits with ideal transformer
- Learn about load and source reflections
- Learn about impedance matching via transformers
- Learn about electric power transfer via transformers

Objectives of Section 12.3:

- Derive equations for useful transformer types—autotransformer, multiwinding
transformer, and center-tapped transformer—from the first principles
- Understand the role of the center-tapped transformer for single-ended to differential
transformation and for power division

Objectives of Section 12.4:

- Understand the physical background of the Steinmetz model and relate the model
parameters to real transformers
- Be able to analyze the nonideal transformer model
- Define transformer voltage regulation and power efficiency
- Briefly discuss the high-frequency transformer model

© Springer Nature Switzerland AG 2019 XII-589

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2_12
Chapter 12 Electric Transformer and Coupled Inductors

Objectives of Section 12.5:

- Introduce the model of two coupled inductors from the first principles
- Learn how to analyze electric circuits with coupled inductors
- Learn about the useful conversion to the T-network of uncoupled inductances
- Obtain the basic exposure to wireless inductive power transfer including its major
features and challenges

Application examples:
- Electric power transfer via transformers
- Wireless inductive power transfer
- Coupling of nearby magnetic radiators

Keywords:
Electric transformer (primary winding, secondary winding, circuit symbol, isolation
transformer, instrumentation transformer, current transformer, clamp on ammeter, potential
transformer, exciting current, magnetizing current, magnetizing inductance, magnetizing
reactance, power conservation, stored energy, turns ratio, step-up transformer, step-down
transformer, high-voltage side, low-voltage side, transformer rating, dot convention, dotted
terminals, voltage polarity, current reference directions, summary of reference directions,
mechanical analogies), Ideal transformer model (ideal magnetic core, ideal open-circuited
transformer, ideal transformer equations, ideal transformer equations in phasor form, power
conservation, stored energy, model in terms of dependent sources), Ampere’s law (linked
current, for ideal magnetic core, for multiwinding transformer), Referred (reflected) source
network in the secondary, Referred (reflected) load impedance in the primary, Load reflection,
Source reflection, Reflected resistance, Reflected inductance, Reflected capacitance, Transformer
as a matching circuit, Matching real-valued impedances, Matching arbitrary complex impedances,
Partial matching condition, Power transfer via transformers (for fixed load voltage, sending-
end voltages, for fixed source voltage), Autotransformer (step down, step up, circuit symbol,
ideal transformer equations), Multiwinding transformer (ideal circuit equations, Ampere’s law,
telephone hybrid circuit), Center-tapped transformer (ideal transformer equations, single-ended
to differential transformation, 180
power divider, 180
power splitter), Real transformer
(nonideal low-frequency model, Steinmetz model, Steinmetz parameters (magnetizing
reactance, core loss resistance, primary leakage reactance, secondary leakage reactance, primary
ohmic resistance, secondary ohmic resistance), nonideal transformer model terminology/analysis,
voltage regulation, transformer power efficiency, nonideal high-frequency model), Model of two
coupled inductors (equations, circuit symbol, circuit analysis, solution for N coupled inductors,
stored energy, conversion to T-network), Mutual inductance (definition, of ideal transformer, of
two coaxial coils), Coupling coefficient (definition, largest possible value, trend), Wireless
inductive power transfer (application, basic model, examples of), Mutual coupling for
nearby magnetic radiators (arrays of magnetic radiators), Lenz’s law

XII-590
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

Section 12.1 Ideal Transformer as a Linear Passive

Circuit Element
12.1.1 Electric Transformer
General
An electric transformer is a simple and versatile device, which targets AC power transfer
from one electric circuit to another. These circuits are coupled via a time-varying
magnetic flux linking two or more coils. There is no direct electric connection between
the coils. The transformer cannot transfer power between the DC circuits. Analysis of
transformers involves many principles that are basic to the understanding of electric
machines. Transformers are primarily used to:
1. Change the voltage level in power electronics AC circuits.
2. Insulate one AC (or RF) circuit from another (isolation transformers).
3. Match the impedance of the source and the load in electronic circuits.
4. Measure AC voltages and currents (instrumentation transformers).
As another everyday application example, we mention various DC power supplies
(AC to DC converters or adapters), both switching and linear. These DC supplies power
PCs, printers, modems, cordless phones, video game consoles, etc. at your home.
Low-frequency (bulky) or high-frequency (smaller) transformers are very important
parts of these supplies, irrespective of their particular construction.

Function
Although the transformer typically consists of two coupled inductors—see Fig. 12.1—its
function is principally different from that of the familiar inductance. While the inductance
is an energy-storage (and energy-release) circuit element, the ideal transformer, as a new
circuit element, never stores any instantaneous energy. It does not possess any inductance
(or impedance in general) either.

Approach
The model a two-winding electric transformer introduced in this section does yet not use
the concept of a magnetic circuit. Instead, we accurately formulate and employ Faraday’s
law and Ampere’s law directly. The same transformer model in the framework of
magnetic circuits is revisited in the next chapter. Through the text, we use opposite
reference directions for the transformer currents; the equal directions are also discussed in
the text.

12.1.2 Ideal Open-Circuited Transformer: Faraday’s Law of Induction

We will perform transformer analysis in several steps illustrated in Fig. 12.1.

XII-591
Chapter 12 Electric Transformer and Coupled Inductors

a) cross-section A b)

ideal-transformer model

Lm
vS(t) + +
+ v1(t) N1 +
- v1
-
- -
B

c) cross-section A d)
dot convention

+ ideal-transformer model
v2(t)

vS(t) + + +
+ N1 N2 + + N2
v1(t) v1 - v v2
- - N1 1
- - -
B
-

e) cross-section A f)
i2(t) ideal-transformer model

+ i1 i2
i1(t)
v2(t)
N2 N2
i v
vS(t) + + N 1 2 N1 1
+ v1(t) N1 N2 + v1 + R
- - -
-
-
B
-

Fig. 12.1. Construction of an ideal electric transformer. The negligibly small exciting current is not
shown. In all three cases, the magnetic flux Φ(t) in the core remains the same.

Using Faraday’s Law of Induction

Transformer construction can start with a single lossless inductor with a closed magnetic
core directly connected to an AC voltage source, see Fig. 12.1a. The inductor is a coil
with N1 turns—the primary winding of the transformer. Its inductance neglecting the flux
leakage effect will be denoted here by Lm. By KVL,

υ1 ðt Þ ¼ υS ðt Þ ð12:1Þ

The primary winding establishes a finite time-varying magnetic flux Φðt Þ ¼ ABðt Þ in the
core. Here, B is the flux density and A is the core cross section. According to Faraday’s

XII-592
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

law of induction—see Eq. (6.19) and (6.33)—this flux is uniquely determined by the
winding voltage (which is equal to the source voltage) in the form

dΦðt Þ
υ 1 ðt Þ ¼ N 1 ð12:2Þ
dt
Exciting Current and Inductance for the Ideal Magnetic Core
An inductor current iΦ(t) which would be present in Fig. 12.1a is called the exciting
current (or magnetizing current). Accordingly, inductance Lm is called the magnetizing
inductance. The magnetizing inductance is found using Eq. (6.23). The exciting current is
then found using Eq. (6.19) of the same chapter, which is valid in both static and dynamic
cases. This gives

μ0 μr AN 21 Φ ðt Þ
Lm ¼ , i Φ ðt Þ ¼ N 1 ð12:3Þ
l Lm

where A is the core cross section shown in Fig. 12.1. When the relative magnetic
permeability μr of the core is very high, the coil inductance Lm is very large. Therefore,
the corresponding inductor current iΦ(t) is quite small. An ideal magnetic core assumes
that μr ! 1. Therefore, according to Eq. (12.3),

Lm ! 1, iΦ ðt Þ ¼ 0, μr ! 1 ð12:4Þ

Equation (12.4) corresponds to the ideal transformer model. The primary winding
becomes an open circuit of infinite inductance as shown in Fig. 12.1b. However, the
finite magnetic flux Φ(t) is still established in the core. There is no contradiction here
since a negligible exciting current iΦ(t) is necessary to establish the finite flux Φ(t) in a
core with the infinitely high permeability (infinitely high inductance). The situation is
somewhat similar to an operational amplifier with the negative feedback where the
negligible input different voltage controls the large-signal amplifier operation.

Using Faraday’s Law a Second Time: Relation Between Transformer Voltages

As a next step, another coil with N2 turns—the secondary winding of the transformer—
can be added as shown in Fig. 12.1c. The core flux Φ(t) also links the secondary winding
and creates an open-circuit voltage at its terminals. No flux leakage in air is permitted in
the ideal model. Faraday’s law is applied a second time, which yields

dΦðt Þ
υ 2 ðt Þ ¼ N 2 ð12:5Þ
dt
The plus sign implies the dot convention to be discussed shortly. From Eqs. (12.2) and
(12.5), the voltage ratio becomes

XII-593
Chapter 12 Electric Transformer and Coupled Inductors

υ2 N 2
¼ ð12:6Þ
υ1 N 1

at any time instant. Thus, the voltage ratio is the ratio of respective turns. The equivalent
circuit of the open-circuited transformer is that of Fig. 12.1d. The secondary winding
operates as a dependent voltage-controlled voltage source. Thus far, it has zero effect on
the circuit connected to the primary winding.

Exercise 12.1: You are given the source voltage in the form υS ðt Þ ¼ 170 cos ð2π60tÞ ½V,
the number of turns of the primary winding N 1 ¼ 200, the finite relative permeability of the
magnetic core, μr ¼ 5000, the coil length of the primary winding of 0.1 m, and the core
cross-section of A ¼ 0:001 m2 .
A. Find the exciting current (no-load current) in the primary winding of this non-ideal
transformer. Justify the ideal-transformer approximation Eq. (12.4).
B. Reduce N1 from 200 to 20 and repeat the solution.
Answers:
A. iΦ ðtÞ ¼ 179 sin ð2π60t Þ ½mA. The current amplitude and its rms value are much
less than typical current amplitudes of 5 A (3.54 A rms) observed in residential AC
circuits. The ideal-transformer approximation is justified.
B. iΦ ðtÞ ¼ 17:9 sin ð2π60t Þ ½A. The current amplitude is quite high; the ideal-
transformer approximation is severely violated.

Dot Convention
Windings of transformers are marked to indicate the relative voltage polarities of
voltages υ1 and υ2. We indicate the relative polarities by the dot convention.
Namely, voltages υ1 and υ2 with the positive polarity at the dotted terminals will be
strictly in phase, see Fig. 12.1c. It does not matter where the dots are exactly located.
Yet another meaningful definition of the dotted terminals is as follows. Currents entering
the dotted terminals (which means the passive reference configuration not to be
confused with Fig. 12.1e) would produce fluxes in the same direction in the core that
forms the common magnetic path. If the windings are visually seen, the polarities are
determined by examining clockwise or counterclockwise coil winding directions. Other-
wise, an experiment could be employed, with a function generator and an oscilloscope.
Emphasize that the dot convention only determines the voltage polarity; it has nothing
in common with large currents which can flow in a loaded transformer to be discussed
shortly.

XII-594
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

Exercise 12.2: In Fig. 12.1c, the winding direction of the secondary winding is reversed.
Where should the dot be placed?
Answer: To the lower end of the secondary winding.

12.1.3 Appearance of Transformer Currents

Finally, a load resistance R is connected across the terminals of the secondary winding,
this resistance should be present in Fig. 12.1e, f. As a result, a large current i2(t) will flow
in the secondary winding, which is given by Ohm’s law:

υ 2 ðt Þ
i 2 ðt Þ ¼ ð12:7Þ
R
In this case, the transformer performs its major function of power transfer from the source
to the load. When a large current i2(t) starts to flow, it will drastically change the magnetic
field in the core, core flux Φ(t), and consequently voltage υ1(t) according to Eq. (12.2).
However, any extra change in υ1(t) is in contradiction with KVL stated by Eq. (12.1).
Therefore, another large current i1(t) in the primary winding will immediately start to flow
too, in order to undo this change. In other words, current i2 “calls current i1 into existence.”
Ampere’s law is used to establish the quantitative relation between transformer currents i1
and i2 at any time instant and in any circuit, not necessarily the circuit from Fig. 12.1.

12.1.4 Ampere’s Law

Ampere’s law does not operate in terms of the magnetic flux Φ (or flux density ~ B), but
~ defined in Section 6.1.7. Consider a closed contour
rather in terms of the magnetic field H
abcd in Fig. 12.2. Ampere’s law expresses the field created by a linked current i via a
contour integral:
þ
~  d~
H l¼i ð12:8aÞ
abcd

It does not matter whether the magnetic core is present or not. The “linked current” i is the
total current that passes through a surface bounded by the contour in Fig. 12.2. The
direction in which the current traverses the contour and reference direction of the loop
abcd (clockwise in Fig. 12.2) are related by the right-hand rule. For the contour abcd in
Fig. 12.2, the field in the magnetic core is a constant and is parallel to the contour. In this
case, Eq. (12.8a) is simplified as

lH ¼ i ð12:8bÞ

XII-595
Chapter 12 Electric Transformer and Coupled Inductors

cross-section A
i2(t)
b c
i1(t) H +

+
N1 N2

H
-
a d

Fig. 12.2. Ampere’s law for a transformer.

where l ¼ l ab þ l bc þ l cd þ l da is the total contour length. Ampere’s law is further

simplified for time-varying currents linking the ideal magnetic core with μr ! 1. Note
the magnetic flux in the core is given by Φðt Þ ¼ ABðt Þ ¼ Aμ0 μr H ðt Þ. When μr ! 1, any
finite magnetic field H(t) would create an infinite flux in the core and thus infinite voltages
across the coils, which is impossible. Therefore, we must set H ¼ 0 in Eq. (12.8b) and
obtain

i ¼ 0 f or μr ! 1 ð12:8cÞ

In order words, for the ideal magnetic core, the time-varying linked current must be zero.

12.1.5 Ideal Loaded Transformer

For the transformer in Fig. 12.2, the linked current is given by

i ¼ N 1 i1  N 2 i2 ð12:9Þ

since current i1 is inflowing and current i2 is outflowing through the surface bounded by
the contour. Therefore, according to Eq. (12.8c)
i1 N 2
N 1 i1  N 2 i2 ¼ 0 ) ¼ ð12:10Þ
i2 N 1

at any time instant. Equation (12.9) completes the analysis of the ideal transformer. It
states that the current ratio is the inverse turns ratio. This result does not depend on the
particular value of the load resistance in Fig. 12.1f. It is perhaps even more important that
Eq. (12.10) does not require the source to be directly connected to the primary. Thus, the
ideal transformer as a new passive linear circuit element is completely described by
Eqs. (12.6) and (12.10) (ideal transformer equations), respectively. The corresponding
circuit model may be expressed in terms of two dependent sources—the voltage-
controlled voltage source in the secondary and the current-controlled current source in
the primary—see Fig. 12.1f.

XII-596
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

Typical Transformer Circuit

Figure 12.3a shows a typical single-phase transformer circuit along with the ideal
transformer circuit symbol (with a magnetic core). The transformer acts as a linear
interface between an arbitrary AC source circuit and an arbitrary AC load circuit. Both
the source and the load do not have to be linear. Figure 12.3b shows the corresponding
dependent-source model. Such a model may be quite useful in transformer circuits where
extra coupling between the source and the load exists, for example, via a capacitor. This
may occur in high-frequency transformers.

a) i1(t) i2(t)

+ +
source v1(t) v2(t) load

- -

b) i2

N2 N2
i v
+ N1 2 N1 1
source v1 + load
-
-

Fig. 12.3. Transformer interfacing a source and a load and its model in terms of dependent sources.

Power Conservation
Combining Eqs. (12.6) and (12.10), we obtain for the instantaneous power delivered to
the load
N2 N1
p2 ¼ υ2 i2 ¼ υ1 i1 ¼ υ1 i1 ¼ p1 ð12:11Þ
N1 N2

Equation (12.11) states that, in the absence of losses, instantaneous (and indeed average)
power is conserved during voltage/current transformation; there is no net gain or loss
of power for an ideal transformer. As soon as power is consumed by the load, the same
power is drawn from the source. The transformer, therefore, provides a physical isolation
between the load and the source while maintaining electric power transfer. Note that
Fig. 12.3b has similarities with the equivalent amplifier circuit model. If the transformer
may be used as a voltage booster, why do we need the semiconductor amplifier then?
The answer is that it is the power that counts. When the transformed voltage increases,
the transformed current decreases so that the output power remains the same. An
amplifier, on the other hand, boosts the total power, which is true amplification of the
source signal. Still, the transformer may be used at the amplifier output to provide a
significant extra voltage (but not power) gain.

XII-597
Chapter 12 Electric Transformer and Coupled Inductors

Stored Energy
The ideal loaded transformer in Fig. 12.1e, f does not possess any stored magnetic-field
energy. The instantaneous magnetic-field energy stored in a magnetic core of volume V is
given by Eq. (6.18):
ð
1 ~ Al 2 l
E ðt Þ ¼ Bðt Þ  ~
Bðt Þdυ ¼ B ðt Þ ¼ Φ 2 ðt Þ ð12:12Þ
μ0 μr μ0 μr Aμ0 μr
V

The energy becomes exactly zero when μr ! 1. Here, l is the length of the centerline of
the magnetic core and A is its cross section. The same proof is valid for the open-circuited
ideal transformer.

Exercise 12.3: Under conditions of Exercise 12.1, determine the instantaneous stored
energy of a non-ideal transformer given that the core centerline length is 30 cm.
Answers:
A. E ðtÞ ¼ 0:24 sin 2 ð2π60tÞ ½J, which is a small value.
B. E ðt Þ ¼ 24 sin 2 ð2π60t Þ ½J, which is a quite significant number.

Exercise 12.4: In the circuit in Fig. 12.1f, υS ðt Þ ¼ 325 cos ð2π50t Þ ½V. For a 400 Ω
resistive load, determine load voltage, load current, and average power delivered to the
load when the turns ratio is equal to 10:1, 1:1, and 1:10.
Answers:
For 1:10 turns ratio:
υR ðt Þ ¼ 32:5 cos ð2π50t Þ ½V, iR ðt Þ ¼ 0:08125 cos ð2π50t Þ ½A, P ¼ 1:320 W
For 1:1 turns ratio:
υR ðt Þ ¼ 325 cos ð2π50tÞ ½V, iR ðtÞ ¼ 0:8125 cos ð2π50t Þ ½A, P ¼ 132:0 W
For 10:1 turns ratio:
υR ðt Þ ¼ 3250 cos ð2π50tÞ ½V, iR ðtÞ ¼ 8:125 cos ð2π50t Þ ½A, P ¼ 13:2 kW

12.1.6 Ideal Transformer Versus Real Transformer: Transformer

Terminology
Ideal Transformer Versus Real Transformer
The major assumptions of the ideal transformer discussed above are:
1. Infinite permeability of the magnetic core and no magnetic flux leakage from the
core in air.
2. No coil resistance loss (through using copper wires of a very small resistance).

XII-598
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

3. No other loss in the core called the iron loss. The iron losses would include
hysteresis loss and eddy current loss.
Real transformers studied in Section 12.4 deviate from this ideal circuit model often
very significantly. The ratio of output power to input power is called the efficiency of the
transformer. For large power transformers, the efficiency can be in excess of 98 %. For RF
(radio-frequency) transformers, the efficiency is typically much lower. Two methods of
analysis can be used to study realistic transformers:
1. An extended equivalent circuit model that includes the present ideal transformer
model plus extra inductances and resistances, see Section 12.4.
2. A different mathematical model of magnetically coupled circuits with self- and
mutual inductances, see Section 12.5.

Terminology
Engineers have adopted a special terminology when dealing with transformers:
A. The ratio N1 : N2 is the turns ratio of the transformer. A transformer with a primary
winding of 100 turns and a secondary winding of 200 turns has a turns ratio of
1:2. A transformer with a primary winding of 200 turns and a secondary winding of
150 turns has a turns ratio of 4:3.
B. When N 2 > N 1 , the transformer increases the input AC voltage; it is called a step-
up transformer.
C. When N 2 < N 1 , the transformer decreases the input AC voltage; it is called a
step-down transformer.
D. The winding with a higher number of turns is the high-voltage (HV) side of
the transformer.
E. The winding with a smaller number of turns is the low-voltage (HV) side of the
transformer.
In Figs. 12.1, 12.2, and 12.3, we have used the opposite current reference directions
for the two dotted terminals. Quite often, the same reference directions are employed.
This is to underscore the fact that either winding may serve as the input of the transformer.
Sign minus should then be inserted into Eq. (12.10) which relates i1 and i2.

Transformer Rating
Power transformers seldom drive purely resistive loads. Therefore, their power rating is given
in VA (volt-amperes) or kVA instead of watts, identical to the complex power defined in
Section 11.1. More precisely, this is the apparent load power defined by Eq. (11.21d).
Consider a popular example of a transformer that carries the following information on a
nameplate or in a reference manual: 10 kVA, 1100:110 V. The voltage rating means the one
transformer winding (high-voltage side) is rated for 1100 V, whereas another (low-voltage
side) for 110 V. The turns ratio is the voltage ratio, N 1 : N 2 ¼ 10. The corresponding current
ratings are 9.09 A rms and 90.9A rms, respectively.
XII-599
Chapter 12 Electric Transformer and Coupled Inductors

Exercise 12.4: A power transformer is rated as 1 kVA, 120:480 V. Determine transformer

type, turns ratio, and the rated current on the low-voltage side.
Answers: This is a step-up transformer with turns ratio 1:4 and a current on the
low-voltage side of 8.333 A rms.

Example 12.1: In the circuit shown in Fig. 12.4, υS ðt Þ ¼ 170 cos ð2πωt Þ ½V. The ideal-
transformer model is used. Determine all circuit currents and voltages when:
A. The switch is open.
B. The switch is closed.
C. The switch is closed and ω ! 0.
Solution A: The current through the secondary winding is zero, so is the current through
the primary. The open-circuit condition thus passes through the transformer. The physical
background for this is an infinite permeability of the magnetic core and hence an infinite
inductance of the primary coil. You should know that the infinite inductance is an open
circuit for an AC signal. By KVL, the voltage of the primary winding υ1(t) is equal to the
source voltage. Open-circuit voltage of the secondary winding υ2(t) is exactly five times
smaller.
Solution B: Both winding voltages remain identical to the voltages in the previous case.
The load voltage is now the voltage of the secondary winding. The load current, by Ohm’s
law, is iR ¼ i2 ¼ υR =RL ¼ 0:850 cos ωt ½A. The primary-winding current is exactly five
times smaller than the load current.
Solution C: When the frequency approaches zero, the transformer loses its functionality.
The induced voltages υ1 and υ2 tend to zero. Moreover, the primary winding of the
transformer becomes a short circuit, which prohibits using transformers loaded with DC
power sources.

i1(t) i2(t)

+ +
Vmcos( t)=vS(t) + v1 v2 RL=40
-
- -
N1:N2=5:1

Fig. 12.4. A transformer circuit with a load and a switch.

Exercise 12.5: In the circuit shown in Fig. 12.4, the current through resistance is given by
i2 ðt Þ ¼ 1 cos ð2π60t Þ ½A. With the switch closed, find the source voltage.
Answer: υS ðtÞ ¼ 200 cos ð2π60t Þ ½V.

XII-600
Chapter 12 Section 12.1: Ideal Transformer as a Linear Passive. . .

12.1.7 Mechanical Analogies of a Transformer

An electric transformer operates with alternating currents. One mechanical analogy of the
transformer is a gear transmission or gearbox, see Fig. 12.5a.

a)

primary secondary D1 D2

b)

N N
pivot

Fig. 12.5. Analogies for a transformer.

In terms of angular speed ω [rad/s] and developed torque T ½N  m, one has
D2 D1
T2 ¼ T 1, ω2 ¼ ω1 ð12:13Þ
D1 D2

where D1,2 are pitch diameters of gear wheels. Here, torque is the voltage and speed is the
current. D1,2 are similar to the number of turns, N1,2, of the primary and secondary windings
of the transformer, respectively. Emphasize that the present transformer analogy still ignores
the field effect—magnetic coupling between the coils. Therefore, it will fail in the DC case. A
more realistic hydraulic transformer analogy is shown in Fig. 12.5b. The model with four
pistons transforms power from one circuit to another in the AC case only. It is drawn for a
transformer whose turns ratio is 1:1. When, for example, a transformer with a turns ratio of
2:1 is required, the area of output pistons is doubled. This doubles the output current, but the
output voltage (the force) will be halved.

Historical: The transformer action was independently invented and utilized by many
researchers, starting with Michael Faraday, Joseph Henry, Nicholas Callan of Ireland,
and Pavel Yablochkov of Russia. Hungarian engineer Otto Titusz Bláthy introduced the
name “transformer.” A power transformer developed by Lucien Gaulard of France and
John Dixon Gibbs of England was demonstrated in London in 1881 and then sold to the
American company Westinghouse. In 1886 William Stanley, working for Westinghouse,
built the refined, commercially used AC transformer used in the electrification of Great
Barrington, Massachusetts.

XII-601
Chapter 12 Electric Transformer and Coupled Inductors

Section 12.2 Analysis of Ideal Transformer Circuits

12.2.1 Circuit with a Transformer in the Phasor Form
Consider a generic transformer circuit shown in Fig. 12.6a in the frequency domain. The
circuit is given in the phasor form assuming a harmonic signal source. A source circuit with
the phasor voltage VS and impedance ZS is connected to a load with impedance ZL via an
ideal transformer studied in the previous Section. Since any linear AC source network can
be represented in the form of its Thévenin equivalent, and any linear passive source can
be replaced by the equivalent impedance, Fig. 12.6a is a rather general interpretation of
the transformer setup with linear networks in Fig. 12.4. When written in the phasor form,
the ideal transformer model given by Eqs. (12.6) and (12.10) does not need a special
treatment. We simply replace the real-valued voltages and currents by phasors:
N2 N2
υ2 ¼ υ1 ) V2 ¼ V1 ð12:14aÞ
N1 N1
N1 N1
i2 ¼ i1 ) I2 ¼ I1 ð12:14bÞ
N2 N2

In power electronics, phasor voltage and phasor current in Fig. 12.6a are often expressed
in terms of rms values times the phasor (the complex exponent). This is in contrast to the
previous analysis where we have used the amplitude of a sinusoidal function times
the phasor. The circuit analysis remains the same, but the factor of 2 in the expressions
for the power disappears. We will mention this convention every time when required.

12.2.2 Referred (Or Reflected) Source Network in the Secondary Side

What voltage and impedance does the load see in Fig. 12.6a? In other words, what is the
Thévenin equivalent circuit of the source and the transformer combined? To answer this
question, we find Thévenin equivalent voltage VT of the one-port network with terminals
a and b in Fig. 12.6a as its open-circuit voltage. Using Eqs. (12.14a, b) and setting ZL
¼ 1 in Fig. 12.6a yields
N2 N2
I2 ¼ 0 ) I1 ¼ 0 ) VT  V2 ¼ V1 ¼ VS ð12:15Þ
N1 N1

The Thévenin equivalent impedance ZT is found by dividing the open-circuit voltage

by the short-circuit current ISC ¼ I2 . Setting ZL ¼ 0 in Fig. 12.6a, one finds the short-
circuit current
N1 N 1 VS
ZL ¼ 0 ) V2 ¼ V1 ¼ 0 ) ISC ¼ I2 ¼ I1 ¼ ð12:16Þ
N2 N 2 ZS

XII-602
Chapter 12 Section 12.2: Analysis of Ideal Transformer Circuits

a)
ZS I1 I2
a c

+ + +
VS + V1 V2 VL ZL
-
- - -
b N1:N2 d

b) 2

ĩNN ĩ Z
2

1
S I2
c

+ +
N2 +
V V2 VL ZL
N1 S -
- -
d

c)
ZS I1 I2
a c

+ + +
VS + V1 V2 VL ZL
-
- - -
b N1:N2 d

d) ZS I1
a

+
2
VS +
- V1 ĩNN ĩ Z
1

2
L

-
b

Fig. 12.6. (a) A source circuit and a load circuit connected to a transformer. (b) Equivalent circuit
of the source with the transformer when looking from the load. (c) The same as (a). (d) Equivalent
circuit of the load with the transformer when looking from the source.

In summary, the equivalent circuit of the source and the transformer combined is shown
in Fig. 12.6b. It is characterized by
 2
N2 N2
VT ¼ VS , ZT ¼ ZS ð12:17Þ
N1 N1

We conclude that, from the viewpoint of the load, the ideal transformer boosts the source
voltage by a factor of N2/N1 and multiplies the source impedance by a factor of (N2/N1)2.
The meaning of Eq. (12.17) is commonly expressed in the following form:

XII-603
Chapter 12 Electric Transformer and Coupled Inductors

1. The source is reflected to the secondary side by the inverse turns ratio.
2. The source impedance is reflected to the secondary by the square of the inverse turns
ratio.
An inspective student may notice a contradiction in Eq. (12.16): as long as
V2 ¼ V1 ¼ 0, the transformer apparently should not function at all. In fact, the equality
V2 ¼ V1 ¼ 0 is never exactly satisfied; a (infinitesimally small) voltage across the
transformer is assumed that supports its operation. Mathematically, one considers a finite
but small load impedance and then obtains Eq. (12.16) in the limit ZL ! 0.

12.2.3 Referred (Or Reflected) Load Impedance to the Primary Side

What is the equivalent impedance of the load and the transformer combined? With
reference to Fig. 12.6c, one has
 2  2
V1 N 1 V2 N1
ZT  ¼ ¼ ZL ð12:18Þ
I1 N 2 I2 N2

The equivalent circuit of the load and the transformer combined are shown in Fig. 12.6d.
The meaning of Eq. (12.18) is commonly expressed in the following form: the load
impedance is reflected to the primary side by the square of the turns ratio.

Exercise 12.6: Load impedance is 12 Ω. Find the equivalent impedance of the load
combined with a 1:2 step-up transformer to the primary side.
Answer: RT ¼ 3 Ω.

Exercise 12.7: Source voltage is given by υS ðt Þ ¼ 1 cos ωt ½V and the source impedance
is 12 Ω. Find the equivalent circuit of the source combined with a 1:2 step-up transformer
in the secondary side. Express your result both in frequency domain and in time domain.
Answer: RT ¼ 48 Ω, VT ¼ 2∠0
½V or υT ðtÞ ¼ 2 cos ωt ½V.

Both source and load reflections are of great practical value since they eliminate
the transformer from the circuit analysis and thereby simplify the overall circuit
analysis and design. The analysis with Norton equivalent circuits is developed in a
similar fashion.

XII-604
Chapter 12 Section 12.2: Analysis of Ideal Transformer Circuits

Exercise 12.8: To bring out an important application, let the load be a series RLC network
with resistance R, inductance L, and capacitance C. An equivalent impedance of the load
combined with a N1 : N2 step-up transformer at the primary is again a series RLC network.
Find the reflected (or “transformed”) values of the resistance, inductance, and capacitance.
 2  2  2
Answer: R ! NN 12 R, L ! NN 12 L, C ! NN 21 C.

12.2.4 Transformer as a Matching Circuit

Matching Real-Valued Impedances
Consider a transformer circuit with real-valued impedances shown in Fig. 12.7. A given
practical voltage source with resistance RS is to be connected to a given load with
resistance RL 6¼ RS . The principle of maximum power transfer states that there should
be RL ¼ RS for the maximum power transfer from the source to the load. Could we still
achieve the maximum available power from the source without changing the source and
the load? The answer is yes; such a procedure is known as impedance matching with a
transformer. A transformer used for this purpose is known as a matching transformer.
The idea is to use a transformer with the turns ratio:
pffiffiffiffiffiffiffiffiffiffiffiffiffi
N 1 : N 2 ¼ RS =RL ð12:19Þ

To prove this fact we consider a transformer with an arbitrary turns ratio in the circuit of
Fig. 12.7a. The load resistance is reflected to the primary side by the square of the turns
ratio. The resulting equivalent circuit is a voltage divider. It includes the voltage source
and two resistances in series. Instantaneous power delivered to the load is given by
( )  2
xRL N1
pðt Þ ¼ 2
υS ðt Þ, x ¼
2
ð12:20Þ
ðRS þ xRL Þ N2

The expression in curly brackets is maximized with regard to parameter x, which may
attain any positive values. Its maximum (and the maximum power) is achieved when

x ¼ RS =RL ð12:21Þ

or when the reflected load resistance is exactly RS, see Fig. 12.7b. The same circuit may
be equally well solved by reflecting the source to the primary side; the related homework
problems provide several examples.

XII-605
Chapter 12 Electric Transformer and Coupled Inductors

a) iS(t)
RS

vS(t)
+ RL
-

N1 : N2 = RS / RL

b) iS(t)
RS

vS(t) + RS
-

Fig. 12.7. (a) A matching transformer for real impedances. (b) Equivalent circuit of the load with
the transformer when looking from the source.

Example 12.2: A passive RFID tag is modeled by a load resistance of 9 Ω. The tag is
wirelessly powered; it is augmented with a small collecting antenna, which has a radiation
resistance of 1 Ω. Find the ratio of the average received powers from the antenna with and
without the matching transformer.
Solution: We use Fig. 12.7a to visualize the circuit. The antenna is the source, the tag is
the load. For the received power without the transformer, one has

1 RL jVS j2
PL ¼ ¼ 0:045jVS j2 ½ W ð12:22Þ
2 ðRS þ RL Þ2
where VS is a phasor voltage at the antenna terminals. With the matching transformer
(1:3 turns ratio), the result has the form (after reflecting the source to the secondary side):

1 RL =9jVS j2
PL ¼ ¼ 0:125jVS j2 ½W ð12:23Þ
2 ðRS þ RL =9Þ2
The ratio of the two powers is 2.78 or 10log10 ð2:78Þ ¼ 4:4 dB. This is an improvement
of the tag performance which results in a greater reading range.

Exercise 12.9: A 4-Ω speaker is connected to an AC voltage source with the voltage
amplitude of 10 V. The source’s impedance is 1 Ω.
A. Determine the average power delivered to the speaker.
B. Given a 1:2 matching transformer, determine power delivered to the speaker.
Answer: A. 8 W; B. 12.5 W.

XII-606
Chapter 12 Section 12.2: Analysis of Ideal Transformer Circuits

Matching Arbitrary Complex Impedances

Can a transformer match two arbitrary complex impedances? Unfortunately, it cannot.
The transformer operates as an impedance multiplier; it multiplies (or divides) by a real
number. On the other hand, the complex impedance match requires two complex
conjugate impedances. A transformer could often provide a “better” match (see the
summary to this chapter) but cannot perform impedance matching in full. Other circuit
elements (capacitance or inductance) may be necessary to complete the task.

12.2.5 Application Example: Electric Power Transfer via Transformers

Circuit with a Fixed Load Voltage
Figure 12.8a shows a circuit for transmitting electric power over a long transmission line
with the total resistance R and the total inductance L. The circuit in Fig. 12.8a is converted
to a phasor form first. We consider the phasors in terms of rms values. Two competing
schemes are studied: transmission without transformers (see Fig. 12.8b) and a transmis-
sion scheme with two 1:20 and 20:1 ideal transformers—see Fig. 12.8c. In order to
compare the performance of two circuits (with and without transformers), it is assumed
that the load phasor voltage VL (V rms) and the load phasor current IL (V rms) have the
same values in both cases. This guarantees us the same average power delivered to the
load. The power loss (ohmic loss) in the line resistance in Fig. 12.8b is

Ploss ¼ RjIL j2 ð12:24Þ

The power loss in the line resistance in Fig. 12.8c is decreased by a factor of 400:
1
Ploss ¼ RjIL j2 ð12:25Þ
400
since the line current is exactly 20 times less than in the first case. This result is
independent of the particular values of R and L. Simultaneously, the line voltages increase
by a factor of 20, but the load voltage still remains the same due to the step-down
transformer. Thus, using a pair of transformers allows us to choose an economically
optimum voltage for transmitting a given amount of power. The line sees a high voltage
of the secondary of the first transformer while the load essentially sees the source voltage.
Not only does the use of transformers greatly decrease the line loss, but it also potentially
allows us to use smaller source voltages (sending-end voltages). The required source
power also decreases.

XII-607
Chapter 12 Electric Transformer and Coupled Inductors

a) transmission line (TL)

iL(t)
vS(t)
+ source load RL
-
iL(t)

phasor model IL
b) R Z=jwL

+
VS1 + source load RL VL
-
-
c) step up transformer step down transformer
IL
1:20 R Z=jwL 20:1
+ + +
VS2 + 20VS2 20VL RL VL
-
- - -

Fig. 12.8. (a) Schematic of a transmission line. (b) Transmission line without a transformer
(phasor form). (c) Transmission line with step-up and step-down transformers (phasor form).

Circuit with a Fixed Source Voltage

Next, the performance of the circuit in Fig. 12.9 with a fixed source voltage is to be
analyzed, with and without 1:10 step-up transformer and 10:1 step-down transformer,
respectively. All phasors are given in terms of the rms values. We are interested in source
current IS and active source power:
 
PS ¼ Re VS  I*S ð12:26Þ

We are also interested in load parameters VL, IL and active load power:
 
PL ¼ Re VL  I*L ð12:27Þ

and, finally, in line power loss:

Ploss ¼ RjIline j2 ð12:28Þ

The circuit in Fig. 12.9a without the transformers is solved by finding the equivalent
impedance and then the source current. All circuit currents will coincide, IS ¼ Iline ¼ IL .
The circuit with the transformers is solved by reflecting the load impedance to the primary
side twice, see Fig. 12.9b and c, respectively.
The simplified circuit in Fig. 12.9c allows us to find the source current. Then, we return
to the circuit in Fig. 12.9b and find the line current following the ideal transformer model
Eq. (12.14b). Finally, we return to the circuit in Fig. 12.9a and restore the load current

XII-608
Chapter 12 Section 12.2: Analysis of Ideal Transformer Circuits

Fig. 12.9. Analysis of a particular power-transfer circuit.

from the line current. Table 12.1 summarizes the corresponding numerical values. It
illustrates the reduction in the line loss, which is again approximately proportional to the
square of the turns ratio.
Table 12.1. Circuit parameters (V rms, A rms, and W) for the power-transfer circuit in Fig. 12.9.
Circuit VS IS PS Iline Ploss VL IL PL
No tr. 480∠0
90∠  38
34,312 90∠  38
1634 452∠  1
90∠  38
32,678
W tr. 480∠0
96∠  37
36,837 9:6∠  37
18.4 480∠0
96∠  37
36,818

Historical: The use of AC in favor of DC is based on the efficiency of distributing

electrical power. In DC systems, heavy conductors must be used to deliver electrical
power over long distances. However, in AC systems, transformers boost voltages to high
levels so that electricity can be distributed across country on relatively thin wires (high
voltage, low current), and then converted back to low voltage, high current where needed.
This is not possible with DC.
In the late 1880s, USA experienced what many citizens termed the “Battle of the
Currents” where George Westinghouse and Nikola Tesla prevailed with their AC power
distribution system over world-famous Thomas Edison who advocated DC power and tried
to discredit AC by saying it was more dangerous. To sway the public opinion, Edison went to
the extreme of secretly acquiring an AC generator and designing the first electric chair with
it. The first public execution of a prisoner in this electric chair was William Kemmler in 1890
using 2000V AC. The electrocution required two attempts. This incident supposedly
prompted Westinghouse to state that an axe would have been a more humane execution.

XII-609
Chapter 12 Electric Transformer and Coupled Inductors

Section 12.3 Some Useful Transformers

12.3.1 Autotransformer
A transformer in which a part of the winding is common to both primary and secondary
circuits—see Fig. 12.10—is known as an autotransformer. This common part may be
controlled by a sliding contact as shown in the same figure.

a) i1(t) b)
sliding contact i1
F
+
N1 i2(t)
+ i2
N1
v1(t)
+ = v1(t)
N2 +
N2 v2(t) v2(t)
-
- - -
B

Fig. 12.10. Step-down autotransformer along with the corresponding circuit symbol. Circuit
symbols for the autotransformer may vary.

An autotransformer is typically a step-down transformer. Autotransformers are used

widely and come in many sizes. The autotransformer is also known as a Variac (variable
AC transformer). Such autotransformers are used in the laboratory to obtain a variable
output voltage by means of a movable contact in Fig. 12.10 that changes N2 and
consequently the turns ratio N1 : N2. An autotransformer will again be analyzed using
the ideal transformer model. Primary winding N1 and secondary winding N2 in Fig. 12.10
share the same flux Φ(t). Both induced voltages υ1(t) and υ2(t) in Fig. 12.10 obey
Faraday’s law given by Eqs. (12.2) and (12.5), respectively. One has

dΦ dΦðt Þ
υ1 ðt Þ ¼ ðN 1 þ N 2 Þ , υ 2 ðt Þ ¼ N 2 ð12:29Þ
dt dt
Therefore,
N2
υ2 ¼ υ1 ð12:30Þ
N1 þ N2

Using Ampere’s law for the ideal magnetic core Eq. (12.8c), we obtain an expression that
is very similar to Eq. (12.10) of Section 12.1,
N1 þ N2
ðN 1 þ N 2 Þi1  N 2 i2 ¼ 0 ) i2 ¼ i1 ð12:31Þ
N2

XII-610
Chapter 12 Section 12.3: Some Useful Transformers

Thus, the model of the ideal autotransformer is indeed indistinguishable from the ideal
transformer model. Advantages of the autotransformer are generally lower losses and
variable output voltage. However, there is now a direct electric connection between the
primary and secondary sides, which potentially enables direct current flow.

Exercise 12.9: A voltage source connected to the primary winding of an autotransformer

has the form υS ¼ 10 cos ωt þ 1 ½V. Ohmic resistance of the primary winding is 1 Ω.
Magnetic losses are ignored. Turns ratio established by a sliding contact is given by
N 1 þ N 2 : N 2 ¼ 5. Determine the voltage in the secondary.
Answer: Using the superposition theorem for DC and AC voltages, one has
υS ¼ 2 cos ωt þ 0:2 ½V.

12.3.2 Multiwinding Transformer

A multiwinding transformer shown in Fig. 12.11 is one in which two or more secondary
windings are placed on the same core.

a) i2(t) b)
i1(t) F + multiple secondary windings

N2 v2(t) N1
+
N1
= N1
v1(t)
- i3(t)
N3
- +
N3
B v3(t)
-

Fig. 12.11. A multiwinding transformer along with the corresponding circuit symbol. Circuit
symbols for the multiwinding transformer may vary.

For example, a three-winding transformer shown in Fig. 12.11 has a primary winding
N1 and two secondary windings, N2 and N3, respectively. It is useful when lossless power
division is required. The multiwinding transformer is also analyzed using the ideal
transformer model. All induced voltages υ1(t), υ2(t), and υ3(t) in Fig. 12.11 obey Fara-
day’s law of induction. Therefore,
υ1 υ2 υ3
¼ ¼ ð12:32Þ
N1 N2 N3

Using Ampere’s law for the ideal magnetic core Eq. (12.8c), we obtain

N 1 i1  N 2 i2  N 3 i3 ¼ 0 ð12:33Þ

XII-611
Chapter 12 Electric Transformer and Coupled Inductors

Example 12.3: In the circuit shown in Fig. 12.11, N 1 ¼ N 2 ¼ N 3 . How is the instanta-
neous power partitioned between the two secondary windings?
Solution: According to Eqs. (12.32) and (12.33),
υ1 ¼ υ2 ¼ υ3 , i1 ¼ i2 þ i3 ð12:34Þ
If two secondary windings are terminated into the same load resistances, then
i2 ¼ i3 ¼ 0:5i1 . Therefore, the input power is divided equally. However, if one of the
loads is an open circuit, all the input power is transferred to another load.

A multiwinding transformer may be used, for example, in a telephone hybrid circuit

which is designed to convert a two-wire interface to a four-wire interface. A telephone
hybrid is the circuit which separates the transmitted and received audio which are initially
sent both at the same wire pair in two-wire normal telephone interface.

12.3.3 Center-Tapped Transformer: Single-Ended to Differential

Transformation
Figure 12.12 shows a particular, yet widely used modification of the multiwinding
transformer in Fig. 12.11. Two equal secondary windings, with N2/2 turns each, are
connected to each other and then to ground. The center tap of the secondary coil may be
grounded. The output voltages of the two secondary windings versus the dotted terminals
are still equal to each other and equal to voltage υ2. However, since the center tap is
grounded, the absolute voltages at the output to the transformer are
N2
þυ2 ðt Þ, 0 V, and  υ2 ðt Þ where υ2 ¼ υ1 ð12:35aÞ
2N 1
versus ground. The output voltage is thus identical to a dual-polarity AC voltage source.
More generally, this design is intended to convert a single-ended (two-wire) voltage signal
υ1, i.e., a signal that is initially referenced to ground, to a balanced differential (three-wire)
voltage signal with equal voltages with respect to ground but of opposite polarity.
a) i2(t) b)

i1(t)
F +
N1:N2
v2(t)
+
v1(t) N1 N2 - =
CT
+ (center tap)
- v2(t) 0V

0V B -
i3(t)

Fig. 12.12. Center-tapped transformer intended for single-ended to differential signal transforma-
tion (differential transformer) and the corresponding circuit symbol.
XII-612
Chapter 12 Section 12.3: Some Useful Transformers

The differential signal has significant advantages; it can be used in different applica-
tions including power electronics, instrumentation, communication circuits, and high-
speed digital circuits. If, and only if, the center-tapped transformer is connected to the
two identical loads, the center tap carries zero current and with reference to Fig. 12.12,
N2
i2 ðt Þ ¼ i3 ðt Þ where i1 ¼ i2 ð12:35bÞ
N1

The three-wire model described by Eq. (12.35a) is often called a three-wire single-phase
system. Note that all of the AC power that is supplied to your house by a residential AC
distribution network is a 120/240 V rms three-wire single-phase system, see Fig. 11.13.

Example 12.4: Common home appliances use a single-phase 120-V rms AC line
(a refrigerator, TV, microwave, etc.) and a 240-V rms three-wire differential AC line
(electric range, dryer, etc.). A house located in a remote area is powered from a local
micro hydropower generator located a couple of miles away. Suggest a way to power all
appliances when the input power is a not a residential AC distribution network, but a
single-phase two-wire 240 V rms AC line from the generator.
Solution: A solution is shown in Fig. 12.13. The center-tapped transformer is used to
accomplish the task. All loads are to be connected in parallel.

neutral (center tap)

N1:N2
+

120 V rms
+ TV
240V rms 240 V rms
- 120 V rms 120 V rms
Stove, Dryer
MW Refr. -

Fig. 12.13. Powering home appliances in remote areas using a single-phase two-wire 240 V rms
AC line as an input. Note that the dot locations are less important for this case.

Example 12.5: A dipole antenna with two wings is to be fed as follows: one wing is fed
with a voltage signal þυðtÞ versus ground plane and another wing is fed with a voltage
signal υ1 ðtÞ versus ground plane (180
phase shift). At the same time, the input from a
transmitter is a single-phase coaxial line. Suggest a way of connecting the antenna to a
transmitter.
Solution: A solution is shown in Fig. 12.14. The center-tapped transformer from Fig. 12.12
is again used to accomplish the task. From the viewpoint of an RF circuit design, the
transformer in Fig. 12.14 is sometimes called a balun transformer; the word balun is an
acronym for balanced-to-unbalanced converter. These balun structures are employed in such
common RF and microwave components as mixers, antenna-feed networks, and frequency
multipliers.

XII-613
Chapter 12 Electric Transformer and Coupled Inductors

Electrically connected current

ground plane

Grounded case

Port 1
Feeding
g port

Port 2

Fig. 12.14. A balun (center-tapped) transformer attached to a dipole antenna. The metal case of the
transformer is grounded.
Note that the transformer in Fig. 12.12 is simultaneously a power divider, namely, the
so-called 180
power divider or 180
power splitter. It is amazing to see how many
different applications rely on the “good old” transformer.

12.3.4 Current Transformer

There is a device that has been around for a long time and which allows the engineer to
monitor the electrical power system. It is called the current transformer and is shown in
Fig. 12.15. The current transformer is a main measuring tool to determine current flow in
a power system. The corresponding hardware device is known as a clamp on ammeter.

a) b)

i1
F F power conductor

i2 N2

Fig. 12.15. A current transformer and the corresponding circuit symbol. Circuit symbols for the
current transformer may vary.

Consider a straight conductor in Fig. 12.15 that carries current i1 to be measured. This
conductor effectively forms one turn of the primary coil. The secondary coil has N2 turns.
We again assume the ideal transformer model and apply Ampere’s law for the ideal
magnetic core Eq. (12.8c). The current through the secondary coil then follows the
equality (note current directions in Fig. 12.15)

XII-614
Chapter 12 Section 12.3: Some Useful Transformers

1
i1  N 2 i2 ¼ 0 ) i2 ¼ i1 ð12:36Þ
N2

which is a particular case of the ideal transformer equation with N 1 ¼ 1. Thus, by

measuring current i2 or the associated voltage, the unknown current i1 can be established.
Note that Fig. 12.15 and Eq. (12.36) only demonstrate the very basic concept; the
practical current transformer design is significantly more complicated. Another type of
instrumentation transformers—potential transformers—is used for accurate AC voltage
measurements.

XII-615
Chapter 12 Electric Transformer and Coupled Inductors

Section 12.4 Real-Transformer Model

12.4.1 Model of a Nonideal Low-Frequency Transformer
Figure 12.16 shows a linear circuit model of a practical low-frequency (60 or 50 Hz)
two-winding transformer. This circuit is also known as the Steinmetz model and its
parameters as Steinmetz parameters. The complete model will be explained in several
steps:
1. Consider the open-circuited transformer first. The primary winding is characterized
by a large but finite magnetizing inductance, Lm, which was defined in Eq. (12.3).
This is the standard inductance expression for a long solenoid with the magnetic
core. It is shown in Fig. 12.16a.
2. A small amount of magnetic flux is still situated outside the core so that Eq. (12.3)
needs to be refined. The total inductance as in Fig. 12.16 of the primary winding is
therefore somewhat larger:

L1 ¼ Lm þ Ll1 ð12:36Þ

where a small addition Ll1 is called the leakage inductance, see Fig. 12.16a.
3. A practical primary winding has a certain ohmic resistance, R1, which is placed in
series with L1. Simultaneously, the core loss (hysteresis and eddy current loss) in the
magnetic material consumes some extra current. It is modeled by an equivalent
resistance Rc, which is placed in parallel with Lm. Rc is often called the core loss
resistance. The resulting equivalent circuit in Fig. 12.16b is also an equivalent
circuit of a nonideal inductor with the magnetic core.
4. Finally, the secondary winding is added and a load is connected to the transformer.
This results in the complete equivalent circuit model of Fig. 12.16c. Two new circuit
parameters Ll2, and R2 are the leakage resistance and the ohmic resistance, respec-
tively, of the secondary winding.
In general, all model parameters, and especially Lm and Rc, are frequency dependent.

12.4.2 Model Parameters and Their Extraction

Table 12.2 lists typical equivalent circuit values for three distinct power transformers
of different power ratings and compares them with the ideal transformer model.
Not the inductances themselves but the rather reactance values are given at angular
frequency ω ¼ 2πf where f ¼ 50 or 60 Hz.

XII-616
Chapter 12 Section 12.4: Real-Transformer Model

a) cross-section A

lossless case
i1 i1
Ll1

+
N1
+
v1
v1 Lm
- -
B

b) cross-section A

lossy case
i1
i1 R1 Ll1

+ +
v1 N1 v1
Rc Lm
- -
B

c) cross-section A
i2
full model
i1 i2
i1 + R1 Ll1 Ll2 R2
v2
i
+ +
+
N1 N2 v1 v2
v1
Rc Lm
- - -
B
- ideal transformer

Fig. 12.16. Circuit model of a low-frequency transformer—the Steinmetz model.

It follows from Table 12.2 that the real-transformer data is reasonably well approxi-
mated by the ideal model. In particular, we may observe that

R1, 2  Rc , X l1, 2  X m ð12:37Þ

in all three cases considered. In a well-designed transformer, the following equalities

should be satisfied:

XII-617
Chapter 12 Electric Transformer and Coupled Inductors

Table. 12.2. Equivalent circuit values for three different practical power transformers as compared
to the ideal transformer model.
#1—2 kVA #2—10 kVA #3—100 kVA
Ideal 230:115 V 2300:230 V 11,000:2200 V
Element nameplate model 50 Hz 60 Hz 60 Hz
Magnetizing reactance X m ¼ ωLm , Ω 1 1437.5 69,400 57,300
Core loss resistance, Rc, Ω 1 294.2 75,600 124,000
Primary leakage reactance X l1 ¼ ωLl1 , Ω 0 0.430 12 31.2
Secondary leakage reactance, 0 0.006 0.12 1.25
X l2 ¼ ωLl2 , Ω
Primary ohmic resistance R1, Ω 0 0.428 5.80 6.1
Secondary ohmic resistance R2, Ω 0 0.123 0.0605 0.29

 2  2
N1 0 N1 0
R1  R2 ¼ R2 , X l1  X l2 ¼ X 2 ; ð12:38Þ
N2 N2

along with Eq. (12.37). Here, the prime denotes circuit parameters of the secondary
winding referred to the primary side.

Exercise 12.10: For three transformers from Table 12.2, find the corresponding turns ratio
and prove whether or not equalities given by Eq. (12.38) are approximately satisfied.
Answer:
- Model #1—turns ratio 2:1, first Eq. (12.38) is satisfied, but the second is not.
- Model #2—turns ratio 10:1, both Eq. (12.38) are satisfied.
- Model #3—turns ratio 5:1, both Eq. (12.38) are satisfied.

If the complete geometry and design characteristics of a transformer are available, all
parameters of the equivalent circuit model can be calculated theoretically.
Most complicated is the calculation of leakage resistances. At the same time, these
parameters are directly and more easily found using measurements. Open-circuit trans-
former test and short-circuit transformer test (and often the DC bridge test) are performed
to find all parameters of interest.

12.4.3 Analysis of Nonideal Transformer Model

When a harmonic voltage source is applied to a nonideal transformer and a linear load is
connected to it, the corresponding equivalent circuit follows Fig. 12.17a. Although only a
resistive load is indicated, the same circuit applies to an arbitrary RLC linear load.

XII-618
Chapter 12 Section 12.4: Real-Transformer Model

The corresponding circuit in frequency domain is shown in Fig. 12.17b. Despite its
apparently complicated nature, the circuit analysis is quite straightforward.

a) transformer model
i1(t)=iS(t) i2(t)
R1 Ll1 a=N1/N2 Ll2 R2

vS(t) + RL
- RC Lm

b) impedance/phasor form
-1
I1=IS a I2 I2
R1 jXl1 a=N1/N2 jXl2 R2

+ + +
VS + aE2 E2 VL ZL
- RC jXm - - -

Fig. 12.17. Equivalent circuits for real transformers with harmonic source and linear load.

Terminology
1. In power electronics, phasor voltage and phasor current in Fig. 12.17b are expressed
in terms of rms values times the phasor (the complex exponent). This is in contrast
to the previously analyzed circuits where we have mostly used the amplitude of the
sinusoidal function times the phasor. The analysis remains the same, but the factor
of 2 in the expressions for the power disappears. For example, the input (source) and
output (load) active powers are given by
   
PL ¼ Re VL  I*L , PS ¼ Re VS  I*S ð12:39Þ

2. The rated load is that which has the rms voltage corresponding to the transformer
rating. For example, the rated load for transformer #1 in Table 12.2 has

VL ¼ 115∠0
½V rms ð12:40Þ

It is customary to choose the phase of the load voltage equal to zero.

3. An impedance load has a power factor PF, lagging or leading. It means that the
rated-load current is defined by
S
IL ¼ ∠ arccosðPF Þ ½A rms ð12:41Þ
VL

where S is transformer power rating (the apparent load power). For example,
for transformer #1 in Table 12.2, S ¼ 2 kVA. In Eq. (12.41), the minus sign

XII-619
Chapter 12 Electric Transformer and Coupled Inductors

corresponds to the lagging power factor and the plus sign to the leading power
factor, see Section 11.1.

Example 12.6: In the circuit in Fig. 12.17b, the load voltage VL and the load current IL are
both known. Determine the source phasor voltage VS and the source phasor current IS in
the general form. Use the notation a ¼ N 1 =N 2 .
Solution: Applying KVL to the circuit connected to the secondary winding of the ideal
transformer yields

E2 ¼ VL þ ðR2 þ jX l2 ÞIL ð12:42Þ

Next, we apply KCL to the circuit connected to the primary winding of the ideal
transformer. This gives (I2 ¼ IL )
 
IL 1 1
IS ¼ I1 ¼ þ þ aE2 ð12:43Þ
a RC jX m

Finally, KVL for the primary circuit branch leads to

VS ¼ aE2 þ ðR1 þ jX l1 ÞIS ð12:44Þ

Example 12.7: Given the rated load with the power factor of 1, determine the source
phasor voltage VS and the source phasor current IS for transformer #1 in Table. 12.2.
Solution: For the rated load, VL ¼ 115∠0
½V rms. The load phasor current is found
from Eq. (12.41), IL ¼ 17:39∠0
½A rms. Further, we use the solution of the previous
example with a ¼ 2, plug in the model parameters from Table 12.2, and obtain

VS ¼ 238:5∠ þ 1:0
, IS ¼ 9:49∠  1:0
ð12:45Þ

Emphasize that the source voltage of 238.5 V rms is higher than the rated or nameplate
value of 230 V rms. The additional potential of 8.5 V is needed to overcome the finite
impedance of the non-ideal transformer. Indeed, for the ideal-transformer model, the
agreement would be perfect.

Exercise 12.11: Repeat the previous example for transformer #2 and transformer #3 from
Table. 12.2.
Answer:
- Model #2—VS ¼ 2354∠ þ 2:5
, IS ¼ 4:38∠  0:4
.
- Model #3—VS ¼ 11142∠ þ 2:9
, IS ¼ 9:19∠  1:2
.

XII-620
Chapter 12 Section 12.4: Real-Transformer Model

Exercise 12.12: Repeat the previous example for every transformer model from Table 12.2
given the rated load with the power factor of 0.9 lagging (power angle of 25.8
). Does the
deviation from the rated input voltage increase?
Answer:
- Model #1—VS ¼ 239:3∠ þ 0:1
, IS ¼ 9:49∠  24:7
.
- Model #2—VS ¼ 2394∠ þ 1:7
, IS ¼ 4:39∠  26:1
.
- Model #3—VS ¼ 11373∠ þ 2:3
, IS ¼ 9:26∠  26:7
.
In all three cases, the deviation from the rated input voltage slightly increases when
compared to the case of the rated resistive load.

12.4.4 Voltage Regulation and Transformer Efficiency

Voltage Regulation
Since the transformer model is nonideal, the voltage delivered to the load does depend on
the load current. This means, for example, that the voltage delivered to lights will
somewhat decrease when another parallel high-current load (a dryer) is turned on. This
undesired effect is characterized by the regulation of a transformer (or percentage
regulation):

jVno-load j  jVL j
Regulation ¼
100 % ð12:46Þ
jVL j

Here, |Vno‐load| is the rms voltage for the open-circuited load, and |VL| is the rms voltage
of the rated load. The regulation is determined by calculating VS for the rated load and
then using the same VS to find the voltage across the open-circuited load. With reference
to Fig. 12.17b, a simple yet accurate approximation for the no-load voltage is given by
VS
Vno-load  ð12:47Þ
a
To prove this fact we observe that I2 ¼ 0 for the open-circuited load. Furthermore,
almost all of VS is applied to the primary.

Exercise 12.13: Given the rated load with the power factor of 1.0, determine the percent-
age regulation for three transformer models in Table. 12.2.
Answer:
- Model #1—Regulation ¼ 3:8 %.
- Model #2—Regulation ¼ 2:4 %.
- Model #3—Regulation ¼ 1:3 %.

XII-621
Chapter 12 Electric Transformer and Coupled Inductors

Exercise 12.14: Repeat the previous exercise for the power factor of 0.9 lagging.
Answer:
- Model #1—Regulation ¼ 4:1 %.
- Model #2—Regulation ¼ 4:1 %.
- Model #3—Regulation ¼ 3:4 %.

Transformer Efficiency
The transformer equivalent circuit in Fig. 12.17 includes parasitic resistances and thus
implies power loss in the transformer itself. This power loss is characterized by trans-
former power efficiency (percentage efficiency), η, which is defined by
PL
η¼
100 % ð12:48Þ
PS

where the input active power PS and the output active power PL are given by Eq. (12.39).
In well-designed power transformers, the efficiency may approach 99 %.

Exercise 12.15: Given the rated load with the power factor of 1.0, determine the percent-
age efficiency for three transformer models in Table. 12.2.
Answer:
- Model #1—η ¼ 88:4 %.
- Model #2—η ¼ 97:1 %.
- Model #3—η ¼ 97:9 %.

Exercise 12.16: Repeat the previous exercise for the power factor of 0.9 lagging.
Answer:
- Model #1—η ¼ 87:3 %.
- Model #2—η ¼ 96:8 %.
- Model #3—η ¼ 97:7 %.

12.4.5 About High-Frequency Transformer Model

At medium (in the audio range) and higher frequencies, the effect of winding capaci-
tances and other capacitances shown in Fig. 12.18 becomes important. When frequency
increases, the transformer, as an RLC circuit, may ultimately exhibit a resonance that will
cause the effective parameter a ¼ N 1 =N 2 to change. A number of capacitances may be
added to the equivalent circuit model to model this effect.

XII-622
Chapter 12 Section 12.4: Real-Transformer Model

Winding

Tank
Core wall

Fig. 12.18. Internal winding capacitances of a realistic high-frequency transformer.

XII-623
Chapter 12 Electric Transformer and Coupled Inductors

Section 12.5 Model of Coupled Inductors

12.5.1 Model of Two Coupled Inductors
Definitions
Now it is time to examine a model of two coupled inductors L1 and L2. This model is
perhaps more general than the transformer model. What does coupled mean? For
simplicity, we start with a “broken” transformer model with a large air gap shown in
Fig. 12.19. Primary and secondary windings have currents i1 and i2, respectively.
However, in contrast to the previous study, both currents will now enter the dotted
terminals. This means that the primary winding, inductor #1, as well as the secondary
winding, inductor #2, will both satisfy the passive υ-i reference configuration for the
inductance. Therefore, when the air gap is very large, we will have the familiar inductor
laws for both voltages υ1 and υ2 in Fig. 12.19, respectively:
di1 di2
υ1 ¼ þL1 , υ2 ¼ þL2 ð12:49Þ
dt dt

i2

i1 i1 M i2
1 +

+ + +
v1 L1 L2 v2
= L1 L2
v1 v2

- - -
N1

2
-
N2

Fig. 12.19. Two coupled inductors. Note that we no longer use the transformer symbol. Also note
the passive reference configuration for each inductance.

When the air gap is not very large, inductor #1 will be subject to a certain time-varying
magnetic flux Φ2(t) created by current i2(t) in the second inductor. According to Faraday’s
law, voltage υ1(t) will therefore additionally depend on current i2(t), more precisely, on its
time derivative. Quite similarly, inductor #2 will be subject to a certain time-varying
magnetic flux Φ1(t) created by current i1(t) in the first inductor. According to Faraday’s
law, voltage υ1(t) will also depend on time derivative of current i1(t). As a result, instead
of the inductance laws given by Eq. (12.49) for two uncoupled (literally noninteracting)
inductors, we have the model

XII-624
Chapter 12 Section 12.5: Model of Coupled Inductors

di1 di2
υ1 ¼ þL1 þM
dt dt ð12:50Þ
di1 di2
υ2 ¼ þM þ L2
dt dt
of two coupled inductors. Here the coefficient M > 0 is the mutual inductance of two
coils, which is also measured in henries. You can see from the second expression in
Eqs. (12.50) that the mutual inductance determines the voltage induced in inductor #2 due
to changes of the electric current in inductor #1. Alternatively, from the first expression in
Eqs. (12.50), the same mutual inductance determines the voltage induced in inductor #1
due to changes of the electric current in inductor #2. The signs in Eq. (12.50) are
important. They follow a few rules:
1. Eqs. (12.50) corresponds to the dot convention and voltage polarities/current
directions shown in Fig. 12.19.
2. If one of the current reference directions, say the direction of i2, is selected
oppositely, we will have to use the minus sign in the respective terms in
Eqs. (12.50).
3. If one of the voltage polarities changes, we have will have to use the minus sign
where required in Eqs. (12.50).
Figure 12.19 also shows the circuit symbol for two coupled inductors. This symbol
does not include the magnetic core; event if it is present in reality. Also, the mutual
inductance is shown by an arrow.

12.5.2 Analysis of Circuits with Coupled Inductors

Solving circuits with the coupled inductors requires care. Consider, for example, a
simple circuit shown in Fig. 12.20 in frequency domain. We cannot apply the
impedance relations following from Eq. (12.49). For two coupled inductors,
V1 6¼ þjωL1 I1 and V2 6¼ jωL2 I2 . Instead, we should convert Eqs. (12.50) to the
phasor form first and obtain

V1 ¼ jωL1 I1 þ jωM I2
ð12:51Þ
V2 ¼ jωM I1 þ jωL2 I2

These are two equations for four unknowns V1, I1, V2, I2. The two remaining equations
are KVL for the left part of the circuit and KVL for the right part of the circuit, i.e.,

V1 ¼ þVS and V2 ¼ RL I2 ð12:52Þ

XII-625
Chapter 12 Electric Transformer and Coupled Inductors

j2
I1 I2

+ +
VS + V1 j5 j4 V2 RL= 10
-
- -

Fig. 12.20. Circuit with two coupled inductors in the phasor form.

Example 12.8: Solve the circuit with two coupled inductors in Fig. 12.20—determine the
phasor load current given that VS ¼ 10∠0
½V.
Solution: First, we substitute into Eqs. (12.51) the expressions for the reactances
jωL1 ¼ j5 Ω, jωL2 ¼ j4 Ω, jω M ¼ j2 Ω and obtain

V1 ¼ j5I1 þ j2I2 ½V

ð12:53Þ
V2 ¼ j2I1 þ j4I2 ½V

Since, by KVL, V2 ¼ RL I2 ¼ 10I2 , from the second Eq. (12.53), one has
I1 ¼ ð2 þ j5ÞI2 ½A ð12:54Þ

Since by KVL, V1 ¼ VS ¼ 10 ½V, from the first Eq. (12.53), one has

10 ¼ ð25  j8ÞI2 ) I2 ¼ 0:381∠162
½A ð12:55Þ

The phasor source current I1 is given by I1 ¼ 2:052∠  86
½A.

A general solution of the four Eqs. (12.51) and (12.52) is of practical interest.
Following the method of Example 12.8, it is obtained in the form:
 
VS L2 RL
I2 ¼ ½A, I1 ¼  þ I2 ½A
M 2  L1 L2 L1 RL M jωM ð12:56Þ
jω 
M M

Exercise 12.17: Using Eqs. (12.56) solve the previous example

pffiffiffiffiffiffiffiffiffigiven
ffi that jωL1 ¼ j4 Ω,
jωL2 ¼ j1 Ω, and the mutual inductance is given by M ¼ L1 L2.
Answer: I1 ¼ 2:513∠  84
½A, I2 ¼ 0:500∠180
½A,
V1 ¼ 10:000∠0
½V, V2 ¼ 5:000∠0
½V:

XII-626
Chapter 12 Section 12.5: Model of Coupled Inductors

Solution for N Coupled Inductors

The solution method for two coupled inductors may be straightforwardly extended to the
case of N coupled inductors. In this case, Eqs. (12.51) will involve a matrix of self- and
mutual inductances (the inductance matrix L) on the size N
N. KVL applied to every
individual inductor circuit gives N remaining algebraic equations. The coupled system
of 2N algebraic equations is then solved for unknowns I1, . . ., IN and V1, . . ., VN. The
systems of coupled inductors are used in many applications, mostly for (multiple output)
switched-mode power conversion. They could also be used for sensor and other purposes
in bioelectromagnetics and other disciplines—see Fig. 12.21.

Fig. 12.21. An array of coupled inductors (small coils) located on top of a human-head phantom.
This hypothetical setup was tested for applications related to brain stimulation.

Finding Mutual Inductance(s)

Despite the dynamic nature of Eqs. (12.50), the mutual inductance defined previously by
Eq. (6.16) is inherently a static quantity. It may be computed from the corresponding 3D
magnetostatic analysis (often quite complicated). We will briefly review this question at
the end of this section. Right now, however, we will establish the highest possible value
of M, which is achieved for the ideal transformer.

Conversion to T-Network
It may be convenient to replace a circuit with two coupled inductors by a circuit without
magnetic coupling. This can be done using either a T-network or a Π-network of three
inductances—see Section 3.3 of Chapter 3. Figure 12.22 illustrates a conversion to the
T-network for the circuit from Fig. 12.20. For leftmost inductance La, rightmost induc-
tance Lb, and shunt inductance Lc of the T-network, one has

La ¼ L1  M , Lb ¼ L2  M , Lc ¼ M ð12:57Þ

The proof is based on establishing υ–i relationships for both two-port networks in
Fig. 12.22 and demonstrating their identity.

XII-627
Chapter 12 Electric Transformer and Coupled Inductors

I1 j2 I2

+ +
VS + V1 j5 j4 V2 RL
-
- -

I1
= I2
j3 j2

+ +
VS + V1 j2 V2 RL
-
- -

Fig. 12.22. Conversion of two coupled inductors to the T-network of three inductances.

12.5.3 Coupling Coefficient

Energy in a Coupled Circuit
Multiplying Eqs. (12.50) by i1 and i2, respectively, and combining them together, one can
find expressions for the instantaneous power p(t) and instantaneous energy E(t) stored in
the system of two coupled inductors.
9
di1 di2 >

υ 1 i 1 ¼ L1 i 1 þ M i1 = d 1 2 1 2
dt dt ) p ð t Þ ¼ υ i
1 1 þ υ i
2 2 ¼ L 1 i þ L 2 i þ M i i
1 2 )
di1 di2 >
; dt 2 1 2 2
υ 2 i 2 ¼ M i2 þ L2 i 2
dt dt ð12:58Þ
ðt  
0 0 1 1
Eðt Þ ¼ p t dt ¼ L1 i21 þ L2 i22 þ M i1 i2  0
2 2
0

To complete the square in the expression for the energy, we both add and subtract the term
pffiffiffiffiffiffiffiffiffiffi
L1 L2 i1 i2 , which gives
1pffiffiffiffiffi pffiffiffiffiffi 2  pffiffiffiffiffiffiffiffiffiffi
E ðt Þ ¼ L1 i1 þ L2 i2 þ M  L1 L2 i1 i2  0 ð12:59Þ
2

Mutual Inductance of the Ideal Transformer

For the ideal transformer, the energy stored in the transformer is zero, for any values of i1,
i2. This fact has been proved in Section 12.1. Given arbitrary i1, i2, the condition E ðt Þ ¼ 0
in Eq. (12.58) is satisfied if and only if,
pffiffiffiffiffiffiffiffiffiffi
M ¼ L1 L2 ½H ð12:60Þ

XII-628
Chapter 12 Section 12.5: Model of Coupled Inductors

and
pffiffiffiffiffi
i1 L2 N2
¼  pffiffiffiffiffi ¼  ð12:61Þ
i2 L1 N1

Equation (12.60) is the definition of the mutual inductance for the ideal transformer.
It gives the largest possible value of the mutual inductance. Equation (12.61) uses
Eq. (6.23) for the coil inductance. We note that this is indeed the ideal-transformer law
(12.10), but with the minus sign due to the opposite current direction in Fig. 12.19.

Example 12.9: A system of two coupled inductors is characterized by Eq. (12.60). Is this
system equivalent to an ideal transformer?
Solution: We check the voltage relation first. From Eq. (12.50),
 
pffiffiffiffiffi pffiffiffiffiffi di1 pffiffiffiffiffi di2
L2 L1 þ L2 rffiffiffiffiffi
υ2 dt dt L2 N 2
¼  ¼ ¼ ð12:62Þ
υ1 pffiffiffiffiffi pffiffiffiffiffi di1 pffiffiffiffiffi di2 L1 N 1
L1 L1 þ L2
dt dt

Thus, the ideal-transformer law for voltages is satisfied. However, the ideal-transformer law
for currents is not satisfied. An example is given by Eqs. (12.56). In order to obtain the ideal-
transformer model, we must additionally assume that

L1 ! 1, L2 ! 1, M !1 ð12:63Þ

Otherwise, the coupled-inductor model will additionally take into account the magnetizing
inductance, which is already a part of the nonideal transformer model studied in the
previous section.

Coupling Coefficient
The mutual inductance in the general case is expressed in terms of the two coils’ self-
inductances L1 and L2, in the form
pffiffiffiffiffiffiffiffiffiffi
M ¼ k L1 L2 ½H  ð12:64Þ

where k is a so-called coupling coefficient, 0 < k  1. The coupling coefficient deter-

mines the amount of total magnetic flux linkage from the first coil shared by the second
coil and vice versa. For the ideal transformer, the flux is entirely concentrated within the
common magnetic core so that k ¼ 1. Figure 12.23 illustrates different values of the
coupling coefficient.

XII-629
Chapter 12 Electric Transformer and Coupled Inductors

a) k~1 b) k<1 c) k<<1

Fig. 12.23. Different configurations and coupling coefficient values for two coupled inductors.

12.5.4 Application Example: Wireless Inductive Power Transfer

Next, we shall consider two arbitrary coils in spatial proximity to each other. Equations
(12.50) will always hold for any such coil configuration and for any distance, d, between
the coils. However, the coupling coefficient, k, in Eq. (12.64) has to be determined
separately for every particular geometry. The coupling coefficient may now be much
less than one. Nonetheless, for any non-zero k, the variable current in coil #1 will create a
variable voltage in coil #2. In other words, we could create wireless power or data
transfer between two, or even more than two, inductors. This is due to the very distinct
property of an inductor’s magnetic near field. We are going to briefly discuss two
examples of wireless data and/or power transfer with two coupled inductors. One
approach is radio-frequency identification (RFID) which employs the use of a patch,
known as an RFID tag, that is attached to the person or machinery to be monitored. The
RFID technology has created a rapidly growing industry following the first patent issued
in 1983 to Charles Walton (US Patent 4,284,288).

Example 12.10: The mutual inductance and the associated magnetic coupling provide a
fundamental example of wireless data transfer between two physically distant circuits
containing two coupled inductors. We should keep in mind that even though this is
considered wireless coupling, it is quite different from true radio-frequency
(RF) wireless data transmission over large distances. Still, many of the modern circuits,
including the first two RFID standards in Table 12.3, use magnetic coupling and mutual
inductance to enable wireless power transfer or data transfer from the tag to the reader and
vice versa.

XII-630
Chapter 12 Section 12.5: Model of Coupled Inductors

Example 12.10 (cont.):

Table 12.3. Standard frequencies for RFID tags (RFID TX/RX
systems).
Low frequency (LF) 125–134 kHz
High frequency (HF) 13.56 MHz
Ultra high frequency (UHF) 868–930 MHz
Microwave 2.45 GHz
Microwave 5.80 GHz
Circuits with coupled inductors are highlighted.

Example 12.11: The mutual inductance and the associated magnetic coupling also pro-
vide a basic example of a wireless power (not data) transfer between two physically
separated circuits. Let us assume that a power transfer is needed between two circuits, but
without mechanical contact. For example, one circuit may be fixed, and another contained
in a rotating or indexing machine. Clearly, we cannot use a wired wall plug for such a
situation. An important example of near field wireless power transfer is a battery of an
implanted device in a human body, which needs to be charged from time to time. A
solution to the problem involves a pair of coupled inductors. Figure 12.24 depicts a 120-W
inductive power transfer system from Mesa Systems Co., Medfield, MA. This system can
be powered by any 12 V battery and also includes DC to AC converters.

Fig. 12.24. A 120-W inductive power-transfer system powered by a 12-V battery from Mesa
Systems Co., Medfield, MA.

Basic Model of Inductive Power Transfer at Large Separation Distances

Two coaxial coupled inductors L1 and L2 in the wireless link configuration are shown in
Fig. 12.25. One of them is the transmitter (TX) and the second is the receiver (RX). The
separation distance between the inductors is d. The coupling coefficient k is usually
much less than one, and the calculation of the mutual inductance requires care. It is

XII-631
Chapter 12 Electric Transformer and Coupled Inductors

prudent to calculate the mutual inductance of two single coaxial loops of current
separated by d first. The result has the form

πμ0 r21 r22

M loop  ½H, d r1 , r2 ð12:65Þ
2 d3

where r1, r2 are the loop radii. Note that d3 is in the denominator indicating that the mutual
inductance decreases very rapidly when the separation distance d increases. This is
somewhat discouraging, but an unavoidable result for magnetic near-field calculations.
The mutual inductance between the two air-core coils having radii r1, r2, and N1 and N2
turns, respectively, is obtained from Eq. (12.65) in the form

+ v1 - - v2
+

L1 L2

N1 N2
i1 d i2

Fig. 12.25. Coupling between two coils in a typical near-field wireless link configuration.

N 1 N 2 πμ0 r21 r22

M ¼ N 1 N 2 M loop  ð12:66Þ
2 d3
Knowing the mutual inductance, we can find the received current or voltage (and the
received power) in the second coil once the transmitted voltage or the current is known in
the first coil.

Example 12.12: Two coaxial air-core coils with r1 ¼ r2 ¼ 1:0 cm and with N 1 ¼ N 2
¼ 100 are separated by 1 m. What is the voltage signal induced in the second coil (RX) if
the current in the first coil (TX) is given by i1 ðt Þ ¼ 100 mA
sin ðωt Þ?
Solution: We find the mutual inductance first, using Eq. (12.67):

100
100
4π 2
107 108
M¼  2:0
1010 ½H ð12:67Þ
2 1
A voltage induced in coil #2 according to the second equation in Eq. (12.50) is given by

di1
υ2 ¼ M ¼ 2:0
1010
101 ω cos ðω t Þ ð12:68Þ
dt
One can see that the higher the transmission frequency, the greater response we will obtain!
Let us assume that the transmission frequency is given by f ¼ 120 kHz. Then, ω ¼ 2πf
 0:75
106 rad=s and Eq. (12.68) yields

XII-632
Chapter 12 Section 12.5: Model of Coupled Inductors

Example 12.12 (cont.):

υ2 ¼ 2:0
0:75
1011
106 cos ðωtÞ ½V ¼ 15 cos ðωt Þ ½μV ð12:69Þ

Such a weak voltage signal can hardly be observed on an oscilloscope, extracted from
noise, and then amplified in a class laboratory setup. The most critical parameter is the coil
radius; if it increases by the factor of 2, the received voltage will increase by a factor of 16.
We see from this example that the near-field wireless link is tricky: the weak received
signal must be carefully managed before conducting any experimentation.

The above analysis implies that the separation distance between the two coils is much
greater than any of the coil dimensions. A finite magnetic core is not yet taken into
account in Eq. (12.67). The equation is only valid for ceramic-core coils. Also note that
the presence of steel conductors (armatures) nearby may greatly increase the efficiency of
the wireless link. Such an effect is frequently observed in the laboratory. If the receive coil
is terminated into a very small resistance, infinite received current and power may be
obtained, even if the received voltage is small. Where is the contradiction? The point is
that the received current creates its own magnetic field that opposes the changes in the
transmitted magnetic field. As long as the received current is small, this is not an issue.
The rule of thumb is to request the magnitude of the magnetic field created by the
received current to be at most 10 % of the transmitted one at the receiver location.

Example 12.13: Compile a MATLAB script that will estimate the voltage signal (voltage
amplitude) induced in the second coaxial coil (RX) if the periodic current (current
amplitude) in the first coaxial coil (TX) is known. The signal strength (amplitude of the
receiver voltage) is to be plotted as a function of the distance between two coils, d. We
know coil radius, r, and number of turns, N (the same for both coils).
Solution: The text of the corresponding MATLAB script is given below. It uses
Eq. (12.67). Next, it plots the voltage as a function of the separation distance. The plot
for the previous example at 1 MHz is shown in Fig. 12.26.
mu0 = 4*pi*1e-7; % permeability of vacuum (air)
omega = 2*pi*1e6; % angular frequency, rad/s
i1 = 0.1; % amplitude of exciting current i1, A
r = 1e-2; % coil radius, m
l = 0.1; % coil length, m
N = 100; % number of turns
d = [0.1:0.01:2]; % separation distance, m
M0 = pi*mu0*r^4*N^2./(2*d.^3); % mutual inductance
v2 = M0*omega*i1; % received voltage, V
semilogy(d, v2*1000); grid on;
xlabel('distance d, m'); ylabel('received voltage, mV')

XII-633
Chapter 12 Electric Transformer and Coupled Inductors

received voltage, mV
2
10

1
10

0
10

-1
10

-2
10
0 0.4 0.8 1.2 1.6 2
distance d, m

Fig. 12.26. Voltage amplitude in coil #2 when the current amplitude in coil #1 is 100 mA.

12.5.5 Application Example: Coupling of Nearby Magnetic Radiators

The mutual coupling between the two inductors (a transmitter and a receiver) is key for
the wireless inductive power transfer. At the same time, the mutual coupling may have a
negative effect when the transmitter includes two or more independently driven coils
assembled in a coil array. The mutual coupling between the individual transmit coils may
reduce the individual coil current, i.e., reduce the resulting total magnetic field.

Circuit with Two Identical Radiators

The circuit shown in Fig. 12.27 formalizes the problem. It models the coupling effect
between two nearby identical magnetic radiators. This circuit is important for near-field
wireless power transfer with coil arrays including medical applications. Our goal is to
express the source phasor current IS through the circuit parameters for two distinct cases:
A. Mutual coupling is absent.
B. Mutual coupling is present; the mutual reactance is X M > 0.

jXM
RS RS

IS IS

VS + jX1 jX1 + VS
- -

Fig. 12.27. Modeling mutual coupling of two transmit coils.

To solve the circuit, we convert the two coupled inductors to a T-network. The
resulting circuit is shown in Fig. 12.28.

XII-634
Chapter 12 Section 12.5: Model of Coupled Inductors

RS j(X1-Xm) j(X1-Xm) RS

IS IS

VS + jXM + VS
- -

Fig. 12.28. Equivalent circuit with uncoupled inductors.

When the mutual coupling is absent, the central inductor becomes a wire; both currents
are given by
VS
IS ¼ ð12:70Þ
RS þ jX 1

When the mutual coupling is present, we can use KVL for the either circuit loop, which
gives

VS þ RS IS þ jðX 1  X M ÞIS þ 2jX M IS ¼ 0 ð12:71Þ

Therefore,
VS
IS ¼ ð12:72Þ
RS þ jðX 1 þ X M Þ

Note that the sign in front of XM may vary depending on coil orientation. For example, it
is positive for two coaxial coils (or loops) with in-phase currents (or in-phase fluxes) and
is negative otherwise.

Tuned Radiators
The individual radiator circuit should be tuned to the operating frequency to maximize the
circuit current/magnetic field. The tuning is typically achieved by a series capacitor with
reactance X 1 , which exactly cancels the inductor’s reactance þX 1 . If this is the case,
Eq. (12.72) is transformed to

VS jVS j
IS ¼ ) jIS j ¼ qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ð12:73Þ
RS þ jX M
R2S þ ðX M Þ2

Thus, for the tuned radiators, the mutual coupling reduces the circuit current and the
magnetic field, irrespective of the coil orientation. When XM is small compared to the
source resistance, this effect is of little value. Altogether, it can be eliminated by adjusting
the value of the tuning capacitors.

XII-635
Chapter 12 Electric Transformer and Coupled Inductors

Exercise 12.18: Determine the relative reduction in the emanating magnetic field at any
point in space for two tuned identical coupled radiators when RS ¼ 10 Ω and X M ¼ 1 Ω.
Answer: 0.5 %.

Exercise 12.19: Which reactance should the tuning capacitors have in order to eliminate
the effect of the mutual inductance in Fig. 12.28?
Answer: X 1  X M .

XII-636
Chapter 12 Summary

Summary
Ideal-transformer model
Concept Circuit Formulas/Meaning
Faraday’s law Primary winding:
Dependent voltage source in the dΦðt Þ
secondary winding υ1 ðt Þ ¼ N 1
dt
Secondary winding:
dΦðt Þ
υ2 ðt Þ ¼ N 2
dt
Conclusion:
υ2 N 2
¼
υ1 N 1
Ampere’s law l ¼ l ab þ l bc þ l cd þ l da
Dependent current source in the General form:
primary winding lH ¼ i
Ideal core:
0 ¼ i ¼ N 1 i1  N 2 i2
Conclusion:
i1 N 2
¼
i2 N 1

Ideal-transformer model with two Ideal-transformer

dependent sources Equivalent circuit model of the equations:
ideal transformer υ2 N 2
¼
υ1 N 1
i1 N 2
¼
i2 N 1
Power conservation:
υ 1 i1 ¼ υ 2 i2
Stored energy: 0

Simple transformer circuit Load voltage:

Circuit with transformer symbol N2
υR ¼ υS
N1
Load current:
υR
iR ¼
R
Source power:
 2 2
N 2 υS ðt Þ
pS ð t Þ ¼
N1 R
(continued)

XII-637
Chapter 12 Electric Transformer and Coupled Inductors

Terminology
Turns ratio N1 : N2 ¼ a
Step up transformer N 2 > N 1, a < 1
Secondary winding is the high-voltage side
Step down transformer N 2 < N 1, a > 1
Primary winding is the high-voltage side
Ratings Apparent load power and
Primary voltage (rms) : Secondary voltage (rms)
Some useful transformers
Autotransformer Circuit symbol Step down setup:
Voltage relation:
(from Faraday’s law)
N2
υ2 ¼ υ1
N1 þ N2
Current relation:
(from Ampere’s law)
N1 þ N2
i2 ¼ i1
N2
Multiwinding transformer Circuit symbol Voltage relation:
(from Faraday’s law)
υ1 υ2 υ3
¼ ¼
N1 N2 N3
Current relation:
(from Ampere’s law)
N 1 i1 ¼ N 2 i2 þ N 3 i3

Center-tapped transformer Circuit symbol Voltage relation:

(from Faraday’s law)
N2
υ2 ¼ υ1
2N 1
Current relation:
(from Ampere’s law for two
identical loads)
N1
i1 ¼ i2 ¼ i3
N2
Current transformer Circuit symbol Instrumentation transformer
(to measure current)
1
i2 ¼ i1
N2

(continued)

XII-638
Chapter 12 Summary

Phasor form of a transformer circuit

Particular types of dot convention

V2 N2 I2 N1
¼þ , ¼þ
V1 N1 I1 N2

V2 N2 I2 N1
¼þ , ¼
V1 N1 I1 N2

V2 N2 I2 N1
¼ , ¼þ
V1 N1 I1 N2

V2 N2 I2 N1
¼ , ¼
V1 N1 I1 N2

Reflecting the source to the secondary side

Transformer circuit Circuit equivalent (load voltage is the same)

Reflecting the load to the primary side

Transformer circuit Circuit equivalent (source current is the same)

(continued)

XII-639
Chapter 12 Electric Transformer and Coupled Inductors

Transformer-based impedance matching

Exact matching with real impedances
pffiffiffiffiffiffiffiffiffiffiffiffiffi
Transformer circuit Circuit equivalent with N1 : N 2 ¼ RS =RL

Approximate matching for maximum power transfer from source to load

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
  ffi
Transformer circuit N : N ¼ 4 R2 þ X 2 =R2
1 2 S S L

Non-ideal low-frequency transformer model (Steinmetz model)

Primary winding and magnetic core:
Magnetizing inductance Lm (X m ¼ ωLm );
Core loss equivalent resistanceRc;
Leakage inductance Ll1 (X l1 ¼ ωLl1 );
Ohmic resistanceR1;
Secondary winding:
Leakage inductance Ll2 (X l2 ¼ ωLl2 );
Ohmic resistance R1
Model of coupled inductors
Circuit symbol Physical counterpart

Reduction to ideal-transformer model Ideal-transformer model

(continued)

XII-640
Chapter 12 Summary

Phasor form Constitutive relations:

V1 ¼ jω L1 I1 þ jω M I2
V2 ¼ jω MI1 þ jω L2 I2
Mutual inductance:
pffiffiffiffiffiffiffiffiffiffi
M ¼ k L1 L 2
Dimensionless coupling coefficient:
0k1
Simple circuit To solve the circuit use the constitutive relations
and KVL for each loop. Solution:
VS
I2 ¼ ½A
M 2  L1 L2 L1 RL
jω 
M M
 
L2 RL
I1 ¼  þ I2 ½A
M jωM
Conversion to a T-network without Equivalent T-network of uncoupled inductances
coupling

Mutual inductance for two short ceramic

coaxial coils N 1 N 2 πμ0 r21 r22
M
2 d3
r1, r2 are the coil radii;
d >> r1 , r2 is the coil separation distance

Coupling between two nearby near-field

magnetic radiators Mutual coupling between two radiators generally
reduces the amplitude of the source current in every
VS
radiator: IS ¼
R S þ j ðX 1 þ X M Þ
and the associated magnetic field for both of them

XII-641
Chapter 12 Electric Transformer and Coupled Inductors

Problems Problem 12.2. Are the following figures

correct?
12.1 Ideal Transformer
as a Linear Passive Circuit a)
Element
12.1.2 Ideal Open-Circuited Trans-
former: Faraday’s Law of Induction
Problem 12.1. The following figure shows two
open-circuited coils subject to a time-varying
flux, Φ(t), in the core.

a) a b a b
b)

#1 #2

b) a b a b

c)

#1 #2

Faraday’s law does not explicitly say anything

about the polarity of the induced voltage in the
coils. Nevertheless, the induced voltage does
have a preferred polarity determined by Lenz’s
law (Heinrich Friedrich Emil Lenz (1804–
1865), a Russian physicist of German ethnicity,
taught at the University of St. Petersburg). Problem 12.3. For the circuit shown in
Lenz’s Law states that the circuit current due the following figure, determine voltage υ2(t) if
to induced voltage υ produces a flux in such a υðt Þ ¼ 100 cos ωt ½V using Faraday’s law of
direction as to oppose the change of the flux. induction. Count the number of turns. Assume
In other words, the induced current always no flux leakage.
seeks to maintain the status quo of the magnetic
field. To find the direction of the anticipated
circuit current, you may imagine a resistance
+
connected across the coil terminals. Now,
assume that the magnetic flux in the core, Φ,
is increasing in time. Label the polarity of the v(t)
induced voltages for coils #1 and #2, respec- +
tively, by  and show the direction of the v2(t)

anticipated circuit current. - -

XII-642
Chapter 12 Problems

Problem 12.4. For the circuit shown in

the figure below, determine voltage υ2(t) if
υðt Þ ¼ 70 cos ω t ½V using Faraday’s law of
induction. Count the number of turns. Assume
no flux leakage.

i1(t)
+
v2(t)
+ i2(t)
-
v(t)

- Problem 12.8. For the circuit shown in the

following figure, establish the relation between
currents i1(t), i2(t), and i3(t). Assume an ideal
magnetic core. Count the number of turns.
Problem 12.5. Explain the meaning of the ideal
magnetic core. What is the inductance value for
an inductor which uses the closed-loop ideal i3(t)
magnetic core?

Problem 12.6. You are given the source volt-

age in the form υS ðt Þ ¼ 340 cos ð2π60t Þ ½V ,
i1(t)
the number of turns of primary winding
N 1 ¼ 250, the finite relative permeability of
the magnetic core, μr ¼ 7000, the coil length
of the primary winding of 9 cm, and the core i2(t)

cross section of A ¼ 0:0015 m2 .

A. Find the exciting current iΦ(t) (no-load Problem 12.9. For the circuit shown in the
current) in the primary winding of this figure below, determine current i2(t) given that
nonideal transformer. i1 ðt Þ ¼ 1 cos ωt ½A. Count the number of
B. Reduce N1 by the factor of 10 and repeat turns.
the solution.
H= 100 cos t [A/m]
12.1.4 Ampere’s Law
12.1.5 Ideal Loaded Transformer R= 3 cm
i1
12.1.6 Ideal Transformer Versus
Realistic Transformer: Transformer
Terminology
Problem 12.7. For the circuit shown in the i2
following figure, establish the relation between
currents i1(t) and i2(t). Assume the ideal mag-
netic core. Count the number of turns.

XII-643
Chapter 12 Electric Transformer and Coupled Inductors

Problem 12.10. For the circuit shown in the current iS(t), load voltage υR(t), and load current
figure below, establish the relation between iR(t) when:
currents i1(t), i2(t), and i3(t). Assume an ideal A. The switch is open.
magnetic core. Count the number of turns. B. The switch is closed.
Hint: Ampere’s law for the ideal core applies C. The switch is closed and ω ! 0.
to every closed path.
iS(t)
2:1
iR(t)
i1(t) i3(t)
+
vS(t) + vR 20
i2(t) -
-

Problem. 12.14. In the circuit shown in the

figure below, the current through resistance
is given by iR ðt Þ ¼ 50 cos ð2π60t Þ ½A.
Determine:
Problem 12.11. In Problem 12.6, determine the
instantaneous stored energy of a nonideal trans- - Source voltage υS(t)
former given a core centerline length of 30 cm. - Source current iS(t)

Problem 12.12. In the circuit shown in the iS(t)

10:1
figure below, υS ðt Þ ¼ 170 cos ð2π60tÞ ½V. iR(t)
For a 100 Ω resistive load, determine:
vS(t) + 20
- Source current iS(t) -
- Load voltage υR(t)
- Load current iR(t)
- Average power delivered to the load P
Problem 12.15. A power transformer is rated
when the turns ratio is equal to 10:1, 1:1, and as 100 kVA, 11,000:2200 V, 60 Hz. Determine:
1:10.
- Transformer type
iS(t)
- Turns ratio
N1:N2 - The rated current on the low-voltage side
iR(t)
+
Problem 12.16. Repeat the previous problem
vS(t) + vR RL
- for a transformer rated as 10 kVA, 2300:230 V,
- 60 Hz.

Problem 12.13. In the circuit shown in the figure Problem 12.17. Repeat Problem 12.15
below, υS ðt Þ ¼ 325 cos ð2πωtÞ ½V. The ideal for a transformer rated as 2 kVA, 230:115 V,
transformer model is used. Determine source 50 Hz.

XII-644
Chapter 12 Problems

ZS
12.2 Analysis of Ideal- N1:N2

Transformer Circuits +
VS + VL ZL
12.2.1 Circuit with a Transformer in the -
Phasor Form -
12.2.2 Referred (or Reflected) Source
Network in the Secondary Side Problem 12.23. In the circuit shown in the
figure below, υS ðtÞ ¼ 2:5 cos ðωt þ 45
Þ ½V
12.2.3 Referred (or Reflected) Load where ω ¼ 1000 rad=s, C ¼ 10 μF,
Impedance in the Primary Side R1 ¼ 25 Ω, RL ¼ 200 Ω. Furthermore,
Problem 12.18. The load impedance is N 1 : N 2 ¼ 1 : 2. Find voltage across the load
12:5  j2:5 Ω. Find the equivalent impedance υL(t) in time domain. Assume the ideal
of the load ZT combined with a 2:1 step-down transformer.
transformer in the primary side.
R1 C1 N :N
Problem 12.19. Source voltage is given by 1 2

υS ðt Þ ¼ 2:5 cos ωt þ 45
½V and the source +

impedance is 5  j0:5 Ω. Find the equivalent vS(t)
+ vL(t) RL
circuit of the source (find VT and ZT) combined -
with a 1:4 step-up transformer in the secondary -
side. Express your result both in frequency
domain and in time domain. Problem 12.24. For the circuit shown in
the following figure, VS ¼ 10∠45
½V,
Problem 12.20. A large capacitance value of ZS ¼ 5 þ j5 Ω, ZL ¼ 5 þ j5 Ω, and
1.6 mF is required. The available physical com- N 1 : N 2 ¼ 1 : 2. Find phasor current of the
ponent is a 100-μF capacitor. You are given an source, IS. Express your result in polar form.
ideal transformer with an arbitrary turns ratio. Assume the ideal transformer.
Design the equivalent circuit for the 1.6 mF
capacitance, specify the transformer turns ZS N1:N2
ratio, and draw the corresponding circuit
IS
diagram.
+ VS ZL
-
Problem 12.21. For the circuit shown in the
figure below, determine transformer’s turns
ratio.
Problem 12.25. In the circuit shown in the
N1:N2 following figure, υS ðt Þ ¼ 2:5cos ðωt þ 45
Þ ½V
where ω ¼ 2000 rad=s, C ¼ 10 μF, R1 ¼ 50 Ω,
RL ¼ 20 Ω. Furthermore, N 1 : N 2 ¼ 2 : 1. Find
Leq=9 mH 1 mH
source current iS(t) in time domain. Assume
the ideal transformer.

iS(t)
Problem 12.22. For the circuit shown in the R1 N1:N2 C2
following figure, VS ¼ 10∠45
½V,
ZS ¼ 5 þ j5 Ω, ZL ¼ 5 þ j5 Ω, and
+
N 1 : N 2 ¼ 1 : 2. Find phasor voltage across vS(t) - RL
the load, VL. Express your result in polar
form. Assume the ideal transformer.

XII-645
Chapter 12 Electric Transformer and Coupled Inductors

12.2.4 Transformer as a Matching modeled as a resistance RL. Express the turns

Circuit ratio of a matching transformer, which is nec-
Problem 12.26. A 16-Ω load is connected to an essary for maximum average power transfer
AC voltage source with a voltage amplitude of from the antenna to the load, in terms of three
10 V and series resistance of 1 Ω. A matching given problem parameters. Note that this prob-
transformer is used with the turns ratio N1 : N2. lem has an elegant analytical solution.
Determine the average power delivered to the
12.2.5 Application Example: Electric
load when
Power Transfer via Transformers
A. N1 : N2 ¼ 1 : 1 Problem 12.29. The performance of a trans-
mission line circuit in the following figure is to
B. N1 : N2 ¼ 1 : 2 be analyzed, with and without 1:10 step-up
transformer and 10:1 step-down transformer,
C. N1 : N2 ¼ 1 : 4 respectively. All phasors are given in terms of
the rms values. Solve the circuit with and with-
D. N1 : N2 ¼ 1 : 5
out transformers and fill out the table that fol-
lows including
  active load power
PL ¼ Re VL  IL * , active source power,
Problem 12.27. A passive RFID tag circuit  
measuring temperature is modeled by a load PS ¼ Re VS  IS * , and the line power loss
resistance of 16 Ω. The tag is wirelessly Ploss ¼ RjIline j2 . All powers are to be reported
powered; it is augmented with a small in watts. You are given ZL ¼ 4 þ j2 Ω.
collecting antenna, which has the radiation
resistance of 1  j2 Ω. The negative antenna IS Iline IL
reactance is an inherent part of the small dipole 1:10 10:1
antenna design. Find the ratio of the average 0.4 j0.5 +
received powers from the antenna with and + VS VL ZL
without the matching transformer. Assume the -
load matching condition for the real part of the -
impedances.
Param. VS IS PS Iline
ZS tag
N1:N2
No tr. 240∠0

+ W tr. 240∠0

VS + VL RL Param. Ploss VL IL PL
-
- No tr.
W tr.

Problem 12.28. The previous problem may be

Problem 12.30. Repeat the previous problem
generalized as follows. An antenna operates as
when the transformer setup changes to a 1:5 step-
an energy-harvesting source. The antenna is
up transformer and a 5:1 step-down transformer.
modeled as an impedance ZS ¼ RS þ jX S in
series with a fixed ideal voltage source. The Problem 12.31. An AC-direct micro-hydro-
nonzero antenna reactance is an inherent part power system is illustrated in the following
of the small antenna design. The load is figure.

XII-646
Chapter 12 Problems

transformer shown in the following figure has

the form υS ðtÞ ¼ 5 cos ωt ½V. The source cur-
rent into the dotted terminal is iS ðt Þ ¼
10 cos ωt ½A. You are given N 1 ¼ 200,
N 2 ¼ 50. Determine voltage υ2(t) and current
i2(t) in the secondary.

i2

+
i1
Reprinted from Micro-Hydropower Systems N1
v2(t)
Canada 2004, ISBN 0-662-35880-5 + N2
v1(t)
-
-
The system uses a single phase induction gen-
erator with the rms voltage of 240 V. The sys-
tem serves four small houses, each connected to Problem 12.34. Solve the previous problem for
the generator via a separate transmission line the circuit shown in the following figure.
with the same length of 3000 m. Each line uses
i1
AWG#10 solid aluminum wire with a diameter
of 2.59 mm. The house load in every house is
+
an electric range with the resistance of 20 Ω. i2
N1
Determine total active power delivered by the v1(t)
generator, PS, total power loss in the trans- N2 +
v2(t)
mission lines, Ploss, and total active useful -
-
power, PL:
1. When no transformers are used;
2. When a 1:5 step-up transformer is used in Problem 12.35. For the circuit shown in the
powerhouse and a 5:1 step-down trans- following figure, VS ¼ 100∠45
½V,
former is used at home. ZS ¼ 5 þ j5 Ω, ZL ¼ 5 þ j5 Ω, and
N 1 : N 2 ¼ 4 : 1. Find phasor current of the
Problem 12.32. Solve the previous problem source, IS. Express your result in polar form.
when the distributed line inductance is addi- Assume the ideal autotransformer.
tionally taken into account. The inductance
ZS
per unit length
 of a two-wire line is given by
μ0
π lnd
a þ 1
4 where a is the wire radius and d is IS
the separation distance. Assume the separation N1
distance of 1 m. The operation frequency is +
VS N2
50 Hz. -
ZL

12.3 Some useful

transformers Problem 12.36. Find the equivalent input
impedance, Zin, for two autotransformer cir-
12.3.1 Autotransformer
cuits shown in the following figure.
Problem 12.33. A voltage source connected to
the primary winding of an ideal step-up auto-

XII-647
Chapter 12 Electric Transformer and Coupled Inductors

a) for every individual load are shown in the fig-

ure. Determine:
N1
Zin
A. Turns ratio, N1 : N2 of the transformer
N2 B. rms value of input current i1
ZL
i1
N1:N2
b)
200 W
+ 120 V
TV
N1 2400 V 10kW
ZL - 120 V 1500 W
Range
N2 MW
Zin

Problem 12.41. Determine phasor currents IR1,

IR2, and IR0 for the center-tapped balanced
12.3.2 Multiwinding Transformer transformer circuit shown in the following fig-
12.3.3 Center-Tapped Transformer: ure. You are given the source phasor voltage
Single-Ended to Differential VS ¼ 10∠0
½V, resistance values R ¼ 50 Ω,
Transformation and turns ratio N 1 : N 2 ¼ 1.
Problem 12.37. In the circuit shown in the IR1
following figure, N 1 ¼ 2N 2 ¼ N 3 . How is the N1:N2

instantaneous power partitioned between R

the two secondary windings if both of them VS +
- IR0
are terminated into the same load resistances? R
To answer this question, express both p2(t) and
p3(t) in terms of p1(t). IR2

i2
i1 + Problem 12.42. Solve the previous problem in
+
N2 v2(t) a general form, i.e., express phasor currents IR1,
v1(t) N1 i3 - IR2, and IR0 in the (generally unbalanced) cir-
+
- N3 v3(t) cuit shown in the following figure in terms of
- given circuit parameters VS, Z1, Z2, and turns
ratio a ¼ N 1 : N 2 .
Problem 12.38. In the circuit of the previous
problem, N 1 ¼ 2N 2 ¼ 2N 3 . How is the instan-
IR1
taneous power partitioned between the two sec- N1:N2
ondary windings if winding #2 is terminated Z1
into resistance R and winding #3 is terminated VS +
into resistance 3R? - IR0
Z2

Problem 12.39. Determine the turns ratio for

IR2
the center-tapped transformer in Fig. 12.13.

Problem 12.40. A household is using an ideal

center-tapped distribution transformer shown in 12.3.4 Current Transformer
the following figure. All voltage values are the Problem 12.43. Determine current i2(t) in an
rms values. The resistive loads include a TV, a ideal current transformer shown in the figure
microwave, and a kitchen range. The powers given that i1 ðt Þ ¼ 10 cos ωt ½A. Count the
number of turns.

XII-648
Chapter 12 Problems

i1 Element nameplate 20 kVA

2400:240 V
60 Hz
Magnetizing reactance 15,000
X m ¼ ωLm , Ω
Core loss resistance, Rc, Ω 100,000
i2 Primary leakage reactance 6.5
X l1 ¼ ωLl1 , Ω
Secondary leakage reactance, 0.07
12.4 Real-Transformer X l12 ¼ ωLl12 , Ω
Primary ohmic resistance R1, Ω 3.0
Model
Secondary ohmic resistance 0.03
12.4.1 Model of a Nonideal R2, Ω
Low-Frequency Transformer
12.4.2 Model Parameters and Their Check whether or not equalities (12.38) for the
well-designed transformer are satisfied.
Extraction
12.4.3 Analysis of Nonideal Transformer Problem 12.48. In a well-designed trans-
Model former, the number of turns of both primary
12.4.4 Voltage Regulation and Trans- and secondary windings is usually quite large.
former Efficiency Why is it so? Can we just design a 10:1 trans-
former with 10 and 1 turns, respectively?
Problem 12.44
A. Name all seven components of the Problem 12.49. The nonideal transformer
low-frequency nonideal transformer model is shown in the following figure in fre-
model. quency domain. Given the rated load with a
B. Draw the corresponding circuit diagram. power factor of 0.8 lagging, determine the
source phasor voltage VS and the source phasor
Problem 12.45. A practical power transformer current IS for transformer #1 model from
is characterized by the following nameplate Table 12.2.
information: 20 kVA 2400:240 V 60 Hz.
I2
-1
A. Determine rated-load phasor voltage VL a I2
(show units and assign phase zero). a=N1/N2
B. Determine rated-load phasor current IL R1+jXl1 R2+jXl2
IS
for power factor of 0.8 lagging (show + + +
units). VS + aE2 E2 VL ZL
- RC jXm
C. Determine rated-load active power PL - - -
(show units).

Problem 12.46. Repeat the previous problem Problem 12.50. Repeat the previous problem
for the power transformer characterized by for transformer #2 model from Table 12.2.
2 kVA 230:115 V 50 Hz.
Problem 12.51. Repeat Problem 12.49 for
Problem 12.47. A practical power transformer transformer #3 model from Table 12.2.
is characterized by the following information:
Problem 12.52. The nonideal transformer
model is shown in the following figure in

XII-649
Chapter 12 Electric Transformer and Coupled Inductors

frequency domain. Given the rated load with a Problem 12.57. Solve Problem 12.55 when the
power factor of 0.7 lagging, determine: mutual reactance is equal to 3 Ω.
A. Percentage regulation
B. Percentage efficiency Problem 12.58. In the circuit shown in the
following figure, determine source phasor cur-
for transformer #1 model from Table 12.2 rent IS given that RL ¼ 1 Ω, RS ¼ 2 Ω, and
VS ¼ 15∠0
½V.
I2
-1
a I2
j3
a=N1/N2 RS
R1+jXl1 R2+jXl2
IS
+ + + IS
VS + aE2 E2 VL ZL
- VS + j5 j4 RL
RC jXm - - - -

Problem 12.53. Repeat the previous problem

for transformer #2 model from Table 12.2.
12.5.4 Application Example: Wireless
Problem 12.54. Repeat Problem 12.52 for Inductive Power Transfer
transformer #3 model from Table 12.2. 12.5.5 Application Example: Coupling
of Nearby Magnetic Radiators
Problem 12.59. Two small coaxial ceramic-
12.5 Model of Coupled core coils with r1 ¼ r2 ¼ 2:0 cm and with N 1
Inductors ¼ N 2 ¼ 100 are separated by 1 m. What is the
voltage signal induced in the second coil (RX) if
12.5.1 Model of Two Coupled Inductors
the current in the first coil (TX) is given by
12.5.2 Analysis of Circuits with Coupled i1 ¼ 100 mA
sin ðωtÞ ? The operation fre-
Inductors quency is 1 MHz.
12.5.3 Coupling Coefficient
Problem 12.55. Solve a circuit with two Problem 12.60. Repeat the previous problem
coupled inductors in the figure below in fre- when the operation frequency changes to
quency domain: 10 MHz.
A. Determine the load phasor current and
the source phasor current. Problem 12.61. In the circuit shown in the
B. Determine phasor voltages V1 and V2 following figure, find source phasor current IS
in time domain iS(t) given that υS ðt Þ ¼
given that VS ¼ 10∠0
½V and RL ¼ 10 Ω. 10 cos ωt ½V.
I1 j4 I2 j1
4 -j4 -j4 4

+ + IS IS
+ VS V1 j4 j4 V2 RL VS + + VS
- - j3 j3
-
- -

Problem 12.56. Solve the previous problem

when the mutual inductance is exactly zero.

XII-650
Index

A maximum power efficiency, principle of, 554–555

AC circuit analysis, steady-state AC circuit maximum power transfer, principle of, 556–557
complete solution for, 424–425, 429, 442–443 power expressions, 553
KVL, KCL and equivalent impedances, 423–424, types
442–443 apparent power, 545–547
at single frequency, 429, 443–444 arbitrary periodic AC signals, 541–542
source transformation, 425–427, 443–444 average power, 545–547
superposition theorem, 429–430, 443–444 instantaneous, 537
Thévenin and Norton equivalent circuits, 427–429, power angle, 543–544
443–444 power triangle, 547–549
AC coupled, 408 reactive power, 545–547
AC power resistor, capacitor, and inductor, average power for,
balanced three-phase systems 544–545
apparent power, 570 rms voltages and AC frequencies in world,
average power, 570 540–541
balanced delta-connected load, 572 time-averaged, 538–540
balanced delta-connected source, 572–575 wattmeter, 550
instantaneous power, 568–569 AC voltage source, 19
material consumption, 570–572 Aliasing frequency, 474
reactive power, 570 Alternating-current circuits, 19–20
distribution Alternating current power supplies, 403
automotive alternator, 562 American Telephone & Telegraph Company (AT&T), 454
long-distance power transmission, neutral wire in, Ampere’s law, 14–16, 595–596, 610
565–567 Amplifier bandwidth, 467–468
neutral conductor, 558–559 Amplifier circuit model, 207–209, 221, 254–255
phase voltages, 559–561 amplifier DC imperfections and cancellation, 229,
residential household, 563 261–263
single-phase three-wire power distribution system, input bias and offset currents, 231–232
558–559 input offset voltage, 229–230
single-phase two-wire power distribution system, output offset voltage, cancelling, 230–231
558–559 cascading amplifier stages, 227–229, 260–261
split-phase distribution system, 558–559 current flow in, 218–219
synchronous three-phase AC generator, 562 resistance values, choosing, 259
synchronous three-phase AC motor, 562–563 discrete resistance values and potentiometers, 222
three-phase four-wire power distribution system, gain tolerance, 222–223
558–559 non-inverting and inverting configurations,
three-phase wye-wye circuit, 563–565 221–222
wye (Y) source and load configurations, 561 virtual-ground circuit, 232–233
power factor correction voltage amplifier vs. matched amplifier
automatic power factor correction system, 554 input load bridging, 224–226
capacitance value, 552 input load matching, 226–227
definition, 551

© Springer Nature Switzerland AG 2019 651

S. N. Makarov et al., Practical Electrical Engineering, https://doi.org/10.1007/978-3-319-96692-2
652 Index

Amplifier circuit model (cont.) Binary-weighted-input digital-to-analog converter, 219

sensor’s equivalent circuit and amplifier’s Bipolar junction transistor (BJT), 20
equivalent circuit, 224 Bistable amplifier circuit, 342–344
whole voltage amplifier circuit, model of, 223–224, Blocking capacitor, 301, 315
259–260 Bode plots, 453–456, 505
Amplifier DC imperfections Boundary element method, 11
and cancellation, 229, 261–263 Break frequency, 452
input bias and offset currents, 231–232 Buffer amplifier, 216–217, 347
input offset voltage, 229–230 Butterworth response, 513
output offset voltage, cancelling, 230–231 Bypass capacitor, 299–300, 315
Amplifier feedback loop, 211
Amplifier IC, 468–469
Amplifier operation, 256–257 C
amplifier circuit model, 207–209 Capacitance, 19, 417
ideal-amplifier model definitions, 273, 309
concise form, operational amplifier in, 209 dynamic behavior, 290–292, 312–313
first summing-point constraint, 209–210 electric field energy, 274
input/output resistances and output current, electrostatic discharge, 275–276
realistic values of, 210 1-μF Capacitor
open-circuit/open-loop voltage gain, 204 blocking capacitor, 301, 315
open-loop configuration, power rails and voltage capacitive touchscreens, 282–283
transfer characteristic in, 205 capacitor marking, 281–282
operational amplifier comparator, 206–207 ceramic capacitor, 281
power rails, practice, 205–206 dielectric breakdown effect, 280
symbol and terminals, 203–204 dielectric strength, 280
Amplifier’s equivalent circuit, 223 electrolytic capacitor, 281
Amplitude, 403, 404, 408 normalized breakdown voltage, 280
transfer function, 452–453, 503 relative permittivity, 280
Analog filter, 447 tantalum capacitor, 281
Analog low-pass RC filter, 449 finger capacitance, 282
Analog-to-digital converter (ADC), 228, 241 to ground, 273–274, 309
Angle notation, 416 in parallel and in series, 278–279
Angular frequency, 403 parallel-plate capacitor, 276–278
Antenna transmission, 171–172 parasitic capacitance, 282
Apparent power, 545–547, 570, 576 self-capacitance, 273, 309
Arbitrary load, 576 of two conductors, 273
Astable multivibrator, 342 of two equal conductors, 274
Asymptotic expansion, 217 Capacitance value, 552
Automatic gain control, 222 Capacitive coupling, 221
Automatic power factor correction system, 554 Capacitive touchscreens
Automatic traffic light, 524 mutual-capacitance method, 283
Average power, 538–540, 545–547, 570, 576 self-capacitance method, 282–283
Capacitor voltage, continuity of, 325–326
Cascading amplifier stages, 227–229
B Center frequency of the band-pass filter, 509
Balanced circuit, 565 Channels, 408
Balanced delta-connected load, 572, 578 Characteristic equation, 366–367
Balanced delta-connected source, 572–575, 578 Charging, concept of, 337
Balanced phase voltages, 560 Circuit analysis method, 167
Balanced three-phase load, 561 Circuit current, 360
Balun transformer, 613 Circuits with resistances and capacitances, 349–351
Band-pass filter, 459, 514, 527 Circuits with resistances and inductances, 351–353
Band-reject filter, 514, 527 Circuit with bypass capacitor, 355–357
Band-stop, 515 Clamp on ammeter, 614
Bandwidth of the series resonant RLC circuit, 502–505 Clock frequency, 342
Bell Telephone Company, 454 Clock signal, 342
Binary counter, 220 Closed-loop AC gain, 465–467
Index 653

Closed-loop configuration, 204 Coupled inductors model, 640–641

Closed-loop DC gain, 465 conversion to T-network, 627–628
Closed-loop gain, 214, 216, 242–243 coupling coefficient, 628–630
Common-mode amplifier circuit gain, 236 mutual inductance(s), 627
Common-mode gain, 235–237 N coupled inductors, 627
Common-mode input signal, 209 phasor form, 625–626
Common-mode rejection ratio (CMRR), 236 tuned radiators, 635–636
Common-mode voltage, 234 two coupled inductors, 624–625
Comparator, 206–207, 344 two identical radiators, 634–635
Compensated Miller Integrator, 303–304, 316 wireless inductive power transfer, 630–634
Complementary solution, 356, 366 Critical damping, 367
derivation of, 366–367 Cross product, 17
Complex conjugate, 544 Current amplifier, 245, 246
Complex impedance, magnitude and phase of, 420–422 Current divider circuit, 111–113, 135–136
Complex power, 547, 576 Current flow
Complex transfer function in amplifier circuit, 218–219
high-pass filter, 457 model, 11–13
low-pass filter, 457 Current limiter, 111, 135
next-stage filter load, 459–460 Current source, 19
phase Bode plots, 458 Current transformer, 614
Compliance, 19 Cutoff frequency, 508
Conductors, electrostatics, 7–8
charges, Coulomb force, and electric field, 3–4,
23–24 D
Coulomb’s law, 9–10, 23–24 Damping coefficient, 360
electric potential and electric voltage, 4–5, 23–24 Damping ratios, 367
electric voltage vs. ground, 5–7, 23–24 solution behavior for, 368–369
equipotential conductors, 7–8, 23–24 3-dB frequency, 455
Conservative field, 5 20-dB-per-decade roll-off, 455
Constant-speed water pump, 19 DC coupled, 408
Constant-torque water pump, 18–19 DC-coupled amplifiers, 220
Continuous and discrete Fourier transform DC-coupled single-supply amplifier, 232–233, 263
applications of, 476 DC steady state, 18–19, 296
bandlimited, 471, 473 Decade, 455
definition of, 470, 474, 483 Decoupling inductor, 301, 315
direct Fourier transform, 470 Dependent sources
FFT, 473–474 arbitrary time-varying voltage, 64
Fourier spectrum, 470 current-controlled current source, 66
Fry’s Electronics store, 472 current-controlled voltage source, 66
fundamental frequency, 473 definition, 64–65
of Gaussian pulse, 483 Thévenin’s theorem, 164–165
input pulse signal, 478–479 time-varying sources, 84–85
inverse DFT, 473–474 AC source polarity, 68
inverse Fourier transform, 470 source amplitude, 67
mathematical properties of, 472–473 transfer characteristics, 66–67, 84–85
numerical differentiation, 476–478 voltage-controlled current source, 65–66
properties of, 483 voltage-controlled voltage source, 65
of rectangular pulse, 483 Dependent vs. independent sources, 64, 84
sampling points, 473, 483 Difference amplifier, 235–237, 264–265
sampling theorem, 475–476 differential gain and common-mode gain,
sinc function, 471 235–237, 264
structure of, 475 differential input signal, 234–235
time-domain computational solution, 479–480 Differential amplifier circuit gain, 236
Contour integral, 5 Differential gain, 235–237
Controls block diagram, 242 Differential input signal, 209
Core loss resistance, 616 Differential input voltage, 204
Corner frequency, 455, 456 Differential-mode input resistance, 238
Coulomb force, 3–4 Differential sensor, 235
Coulomb’s law, 9–10 Differential voltage, 234
654 Index

Differentiator amplifier, 304–305 Dynamic random-access memory (DRAM), 329

Digital memory cell, 329–330 Dynamic undershoot, 372
Digital memory element, 344
Digital repeater, 206, 207
Digital signal processing (DSP), 476 E
Digital voltage, 206 Eddy current loss, 599
Diode, 20 Electric circuits
Direct current (DC), 13 conductors, electrostatics of
Discrete cosine transform, 476 charges, Coulomb force, and electric field, 3–4,
Discrete Fourier transform (DFT), 470, 483 23–24
bandlimited, 473 Coulomb’s law, 9–10, 23–24
definition of, 474 electric potential and electric voltage, 4–5, 23–24
FFT, 473–474 electric voltage vs. ground, 5–7, 23–24
fundamental frequency, 473 equipotential conductors, 7–8, 23–24
inverse DFT, 473–474 hydraulic and fluid mechanics analogies
sampling points, 473 for alternating-current circuits, 19–20, 25
structure of, 475 in DC steady state, 18–19, 25
Discrete resistance values, 222 for semiconductor circuit components, 20, 25
Dotted terminals, 594 steady-state current flow
Dual in-line (DIP-N) package, 203 current flow model and electrostatics, 11–13,
Duality, 505–507 24–25
parallel connected second-order RLC circuit, 365 electric current, 11, 24–25
Duality of series/parallel RLC electric circuits, 365 electric power transfer, origin of, 16–17
Dynamic circuit elements magnetostatics and Ampere’s law, 14–16
amplifier circuits, 302–303 physical model of, 13–14, 24–25
blocking capacitor, 301, 315 Electric current density, 11
bypass capacitor, 299–300, 315 Electric field, 3–4
Capacitance and electric potential, 7
dynamic behavior, 290–292, 312–313 Electric field intensity, 3
electric field energy, 274 Electricity, 3
electrostatic discharge, 275–276, 309 Electric load, 13
1-μF capacitor, 281–283 Electric permittivity, 9
finger capacitance, 282 Electric potential, 4–5
to ground, 273–274 electric field and, 7
in parallel and in series, 278–279, 310 Electric power transfer, origin of, 16–17
parallel-plate capacitor, 276–278, 309 Electric transformer, 19–20
parasitic capacitance, 282 autotransformer, 610–611
self-capacitance, 273 center-tapped transformer, 612–614
of two conductors, 273 coupled inductors model
of two equal conductors, 274 conversion to T-network, 627–628
compensated Miller Integrator, 303–304, 316 coupling coefficient, 628–630
decoupling inductor, 301, 315 mutual inductance(s), 627
differentiator amplifier, 304–305 N coupled inductors, 627
inductance phasor form, 625–626
dynamic behavior, 293–295, 312–313 tuned radiators, 635–636
magnetic-field energy, 285 two coupled inductors, 624–625
magnetic flux density, 284 two identical radiators, 634–635
1-mH Inductor, 288–289 wireless inductive power transfer, 630–634
mutual inductance, 285 current transformer, 614–615
in parallel and in series, 287–288, 312 ideal transformer
self-inductance, 284, 311 AC power transfer, 591
of solenoid, 285–287, 311–312 Ampere’s law, 595–596
instantaneous energy and power, 295–296, 313 Faraday’s law of induction, 591–595
Miller integrator, 302–303, 315–316 ideal loaded transformer, 596–598
steady state, 296–297, 314–315 magnetic circuit, 591
at very high frequencies, 297–298 mechanical analogies of, 601
Dynamic process, 212–213 principally different, 591
Index 655

vs. real transformer, 598–600 ceramic capacitor, 281

transformer currents, appearance of, 595 dielectric breakdown effect, 280
ideal transformer circuits dielectric strength, 280
fixed load voltage, 607–608 electrolytic capacitor, 281
fixed source voltage, 608–609 normalized breakdown voltage, 280
matching circuit, 605–607 relative permittivity, 280
phasor form, 602 tantalum capacitor, 281
referred/reflected load impedance, 604–605 Feedback factor, 242
referred/reflected source network, 602–604 Feedback gain, 242
multiwinding transformer, 611–612 Feedback system
real-transformer model application of, 244–245
high-frequency transformer model, 622–623 closed-loop gain and error signal, 242–244, 266–267
model parameters and extraction, 616–618 signal-flow diagram of, 242
nonideal low-frequency transformer, 616 voltage, current, transresistance, and transconductance
nonideal transformer model, 618–621 amplifiers with negative feedback, 245–246,
transformer efficiency, 622 267
voltage regulation, 621–622 Field magnitude, 3
Electric voltage, 4–5 Filter circuits
vs. ground, 5–7 amplitude transfer function, 452–453
Electromagnetic forming, 328–329 bode plot, 453–456
Electromagnetic material processing, 324, 328–329 complex transfer function
electromagnetic forming, 328–329 high-pass filter, 457
self-induced Lorentz force, 329 low-pass filter, 457
Electromagnetic railgun, 326–328 next-stage filter load, 459–460
Electronic article surveillance (EAS), 518 phase Bode plots, 458
Electronic ignition system, 336, 340 continuous and discrete fourier transform
Electronic oscillators, 342 applications of, 476
Electrostatic discharge (ESD), 9, 275–276, 309 bandlimited, 471, 473
Electrostatics of conductors, 3, 11–13 definition of, 470, 474, 483
charges, Coulomb force, and electric field, 3–4, 23–24 direct Fourier transform, 470
Coulomb’s law, 9–10, 23–24 FFT, 473–474
electric potential and electric voltage, 4–5, 23–24 Fourier spectrum, 470
electric voltage vs. ground, 5–7, 23–24 Fry’s Electronics store, 472
equipotential conductors, 7–8, 23–24 fundamental frequency, 473
Energy-accumulating capacitor circuit, 330–332 of Gaussian pulse, 483
Energy-accumulating inductor circuit, 337–339 input pulse signal, 478–479
Energy-release capacitor circuit, 322–324 inverse DFT, 473–474
Energy-release inductor circuit, 333–336 inverse Fourier transform, 470
Energy-release RL circuit, 339–340 mathematical properties of, 472–473
Equipotential conductors, 7–8 numerical differentiation, 476–478
Equipotential lines, 6 properties of, 483
Equipotential surface, 6, 8, 12 of rectangular pulse, 483
Equivalent circuit values, 618 sampling points, 473, 483
Equivalent impedances, 423–424 sampling theorem, 475–476
Equivalent resistance, 103 sinc function, 471
Error signal, 243–244 structure of, 475
Euler’s formula, 409, 410 time-domain computational solution, 479–480
decibel, and roll-off, 453–456
first-order filter circuits
F analog filter, 447
Falling/trailing edge, 372 analog low-pass RC filter, 449
Faraday’s law, 520, 591–595 DC circuit, 448
of induction, 59 high-pass filter, 449, 451
Fast Fourier transform (FFT), 473, 474, 483 load connected to filter, 451–452
1-μF capacitor MATLAB, 450–451
capacitive touchscreens, 282–283 qualitative analysis, 448–449
capacitor marking, 281–282 RC voltage divider circuit, 447
656 Index

Filter circuits (cont.) G

real-valued voltages, 448 Gain-bandwidth product, 464–465
resistor voltage and capacitor voltage, 448 Gain tolerance, 222–223
superposition principle, 450 Gauss’ theorem, 7
two-port networks, 449 Generator theorems, 188–195
voltage division yields, 448 active linear networks, 154
half-power frequency, 452–453 circuit analysis method, 167
operational amplifier circuits negative equivalent resistance, 165–166
amplifier bandwidth, 467–468 short/open circuit methods, 154–155
amplifier IC, 468–469 solar panels, 155–156
closed-loop AC gain, model of, 466–467 source transformation theorem, 156–158
frequency bandwidth, 463 Thévenin and Norton equivalents
front-end amplifier, 463 circuit solution, 163–164
open-loop AC gain, model of, 465–466 dependent sources, 164–165
open-loop amplifier gain, 463–464 using Them circuit solution, 161–163
unity-gain bandwidth vs. gain-bandwidth product, without dependent sources
464–465 Norton’s theorem, 159–161
phase transfer function, 456–457 Thévenin’s theorem, 159
RL filter circuits, 460–462 Ground
Filter operation, 478–479 electric voltage vs., 5–7
Filter termination resistance, 451 reference, 5
First-order filter circuits
analog filter, 447
analog low-pass RC filter, 449 H
DC circuit, 448 Half-power bandwidth, 503
high-pass filter, 449, 451 Half-power frequency, 452, 511, 513
load connected to filter, 451–452 Harmonic voltage and current, phasors, steady-state AC
MATLAB, 450–451 circuit
qualitative analysis, 448–449 amplitude, frequency and phase measurements, 408
RC voltage divider circuit, 447 complex exponent, shorthand notation for, 416
real-valued voltages, 448 harmonic voltage and current, 403–405
resistor voltage and capacitor voltage, 448 leading and lagging, 405–407
superposition principle, 450 magnitude, 411–413
two-port networks, 449 operations with phasor diagram, 413–416
voltage division yields, 448 phasor, definition of, 408–410
First-order high-pass filter, 453, 481–482 to real signals, 411
First-order low-pass filter, 453, 467, 481–482 from real signals to, 410–411
First-order ODE, 355 High-frequency asymptotes, 456
First-order transient circuit, 358 High-frequency transformer model, 622–623
Flexible membrane, 19 High-pass filter (HPF), 449, 452–453, 527
Fluid-flow analogy, 336, 337 Homogeneous differential equation, 331
Fluid mechanics, 18 Homogeneous second-order ODE, 353, 360
Fluid mechanics analogies Horseshoe coil, 525
for alternating-current circuits, 19–20, 25 Horseshoe shape, 525
in DC steady state, 18–19, 25 Hydraulic analogies
for semiconductor circuit components, for alternating-current circuits, 19–20, 25
20, 25 in DC steady state, 18–19, 25
Fluid mechanics analogy, 326, 336 for semiconductor circuit components, 20, 25
Fluid valves, 20 Hysteresis loss, 599
Forced response, 356–357, 366
Forcing function, 356, 362
Forward/open-loop gain A, 242 I
Fourier spectrum, 470, 484 Ideal-amplifier model
Frequency, 403, 408 concise form, operational amplifier in, 209
band, 455 first summing-point constraint, 209–210
response, 457 input/output resistances and output current,
Fry’s Electronics store, 472 realistic values of, 210
Index 657

Ideal filter, 508 voltage difference, 52

Ideal magnetic core, 593 Norton’s theorem, 159–161
Ideal transformer practical current source, 57–58, 82–83
AC power transfer, 591 practical voltage source, 54–55, 80–82
Ampere’s law, 595–596 Thévenin’s theorem, 159
Faraday’s law of induction, 591–595 voltage source, 58–59
ideal loaded transformer, 596–598 Inductance, 19, 417
magnetic circuit, 591 dynamic behavior, 293–295, 312–313
mechanical analogies of, 601 magnetic-field energy, 285
principally different, 591 magnetic flux density, 284
vs. real transformer, 598–600 1-mH Inductor, 288–289
transformer currents, appearance of, 595 mutual inductance, 285
Ideal-transformer model, 593, 637–639 in parallel and in series, 287–288
Ideal voltmeter and ammeter, 85–87 self-inductance, 284, 311
absolute voltage, 72–73, 86–87 of solenoid, 285–287, 311–312
digital multimeter, 69 Inductively coupled system, 518
electric ground types, 71 Inductor current, continuity of, 336–337
fluid mechanics analogy, 70 Inductor inertia, 336
ground and return current, 71–72 Inhomogeneous first-order differential equation, 331
voltage drop, 72–73, 86–87 Inhomogeneous particular solution, 356
wrong connections, 69–70 Input bias current, 231–232
Ideal wire, 14 Input load bridging, 224–226
Impedance, 339, 416 Input load matching, 226–227
Impedance bridging, 224 Input offset voltage, 204, 229–230
Impedance matching, 557 Input resistance of amplifier circuit, 207, 223–224
Impedance of the capacitor, 421 Input voltage to the amplifier circuit, 213
Impedance of the inductor, 421 Instantaneous AC power, 537
Impedance of the resistor, 420 Instrumentation amplifier, 265–266
Impedance phasor diagram, 421 building, 238
Impedance, steady-state AC circuit, 339, 416, 419–422 concept of, 238–240
concept of, 417–419, 439–442 in laboratory, 240–241
of human body, 422 motivation for, 237–238
magnitude and phase of, 420–422 Integrated circuit (IC), 9, 203
physical meaning of, 419–420, 439–442 Integration constants for second-order circuits, 368
Independent ideal voltage source, 80–82 Internal compensation, 463
active reference configuration, 52 Inverse fast Fourier transform (IFFT), 473, 474
electric circuit solution, 54 Inverse Fourier transform, 470
ν-i characteristic, 52–53 Inverse stiffness, 19
polar device, 52 Inverting amplifier, 215–216
symbols, 53 Inverting input, 203
voltage difference, 52 Inverting Schmitt trigger, 344
Independent sources Iron loss, 599
chemical battery, 83–84
battery voltage and capacity, 61
characteristics, 63 K
circuit model, 62–63 Kirchhoff’s current law (KCL), 93–95, 127–128, 423–425
DC voltage generator, 59–60 in phasor form, 425
ideal current source, 82 Kirchhoff’s voltage law (KVL), 14, 95–98, 128–129,
active reference configuration, 55–56 423–425
ν-i characteristic, 56 in phasor form, 425
symbols, 56–57
independent ideal voltage source, 80–82
active reference configuration, 52 L
electric circuit solution, 54 Laboratory ignition circuit, 340–341
ν-i characteristic, 52–53 Lagging, steady-state AC circuit, 405–407
polar device, 52 Leading edge, 372
symbols, 53 Leading, steady-state AC circuit, 405–407
658 Index

Leakage inductance, 616 Low-pass filter bandwidth, 511

Light-emitting diodes (LEDs), 324 Low-resistance load, 452
Linear feedback system, 242
application of, 244–245
closed-loop gain and error signal, 242–244, 266–267 M
signal-flow diagram of, 242 Magnetic circuit, 591
voltage, current, transresistance, and transconductance Magnetic field, 14
amplifiers with negative feedback, Magnetic-field intensity, 14
245–246, 267 Magnetic near-field calculations, 632
Linear functions, 349 Magnetizing current, 593
Linear oscillators, 342 Magnetizing inductance, 593
Linear passive circuit element Magnetostatics
components, 31 and Ampere’s law, 14–16
elements, 31 steady-state current flow and
resistance current flow model and electrostatics, 11–13,
current through resistance, 32–33 24–25
fixed resistors, 41–42 electric current, 11, 24–25
ν-i characteristics, 34, 76–78 physical model of, 13–14, 24–25
ohmic conductors, 36–39, 76–78 Magnitude
Ohm’s law, 33 of complex impedance, 420–422
open and short circuits, 34–35 phasors, 411–413
power delivered, 35–36, 78–79 Matched amplifier, amplifier circuit design
power loss, 39–41, 78–79 input load bridging, 224–226
symbols and terminals, 31 input load matching, 226–227
variable resistors, 42 sensor’s equivalent circuit and amplifier’s equivalent
voltage across resistance, 32 circuit, 224
resistive sensors Matching transformer, 605–607
circuit symbols, 45 Material conductivity, 11
photocell, 43–44 MATLAB script, 633
photoresistor, 44 Maximum amplitude, 501
physical component, 79 Maximum power transfer, 577
potentiometric position sensor, 44–45 Maxwell’s minimum heat theorem, 98
sensitivity, 45 Mechanical analogy, 344, 359
strain gauge, 44 Mesh analysis, 187–188
thermistor, 43 linear circuits, 150–152
thermocouple, 43–44 supermesh, 152–153
Line currents, 565 Metal detector, 523
Line impedance, 567 Metal-oxide-semiconductor field-effect transistor, 20
Line integral, 5 1-mH Inductor, 288–289
Lines of force, 6–8 Miller integrator, 302–303, 315–316
Line-to-neutral voltages, 559 Multiple-input amplifier circuit, 219–220
Line voltages, 564–565 Multiple-input and multiple-output (MIMO)
LM148-series amplifiers, 466 communication systems, 123
Load cell, 241
Load line method, Nonlinear circuits, 198–200
with ideal diode, 177, 178 N
iterative method Natural frequency, 367
explicit and implicit schemes, 178 Natural response, 356–357, 366
inverse function, 178 NC terminals. see Not connected terminals
for solar cell, 179–180 Near-field communication (NFC), 518
ν-i characteristic, 176 Near-field communications
Load match, 547 near-field wireless communication, 518
Load voltage, 357 near-field wireless link, 522–523
Load voltage amplitude, 537 receiver circuit, 520–521
Loopstick antenna, 523 transmitter circuit, 519–520
Lorentz force, 59, 327 Near-field wireless link, 518, 522–523
Lower and upper half-power frequencies, 504 Negative feedback
Low-frequency asymptotes, 456 amplifier circuit, current flow in, 218–219, 258
Low-pass filter (LPF), 451, 452, 527 amplifier feedback loop, 211
Index 659

dynamic process, 212–213 ohmic element, 47

idea of, 211 radiating monopole antenna, 46
mathematics, 217–218, 257–258 saturation current, 48
multiple-input amplifier circuit, 219–220, 259 Shockley equation, 48
summing-point constraints, 211 static resistance, 48–49
two summing-point constraints, amplifier circuit Non-ohmic circuit elements, 47–48
analysis using, 254–257 Non-STC transient circuit, 353–354
inverting amplifier, 215–216 Nonzero phase, 405
non-inverting amplifier, 213–215 Norton theorem for steady-state AC circuits, 428
voltage follower/buffer amplifier, 216–217 Not connected (NC) terminals, 204
voltage, current, transresistance and transconductance Numerical differentiation, 476–478
amplifiers with, 245–246
Networking theorems
arbitrary electric network, 98 O
electric network and topology, 127 Octave, 455
branch currents and voltages, 94 ODEs. see Ordinary differential equations
branches, 91 Offset currents, 231–232
loop, 92–93 Offset-null terminals, 230
mesh, 93 Ohmic conductor
nodes, 92 average carrier velocity, 37–39
series and parallel connections, 93 conductivities list, 38
Kirchhoff’s current law, 93–95, 127–128 total current, 36–37
Kirchhoff’s voltage law, 95–98, 128–129 Ohm’s law, 33, 218, 537, 544, 595
Maxwell’s minimum heat theorem, 98 Open-circuit transformer test, 618
network ports, 99 Open-circuit voltage gain, 204
Tellegen’s theorem, 98, 99 Open-loop AC amplifier gain, 465
NMOS transistor, 20 Open-loop amplifier gain, 463–464
Nodal analysis, 145, 185–186 Open-loop configuration, 204
circuit simulators, 145 power rails and voltage transfer
linear circuit characteristic in, 205
absolute node, 146 Open-loop gain behavior, 463–464
absolute values, 145 Open-loop voltage gain, 204
bridge circuit, 146 Operating at the same frequency, 427
current source, 148–149 Operational amplifier circuits
node voltages, 145 amplifier bandwidth, 467–468
voltage source, 147 amplifier IC, 468–469
supernode, 149–150, 186–187 closed-loop AC gain, model of, 466–467
Noise signals, 542 frequency bandwidth, 463
Nonideal digital waveform front-end amplifier, 463
modeling circuit, 370–372 open-loop AC gain, model of, 465–466
solution, 372–374 open-loop amplifier gain, 463–464
Nonideal transformer model, 618–621 unity-gain bandwidth vs. gain-bandwidth product,
Non-inverting amplifier, 213–215, 469 464–465
configuration, 467 Operational amplifier comparator, 206–207
Non-inverting input, 203 Ordinary differential equations (ODEs), 322
Non-inverting Schmitt trigger, 344 first-order, 355
Nonlinear circuits homogeneous second-order, 353
load line method, 198–200 second-order, 366
with ideal diode, 177, 178 Oscillation frequency, 346–347
iterative method, 178–180 Oscilloscope, 405, 408
ν-i characteristic, 176 Output offset voltage, canceling, 230–231
Nonlinear passive circuit elements Output resistance of amplifier circuit, 207, 223–224
dynamic resistance, 49–50 Output short-circuit current, 210
electronic switch, 50–51 Output terminal, 203
equivalent resistance, 46 Overdamping, 367
non-ohmic circuit elements, 47–48 Overshoot, 369–370
660 Index

P Power terminal, 204

Parallel connected second-order RLC circuit Power transfer, 195–198
duality, 365 maximum power theorem, 168–170
independent function, initial conditions power efficiency, 170–171
and choice of, 364 solar panel, 172–175
representation and qualitative operation, 363 transmitting antenna, 171–172
voltage, circuit equation in terms, 363–364 Power transfer function, 455
Parallel resonant RLC circuit, 505–507 Power triangle, 547–549, 576
Parallel RLC circuit, 516–517 Poynting vector, 17
Parallel RLC tank circuit, 505 Primary winding, 592
Particular solution, 356, 366 Proximity sensors, 523–525
Passband, 455, 511 Pulse width modulation (PWM), 370
Passive linear circuit elements, 290
Passive reference configuration, 32–33
Peak-to-peak value, 405 Q
Period, 404, 408 Q-factor, 501
Phase, 403, 408 Quality factor of the nonideal inductor, 521
of complex impedance, 420–422 Quality factor of the series resonant RLC circuit, 500
Phase transfer function, 456–457 Quasi-static magnetic dipole, 518
Phase voltages, 559–561, 563–564
Phasors
current, 409 R
definition of, 408–410 Radii of equivalent conductors, 571
diagram, 413–416 Radio-frequency identification (RFID), 518, 630
KVL and KCL on, 424–425 Railgun, 326–328
method, 521 Rails, 205
shorthand notation, 416 Rail-to-rail amplifiers, 206
steady-state AC circuit RC circuits, 322
amplitude, frequency and phase measurements, capacitor voltage, continuity of, 325–326, 379–381
408 digital memory cell, 329–330
complex exponent, shorthand notation for, 416 electromagnetic material processing, 328–329, 381
harmonic voltage and current, 403–405 electromagnetic railgun, 326–328, 381
leading and lagging, 405–407 energy-accumulating capacitor circuit, 330–332,
magnitude, 411–413 382–384
operations with phasor diagram, 413–416 energy-release capacitor circuit, 322–324, 379–381
phasor, definition of, 408–410 time constant of, 324–325, 379–381
to real signals, 411 RC voltage divider circuit, 447
from real signals to, 410–411 Reactive power, 545–547, 570, 576
voltage, 409, 411, 417 Realistic values, of input/output resistances and output
Pk-Pk value, 405 current, 210
Polar forms, 411–413 Real-transformer model
Positive feedback, 342 high-frequency transformer model, 622–623
bistable amplifier circuit with, 342–344 model parameters and extraction, 616–618
Positive polarity, 594 nonideal low-frequency transformer, 616
Potential energy of a unit charge, 5 nonideal transformer model, 618–621
Potential transformers, 615 transformer efficiency, 622
Potentiometers, 42, 222 voltage regulation, 621–622
Power amplifier, 224 Receiver circuit, 520–521
Power angle, 543–544 Rectangle rule, 473
Power conservation, 597 Rectangular forms, 411–413
Power factor correction, 577 Reduction of resistive networks, 104–105
Power factor correction capacitor, 551 Relaxation constant, 323, 334
Power gain, 460 Relaxation oscillator, 342
Power rails practice, 205–206 Resistances, 417
Power-related theorems, 129–130 and reactance, 546–547
Maxwell’s minimum heat theorem, 98 in series and in parallel, 131–133
Tellegen’s theorem, 98, 99 equivalent circuit element, 103–104
Index 661

parallel connection, 103 second-order band-reject RLC filter, 514–516

series connection, 102–103 second-order high-pass RLC filter, 512–514
Resistance values second-order low-pass RLC filter, 511–512
amplifier circuit design, 259 near-field communications
discrete resistance values and potentiometers, 222 near-field wireless communication, 518
gain tolerance, 222–223 near-field wireless link, 522–523
non-inverting and inverting configurations, receiver circuit, 520–521
221–222 transmitter circuit, 519–520
Resistive load, 576 proximity sensors, 523–525
Resistive sensors theory of
circuit symbols, 45 self-oscillating ideal LC circuit, 495–497
photocell, 43–44 series resonant ideal LC circuit, 497–498
photoresistor, 44 series resonant RLC circuit (see Series resonant
potentiometric position sensor, 44–45 RLC circuit)
sensitivity, 45 Second-order transient circuits, 354
strain gauge, 44 transient circuit fundamentals, 377–378
thermistor, 43 circuit current and capacitor voltage, initial
thermocouple, 43–44 conditions in, 361
Resonant circuit at the receiver (RX), 528 independent function, step response and choice of,
Resonant circuit at the transmitter (TX), 528 362–363, 393–394
Resonant frequency, 499 parallel connected second-order RLC circuit,
Reversal property, 470, 473 363–365, 394
Ringing in second-order circuits, 369 series-connected second-order RLC circuit,
Rise time, 369–370 358–360, 393
Rising edge, 372 types of, 358
RL circuits Second summing-point constraint, 211
energy-accumulating inductor circuit, 337–339, Self-capacitance, 9
386–387 Self-induced Lorentz force, 329
energy-release inductor circuit, 333–336, 384–386 Self-oscillating LC circuit, 496
energy-release RL circuit with voltage supply, Self-resonant, 524
339–340, 388 Semiconductor circuit components, 20
inductor current, continuity of, 336–337, 384–386 Sending-end voltages, 607
laboratory ignition circuit, 340–341, 388 Sensitivity threshold, 225
RL filter circuits, 460–462 Sensor’s equivalent circuit, 224
rms voltage and rms current, 538–539 Series and parallel network
Root-mean-square (rms), 539, 542 capacitance, 278–279, 310
combined voltage and current dividers, 137–138
current divider circuit, 111–113, 135–136
S current limiter, 111, 135
Sampling interval, 473 inductance, 287–288, 312
Sampling theorem, 475, 484 reduction of resistive networks, 104–105
Saturation mechanism, 342–343 resistances, 131–133
Sawtooth/triangular wave, 542 equivalent circuit element, 103–104
Sawtooth wave, 542 parallel connection, 103
Secondary winding, 593–594 series connection, 102–103
Second-order band-pass RLC filter, 508–510 sources, 130–131
Second-order band-reject RLC filter, 514–516 combinations of current sources, 102
Second-order high-pass RLC filter, 512–514 dual-polarity power supply, 101
Second-order low-pass RLC filter, 511–512 parallel battery bank, 101–102
Second-order ODE, 366 series-connected battery bank, 100–101
Second-order RLC circuits series vs parallel connection, 102
construction of voltage divider circuit, 133–134
band-pass and band-reject filters, 508 as actuatorcircuit, 109–110, 134–135
cutoff frequency, 508 KCL and KVL, 105
ideal filter, 508 as sensor circuit, 107–109, 134–135
parallel RLC circuit, 516–517 Wheatstone bridge, 113–114, 136–137
second-order band-pass RLC filter, 508–510 Series-connected PFC capacitor, 555
662 Index

Series-connected second-order RLC circuit Thévenin and Norton equivalent circuits, 427–429,
circuit current, solution, 360 443–444
mechanical analogy, 359 impedance
representation and qualitative operation, 358–359 concept of, 417–419, 439–442
Series resonant RLC circuit of human body, 422
bandwidth of, 502–505 magnitude and phase of, 420–422
duality, 505–507 physical meaning of, 419–420, 439–442
quality factor of, 500–502 phasors
resonance condition, 498–500 amplitude, frequency and phase measurements,
Series RLC circuit 408
complementary solution, derivation of, 366–367, complex exponent, shorthand notation for, 416
394–396 definition of, 408–410, 436–437
different damping ratios, solution behavior for, harmonic voltage and current, 403–405, 434–436
368–369, 396 leading and lagging, 405–407, 434–436
finding integration constants, 368, 396 magnitude, 411–413, 437–439
nonideal digital waveform, 370–374 operations with phasor diagram, 413–416,
overshoot and rise time, 369–370, 396–397 437–439
second-order ODE, 366, 394–396 to real signals, 411, 436–437
Series RLC tank circuit, 500 from real signals to, 410–411, 436–437
Short-circuit transformer test, 618 Steady-state current flow
Short/open circuit methods, 154–155 electric power transfer, origin of, 16–17
Signal amplifier, 224 and magnetostatics
Signal-flow diagram, 242 current flow model and electrostatics, 11–13,
Signals, 403 24–25
sinc function, 471 electric current, 11, 24–25
Single-ended sensors, 235 physical model of, 13–14, 24–25
Single input, 213 magnetostatics and Ampere’s law, 14–16
Source circuit, 602 Steinmetz model, 616, 640
Source frequency, 502 Step-down transformer, 599
Source impedances, 427, 567 Step-up transformer, 599
Sources in series and in parallel, 130–131 Stopband, 455
combinations of current sources, 102 Stored magnetic-field energy, 598
dual-polarity power supply, 101 Summing amplifier, 219–220
parallel battery bank, 101–102 Summing/difference node, 242
series-connected battery bank, 100–101 Summing point, 209
series vs parallel connection, 102 Summing-point constraints, 211
Source transformation, AC circuit analysis, 425–427 Superposition principle, 450, 565
Source transformation in the frequency domain, 425–426 Superposition theorem, 138–141
Source voltage, 427 for cellphone, 119–120
Spark gap radio, 341 character, 123
STC transient circuits, 376–377 independent sources, 116
circuits with resistances and capacitances, 349–351, linear circuit, 115
389–390 linearization procedure, 115
circuits with resistances and inductances, 351–353, linear superposition, 116
389–390 for multifrequency AC circuits, 429–430
circuit with bypass capacitor, 355–357, 390–392 nonlinear circuit, 115–116
non-STC transient circuit, 353–354, 390 open circuit, 116
with two inductances and two resistances, 354, 355, short circuit, 116
390 two-port networks, 122–123
Steady-state alternating current (AC) circuit, 403 voltage sources, 117, 118
AC circuit analysis Y (wye) and Δ (delta) networks, 140–141
complete solution for, 424–425, 442–443 balanced, 121–122, 140–141
KVL, KCL and equivalent impedances, 423–424, conversions, 121
442–443 series/parallel equivalents, 120
at single frequency, 429, 443–444 three-terminal networks, 120
source transformation, 425–427, 443–444 two-terminal networks, 120
superposition theorem, 429–430, 443–444 Surface charge density, 7
Index 663

Switching oscillators, 342 laboratory ignition circuit, 340–341, 388

Switching RC oscillator, 342, 345–346, 376, 389 second-order transient circuits, 377–378
bistable amplifier circuit with the positive feedback, circuit current and capacitor voltage, initial
342–344, 388–389 conditions in, 361
electronic oscillators, 342 independent function, step response and choice of,
oscillation frequency, 346–347, 389 362–363, 393–394
555 timer, 347–348 parallel connected second-order RLC circuit,
triggering, 344, 388–389 363–365, 394
series-connected second-order RLC circuit,
358–360, 393
T types of, 358
Telephone hybrid circuit, 612 series RLC circuit
Tellegen’s theorem, 98, 99 complementary solution, derivation of, 366–367,
Thévenin and Norton equivalents 394–396
circuit solution, 163–164 different damping ratios, solution behavior for,
dependent sources, 164–165 368–369, 396
using Them circuit solution, 161–163 finding integration constants, 368, 396
Thévenin equivalent method, 355–357, 502 nonideal digital waveform, 370–374
Thévenin impedance, 427 overshoot and rise time, 369–370, 396–397
Thévenin’s theorem for steady-state AC circuits, 427 second-order ODE, 366, 394–396
Thévenin voltage, 427 STC transient circuits, 376–377
Three-phase balanced wye-wye circuit, 565 circuits with resistances and capacitances,
Three-wire single-phase system, 613 349–351, 389–390
Threshold voltages, 344 circuits with resistances and inductances,
Time-averaged AC power, 538–540 351–353, 389–390
Time averaging, 537, 538 circuit with bypass capacitor, 355–357,
Time constant, 323, 334, 349, 367 390–392
of RC circuits, 324–325 non-STC transient circuit, 353–354, 390
Time-domain computational solution, 478–479 with two inductances and two resistances, 354,
555 timer, 347–348 355, 390
Timer integrated circuit (Timer IC), 348 switching RC oscillator, 342, 345–346, 376, 389
Time-varying sources, 84–85 bistable amplifier circuit with the positive
AC source polarity, 68 feedback, 342–344, 388–389
source amplitude, 67 electronic oscillators, 342
Total load impedance per phase, 567 oscillation frequency, 346–347, 389
Trailing edge, 372 555 timer, 347–348
Transconductance amplifiers, 245–246 triggering, 344, 388–389
Transcranial magnetic stimulation (TMS), 479 Transmission line, 13
Transfer function multiplication, 484 Transmitter circuit, 519–520
Transformer-based impedance matching, 640 Transresistance, 245–246
Transformer power efficiency, 622 Transresistance amplifier, 66, 245
Transformer rating, 599–600 Triangular wave, 542
Transient circuit fundamentals, 375 Triggering, 344, 388–389
RC circuits, 322 Trigger signal, 344
capacitor voltage, continuity of, 325–326, 379–381 Two independent parameters, 419
digital memory cell, 329–330 Two-port networks, 449, 508
electromagnetic material processing, 328–329, 381 Two summing-point constraints
electromagnetic railgun, 326–328, 381 amplifier circuit analysis using, 254–257
energy-accumulating capacitor circuit, 330–332, inverting amplifier, 215–216
382–384 non-inverting amplifier, 213–215
energy-release capacitor circuit, 322–324, 379–381 voltage follower/buffer amplifier, 216–217
time constant of, 324–325, 379–381 Typical transformer circuit, 597
RL circuits
energy-accumulating inductor circuit, 337–339,
386–387 U
energy-release inductor circuit, 333–336, 384–386 Undamped resonant frequency, 360, 496
energy-release RL circuit with voltage supply, Underdamping, 367
339–340, 388 Unity common-mode gain stage, 239
inductor current, continuity of, 336–337, 384–386 Unity-gain bandwidth, 464–465
664 Index

V W
Variable-gain amplifier, 222 Wattmeter, 550
Variac/variable AC transformer, 610 Well-designed transformer, 617
Vector product, 17 Wheatstone bridge, 113–114, 136–137
Virtual-ground circuit, 232–233 Whole voltage amplifier circuit model, 223–224
Virtual-ground integrated circuits, 233 Winding capacitances, 622
Voltage amplifier, 224, 245 Wireless communications, 17
amplifier circuit design Wye-connected load, 561
input load bridging, 224–226 Wye-connected source, 561
input load matching, 226–227 Wye-wye distribution system, 564
sensor’s equivalent circuit and amplifier’s Wye–wye power distribution system, 577–578
equivalent circuit, 224
Voltage difference, 5
Voltage divider circuit, 133–134 Y
as actuatorcircuit, 109–110, 134–135 Y (wye) and Δ (delta) networks, 140–141
KCL and KVL, 105 balanced, 121–122, 140–141
as sensor circuit, 107–109, 134–135 conversions, 121
Voltage divider rule, 106 series/parallel equivalents, 120
Voltage drop, 5 three-terminal networks, 120
Voltage follower/buffer amplifier, 216–217 two-terminal networks, 120
Voltage multiplication, 460
Voltage multiplier, 522
Voltage source, 18 Z
Volt-amperes (VA), 547 Zero-level detector, 207
Volumetric charge density, 7