Power Electronic System Design Linking Differential Equations Linear Algebra and Implicit Functions Keng C Wu Full Chapter PDF

Power Electronic System Design:
Linking Differential Equations, Linear

Algebra, and Implicit Functions Keng C.
Wu
Visit to download the full and correct content document:
https://ebookmass.com/product/power-electronic-system-design-linking-differential-eq
uations-linear-algebra-and-implicit-functions-keng-c-wu/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Linear Algebra and Partial Differential Equations T

Veerarajan
https://ebookmass.com/product/linear-algebra-and-partial-
differential-equations-t-veerarajan/
Transforms and Partial Differential Equations T.

Veerarajan
https://ebookmass.com/product/transforms-and-partial-
differential-equations-t-veerarajan/
First Course in differential equations (11ed) /

Differential Equations and Boundary Value Problems
(9ed) Solutions manual Dennis Zill Roberto Martinez
https://ebookmass.com/product/first-course-in-differential-
equations-11ed-differential-equations-and-boundary-value-
problems-9ed-solutions-manual-dennis-zill-roberto-martinez/
Differential Equations and Boundary Value Problems:

Computing and Modeling (Edwards, Penney & Calvis,
Differential Equations: Computing and Modeling Series)
5th Edition, (Ebook PDF)
https://ebookmass.com/product/differential-equations-and-
boundary-value-problems-computing-and-modeling-edwards-penney-
calvis-differential-equations-computing-and-modeling-series-5th-
Mathematical Physics with Differential Equations
Yisong. Yang
https://ebookmass.com/product/mathematical-physics-with-
differential-equations-yisong-yang/
Advanced Differential Equations 1st Edition Youssef

Raffoul
https://ebookmass.com/product/advanced-differential-
equations-1st-edition-youssef-raffoul/
Schaum's Outline of Differential Equations, Fifth

Edition Bronson
https://ebookmass.com/product/schaums-outline-of-differential-
equations-fifth-edition-bronson/
Power System Analysis and Design 6th Edition J. Duncan

Glover
https://ebookmass.com/product/power-system-analysis-and-
design-6th-edition-j-duncan-glover/
Algebra Linear 2nd Edition Alfredo Steinbruch
https://ebookmass.com/product/algebra-linear-2nd-edition-alfredo-
steinbruch/
POWER ELECTRONIC
SYSTEM DESIGN
Linking Differential
Equations, Linear Algebra,
and Implicit Functions
KENG C. WU
Switching Power, Inc. Ronkonkoma, NY, United States
POWER ELECTRONIC
SYSTEM DESIGN
Linking Differential
Equations, Linear Algebra,
and Implicit Functions
Elsevier
Radarweg 29, PO Box 211, 1000 AE Amsterdam, Netherlands
The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom 50
Hampshire Street, 5th Floor, Cambridge, MA 02139, United States
Copyright © 2021 Elsevier Inc. All rights reserved.
No part of this publication may be reproduced or transmitted in any form or by any means,
electronic or mechanical, including photocopying, recording, or any information storage
and retrieval system, without permission in writing from the publisher. Details on how to
seek permission, further information about the Publisher’s permissions policies and our
arrangements with organizations such as the Copyright Clearance Center and the
Copyright Licensing Agency, can be found at our website: www.elsevier.com/permissions.
This book and the individual contributions contained in it are protected under copyright
by the Publisher (other than as may be noted herein).
Notices
Knowledge and best practice in this field are constantly changing. As new research and
experience broaden our understanding, changes in research methods, professional practices,
or medical treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described
herein. In using such information or methods they should be mindful of their own safety
and the safety of others, including parties for whom they have a professional responsibility.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or
editors, assume any liability for any injury and/or damage to persons or property as a matter
of products liability, negligence or otherwise, or from any use or operation of any methods,
products, instructions, or ideas contained in the material herein.
British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library of Congress
ISBN: 978-0-323-88542-3
For Information on all Elsevier publications visit our website at

https://www.elsevier.com/books-and-journals
Publisher: Graham Nisbet

Acquisition Editor: Graham Nisbet
Editorial Project Manager: Grace Lander
Production Project Manager: Prasanna Kalyanaraman
Cover Designer: Brian Salisbury
Typeset by Aptara, New Delhi, India

Dedicated to
Grandson
Mitchell Patrick Wu Cayanni
Contents
About the Author xi

Preface xiii
1. Capacitor and inductor 1

1.1 Capacitor equation in differential form 1
1.2 Capacitor equation in integral form 2
1.3 Inductor equation in differential form 3
1.4 Inductor equation in integral form 4
1.5 Definition of inductance and Faraday’s law 4
1.6 Magnetic coupling and mutual inductance 5
1.7 Transformer equation 7
1.8 Nonideal capacitor, nonideal inductor, and equivalent circuit 10
1.9 Transformer equivalent circuits 11
1.10 Physical size of capacitor and inductor 13
1.11 Specifications for capacitor and inductor 15
2 First-order circuits 19
2.1 RC network with periodic drive source 19
2.2 Sawtooth (triangle ramp) generator 30
2.3 Full-wave rectifier with RC load 33
2.4 A brushless DC Motor with permanent magnets rotor 38
2.5 A BLDC motor speed detector 45
References 47
3 Current source 49
3.1 Semiconductor diode equation 49
3.2 Simple current source 50
3.3 Bob Widlar current source 54
3.4 Improved current source 58
3.5 Source impedance 60
3.6 555 timer 64
3.7 Precision current loop 70
3.8 Current-mode laser driver 74
3.9 LED array driver 76
3.10 JFET current source 77
3.11 MOSFET current source 78
vii
viii Contents
4 Second order 81
4.1 Form 81
4.2 Root 83
4.3 Time domain 85
4.4 Frequency domain 89
4.5 Parallel and serial resonance 92
4.6 Eigen value approach 103
4.7 RC filters and Sallen–Key filters 104
4.8 Power filters 111
4.9 Oscillator 113
4.10 Implicit function 120
5 Gain blocks 123

5.1 Class-A direct-coupled bipolar transistor amplifiers 123
5.2 Class-AB, B, C bipolar transistor amplifiers 129
5.3 Transformer-coupled transistor amplifiers 133
5.4 Class-D switch-mode power amplifiers 135
5.5 Pulse width modulator 139
5.6 Digital (clocked) window comparator 140
5.7 Linear operational amplifiers 142
5.8 Tuned amplifiers and implicit function 147
5.9 Composite nonlinear operational amplifiers 150
5.10 Unity-gain bandwidth of op-amp 153
5.11 Large signal gain of op-amp 156
6 Feedback approaches 167

6.1 Voltage feedback 167
6.2 Current feedback 170
6.3 PID feedback 175
6.4 State feedback 178
6.5 Feedback isolation 180
7 Control practices 189

7.1 Level control 189
7.2 Mode control 190
7.3 Zone control 192
7.4 Variable structures 193
7.5 Sensor 196
7.6 Open loop 198
Contents ix
7.7 Close loop 200

7.8 Loop contention 203
7.9 Time control 204
7.10 Sequential time control 206
8 Linear regulator 213

8.1 Bipolar series voltage regulator 213
8.2 MOSFET series voltage regulator 223
8.3 Multiple implicit function approach 227
8.4 Design procedure for loop stability 228
8.5 Design procedure for error amplifiers 230
8.6 Current-mode laser driver design procedure 236
8.7 Shunt regulators 238
9 Switch-mode DC/DC converters 241

9.1 Power filter, inductor, and capacitor 243
9.2 Fundamental topologies 249
9.3 Operational dynamics of basic buck topology 254
9.4 Operational dynamics of basic boost topology 257
9.5 Operational dynamics of basic flyback converter 259
9.6 Cascaded converter—nonisolated 261
9.7 Isolated converter—forward converter 264
9.8 Isolated converter—half-bridge converter 269
9.9 Isolated converter—push–pull converter 272
9.10 Isolated converter—full-bridge converter 272
9.11 Isolated converter—quasi-resonant converter 273
9.12 Analog feedback 275
9.13 Close loop—analog 288
9.14 Close loop—digital 296
10 AC drives, rectification, and inductive loads 299

10.1 Reexamine RC-loaded rectifier 299
10.2 AC drive with unidirectional RL load 301
10.3 Half-wave AC drive with nonpulsating current feeding RL load 304
10.4 Full-wave AC drive with nonpulsating current feeding RL load 305
10.5 Phase-controlled AC drive with RL load 307
10.6 Phase-controlled AC drive with free-wheel diode and RL load 309
10.7 Phase-controlled full-wave AC drive with RL load 311
10.8 Three-phase circuits 313
x Contents
11 Rotation, three-phase synthesis, and space vector concepts 319

11.1 Magnetic field (flux) 319
11.2 Synthesis of three-phase sources and inverters 323
11.3 Vector concept 331
Appendix A Accelerated steady-state analysis for a parallel resonant

network fed by nonsinusoidal, half-wave rectified current 347
Appendix B Matrix exponential 349
Appendix C Example 4.7 MATLAB m-file 351
Appendix D Example 8.1 353
Appendix E A general mass-spring-dashpot second-order system;

first alternative 359
Appendix F A general mass-spring- dashpot second-order system;

second alternative 363
Appendix G A general mass-spring- dashpot second-order system;

third alternative 365
Appendix H Matrix exponential—Jordan form 367
Appendix I A step-by-step primer on digital power-supply design 369

Digital tides 369
Tumble to digital 369
Roadmap to digital 370
Navigate to digital filter 371
Work out a forward converter example 373
Implementation 377
Conclusion 380
References 381
Appendix J Motor winding driven by SCR phase-controlled

sine source 383
Index 385
About the Author
Keng C. Wu, a native of Chiayi( ), Dalin( ), Taiwan, received the B.S.

degree from Chiaotung University, Taiwan, in 1969 and the M.S. degree
from Northwestern University, Evanston, Illinois in 1973.
He was a lead member, technical staff, of Lockheed Martin, Moorestown,
NJ. He has published five books: Pulse Width Modulated DC-DC Converters
Chapman & Hall, January 1997; Transistor Circuits for Spacecraft Power System
Kluwer Academic Publishers, November 2002; Switch-mode Power Converters:
Design and Analysis Academic Press, Elsevier, November 2005; Power Recti-
fiers, Inverters, and Converter Lulu.com November 2008.; Power Converters with
Digital Filter Feedback Control, Elsevier, Academic Press, 2016. He holds a
dozen US patents, was awarded Author of the Year twice (2003 and 2006
at Lockheed Martin), and presented a 3-hour educational seminar at IEEE
APEC-2007 S17.
xi
Preface
Years ago, Prof. Emeritus Chi-Tsong Chen, the author of Linear System
Theory and Design, a very successful textbook (Oxford University Press), met
the author at his Flushing, New York residence. In the meeting, and in the
preface of Signals and Systems – A Fresh Look his last publication (PDF form
free to all globally), Prof. Chen lamented that “Feedbacks from graduates
that what they learned in university is not used in industry prompted me to
ponder what to teach in signals and systems.”
Sadly,and based on long professional career serving RCA/GE/Lockheed
Martin space sector, the author can definitively confirm the fact Prof.
Chen was sad about. The less-than-desirable state had existed, and is still
present,in the form that many degree-holding engineers including electrical,
electronic, mechanical, and other specialties are falling short in applying
mathematical tools they were taught in college. Given electrical schematic
drawings, they were unable to formulate and express systems’ dynamics in
state variables and state transition using the first-order differential equations
and linear algebra technique. As a result, they were unable to boost their
productivity using software such as MATLAB.
This book intends to bridge the gap—what is taught in college and how
it is being applied in industry. In essence, this writing shall be considered
didactic.
It begins with Chapter one giving capacitors and inductors, two indis-
pensable energy storage components, an in-depth examination from the
view point of the first-order derivative, its corresponding integral form,
and its physical implications. Chapter two covers RC- and RL-type net-
works governed by a single differential equation. Key steps moving system
differential equations to Laplace transform in a frequency domain and to
a state-space transition form are introduced. Along the way, unconven-
tional approaches deriving Fourier series, explaining orthogonal property,
or treating boundary value problems are also explored. Chapter three covers
current sourcing circuits including current mirror, the workhorse of analog
integrated circuits, and precision current generator loops critical to instru-
mentation. Chapter four extends Chapter two to networks of second order
governed by two first-order differential equations. Procedures transforming
multiple differential equations to Laplace form, to state-transition form, and
to state-transition solution are shown. Chapter five examines circuit blocks
xiii
xiv Preface
and modules performing amplification, voltage-to-time window, duty cycle

modulation, etc. Chapter six covers feedback practices including voltage,
current, isolation, summative current, subtractive current, and state feedback.
Chapter seven discusses configurations of control loops including single loop,
multiple loop, open loop, closed loop, nested loop, loop contention, etc.
Chapter eight deals with linear regulators including series voltage regulator
and current shunt in parallel. Chapter nine explores switch-mode power
processing. Chapter ten presents complexities arising from inductive load fed
by rectified AC sources of single phase, multiple phases, and phase control.
Employing the concept of electromagnetic vectors in space, Chapter eleven
focuses on the formation of magnetic flux vector placed intentionally along
selected orientation, time-varying flux intensity, and rotational flux vector
that makes motor spin.
Considering the writer’s goal is to bridge materials taught in college and
applications of the material in actual industrial settings, the topics outlined
above and organized in that particular order are suitable for college seniors
and novice professionals in the industry. Following the material, and when
facing a real-world design schematic, readers will be able to (1) assign state
variables (circuit node voltage, inductor branch current), (2) write down
multiple differential equations, (3) place equation set in a state-transition
form, (4) select the approach one is more comfortable and confident, for the
time being, (5) obtain system response solutions corresponding to various
drives in different time frames, (6) stitch together a steady-state response
solution in closed-form analytical expressions.
Given time and practice, and when facing system order exceeding three,
most readers will quickly realize that state-transition equation and solution
invoking matrix operation delineated in linear algebra are more effective,
even elegant, in handling high-order systems.
This writer had definitely experienced that awareness, and expects all
reader to do the same.
As indicated in the subtitle of this writing, along the presentation,
mathematical notes are inserted where appropriateness is not violated. Quite
a few may be considered unconventional. This is done in the spirit of never
taking authority dogmatically—a true open mind respecting the unlimited
possibilities of viewing nature from multiple angles and a belief that what
Preface xv
was said true in the past may not be true in the future when new discoveries
see the daylight.
On the backdrop of the above conviction, this author took additional
efforts to make this writing also available in Chinese language;thanks to pub-
lisher Elsevier for granting such translation right. Thanks are also extended
to Mr. , at ITRI (Industrial Technology Research Institute, Hsinchu Taiwan),
who had performed the translation, a very demanding task considering the
limitations of Chinese language in handling technical subjects.
With the advance of miniaturized electronic hardware and supercom-
puter equipped with mathematical co-processors, engineering design tasks
are now mostly carried out by the simulation and computation. The
implementation of both always requires design formulation in the form
of analytical expressions based on, in most cases, systems of differential
equations with coefficients depending on components/parts values.
In the course of almost four decades‘Ł‘™professional career in aerospace
industries, the author had definitely derived significant benefits from follow-
ing the path outlined above.
You, readers, can certainly do the same.
Keng C. Wu
Princeton, NJ.
Dec. 2020
CHAPTER 1
Capacitor and inductor

Two components, capacitors and inductors, play irreplaceable roles in elec-
trical power processing for their energy-storage properties.By presenting the
analytical equations governing both in either differential or integral forms,
this chapter illuminates the electromagnetic behaviors of those devices and
elucidates its physical significance when working with driving sources.
MATH. NOTE: In most calculus textbooks, derivatives and integrals are

introduced in the forms of f´(x) = dy/dx = df(x)/dx and ∫ydx = ∫f(x)dx,
given y = f(x) a two-dimensional plane curve and x is the independent vari-
able, with little physical meaning attached except the concept of “tangential
slope,” associated with the derivative, and “geometrical area,” associated with
the integral, employing the approach of limit. The independent variable x is
by no means restricted to signifying only space quantity.It certainly can stand
for time,and many other variables as well.The simple act of replacing dx with
dt, an infinitesimal time increment, introduces interesting, and important,
physical meaning to derivative f´(t) = dy/dt = df(t)/dt. As dt appears in the
numerator (inverse of time), derivative against time yields the dimension of
speed, velocity, and/or frequency; the temporal changes of a time-dependent
variable. ♣
1.1 Capacitor equation in differential form

Almost without exception, the action of capacitors is introduced in text-
books in a differential form; which links current through the device and
time rate of voltage change across it with a positive sign as shown.
MATH. NOTE: At a more fundamental level, the current is expressed as

the rate of charge carriers’ changes, i(t) = dQ(t)/dt, in which Q(t) = Cv(t) and
C, the capacitance and a constant within reason, is a function of geometry
and material property. ♣
i
dv(t ) +v
i(t ) C (1.1)
dt
Power electronic system design. Copyright © 2021 Elsevier Inc.
DOI: 10.1016/B978-0-32-388542-3.00004-2 All rights reserved. 1
2 Power electronic system design
v
i
t
Fig. 1.1 Terminal current and voltage of a capacitor.
What does this form tell us about a capacitive element?

1. When dv/dt = 0, that is when the device voltage reaches an extreme,
a maximum or a minimum, the corresponding device current crosses zero
value. Stated differently, the device’s time-domain current waveform makes
a zero-crossing at the time its corresponding voltage waveform peaks, or
bottoms out. In other words, and in a graphical form, terminal voltage and
through current for a capacitor must hold a relation as shown in Fig. 1.1.
2. As the voltage variable in Eq. (1.1) appears as a derivative, the current
variable does not change its value if Eq. (1.1) is rewritten as
d[VDC + v(t )]
i(t ) = C (1.2)
dt
In this form, one important property of capacitor stands out. That is, the
device sustains a DC (direct current) voltage, VDC , which however does not
contribute to its current. The significance of this attribute is that a capacitor
blocks DC current. Or, DC current does not flow through a capacitor. Only
AC (alternating) current does.
3. Capacitor allows the application of a DC voltage within limit; the
breakdown voltage.
1.2 Capacitor equation in integral form

Eq. (1.1) can of course be rewritten as

1 1 t
v(t ) = i(t )dt = V0 (t0 ) + i(τ )dτ (1.3)
C C t0
MATH. NOTE: This is actually a rewording of the previous MATH
NOTE, that is, charge is equal to the time integral of current. ♣
In contrast to the derivative form, the integral form (Eq. 1.3), in particular
the right-hand side, conveys an extremely important effect of the capacitive
state variable: voltage.
4. In a very straightforward manner, it declares the continuous nature of
capacitor voltage.
Capacitor and inductor 3
t
v
Fig. 1.2 Terminal current and voltage of an inductor.
5. In addition, Eq. (1.3) implies that capacitor voltage is lagging its

current in time domain:voltage phase lags (The integral of a time-dependent
variable yields accumulation of its past acts.) Or, capacitor current is leading
its voltage across: current phase lead.
6. Furthermore, in a subtle way, Eq. (1.3) indicates that a capacitor stores
energy in stationary charge space: E-field.
1.3 Inductor equation in differential form

MATH. NOTE: At a more fundamental level, the inductive voltage will
be expressed as the rate of magnetic flux changes, v(t) = d(t)/dt, in which
(t) = Li(t) and L, the inductance and a constant within reason, is a function
of geometry and material property. This will be expounded in one of the
section to follow. ♣
By the same token, inductor and its electrical symbol was always intro-
duced by the following
i
di(t ) +v
v(t ) L (1.4)
dt
With reasoning similar to those given for capacitors, the following

electric behaviors are expected of an inductor.
1. When di/dt = 0, that is when the device current reaches an extreme,
a maximum or a minimum, the corresponding device voltage crosses zero
value. Stated differently, the device’s time-domain voltage waveform makes
a zero-crossing at the time its corresponding current waveform peaks, or
bottoms out. In other words, and in a graphical form, terminal voltage and
through current for an inductor must hold a relation as shown in Fig. 1.2.
2. As the current variable in Eq. (1.4) appears as a derivative, the voltage
variable does not change its value if Eq. (1.4) is rewritten as
d[IDC + i(t )]
v(t ) = L (1.5)
dt
In this form, one important property of inductor stands out. That is, the
device allows a DC current, IDC , which however does not contribute to its
voltage.
3. The allowed DC current is however constrained within a limit; the
magnetic core saturation and winding wire Ampere rating.
1.4 Inductor equation in integral form

Eq. (1.4) can also be rewritten as

1 1 t
i(t ) = v(t )dt = I0 (t0 ) + v(τ )dτ (1.6)
L L t0
Again in contrast to the derivative form, the integral form (Eq. 1.6), in
particular the right-hand side, conveys an extremely important effect of the
inductive state variable: current.
4. It states the continuous nature of inductive current.
5. In addition, (Eq. 1.6) implies that inductive current is lagging its voltage
in time domain: current phase lag. Or, inductive voltage is leading its through
current: voltage phase lead.
6. Furthermore, in a subtle way, Eq. (1.3) indicates that an inductor stores
energy in circulating charge: H-field.
1.5 Definition of inductance and Faraday’s law

Faraday’s law is also generally presented in a differential form and named
“Faraday’s Law of Induction”: An electromagnetic force, emf, is generated by
a time-varying magnetic flux, e = d/dt. This form may best be interpreted
as magnetic induction on a single turn (loop). For a multiple-loop device,
the case of a practical inductor with N turns, the total induced emf sums up
to v = Ne = d(N)/dt.
Placed in an integral form, Faraday’s law of induction with multiple
loops in series becomes λ = N = ∫vdt: Total flux linkage equals the time
integral of a driving (voltage) source. For a practical device consisting of
a magnetic core, which serves to guide and focus flux, and an N-loop
winding, which passes current, each loop experiences an identical flux, .
With known cross-sectional core area, Ac , supplied by core manufacturer,
the total flux linkage is expressed as λ = NAc B, in which B stands for flux
density. It is then understood in physics that, given a core material with
permeability μ = μo μr (μo vacuum permeability, μr relative permeability), flux
density is proportional to the magnetic field intensity, H, generated by a

driving source in the form of current, I, while permeability serves as the
proportional constant. Therefore, the total flux linkage is further expressed
as λ = NAc μo μr H.
Here, another electromagnetic law kicks in. That is Ampere’s Law:
magneto-motive force Ni = ∫Hdlm , the contour integral of magnetic field
intensity along a closed path lm . The integral may be simplified to HLm for
a geometrically isotropic and materially homogeneous core.
This additional attribute allows us to step further and rewrite the total
flux linkage as λ = NAc μo μr Ni/Lm , or λ = (N2 μo μr Ac /Lm )i. Plugging into
Faraday’s law of induction, we have v = (N2 μo μr Ac /Lm )di/dt. Referring to
Eq. (1.4), inductance is therefore L = N2 μo μr Ac /Lm for an inductor with core
material property μo , μr and core geometry Ac ,Lm ; core area and magnetic
path length.
As given, parameters determining inductance value are grouped in two;
(1) inductance index equal to the product of material property and core area
to magnetic path length ratio, and (2) square of winding turn number. In
industrial sector, core inductance index is often assigned a symbol AL .
1.6 Magnetic coupling and mutual inductance

In previous section citing Faraday’s law of induction, “time-varying mag-
netic flux” is mentioned as the key to magnetic induction. However,
“variable magnetic flux” may be more fitting as there are basically three
mechanisms to create variable magnetic flux.
A flux source may be moved mechanically in space while the inductive
reception loop stays stationary. Or, the flux generator stays stationary while
the reception loop is moved mechanically. Either way, a relative spatial
motion exists between two entities and that is actually the bottom line
of magnetic induction. Hydropower, coal-fired power, and other turbine
generators all fall under this category. In conventional physics text, it is
named “generator action.”
But the third way involves no mechanical motion at all. Both the flux
source and the reception loop remain stationary in space while the flux
generation side is driven by a true time-varying electric source. By so doing,
a time-varying flux is also created. In conventional physics, it is named
“transformer action.”
Employing the “Transformer action,” multiple windings, each with
multiple turns, are built wrapping around a common ferrous core. One
i1 i2
v1 v2
N1 N2
Fig. 1.3 Ferrite cup-cored transformer and its electrical symbol.
B-H curve
0.4 0.4
0.2
B( H)
Tesla
0 0
−0.2
−0.4
−0.4
−400 −200 0 200 400
−400 H 400
Oersted
Fig. 1.4 A B-H curve example without hysteresis.
winding, the primary, serves as a flux source driven by an external voltage

while the rest, the secondary, act as output windings. Fig. 1.3 shows a ferrite
cup-cored transformer, a two-winding and a multi-winding transformer
symbols.
Based on the two-winding transformer, Ampere’s law establishes a mag-
netic field intensity (N1 i1 + N2 i2 )/Lm in the core. When only winding 1 is
driven while winding 2 remains open, i2 = 0, Eq. (1.7) holds
N1 i1 Aw1
λ1 = N1 1 = N1 Aw1 B = N1 Aw1 μoμr = N12 μoμr i1 = L1 i1
Lm Lm
(1.7)
where Aw1 stands for winding 1 cross-sectional area.
Under the same driving condition, Eq. (1.8), a cross-coupled flux linkage,
holds for winding 2, assuming Aw2 < Aw1 .
N1 i1 Aw2
λ2 = N2 2 = N2 Aw2 B = N2 Aw2 μoμr = N2 N1 μoμr i1
Lm Lm (1.8)
λ2 = M21 i1
Next, we move driving source to winding 2 and open winding 1, i1 = 0.

Eq. (1.9), again a cross-coupled flux linkage, holds for winding 1. Readers
are cautioned to note the assumption that Aw2 < Aw1 .
N2 i2 Aw2
λ1 = N1 1 = N1 Aw2 B = N1 Aw2 μoμr = N1 N2 μoμr i2
Lm Lm (1.9)
λ1 = M12 i2
And, Eq. (1.10) holds for winding 2
N2 i2 Aw2
λ2 = N2 Aw2 B = N2 Aw2 μoμr = N22 μoμr i2 = L2 i2 (1.10)
Lm Lm
Clearly, a mutual inductance exists between the two windings. That is
Aw2
M = M12 = M21 = N1 N2 μoμr (1.11)
Lm
An interesting effect is also observed, that is the product of L1 and L2 .
Aw1 2 Aw2
L1 L2 = N12 μoμr N2 μoμr
Lm Lm
(1.12)
Aw1 Aw2 Aw1 2
= N1 N2 μoμr N1 N2 μoμr = M
Lm Lm Aw2
or,

Aw2 √
0<k= < 1, M = k L1 L2 (1.13)
Aw1
Here an imperfect attempt indicates the origin of coupling coefficient,
k. Imperfect because perfectionist theoretician may mount a challenge
against the use of winding area in place of core area. However, considering
the imprecise nature of magnetism, practitioners in the industrial sectors
are willing to accept Eq. (1.13) knowing there are other flux leakages
unaccounted for.
1.7 Transformer equation

In previous sections, the linkage, represented by symbol μo , μr , between flux
density and magnetic field intensity was freely invoked without qualification.
Unfortunately, nature always places limitation on everything. Permeability
of material is not exempted either. It is well understood, and documented,
that ferromagnetic materials exhibit highly nonlinear and hysteric properties
when magnetized. The nonlinearity may be expressed in an analytic form

by borrowing the Fermi–Dirac (electron density) distribution, (Eq. 1.14), and
modifying it.
In solid-state physics, the Fermi–Dirac function is given as
1
p(E ) = E−E f
(1.14)
1+ e kT
−23
in which k = 1.380658 × 10 [Joule/degree], T temperature in degree
Kelvin, Ef Fermi energy.
With slight modifications based on Eq. (1.14), Eq. (1.15) describes the
nonlinear magnetic property for ferrous material.
⎛ ⎞
⎜ ⎟
⎜ 2 ⎟
B(H ) = Bsat ⎜ − 1⎟ (1.15)
⎝ −
H
Bsat
⎠
2μoμr
1+e
in which Bsat is the saturation flux density, specific to a material, in Tesla
(MKS unit]; H the driving magnetic field intensity in Oersted (MKS unit].
For example, given μo = 4π × 10−7 [Henry/m, MKS uint], μr = 2500 [unit-
less], Bsat = 0.25 [Tesla], Eq. (1.15) yields Fig. 1.4 plot.
With further modifications, hysteretic behavior is also accounted for,
(Eq. 1.16), depending on either increasing or decreasing magnetic field.
⎛ ⎞
⎜ ⎟
⎜ 2 ⎟
B(H, Ho ) = Bsat ⎜ H −Ho
− 1⎟ (1.16)
⎝ −
Bsat
⎠
2μoμr
1+e
In the case of increasing, the B-H curve traverses the right ascending leg,
while the left descending leg covers decreasing field (Fig. 1.5).
The main point of focus is that the magnetic flux density for any ferrous
material is limited to an upper bound, Bsat , beyond which the material can
no longer support additional flux change. Once a drive condition, that is,
the volt-second integral, exceeds the limit, the device/material permeability
drops to zero. As a result, the device incorporating the material no longer
offers inductance.Eq.(1.4) then states that,with near-zero inductance,a huge
di will also results. Such a surge current, limited only by winding resistance,
will lead to device destruction. And, this is exactly the reason the following,
flux density of transformer core, is discussed.
B-H curve
0.4 0.4
0.2
B(H, 50)
Tesla 0
0
B(H, −50)
−0.2
−0.4
−0.4
−400 −200 0 200 400
−400 H 400
Oersted
Fig. 1.5 A B-H curve example with hysteresis.
MATH. NOTE: What Fig. 1.5 shows is a multivalue function. At a selected

H coordinate, there are two corresponding flux density values. At a selected
B coordinate, there are two corresponding field intensity values. ♣
Refer now to Fig. 1.3 transformer. Suppose an AC source v(t) = Va cos(ωt)
is driving the primary winding N1 . The integral form of Faraday’s law of
induction gives a total flux linkage
√
Va 2 Vrms
λ = Va cos (ωt )dt = sin (ωt ) = sin (ωt ) (1.17)
ω ω
As discussed above and considering the allowable saturation flux density,
the flux linkage magnitude of Eq. (1.17) must satisfy Eq. (1.18)
√ √
2 Vrms 2 Vrms
|λ| = = ≤ N1 Ac Bsat (1.18)
ω 2π f
where f stands for a driving source frequency. In other words, the driving
source magnitude, in RMS, must be constrained to
2π
Vrms ≤ √ f N1 Ac Bsat ≈ 4.443 f N1 Ac Bsat (1.19)
2
Otherwise, the transformer core will be saturated if the drive exceeds the
limit.
In power processing, in addition to sinusoidal drive, symmetrical 50%
square wave drive, dotted trace in Fig. 1.6, is also a very popular source.
Readers are invited to prove that, in this case, the source magnitude is
Va
flux
−Va
Fig. 1.6 Square-wave drive and core flux function.
Va Va
0
−Va −Va
Fig. 1.7 Rectangular wave drives.

constrained by Eq. (1.20) and the core flux (per turn) traces the solid curve
given in Fig. 1.6.
Vrms ≤ 4 f N1 Ac Bsat (1.20)
At this point, a moment of caution is warranted. Both Eqs. (1.19) and
(1.20) invoke drive source in RMS quantity. This practice entails easily mis-
understood mistake. Specifically, Eq. (1.19) is applicable solely for sinusoidal
drive while Eq. (1.20) for symmetrical, 50%, square-wave drive. For drive
waves, such as those given in Fig. 1.7, conversion factors existing between
drive amplitude,Va ,and its equivalent RMS quantity will alter the numerical
factor, 4.443 or 4, that appears in Eqs. (1.19) and (1.20). Readers are again
invited to work out the flux profile for drive waves shown in Fig. 1.7.
1.8 Nonideal capacitor, nonideal inductor, and

equivalent circuit
So far, all discussions concerning capacitors and magnetic devices are carried
out in the context of idealistic, conceptual space on paper. To be of practical
use, both must be built and brought to face real world. Capacitors are made
of bilayer electrodes, which must extend externally via metal contacts, while
inductors of spiral metal wires. Therefore, serial resistances are inevitably
introduced. This is not all; even a short piece of wire has inductance, however
small. What else? Capacitance exists between adjacent wires. Surfaces of
device package also create sneaky, leaky path. Taking in all, a real capacitor
or inductor may be represented by a lumped model (Fig. 1.8).
As shown, a view begins to emerge that neither device would work
as purely as what it is theoretically. Depending on the frequency range it is
immersed in a capacitor/inductor actually acts like a complex RLC network
(Fig. 1.8). We will come back to this point later.
(a) (b)
Fig. 1.8 Lumped model of real (a) capacitors and (b) inductors.
1.9 Transformer equivalent circuits

In the same spirit, two-wind transformer (Fig. 1.3) can be transformed into
equivalent circuits. However, before proceeding, the drawing will be slightly
modified (Fig. 1.9) to include effects of mutual inductance. Readers should
also refer to physics textbooks for the dot convention invoked in the figure.
For the time being and in the following, winding resistances are omitted
to simplify the derivation of equivalence.
One sees easily that N1 loop gives
di1 di2 di1 di1 di1 di2

v1 = L1 +M = L1 −M +M +M
dt dt dt dt dt dt
(1.21)
di1 d(i1 + i2 )
= (L1 − M ) +M
dt dt
And, N2 loop gives
di2 di1 di1 di2 di2 di2

v2 = L2 +M =M +M + L2 −M
dt dt dt dt dt dt
(1.22)
d(i1 + i2 ) di2
=M + (L2 − M )
dt dt
Both lead to the loop T-form equivalent circuit (Fig. 1.10).
i1 M i2
v1 v2
N1 N2
Fig. 1.9 Two-winding transformer with mutual inductance.

1:1
i1 L 1 - M L2 - M i2
v1 M v2
Ideal
Fig. 1.10 Loop T-form model for two-winding transformer.
1:1
i1
(L1L2 - M 2 )/ i2
v1 M
v2
(L1L2 - M 2 )/ (L1L2 - M 2 )/ Ideal
(L2-M ) (L1-M )
Fig. 1.11 Nodal π -form model for two-winding transformer.
Taking Laplace transform for the left most parts of Eqs. (1.21) and (1.22),
we reach a new equation set
v1 = sL1 I1 (s) + sMI2 (s) v2 = sMI1 (s) + sL2 I2 (s) (1.23)
Eq. (1.23) enables us to express I1 and I2 in terms of v1 and v2 .
v1 sM
v2 sL2 sL2 v1 − sMv2 (L2 − M )v1 + M (v1 − v2 )
I1 = = 2 =
sL1 sM s L 1 L2 − M 2 s L 1 L2 − M 2
sM sL2 (1.24)
v1 (v1 − v2 )
= +
−M
s L1 L2M−M
2 2
s L1LL22−M
sL1 v1
sM v2 sL1 v2 − sMv1 (L1 − M )v2 + M (v2 − v1 )
I1 = = 2 =
sL1 sM s L 1 L2 − M 2 s L 1 L2 − M 2
sM sL2 (1.25)
v2 (v2 − v1 )
= +
L1 L2 −M 2
s L1 L2M−M
2
s L1 −M
Eqs. (1.24) and (1.25) indicate a nodal π -form equivalent circuit

(Fig. 1.11).
Both the loop form and the nodal form transformer equivalent circuits
do not find them useful as it calls out all components and parameters.
1:1
i1 (1-k 2)L 1 i2
v1 k 2L 1 v2
Ideal
Fig. 1.12 Hybrid τ -form model for two-winding transformer.
A keen observation changes the situation. By making L2 – M = 0 in

Fig. 1.10 and recognizing M = kæ(L1 L2 ), M = L2 becomes k2 L1 . As a result,
Fig. 1.10 becomes Fig. 1.12; the hybrid τ -form equivalent circuit.
This form has a loop and a node equation with two parameters, coupling
coefficient and primary inductance, involved.
1.10 Physical size of capacitor and inductor

Based on the fundamental principle dealing with electric charge storage
capacity of a pair of metallic plates or metalized films, the capacitance C of
such a mechanical construction is expressed as
A
C=ε (1.26)
d
in which ε represents the permittivity (dielectric constant) of separator
medium embedded between two plates, while A stands for plate area and d
plate separation, both spatial dimensions. Therefore, to a very good estimate,
the volumetric size for a capacitor will be at least equal to, or larger than,
dA; which is a simple product of two geometrical entities.
However, the process for sizing magnetic devices does not enjoy the
simplicity a capacitor is endowed with, as the construct of magnetic devices
always require some ferromagnetic filling a core shaped purposely to guide
magnetic flux such that the latter forms closed loops. In addition, space
must also be provided to accommodate multiple windings made of turns
of insulated metal wires. Fig. 1.13 shows one half of a ferrite core. The other
half is a mirror image.
As shown, the core has a center post giving a cross-sectional area Ac and
a winding window area Aw , within which coil(s) is(are) wound.
For a core wound with multiple windings, each winding in general
sustains different periodic voltage waveform and carries different currents,
also periodic. Eq. (1.18) can be generalized to specify how each winding will
be designed individually considering core flux limitation.
(Vrms ) j = K f f N j Ac Bmax (1.27)

Ac
Aw
Fig. 1.13 Half of a ferrite core; Ac = center post core area; Aw . (dotted line) = winding
window area, small filled circles = coil wires.
Here, Bmax is often chosen to be less than Bsat . Kf is a scaling factor relating
the RMS value of a periodic voltage and its time-domain magnitude.
Next, for each winding with peak current ij , a winding area (Aw )j is
conceptually assigned. Due to wire shape and unavoidable stacking in actual
build, only k(Aw )j is utilized. Given a desired inverse current density J [unit,
length2 /Amp],the winding peak current and its corresponding winding area
is associated by
k · (Aw ) j
ij = (1.28)
JN j
Eqs. (1.27) and (1.28) allow, for jth winding, its area product
(Vrms ) j JN j i j J (Vrms ) j i j
Ac (Aw ) j = · = (1.29)
K f f N j Bmax k kK f f Bmax
The total core cross-sectional area and winding area product covering all
winds is then
J
Ac A w = A c (Aw ) j = (Vrms ) j i j (1.30)
j
kK f f Bmax j
The summation on the right-hand side hints the total power handling
capacity of the device. The mathematical formulation may not be exactly
right, but it does give the flavor.
Therefore, referring to Fig. 1.13 and ignoring unit discrepancy, the
volumetric size of a magnetic device, in numerical term, may be considered
almost twice of Eq. (1.30). Designers must check with core manufacturers
as to the accounting of the area2 number to avoid over, or under, count;
therefore over, or under, sizing a device.
1.11 Specifications for capacitor and inductor

In the following, only key parameters considered essential are tabulated
for both parts. Readers are to be reminded that a good understanding
of part specifications and ratings constitutes a major role in delivering a
good design, selecting reliable components, and making a high-quality final
product.
1.11.1 Capacitor
Parameter Digits, unit Example Comments

Value Numeral, pF, nF, or 100 pF, 100 μF
μF
Value, tolerance Numeral, % 5%, 10%
Value, thermal Numeral, %/°C, 0.02%/°C
ppm/CÚ
Voltage, breakdown Numeral, Volts 25 V, 100 V DCmax + AC peak
magnitude
Voltage, surge rate Numeral, 1 KV/ms Causes transient
(dv/dt) Volts/sec current and
localized
overheating
Current, steady Numeral, A or mA 0.5 A at 360 Hz Frequency
sinusoidal dependent
Current, single pulse Numeral, A μs; A 500 A 3 μs Causes localized
ns overheating
ESR (Equivalent Numeral, mOhm 7 mOhm Not a constant
Serial Resistance)
Reactance Numeral, Ohm = Xc (ω), A curve Frequency
1/(ωC) dependent
Dissipation factor Numeral, A curve ESR/Xc (ω),
dimensionless
Leakage current Numeral, pA, nA Insulation quality
Lead Polarity Not applicable Polarized or Electrolytic or Film
nonpolarized
Packaging Not applicable Radial or axial Mounting
Disc or tubular consideration
A c B–L m H curve
0.4
Weber(volt.second)=AcB
Ψ(H, 50∙500)
Ψ(H, −50∙500)
−0.4
−2×105 H 2×105
AmpereTurn=LmH
Fig. 1.14 A core specific Ac B–Lm H curve.
1.11.2 Inductor
The intricacy for specifying magnetic devices is, in this writer’s view, orders
of magnitude harder than that for capacitors. Here, we will begin with an
important reexamination of the B-H curves given in Fig. 1.4 and Fig. 1.5.
As far as this writer has been able to reach, the most existing literature
including journalistic articles and textbooks dealing with the subject cover
it solely in terms of B(flux density)-H(field strength) parameters. What was
not clearly mentioned in those presentations is the underlying significance
of presenting in B-H form. In a single statement, we proclaim that the
curve in B-H parameter form is “material specific.” It is characterizing a
specific ferrous material in terms of per unit volume. In other words, it is
independent of core geometry.
Therefore, those material property curves can be easily modified to be
“core specific,” Fig. 1.14, in which core geometry is now included; Ac core
cross-sectional area and Lm magnetic path length. The plot x-axis coordinate
is HLm (Ampere) while the y-axis is Ac B (flux, per turn, in Weber) (Readers
should ignore numerical figures in Fig. 1.14. It just shows that the inclusion
of core geometry will alter plot coordinate scales.)
By the same token, it can be further modified to be “device specific” or
“winding specific,” “terminal specific” in which the winding turn number,
N, is included. With that, the plot x-axis coordinate is Ni (Ampere turn)
while the y-axis is NAc B (total flux linkage in volt-second = ∫vdt).
In summary, three aspects are involved in specifying a magnetic device:
core material B-H, core geometry Ac B–HLm , and winding terminal NAc B-
Ni.
1.11.3 Inductor core material

μi , zero-biased Numeral, 126, NIDC = 0 Referenced to μo
Permeability dimensionless = 4π × 10−7
μr , IDC -biased Numeral, 95, NIDC = 40, Referenced to μo
Permeability dimensionless Amp-Turn = 4π × 10−7
μo , free space 4π × 10−7 μ = μr • μo
Permeability
1.11.4 Inductor core Geometry

2
Ac , cross-sectional Numeral, SI (MKS) 0.67 cm
area meter2 or cgs cm2
lm , magnetic path Numeral, SI (MKS) 8.15 cm
length meter or cgs cm
Aw , winding area Numeral, SI (MKS) 2.13 cm2
meter2 or cgs cm2
1.11.5 Inductor winding

Fill factor 0< kf < 1, N/A 0.77, AWG #20 Wire gauge
dependent
lw , mean turn Numeral, m or 1.85 cm, Wire gauge
length cm AWG#24 dependent
N, turn number Numeral 13, AWG#22 Wire gauge
dependent
Many magnetic device manufacturers also often list the area product,
discussed in Section 1.10, to give designer a sense of how large the device
may be in handling a given power. This particular listing may be considered
to belong to the category of core geometry; but not exactly.
There are certainly other parameters, such as core losses unit volume
(a frequency-dependent parameter), thermal conductivity, etc. that are not
listed above; as this section presentation is not meant to be exhaustive.
Actually, this is the reason that efforts specifying magnetic devices properly
are very demanding.
NON-PRINT ITEMS
Abstract
Capacitors and inductors are two key components in power processing circuits. Both
possess the capabilities of storing energy; capacitor in electrical field (voltage form) while
inductor in magnetic field (current form). The combination of both and the exchange of
either form in a controlled manner constitute the core of electrical power processing.
This chapter examines both components from its most fundamental governing equa-
tions, in both differential and integral forms, and deciphers its electrical behaviors in time
domain. It then reviews the more practical aspects, important to designers, including
the definition, operational limitations in actual applications, equivalent circuits, and part
specifications for both.
Keywords
Capacitor; Current lagging; Current leading; Differential form; Inductor; Integral form;
Voltage lagging; Voltage leading; Zero-crossing
CHAPTER 2
First-order circuits
In Chapter 1, specifically Fig. 1.8, capacitor and inductor with multiple
parasitic elements included exhibit second-order behavior. However, for
the frequency range, less than 500KHz, we are to deal with in this writing,
the serial inductance and leakage resistance for capacitor and the parallel
winding capacitance for inductor may be ignored. By so doing, both reduce
to a first order circuit. Much as a first-order sounds simple, its role in
stitching together inexpensive, component-saving applications cannot be
underestimated. The fact merely reflects a plain truth that simplicity is
equivalent neither to uselessness nor to ineffectiveness.
This chapter attempts to uphold the belief.
MATH. NOTE: Textbooks on differential equations and boundary value
problems define “linear first-order differential equations” as
F (x, y, y ) = 0 (M2.1)
where x is the independent variable, and dependent variable y(x) and its first
derivative y = dy/dx are in the form of the first degree; no term of square,
or higher, power {y2 , (y´)2 ..}.
By habitually using the symbol x and y, Eq. (M2.1) tends to lead readers
to the impression of dealing solely with geometry, or space. Of course,
“linear first-order differential equations” are not limited to dealing merely
with space variables. It can as well treat many other physical variables and
involve “time” as an independent variable. In other words, Eq. (M2.1) can
be modified to
F (t, x, x ) = 0 (M2.2)
where x is any time-dependent variable. In this writing, x mostly stands for
current, voltage, or power. ♣
2.1 RC network with periodic drive source

MATH. NOTE: Fourier Series, orthogonal, and Laplace Transform
The topic, Fourier series, has always been introduced in all textbooks
by claiming that a periodic waveform, for example Fig. M2.1, can be
Power electronic system design. Copyright © 2021 Elsevier Inc.
DOI: 10.1016/B978-0-32-388542-3.00002-9 All rights reserved. 19
0 T/2 T
a0 + ∑ (a
n =1.2..
n sin n 2Tp t + bn cos n 2Tp t )
Fig. M2.1 A periodic waveform and its Fourier representation.
1
1
2 0.5
sin( t )
0 0
0 1.57 3.14
0 t π
Fig. M2.2 A 1 × π rectangle and sin2 t curve with period T = 2π t = θ.
represented by a series of the form, shown also in the figure, without

explaining the significance of each coefficient associated with individual
term summed in the series.
2π 2π

a0 + an sin n t + bn cos n t
n=1.2..
T T
In the following, we examine the subject, perhaps the first time in

teaching the topic, from an extremely unique approach without invoking
periodic signal at all.
A 1 × π rectangle, Fig. M2.2, gives geometrical area equal to π .
In an analytical term, it is expressed as
π π

1 · dθ = π = (sin θ )2 + (cos θ )2 dθ (M2.3)
0 0
As cos θ is merely sin θ shifted by π /2 (snapshot at different time),

therefore
π π
π
(sin θ ) dθ =
2
(cos θ )2 dθ = (M2.4)
0 0 2
First-order circuits 21
1
1
sin(t)∙cos(t)
0
0
−1 −1
0 3.14 6.28
0 t 2∙π
Fig. M2.3 sin(t)cos(t); equal positive and negative area.
The fact, in reference to Fig. M2.2, also gives a very interesting geomet-
rical beauty: the sin2 θ , or cos2 θ , curve is also one of many curves bisecting
the 1 × π rectangle. Another view is: under the curve sin2 θ , above the curve
cos2 θ .
Anyway, if we extend the integration boundary, the following results:
2π π
(sin θ ) dθ = 2
2
(sin θ )2 dθ = π (M2.5)
0 0
or
2π
1
(sin θ )2 dθ = 1 (M2.6)
π 0
We went through a series of steps to obtain Eq. (M2.6). What purpose

does it serve? There are two aspects of the effort; one minor, the other major.
The minor aspect is that it is totally feasible to reach Eq. (M2.6) without
performing any calculus. The major aspect requires more explanation.
Eq. (M2.6) is placed in a different form

1 2π
sin θ · sin θ dθ = 1 (M2.7)
π 0
If we insert a cosine term to one of the two sine terms, the integral does
not change.

1 2π
(sin θ + cos θ ) · sin θ dθ = 1 (M2.8)
π 0
As a product of sin(t) and cos(t) engenders equal positive and negative
area (Fig. M2.3).
1
1
sin(t)∙cos(2·t)
0
0
−1 −1
0 3.14 6.28
0 t 2∙π
Fig. M2.4 sin(t) sin(2t); equal positive and negative area.
If we add more cosine terms at a higher frequency, the integral does not
change either; as a product of sin(t) and cos(nt) again generates equal positive
and negative area.

1 2π
sin θ + cos nθ · sin θ dθ = 1 (M2.9)
π 0 n=1,2..
Interestingly, neither adding more sine terms at a higher frequency makes

any difference (Fig. M2.4).

1 2π
sin θ + sin nθ + cos nθ · sin θ dθ
π 0
n=2.3.. n=1,2..
1 2π (M2.10)
= sin nθ + cos nθ · sin θ dθ
π 0 n=1,2.. n=1,2..
=1
Clearly, as long as f(θ ) contains sin θ

1 2π 1 2π
(sin nθ + cos nθ ) · sin θ dθ = f (θ ) · sin θ dθ =1
π 0 n=1,2,... π 0
(M2.11)
In other words, if an unknown function f(θ ) possesses the property of

1 2π
f (θ ) · sin θ dθ = 1 (M2.12)
π 0
It is saying that f(θ ) contains a component resembling sin θ , or one unit
of sin θ .
The interpretation can be understood from a more insightful angle. From
a purely mathematical view, the creation of a product consisting of a known
sinusoid and an unknown, but periodic function under study has the effect
of accentuating the sinusoidal element embedded in the latter and with the
same frequency and phase.
The operation has a hidden effect, the suppression of elements that do
not match the template sin θ .
From a viewpoint of physical effects, the act of forming a product
and performing integration signifies the search for commonality among
two identities; one as a reference. Nonmatching, or nothing in common,
is orthogonal. Orthogonal of course means zero projection too. It also means
mutually inclusive. The presence of one does not exclude the other. They
also coexist simultaneously.
The above operation can be extended to find a cosine component hidden
in an unknown function. That is, if f(θ ) contains cos θ
2π
1
f (θ ) · cos θ dθ = 1 (M2.13)
π 0
2π
One may be tempted to think that π1 0 f (θ ) · (sin θ + cos θ ) dθ will
find both the sine and the cosine component simultaneously. But, it does
not work as the integral gives a single value. There is no indication how the
single value is partitioned among the two components.
The right way to find both the sine and the cosine component in a single
operation is
2π
1
f (θ ) · (cos θ − i sin θ )dθ (M2.14)
π 0
That is by utilizing the quadrature of the real and the imaginary parts of
a complex number.
The above is equivalent, and extended, to
2π
1
f (θ )e− jnθ dθ (M2.15)
π 0
This is equivalent to searching for both sin(nθ ) and cos(nθ ) components

in f(θ ) and is the basis of Fourier integral.
One more step takes it to
∞
f (t )e−st dt (M2.16)
0
v i(t)
R v o(t)
Va
C
DT T
Fig. 2.1 RC network with periodic voltage drive.
This is recognized as the Laplace transformation.

As complex exponential can represent all functional forms, the Laplace
transformation is considered the most generalized integral transform that is
looking for form/shape similarity (contributing factors) among signals. To
some extent, the concept of first moment (linear), second moment (squared),
and higher moments (cubic and higher) in probability theory has the same
flavor. ♣
Fig. 2.1 gives a simple RC network energized by a periodic, rectangular
voltage source with amplitude Va , period T, and duty cycle D. Few circuits
can be simpler than that, at least in appearance. And that is where the
simplicity ends, and complexity arises. So, what is the challenge? Obtaining
the steady-state output.
2.1.1. Fourier analysis approach

It is safe to say that most people, when approaching the problem, opt for the
Fourier analysis. Let us use a numerical example to illuminate the procedure.
Example 2.1 R = 10, C = 470 μF, Va = 4.8 V, T = 0.01 (f0 = 100 Hz),
D = 0.6, ω0 = 2π f0 .
Given the drive source waveform, it has an average term
T
1
a0 = vi (t )dt = 2.878 (2.1)
T 0
In addition, it has numerous sinusoidal components (index n); namely

cosine and sine elements with amplitude
T T
2 2
an = vi (t ) cos(nω0t )dt, bn = vi (t ) sin(nω0t )dt (2.2)
T 0 T 0
Table 2.1 gives the first 10 harmonics, n = 1 to 10.

To make certain those components do represent to some degree the drive
source, we reconstruct the waveform, vir (t), using 21 terms, the average plus
Table 2.1 Input Fourier harmonic, the first 10, amplitudes for Fig. 2.1 drive source.
n= an = bn =
1 −0.895 2.76
2 0.729 0.525
3 −0.482 0.355
4 0.219 0.689
5 −3.516 ×10-3 −1.868 ×10-4
6 −0.147 0.457
7 0.207 0.147
8 −0.183 0.135
9 0.095 0.305
10 −3.518 × 10-3 −3.741 × 10-4
6
6
5
vi ( t ) 4
3
vir ( t )
2
0 1
0
−1 −1
0 t T
Fig. 2.2 Reconstructed drive source for Fig. 2.1 using Eq. (2.3); (Dot, drive source; solid,
reconstruction with limited terms).
10 harmonics, given above.

10
vir (t ) = a0 + [an cos(nω0t ) + bn sin(nω0t )] (2.3)
n=1
Fig. 2.2 shows that Eq. (2.2) does decompose properly the rectangular
drive into a series of individual, contributing harmonics.
With that, we will next work out the output in a steady state.
First, the RC network gives a transfer function
1
H (s) = , H (s) = H ( jω) arg(H ( jω)) (2.4)
RCs + 1
Based on the theory for AC (alternating current) network analysis,
Fig. 2.1 output, vo (t) is the superposition, or sum (Eq. 2.5), of all individual
6
6
5
vi ( t ) 4
3
vo ( t )
2
0
1
0
−1 −1
0 t T
Fig. 2.3 Fig. 2.1 approximated output; (Dot, drive source; solid, output with 10 harmon-
ics).
4
4
3.97
vi ( t ) 3.93
3.9
vo ( t )
3.87
0
3.83
3.8
3.767 3.77
5.208×10−3 t 6.458×10−3
Fig. 2.4 Zoom in near the peak of Fig. 2.3; (Dot, drive source; solid, output with
10 harmonics).
input harmonics subjected, also individually, to the operation of Eq. (2.4).

m
vo (t ) = [an H ( jnωo ) cos(nω0t + arg(H ( jnωo )))
n=0 (2.5)
+ bn H ( jnωo ) sin(nω0t + arg(H ( jnωo )))]
One should note that b0 = 0. And, by picking only 10 harmonics,

m = 10 in Eq. (2.5), we have approximated output (Fig. 2.3).
The output estimate seems to have a deficiency near the peak. It begins
a descent prior to source retreat. Zoom in near the peak confirms the
observation (Fig. 2.4). This behavior violates the continuous property of
capacitor voltage. Capacitor voltage will decrease only after the drive source
is removed, not before.
6 4.08
6 4.083
5 4.03
vi ( t ) 4 vi ( t ) 3.98
3 3.92
vo ( t ) vo ( t )
2 3.87
0 0
1 3.82
0 3.76
−1 −1 3.708 3.71
0 t T 5.486×10−3 t 6.389×10−3
Fig. 2.5 Fig. 2.1 approximated output; (Dot, drive source; solid, output with 20 harmon-
ics); normal view and zoom in near peak.
Could it be the estimate process requires more harmonics? Let us increase

the summing index, m in Eq. (2.5) to Eq. (2.20) and repeat the computation
(Fig. 2.5).
Clearly, more harmonic is not the solution; zoom in near peak still shows
the same sooner than expected inflection.
We therefore come to a conclusion that the effect is an artifact, rather a
fundamental limitation,of the Fourier analysis as only finite-harmonic count
is invoked. ♠
MATH. NOTE: Differential equations and initial values
In the previous MATH. NOTE, Eqs. (M2.1) and (M2.2) define the “lin-
ear first-order differential equation.” For this writing dealing with electrical
circuits, Eq. (M2.2) dominates. A typical Kirchhoff current law/Kirchhoff
voltage law (KCL/KVL) equation at a network node or around a network
loop appears in its primitive form as “a•di/dt + b•i = g(t)” or “a•dv/dt +
b•v = g(t)” wherein a and b are constants attributed to network components
and g(t) time-dependent source function. Easily, either is rearranged in a
parameter form “di/dt + β•i = α•g(t)” or “dv/dt + β•v = α•g(t).” Given
the equation, all textbooks deliver its solution in the following formidable
looking form
t
− βdt
i(t ), or v(t ) = e αe βdτ g(τ )dτ + C(t0 ) (M2.17)
to
Luckily, β in our studies is often a constant. It simplifies to

t
−βt
i(t ), or v(t ) = e αeβτ g(τ )dτ + C(t0 ) (M2.18)
to
The key however is not in the integration. Rather, it is the initial inte-
gration constant C(t0 ), the initial value/condition so to speak in mathematics
books. The fact is that the time-dependent initial value can move a solution
all over the solution domain.
Actually, the nomenclature “initial value” is, in this writer’s view, not
an ideal choice of word. First of all, the integration constant is not really
depending on time alone. It cannot simply come from thin air. It must come
from somewhere and depend on many elements constituting the real world.
It must be the end state of some past process. And it exerts influence on the
future course of another process. In a sense, physical states propagate from
one instant to the other and are continuous. Niagara Falls, or Grand Canyon,
is extremely steep. Still, it is continuous.
In summary, there appear two critical properties of physical world: The
continuous nature of state transitions and the distinctive state contents at the
transition boundary. That could be the reason the word “boundary value” has
been instead invoked.
Furthermore, boundary values exert influences on the evolution of
physical world by two pathways: space (geometry) and time. The latter
actually is equivalent to energy (frequency). The impact of space boundary is
easily observed in the extraordinary efforts put into shaping propeller blades
for aircraft, ship, submarine, wind turbine, etc., to improve performance
efficiency. However cancer treatments based on systemic, nonspecific, some-
time toxic, chemotherapy are beginning to be questioned; more targeted
and location-specific approaches utilizing energetic particle beams, forms of
energy boundary conditions, are becoming better choices.
2.1.1 Continuity of states and boundary condition approach

It turns out that there is luckily a better approach: continuity of states, state
wrap around, and boundary condition.
As shown in Fig. 2.1, the periodic input drive has two states: Va , 0 < t <
DT, interval “a”; and 0, DT < t < T, interval “b.” Therefore, there are two
distinctive equations,(Eq.2.6),describing the circuit function corresponding
to the two time intervals.
dvoa Va − voa dvob 0 − vob
C = , C = (2.6)
dt R dt R
Assuming yet unknown starting conditions Va0 , at time zero, and Vb0 , at
time DT, and taking Laplace transform of Eq. (2.6), two transfer functions
with starting conditions accounted for are obtained. (Readers should watch
the difference of letter “o”and numeral “0”appearing in variable subscripts.)
Va0 Va Vb0 −DT s

Voa (s) = + , Vob (s) = e (2.7)
s + RC1
RCs s + 1
RC
s + RC
1
Inverse Laplace transform yields time-domain outputs, properly gated,

during two time intervals.

voa (t ) = Va0 e− RC + Va 1 − e− RC [u(t ) − u(t − DT )]
t t
(2.8)
vob (t ) = Vb0 e− RC [u(t − DT ) − u(t − T )]
t−DT
As discussed in Chapter 1, capacitor voltage is continuous. In a steady

state and at transition time boundaries, t = DT and t = T, Eq. (2.8a,b) gives

Va0 e− RC + Va 1 − e− RC = Vb0
DT DT
(1−D)T
(2.9)
Vb0 e− RC = Va0
Specifically, the end state of the first interval voltage at time DT serves as
the starting state for the dynamic output during the second time segment,
while the end state of the second interval voltage serves as the starting state
for the first interval voltage. This act, resembling passing the baton in a
10-km relay race (field and track competition), enables us to find both
starting conditions.

Va 1 − e− RC
DT
(1−D)T
Va0 = (1−D)T , Vb0 = Va0 e RC (2.10)
− DT
e RC − e RC
Once both cyclic starting condition are derived, Eq. (2.8) presents the
output in one cycle; sum of voa (t) and vob (t). If so desired for viewing purpose,
multiple cycle output is easily built by a periodic shift of a single cycle
expression.

m
vo1 (t ) = voa (t ) + vob (t ), vo (t ) = vo1 (t − nT ) (2.11)
n=0
For this current example, and using Eq. (2.10), Va0 = 1.677 Vb0 = 3.929.
Eqs. (2.8) and (2.10) then give Fig. 2.6.
Comparing Figs. 2.3–2.6 clearly shows the superiority of “continuity of
states and boundary condition approach.”
6 4.1
6 4.104
5 4.06
vi ( t ) 4 vi ( t ) 4.02
3 3.98
v( t) v( t)
2 3.94
0 0
1 3.9
0 3.85
−1 −1 3.813 3.81
0 t T 5.625×10−3 t 6.389×10−3
Fig. 2.6 Fig. 2.1 exact output; (Dot, drive source; solid, output); normal view and zoom in
near peak.
1
1
0.5
vi ( t )
0
iR ( t)
−0.5
−1 −1
0 t 1∙T
Fig. 2.7 Resistor current for Fig. 2.1 network.
2.2 Sawtooth (triangle ramp) generator

From Figs. 2.3–2.6 the output waveform exhibits periodic ramp-up/ramp-
down time-domain geometry having shape similar to teeth of a cutting saw,
therefore the practical term “sawtooth” favored by electrical engineers in
industrial sector.
One also may notice that those tooth contours (edges, rim) are curved,
either concave inwardly or convex outwardly. It turns out that the curvature
is generated by the nonlinear current through the resistor R. This assertion is
easily verified by computing the current in question, iR (t) = [vi (t) – vo (t)]/R
(Fig. 2.7).
A more insightful view could be provided here. That is resistor element
alone is a very poor, crude current source. Later on, we will devote a whole
chapter dealing with the creation of current sources. For the time being,
assume a constant current is available for improving and modifying (Fig. 2.1)
network.
I
DT T
v o(t)
r
C
Fig. 2.8 Sawtooth generator.
In Fig. 2.8, a command operates a single-pole-double-throw solid-state

relay such that the constant current, I, is charging the capacitor during 0
< t < DT. Then a discharge follows during the rest of a cyclic period.
Eq. (2.12a,b) describes the actions.
dvoa dvob vob
C = I, C + =0 (2.12)
dt dt r
Without repeating the process similar to Eq.(2.7),the sawtooth generator
output in one cycle is given.
I
voa (t ) = Va0 + t [u(t ) − u(t − DT )]
C (2.13)
vob (t ) = Vb0 e− rC [u(t − DT ) − u(t − T )]
t−DT
And applying again the principle of continuity of state, we have the

following
I (1−D)T
Va0 + DT = Vb0 , Vb0 e− rC = Va0 (2.14)
C
At this point, let us pause and define the mission and process going
forward.In general,the goal of generating a sawtooth is a ramp with specified
voltage swings,Va0 and Vb0 ,given chosen clock,T = 1/f,up-ramp duty cycle
D, and preselected C. With that being said, Eq. (2.14) empowers us to find
the required discharge resistor, r, and the current source, I.
(1 − D)T C
r= , I = (Vb0 − Va0 ) (2.15)
C ln VVa0b0 DT
Example 2.2 Generate a sawtooth with a frequency 100 KHz, up-ramp

duty cycle 0.98, amplitude swing from 5/3 V to 10/3 V, preselected capacitor
C = 0.1 μF.
5
5
4
3
vo ( t )
2
1
0 0
0 5∙10−6 1∙10−5 1.5∙10−5 2∙10−5 2.5∙10−5 3∙10−5
0 t 3∙Ts
Fig. 2.9 Sawtooth generator output; shown only three cycles.
Eq. (2.15) yields r = 2.88 and I = 17 mA. Eq. (2.13) gives a sawtooth
output (Fig. 2.9).
This type of waveform served very important functions; in the CRT
(Cathode Ray Tube), now antiquated, display circuits as the beam scanning
controller and in pulse width modulation control for switch-mode power
converters and motor drivers. We will come to the latter later. ♠
MATH. NOTE: Differential equations and force function
It was mentioned in the previous MATH. NOTE that Eq. (M2.2) and
“di/dt + β r i = α r g(t)” or “dv/dt + β r v = α r g(t)” are the prevalent forms
of differential equations in this writing, and many other electric circuits
literature, rely on them. Much as the form looks simple, appearance can
be deadly deceiving. The trap lays in how the driving (source, generator)
function, g(t) on the right side, is specified.
There are more than one key points that must be properly observed,
otherwise attempts to solve the equation will end either in failure or in
error.
First of all, by sketching casually the driving function g(t) into a dif-
ferential equation without qualification, it seems to imply that the driving
function is valid starting at t = 0 and is actively applied all the time afterward.
These two notions are called into question.
Depending on the circumstance, the driving source time profile, the
properties of individual elements comprising the physical system, the cross-
interactions among elements, and the interaction between elements and
driving source, the active application of a driving source may start from
other than zero time. And, even more the important, the drive source may
kick in only sporadically.
Another random document with
no related content on Scribd:
would have been entirely lost had the two Russian admirals been
qualified for such a command. Captain Pélissier, who had served in
Holland, is said to have given Admiral Tchitchakov advice which he
ought to have followed, had he not been too obstinately attached to
his own opinions; Pélissier even pointed out to generals Suchtelen
and Soltikov the places where they ought to have erected their
batteries in order effectually to bar the egress of the Swedish fleet
from the bay; no attention, however, was paid to his advice. The
prince of Nassau-Siegen proved himself to be in no respect superior
as a commander to Tchitchakov. On the other hand, if the advice of
Duke Charles had been adopted, the Russians would have been
victorious without a battle; King Gustavus and Stedingk, however,
rescued the honour of the Swedish name.
The Swedes had now been closely shut up in the bay of Viborg for
three weeks, and at the end of June were reduced to extremities; in
the beginning of July a grand council of war was held. Duke Charles
and many other members of the council recommended a
capitulation, but the king and Stedingk were in favour of making a
desperate effort to force their way through the enemy’s line. The
attempt was accordingly made on the 3rd of July, and through
Tchitchakov’s neglect it was so far successful, as it enabled the
Swedish fleet to bring the blockading squadron to an engagement.
But the Swedes lost in it not only seven ships of the line, three
frigates, and more than thirty galleys and gunboats, but almost the
whole of the royal guards, the queen’s regiment, and that of Upland,
amounting to six thousand or seven thousand men, which had been
put on board the fleet. Whilst the larger Swedish ships thus
endeavoured to gain the open sea, the flotilla had withdrawn for
safety into an arm of the gulf, which runs parallel to the shore and
stretches towards Friedrichsham. This inlet, called the sound of
Suenske, is extremely difficult of access on the side towards
Friedrichsham, in consequence of a group of rocky islands at its
mouth, but it may be safely reached through the open harbour of
Asph. By this way the prince of Nassau-Siegen determined to pass
into the sound with the Russian flotilla, and attack the Swedes in
their place of refuge.
The latter were well protected from the attack of the Russian fleet
by rocks, and when the prince gave orders for the assault, on the
9th, the sailors were so exhausted and his orders for battle were so
unskilful that the king of Sweden gained a splendid victory on that
and the following day. The loss of the Russians was so great as to
have surpassed any which they had suffered since the Seven Years’
War. Fifty-five vessels were captured, a number of others destroyed,
and fourteen thousand Russians either taken prisoners or slain. In
spite of this signal victory, the king of Sweden now awoke from his
dream of humbling the pride and glory of Russia; already he began
to cast his eyes towards France, and in the following year he
dreamed his monarchical dream in favour of the French émigrés.
The idea of becoming the Godefroy de Bouillon of the aristocratic
and monarchical crusade, which Burke at that time proclaimed in the
English parliament and in his work on the French Revolution, had
been awakened in his mind in 1790, and the empress of Russia
found means of confirming him in his visionary projects. Moreover
his means were exhausted, and he therefore lent a favourable ear to
the proposal of Galvez, the Spanish ambassador, who began to
mediate for a peace between Sweden and Russia.
This peace, concluded at Varela on the Kimmene on the 14th of
August, 1790, served to show how empty all Gustavus’ splendour
was, and how unreal and inefficient were all the efforts he had made.
It was now seen that all the blood had been shed to no purpose, and
all the treasures of his very poor kingdom mischievously
squandered, for everything remained on the footing on which it had
been in the spring of 1788.
PROGRESS OF THE AUSTRO-RUSSIAN WAR WITH TURKEY
We now return to the war in which Austria and Russia were jointly
engaged against Turkey. The whole Austrian army was ready to take
the field at the end of the year 1787: it formed an immense cordon
stretching from the mountains on the coast of the Adriatic Sea to the
Carpathians, and consisted of a main body and five divisions.
Unhappily, the emperor Joseph was desirous of commanding the
main army in person, under the unskilful direction of Lacy, his military
Mentor, who, like his pupil Mack, was a good drill-sergeant, but no
general. The main body consisted of 25,000 infantry and 22,000
horse, and the whole of the troops together amounted to 86,000
cavalry and 245,000 foot, accompanied by 898 pieces of artillery.
In February, 1788, Russia and Austria had simultaneously
declared war against the Turks; but in August of that year England
and Prussia entered into an alliance, the main object of which was to
place Prussia in a situation to prevent the aggrandisement of Austria,
if necessary, by force of arms. This, however, was superfluous in
1788, because the diversion effected by the king of Sweden
prevented the Russians from proceeding with their usual rapidity,
and the emperor Joseph by his presence with the army frustrated the
effect of his immense armaments. The dissatisfaction with the whole
conduct of the war became so general that Joseph was at length
obliged earnestly to entreat Laudon, who had been the popular hero
of the Austrians since the time of the Seven Years’ War, and whom
the emperor had hitherto neither employed nor consulted, to assume
the command of the army in Croatia.
Successes of Laudon (1788 A.D.)
Laudon, having made an express stipulation that the emperor was

not to interfere with his plans marched against the Turks, defeated
them under the walls of Dubitza the very day after he joined the
army, and reduced that fortress; then, pushing into the heart of
Bosnia, he compelled Novi to surrender, whilst the emperor himself
was obliged to hasten to the aid of the army in the Bannat, which
was very hard pressed by the Turks. The division under
Wartensleben, which should have supported it, had been driven
back by the Turks, who succeeded, in consequence of an
incomprehensible neglect on the part of the Austrians, in getting
complete possession of the rocky bed through which the Danube
has forced a passage at a distance of six-and-twenty miles above
New Orsova. The pass, which is not more than a pistol-shot in width,
is commanded by a fortified cleft in the rock, called Veterani’s Hole,
and this post the Austrians should and could have maintained when
the main body of the Turks appeared at Old Orsova on the 7th of
August; this, however, they neglected to do. The Austrian general
suffered himself to be defeated and lost thirteen pieces of cannon,
and as his communications with the main army were cut off, he was
obliged to retreat so far that the garrison of this important post was
left to its fate. The Turks sacrificed great numbers of men in order to
seize this fastness, by the possession of which they immediately
became masters of the whole navigation of the Danube as far down
as Belgrade. As soon as the Danube was lost, the imperial army
found itself threatened in the rear.
Nothing but disaster attended the operations of Joseph and
Wartensleben. The army under the prince of Coburg was somewhat
less unfortunate. Khotin, which the Russians had captured in the last
war without firing a shot, was reduced by it after a most heroic
resistance of three months; and this was the last exploit of a
campaign in which thirty thousand Austrians fell in desultory
skirmishes, and forty thousand were swept off by pestilence—losses
but poorly compensated by the capture of Szabatch, Khotin, Dubitza,
and Novi. Circumstances, however, afterwards proved more
favourable. Jassy was taken; in October, the Russians were in
possession of five districts of Moldavia and of several passes in
Wallachia, and the main army was again able to extend the limits of
its operations. Wartensleben sat down with a part of the army before
Mahadia; and the emperor kept possession of the country from
Pantchova to Semlin.
Victories of Suvarov (1788-1789 A.D.)
After the massacre perpetrated by Suvarov upon the Turks on the

promontory of Kinburn, the Russians had remained for a long time
quiet; but by their possession of the coasts they effectually
prevented the Turks from landing any troops, and by the capture of
the island of Beresam wholly excluded them from the mouth of the
Dnieper. It was not till late in the year 1788 that Potemkin summoned
Suvarov from Kinburn to conduct the siege of Otchakov, where,
however, he was wounded, and after his return to Kinburn the siege
made very little progress. The avarice of Potemkin deprived the
soldiers of the necessary supplies; and the dreadful cold and
disease proved far more injurious to them than the attacks of their
enemies.
At length the frost became so intense that the men were obliged to
excavate pits for dwellings, but the same frost also opened up a
means of attacking the fortress and reducing it after the Russian
fashion, that is, without regard to the sacrifice of thousands of men, a
few weeks earlier than they could otherwise have done. The city is
completely protected on the side towards the Black Sea by a marshy
lake called Liman; and now that the lake was frozen, Potemkin
issued orders to storm the fortress from the sea side, where it was
weakest. The Russians were cruelly sacrificed: one regiment was no
sooner mowed down than another was compelled to advance, and
above four thousand men were slain before the storming of
Otchakov was effected (December 16th), an exploit which was
afterwards extolled to heaven. The Russians, having at length borne
down all resistance and forced their way into the city, were
compensated for their losses and sufferings during the siege by
three days’ murder and pillage; they put citizens and soldiers, men,
women, and children to the sword without mercy or distinction. It is
said that twenty thousand Turks perished in this massacre; but this
piece of Russian heroism, which was not performed by Potemkin
himself but by others at his command, was also rewarded after the
Russian fashion. Every soldier who had taken part in the siege
received a medal of honour, whilst Potemkin, who had contributed
nothing to its success, derived the only real advantage. The empress
had previously deprived Razumovski of the office of hetman, which
she now conferred upon Potemkin, who received in addition a
present of 100,000 rubles, besides what he had appropriated to
himself out of the moneys destined for the besieging army, and what
he had seized out of the rich booty which fell into his hands after the
capture of the city.
The death of the sultan Abd-el-Habed in April, 1789, made no
change in the relations between the Turks and Russians. His
successor, Selim, continued to prosecute the war, and Suvarov
having recovered from the effects of his wound again joined
Potemkin’s army, and was put at the head of the division which was
to co-operate with the Austrians. Laudon had now the command of
the whole Austrian army; the prince of Coburg, however, retained
that of the division which was to keep open the communications with
the Russians; and again he gave such numerous proofs of his
incapacity to conduct any great undertakings, or even to help himself
out of trifling difficulties, that the history of the campaign of 1789
alone ought to have prevented the emperor Leopold from entrusting
him with the command against the French, who possessed generals
and soldiers of a very different kind from those of the Turks. Selim III
had succeeded in getting on foot a very considerable force which
was destined to operate on the extreme point of Moldavia, where
that country touches upon Transylvania, and is separated from
Wallachia by a small river, which also divides the little town of
Fokshani into two parts, one belonging to Moldavia, and the other to
Wallachia. Coburg was advancing thither slowly and methodically,
when the Turkish army encamped in the neighbourhood of the town
turned suddenly upon him, and filled him with such apprehensions of
being completely shut in that, instead of boldly doing what Suvarov
afterwards did, he anxiously besought that general’s speedy
assistance.
Suvarov’s army was lying at Belat in Moldavia; when the news
reached him he at once began a march of between forty and fifty
miles in a direct line over mountains, across ravines and pathless
wilds, and in less than thirty-six hours reached the Austrians on the
30th of July, at five o’clock in the evening. At eleven that night he
sent the plan of the attack upon the Turks, which was to commence
at two in the morning, to the astonished prince, who had never heard
of such rapidity of movement, or seen it equalled even on parade.
The bewildered prince went three times to Suvarov’s quarters
without having seen him; in the battle he made no claim to the
supreme command, which should have belonged to him as the
eldest general, but submitted as a subordinate to Suvarov’s orders.
The Turks, to the number of between fifty and sixty thousand men,
were in position at Fokshani when the Russians and Austrians with
forty thousand men passed the river Purna and stormed their fortified
camp, mounting the ramparts and driving them in at the point of the
bayonet, as if they were assaulting ordinary field-works. The camp
was taken in an hour, with the loss of about eight hundred men; the
whole body of the Turkish infantry fell into disorder, their cavalry
galloped off, were scattered in all directions, and pursued for some
miles with the greatest impetuosity and vehement zeal. The whole of
the baggage and artillery, all the stores collected in Fokshani, a
hundred standards and seventy pieces of cannon, fell into the hands
of the victors; the Austrians exhibited the same zeal, perseverance,
and courage as the Russians, and had they possessed such a
commander as Suvarov, they would have reaped immense fruits
from the victory, but they became sensible, as early as August, that
they were in want of a proper leader.
Suvarov returned to Moldavia; Coburg looked quietly on whilst the
Turks were collecting a new army, and suffered the grand vizir to
advance without obstruction in Wallachia. The Turks directed
Hassan Pasha, who lay in Ismail, to make an expedition against
Repnin, whilst the grand vizir was to march against Prince Coburg,
who had taken up a position at Martinesti, on the river Rimnik. The
news of this fresh attack no sooner reached the Austrian camp than
Coburg, instead of attempting to help himself, again had recourse to
Suvarov, who had already drawn nearer to Coburg from Belat. The
grand vizir’s army, which had been estimated at one hundred
thousand men, pushed forward rapidly by Braila (Ibrahil), and
compelled the advanced posts of the prince to retire into their camp.
Suvarov received the prince’s letter on the 16th of September,
immediately gave orders to march, and two days afterwards
succeeded in forming a junction with the Austrians, at the very
moment in which they were to have been attacked by the Turks.
Austrian and Russian Valour; Austria’s Withdrawal (1789-1790 A.D.)
The Austrians then proved anew that they were not to be

surpassed when not commanded as usual by princes and privileged
persons, who become generals whilst they sleep. Coburg, as he had
previously done at Fokshani, totally relinquished the command at
Martinesti to Suvarov, who immediately availed himself of the
oversight of the Turks in not fortifying their camp before they offered
battle, and attacked them by storm in their unfinished trenches. The
issue was as glorious as it had been on the 31st of July at Fokshani;
the contest, however, was more obstinately maintained. On this
occasion the Russians formed the left wing, whilst the centre and
right were occupied by the Austrians, whose admirably served
artillery scattered the Turkish cavalry, which had made an attempt to
surround and cut off the small body of the Russians. The victory in
this dangerous and hard-fought battle was gained not merely by the
courage, activity, and bayonets of the Austrian and Russian infantry,
but especially by the great military skill of the commander. His orders
to avoid the village of Bochsa, and first to drive the Turks out of the
woods by which they were covered before commencing the main
attack, have been greatly admired, and above all his prudence in not
sacrificing the infantry in a blind storm, which was the more
remarkable in a general accustomed to bring everything to a rapid
determination.
The victory was splendid, the booty immense, the Turkish army a
second time utterly dispersed—a necessary consequence of the
nature of its composition—and the number of killed and wounded
much greater than at Fokshani. Prince Coburg, on account of this
victory, in which he was entitled to little share, was created a field-
marshal; Suvarov received the dignity of a count of the empire from
the emperor Joseph, and the empress of Russia for once gave an
honourable surname to a man who had really earned it by his
personal services; she raised him to a level with her Tchesmian
Orlov and her Taurian Potemkin, and called him Rimnikski, from the
name of the river on the banks of which he had been victorious.
The victory of Rimnik and the capture of Belgrade by Laudon on
the 9th of October were the harbingers of greater success. Hassan
Pasha, the Turkish high-admiral and celebrated conqueror of Egypt,
whose confidence in his good fortune had encouraged him to
assume the command of an army, was totally defeated at Tobak, in
Bessarabia, by Prince Potemkin, and his discomfiture was followed
by the surrender of Bender, Akerman, Kilia Nova, and Isatza, and by
the investment of Ismail. At the same time the prince of Coburg took
Bucharest and Hohenlohe, forcing the passes which lead into
Wallachia, made himself master of Rimnik and Krajova. Laudon also
reduced Semendria and Kladova, and blockaded Orsova, which,
being situated in an island of the Danube, was inaccessible to
regular attacks. By these conquests the allies became masters of the
whole line of fortresses which covered the Turkish frontier; the three
grand armies, originally separated by a vast extent of country, were
rapidly converging to the same point, and threatened, by their united
force, to overbear all opposition, and in another campaign to
complete the subversion of the Ottoman empire in Europe.
AUSTRIANS ENTERING BELGRADE
(From the painting by Karl von Blaas in the Ruhmeshalle of

the Arsenal in Vienna)
But in the midst of this successful career, the increasing ferment in

the hereditary states of Austria, the rebellion in the Netherlands, and,
still more, the interposition of the maritime powers and Prussia,
checked the hopes of Joseph at the very moment when his projects
of aggrandisement seemed hastening to their completion. Justly
alarmed at the successes of the two imperial courts, the three
combined powers incited Poland to throw off the yoke of Russia,
delivered the king of Sweden from Danish invasion, and laid the
foundation of a general alliance for reducing the overgrown power of
Austria and Russia. The king of Prussia even encouraged the rising
discontents in Hungary, fomented the troubles which the impolitic
innovations of Joseph had excited in the Netherlands, and, in the
beginning of 1790, opened a negotiation with the Porte for the
conclusion of an offensive alliance, intended not only to effect the
restoration of the dominions conquered during the existing war, but
even of the Crimea, and the territories dismembered by the two
imperial courts from Poland.
The only power to which Joseph might have turned as a
counterpoise to this combination was France, from whose recent
change of system he had flattered himself with hopes of a cordial
support, and from which he had even received private largesses to a
considerable amount. But now France was in the throes of her great
revolution, and Joseph was left without a resource. Worn down by
innumerable calamities and disease, he died in February, 1790; and
his successor, Leopold, was fortunate enough to conclude a
separate peace with the Porte.
Russia Prosecutes the War; the Storm of Ismail (1790 A.D.)
Russia continued to prosecute the war against the Turks without

the aid of Austria. Ismail still held out, and Potemkin, who had been
besieging it for seven months, began to grow impatient. Living in his
camp like one of those satraps whom he even surpassed in luxury,
he was surrounded by a crowd of courtiers and ladies, who exerted
every effort to amuse him. One of these ladies, pretending to read
the decrees of fate in the arrangement of a pack of cards, predicted
that he would take the town at the end of three weeks. Potemkin
answered, with a smile, that he had a method of divination far more
infallible. He instantly sent orders to Suvarov to come from Galatz
and take Ismail in three days. Suvarov arrived and took such
measures as would seem to indicate that he designed a renewal of
the regular siege; he drew together the scattered divisions of the
troops, formed them into a large besieging army of about forty
thousand men, and ordered the small Russian fleet to come into the
neighbourhood of the city; but his real design was to follow the
course he had successfully pursued before Otchakov, take
advantage of the frost, and reduce the fortress by storm.
Had not Ismail, according to ancient usage, been built without
advanced works, even a general like Suvarov would scarcely have
ventured on such an attack, which in the actual condition of the
defences was attended by such murderous consequences. On the
21st of September the city was twice summoned, and on both
occasions the garrison and inhabitants were threatened with the fate
of Otchakov. The Turks, however, did not suffer themselves to be
terrified into submission, and the fearful storm was commenced on
the 22nd, at four o’clock in the morning. The wall was not mounted
till eight o’clock, after an unexampled slaughter; but still the hottest
part of the struggle took place in the city itself. Every street was
converted into a fortress, every house became a redoubt, and it was
twelve o’clock before the Russians, advancing through scenes of
carnage and desperate resistance, reached the market-place, where
the Tatars of the Crimea were collected. The Tatars fought for two
hours with all the energy of despair, and after they had been all cut
to pieces the struggle was still carried on by the Turks in the streets.
Suvarov at length opened a passage for his cavalry through the
gates into the devoted city; they charged through the streets, and
continued to cut down and massacre the people till four o’clock in the
afternoon. At the conclusion of this dreadful butchery the Russians
received the reward which had been promised them when they were
led to the storm and to certain death,—the city was given up for
three days to the mercy of the victorious troops.
Suvarov himself, in his official report of this
[1791 a.d.] murderous enterprise, states that in the course
of four days 33,000 Turks were either slain or
mortally wounded, and 10,000 taken prisoners. He rates the loss of
the Russians at 2000 killed and 2500 wounded: a number which
seems to us as improbably small as the usual accounts, which
assign 15,000 as the Russian loss, seem exaggerated. There were
two French émigrés present at this storm, one of whom afterwards
became celebrated as a Russian governor-general and French
minister, and the other as a Russian general in the war against his
countrymen. The first was the duke de Richelieu, or as he was then
called de Fronsac, and the second the count de Langeron. Kutusov
also served in this affair under Suvarov and led the sixth line of
attack.
European Intervention; the Treaty of Jassy (1792 A.D.)
About this time the whole diplomacy and aristocracy of Europe

were busily employed in endeavouring to rescue the Turks, in order
to check the dangerously rapid progress of the French and Polish
revolutionists. There speedily grew up such a general desire as the
English wished to promote—of two evils to choose the least—to
secure and uphold the empire of the Turks and to let the nationality
of Poland perish. Russia, however, declined the proffered mediation
of England in the war with the Turks, as she had resolved for this
time to give up her conquests in Turkey in order to indemnify herself
in Poland: she accepted merely the intervention of the friendly
Danes.
Potemkin and the empress were not unthankful for Suvarov’s
servility, since he threw himself and all his services at their feet, and
ascribed everything to them alone. Repnin, whom Potemkin left at
the head of the army when he went to St. Petersburg in October,
1790, pursued a very different course, doing more in two months
than Potemkin had done in three years. He crossed the Danube with
his army, pushed forward into Bulgaria, and caused the whole
Turkish army to be attacked and beaten near Badadagh by Kutusov,
after Gudovitch, the brother of him who had been the faithful aide-
de-camp of Peter III, had completely put down the Tatars in the
Kuban in January, 1791. At the head of forty thousand Russians,
Repnin then advanced against one hundred thousand Turks, under
the command of the same vizir, Yussuf, who had fought with such
success against the emperor Joseph in the Bannat.
Potemkin eager to appropriate the impending victory, started with
great expeditiousness from St. Petersburg when both armies were
ready for battle (July, 1791). He took it for granted that Repnin would
certainly await his arrival at the army; but he did no such thing. He
offered battle before the arrival of Potemkin, whose custom it was to
enjoy the fruits in the gathering of which he had no share. The
victory which Repnin gained over the great Turkish army in July at
Matchin led to a violent altercation between him and Potemkin, who
came too late to have any participation in the honours of the day;
Repnin, however, still remained in command of the army. Potemkin
afterwards did everything in his power to prevent the peace for which
Repnin was to negotiate, although he clearly saw that the course of
events required the Russians to give up this wholesale conquest of
Turkish provinces. Happily, his death left Repnin’s hands free, and a
treaty was concluded at Jassy on the 9th of January, 1792, between
Russia and the Porte, by which the former acquired nothing more
than the fortress of Otchakov, the surrounding territory from the
Dniester to the Bug, and the protectorate of Georgia.
THE DEATH OF POTEMKIN (1792 A.D.); SÉGUR’S

CHARACTERISATION
Not long after Potemkin’s arrival at Jassy,
[1792 a.d.] where his headquarters or, to speak more
properly, his capital and his court were
established, he was seized with a malignant fever, and presumed to
treat it with the same haughty contempt with which he had long been
used to treat his fellow men: he laughed at his physicians, and ate
salt meat and raw turnips. His disease growing worse, he desired to
be conveyed to Otchakov, his beloved conquest, but had not
travelled more than a few miles before the air of his carriage seemed
to stifle him. His cloak was spread by the road-side; he was laid on it,
and there expired in the arms of his favourite niece Branicka.
Catherine fainted three times when she heard of his death: it was
necessary to bleed her; she was thought to be dying. She expressed
almost as much grief as at the death of Lanskoi; but it was not the
lover she regretted: it was the friend whose genius assimilated with
her own, whom she considered as the support of her throne and the
executor of her vast projects. Catherine, holding her usurped
sceptre, was a woman and timid: she was accustomed to behold in
Potemkin a protector whose fortune and glory were intimately
connected with her own. The character of this Russian vizir has been
thus sketched by Count Ségur, who, as ambassador to St.
Petersburg, lived long in habits of intimacy with him:
“Prince Gregory Alexandrovitch Potemkin was one of the most
extraordinary men of his times; but in order to have played so
conspicuous a part, he must have been born in Russia and have
lived in the reign of Catherine II. In any other country, in any other
time, with any sovereign, he would have been misplaced; and it was
a singular stroke of chance that created this man for the period that
tallied with him, and brought together and combined all the
circumstances with which he could tally.
“In his person were collected the most opposite defects and
advantages of every kind. He was avaricious and ostentatious,
despotic and popular, inflexible and beneficent, haughty and
obliging, politic and confiding, licentious and superstitious, bold and
timid, ambitious and indiscreet. Lavish of his bounties to his
relations, his mistresses, and his favourites, yet frequently paying
neither his household nor his creditors. His consequence always
depended on a woman, and he was always unfaithful to her. Nothing
could equal the activity of his mind or the indolence of his body. No
dangers could appal his courage; no difficulties force him to abandon
his projects. But the success of an enterprise always brought with it
disgust. He wearied the empire by the number of his posts and the
extent of his power. He was himself fatigued with the burden of his
existence; envious of all that he did not do, and sick of all that he did.
Rest was not grateful to him, nor occupation pleasing. Everything
with him was desultory—business, pleasure, temper, carriage. In
every company he had an embarrassed air, and his presence was a
restraint on every company. He was morose to all that stood in awe
of him, and caressed all such as accosted him with familiarity.
“Ever promising, seldom keeping his word, and never forgetting
anything, none had read less than he—few people were better
informed. He had talked with the skilful in all professions, in all the
sciences, in every art. None better knew how to draw forth and
appropriate to himself the knowledge of others. In conversation he
would have astonished a scholar, an artist, an artisan, or a divine.
His information was not deep, but it was very extensive. He never
dived into a subject, but he spoke well on all subjects.
“The inequality of his temper was productive of an inconceivable
oddity in his desires, his conduct, and his manner of life. One while
he formed the project of becoming duke of Courland; at another he
thought of bestowing on himself the crown of Poland. He frequently
gave intimations of an intention to make himself a bishop or even a
simple monk. He built a superb palace, and wanted to sell it before it
was finished. One day he would dream of nothing but war; and only
officers, Tatars, and Cossacks were admitted to him: the next day he
was busied only with politics; he would partition the Ottoman Empire,
and put in agitation all the cabinets of Europe. At other times, with
nothing in his head but the court, dressed in a magnificent suit,
covered with ribbons presented to him by every potentate, displaying
diamonds of extraordinary magnitude and brilliance, he was giving
superb entertainments without any cause.
“He was sometimes known for a month, and in the face of all the
town, to pass whole evenings at the apartments of a young woman,
seeming to have alike forgotten all business and all decorum.
Sometimes also, for several weeks successively, shut up in his room
with his nieces and several men whom he honoured with his
intimacy, he would lounge on a sofa, without speaking, playing at
chess, or at cards, with his legs bare, his shirt collar unbuttoned, in a
morning gown, with a thoughtful front, his eyebrows knit, and
presenting to the view of strangers, who came to see him, the figure
of a rough and squalid Cossack. These singularities often put the
empress out of humour, but rendered him more interesting to her. In
his youth he had pleased her by the ardour of his passion, his valour,
and his masculine beauty. Being arrived at maturity, he charmed her
still by flattering her pride, calming her apprehensions, confirming
her power, and caressing her fancies of oriental empire, the
expulsion of the barbarians, and the restoration of the Grecian
republics.
“Potemkin began everything, completed nothing, disordered the
finances, disorganised the army, depopulated his country, and
enriched it with other deserts. The fame of the empress was
increased by his conquests. The admiration they excited was for her;
and the hatred they raised, for her minister. Posterity, more
equitable, will perhaps divide between them both the glory of the
successes and the severity of the reproaches. It will not bestow on
Potemkin the title of a great man; but it will mention him as an
extraordinary person; and, to draw his picture with accuracy, he
might be represented as the real emblem, as the living image of the
Russian Empire. For, in fact, he was colossal like Russia. In his
mind, as in that country, were cultivated districts and desert plains. It
also partook of the Asiatic, the European, the Tatar, and the
Cossack; the rudeness of the eleventh century, and the corruption of
the eighteenth; the surface of the arts, and the ignorance of the
cloisters; an outside of civilisation, and many traces of barbarism.”j
THE QUESTION OF THE IMPERIAL SUCCESSION

Some time before the death of Potemkin, Catherine had begun
proceedings intended to bar the czarevitch Paul from the imperial
succession.a She was by no means the cruel, heartless mother that
many writers are inclined to represent; but she knew her son
thoroughly well, and foreseeing how destructive of all good his reign
would be she could not think without fear of how the empire, which
under her rule had made such rapid strides in the path of prosperity,
glory, and civilisation, would after her remain without any guarantee
for the stability and durability of its existence. With the intention of
preserving the country from such a misfortune, Catherine wished to
make over the throne to the grand duke Alexander Pavlovitch and
therefore the setting aside of the czarevitch appeared in her eyes a
state necessity. Meanwhile it is sufficiently well known that Catherine
had long been accustomed to place the interests of the state above
everything and to sacrifice to them all other considerations and
feelings; therefore the difficulties with which so daring an
administrative step was doubtless accompanied could not stop the
creator of the changes of the year 1762. “Obstacles are created in
this world,” Catherine once wrote, “in order that persons of merit may
set them aside and thus add to their reputation; that is the meaning
of obstacles.” Circumstances were also favourable to this new
change contemplated by Catherine, for at that time no law existed
that exactly established the order of succession to the throne. The
statute of Peter the Great of the year 1722 was still maintained in full
power, and by this statute the reigning Russian sovereigns had the
right of naming anyone they liked as their successors to the throne
according to their own judgment, without being restrained by any
ancient right of primogeniture; and in cases where the heir already
designated showed himself incapable, he could be removed from the
throne.
The diary of Krapovitski can serve as a proof that in the year 1787,
after Catherine’s return from her travels in the south of Russia, the
question as to the necessity of changing the succession to the
throne had already matured in the mind of the empress; she entered
upon the historical study of the matter and read “the right of will of
monarchs.” On the 20th of August, in connection with this same
question, Catherine discussed with her secretary the extent to which
the misfortunes of the czarevitch Paul Petrovitch had been caused
by the false opinion that as eldest son the throne must belong to him.
Further, on the 25th of August, Krapovitski writes: “Ukases as to the
heirs to the throne, named since the time of Catherine I, have been
asked for, and in the explanations a sort of displeasure was
manifested.” To what conclusions the historical study of the
measures taken by Peter the Great led Catherine may be seen from
the context of the following remarks, written by the empress’ own
hand:
“It must be acknowledged that the parent is unhappy who sees
himself obliged for the safeguard of the public good to remove his
offspring. This is a condition which accompanies or is joined to the
autocratic and parental power. And thus I esteem that the most wise
monarch Peter I had doubtlessly the strongest reasons for the
removal of his ungrateful, disobedient, and incapable son, who was
filled with hatred, malice, and viperous envy against him. He sought
to find some particle of evil in his father’s deeds and actions which
were conceived in the spirit of good, he listened to flatterers, shut his
ears to the truth, and nothing was so pleasing to him as to hear his
most glorious father defamed and spoken evil of. He himself was a
sluggard, a coward, double-faced, unstable, gloomy, timid, drunken,
passionate, obstinate, bigoted, ignorant man, of most mediocre
intelligence and of weak health.”
Independent of these remarks, Catherine’s ideas are even more
clearly expressed in other rough draughts concerning the Greek
project and written in her own hand. She writes as follows: “Should
the successes of the war give Russia the means and occasion to
drive out completely the enemies of the name of Christ from the
European frontiers, then Russia, in return for such an entirely
Christian service rendered to the human race, would reserve to
herself the restoration on the ruins of the barbaric power, of the
ancient Greek Empire. Russia would promise to leave such an
empire incomplete independence, to entrust and give it up to the
young Russian grand duke Constantine Pavlovitch, who must then
give his promise not to make in any case any hereditary or other
pretensions to the succession of all the Russias, as equally his
brother must do in regard to the Greek succession.” All these
writings clearly testify that at the time of the second Turkish war the
empress Catherine had definitively come to the conclusion that the
welfare of the state required the setting aside from the succession of
the czarevitch Paul Petrovitch and his replacement by the grand
duke Alexander Pavlovitch.
Meanwhile the czarevitch on his part did all that was possible to
justify in the eyes of Russia Catherine’s intentions to exclude him
from the throne. A contemporary who was in close relations with him,
T. V. Rostopschin writes as follows: “It is impossible to see without
shuddering and pity what the grand duke’s father does; it is as if he
sought for every means of inspiring hatred and disgust. He has taken
it into his head that disrespect and neglect are shown to him;
therefore for this reason, he catches and cavils at everything and
punishes without distinction. Every day one only hears of violence, of
quarrels about trifles of which any private individual would be
ashamed. He sees a revolution everywhere; he sees Jacobite in
everything.”
Catherine’s correspondence shows that already in the year 1791
the plan of excluding the czarevitch Paul from the throne was no
secret to those who were in her intimacy. On the 1st of September,
1791, the empress in a letter to Grimm expresses herself quite
definitely on the matter; in relating her supposition as to the
consequences of the French Revolution, she writes: “But this will not
be in my time and, I hope, not in the time of Alexander.” Finally on
the 14th of August, 1792, Catherine communicates to Grimm
considerations which allow the nomination of Alexander as heir to be
regarded as a matter settled. “Why should the coronation be hurried
on?” writes she; “in the words of Solomon there is a time for
everything. First we will marry Alexander, and then we will crown him
with all possible ceremonies, solemnities, and popular festivities. Oh,
how happy he will be himself, and how happy others will be with
him!” The following letter addressed by Catherine to Count V. P.
Mussin-Pushkin on the 14th of September, 1792, written by the

Power Electronic System Design Linking Differential Equations Linear Algebra and Implicit Functions Keng C Wu Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Power Electronic System Design Linking Differential Equations Linear Algebra and Implicit Functions Keng C Wu Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Power Electronic System Design:

Linking Differential Equations, Linear

Linear Algebra and Partial Differential Equations T

Transforms and Partial Differential Equations T.

First Course in differential equations (11ed) /

Differential Equations and Boundary Value Problems:

Advanced Differential Equations 1st Edition Youssef

Schaum's Outline of Differential Equations, Fifth

Power System Analysis and Design 6th Edition J. Duncan

Algebra Linear 2nd Edition Alfredo Steinbruch

Copyright © 2021 Elsevier Inc. All rights reserved.

British Library Cataloguing-in-Publication Data

Library of Congress Cataloging-in-Publication Data

For Information on all Elsevier publications visit our website at

Publisher: Graham Nisbet

Typeset by Aptara, New Delhi, India

About the Author xi

1. Capacitor and inductor 1

5 Gain blocks 123

6 Feedback approaches 167

7 Control practices 189

7.7 Close loop 200

8 Linear regulator 213

9 Switch-mode DC/DC converters 241

10 AC drives, rectification, and inductive loads 299

11 Rotation, three-phase synthesis, and space vector concepts 319

Appendix A Accelerated steady-state analysis for a parallel resonant

Appendix B Matrix exponential 349

Appendix C Example 4.7 MATLAB m-file 351

Appendix D Example 8.1 353

Appendix E A general mass-spring-dashpot second-order system;

Appendix F A general mass-spring- dashpot second-order system;

Appendix G A general mass-spring- dashpot second-order system;

Appendix H Matrix exponential—Jordan form 367

Appendix I A step-by-step primer on digital power-supply design 369

Appendix J Motor winding driven by SCR phase-controlled

Keng C. Wu, a native of Chiayi( ), Dalin( ), Taiwan, received the B.S.

and modules performing amplification, voltage-to-time window, duty cycle

Capacitor and inductor

MATH. NOTE: In most calculus textbooks, derivatives and integrals are

1.1 Capacitor equation in differential form

MATH. NOTE: At a more fundamental level, the current is expressed as

Fig. 1.1 Terminal current and voltage of a capacitor.

What does this form tell us about a capacitive element?

1.2 Capacitor equation in integral form

Fig. 1.2 Terminal current and voltage of an inductor.

5. In addition, Eq. (1.3) implies that capacitor voltage is lagging its

1.3 Inductor equation in differential form

With reasoning similar to those given for capacitors, the following

1.4 Inductor equation in integral form

1.5 Definition of inductance and Faraday’s law

density is proportional to the magnetic field intensity, H, generated by a

1.6 Magnetic coupling and mutual inductance

Fig. 1.3 Ferrite cup-cored transformer and its electrical symbol.

Fig. 1.4 A B-H curve example without hysteresis.

winding, the primary, serves as a flux source driven by an external voltage

Next, we move driving source to winding 2 and open winding 1, i1 = 0.

1.7 Transformer equation

when magnetized. The nonlinearity may be expressed in an analytic form

Fig. 1.5 A B-H curve example with hysteresis.

MATH. NOTE: What Fig. 1.5 shows is a multivalue function. At a selected

Fig. 1.6 Square-wave drive and core flux function.

Fig. 1.7 Rectangular wave drives.